Page 1

Vol. 21, No. 2 April 2018

I&M around the World: Africa

contents table of

April 2018 VOL. 21, NO. 2

Instrumentation & Measurement I&M society web site

—Patrick Kapita Mvemba, Simon Kidiamboko Guwa Gua Band, Aimé Lay-Ekuakille, and Nicola Ivan Giannoccaro


Wendy Van Moer University of Gävle Department of Electronics, Mathematics and Natural Sciences SE-801 76 Gävle, Sweden


Advanced Acoustic Sensing System on a Mobile Robot: Design, Construction and Measurements

I&M magazine web site


Ultrasound-Guided Minimally Invasive Grinding for Clearing Blood Clots: Promises and Challenges


—Dalia Mahdy, Ramez Reda, Nabila Hamdi, and Islam S. M. Khalil

associate editor-in-chief Simona Salicone

Experimentally Driven Demystification of System Identification for Nonlinear Mechanical Systems

senior editorial assistant

—Mark Vaes, Yves Rolain, Johan Pattyn and Gerd Vandersteen

Bruno Ando

I&M editorial board Ruth A. Dyer Alessandro Ferrero Mark Yeary Salvatore Baglio Zheng Liu Ruqiang Yan Veronica Scotti Charles Nader Lee Barford Kevin Bennet

A New Low Cost Power Line Communication Solution for Smart Grid Monitoring and Management

advertising sales manager Onkar Sandal +1 800 627 0932 x218 Fax: +1 785 843 1853


Editorial 2 Guest Editorial 3 Life After Graduation 15 Basic Metrology 26

Future Trends in I&M Society News I&M Society Awards

34 36 37


on the cover:

Credit: iStock photo. To all Companies, please send your “New Products” information for possible inclusion in the IEEE I&M Magazine to: Robert M. Goldberg 1360 Clifton Ave. PMB 336 Cifton, NJ 07012, USA E-mail:


—Giovanni Artale, Antonio Cataliotti, Valentina Cosentino, Dario Di Cara, Riccardo Fiorelli, Salvatore Guaiana, Nicola Panzavecchia, and Giovanni Tinè

managing editor

Beverly Lindeen


Calendar 35

New Products


IEEE INSTRUMENTATION & MEASUREMENT MAGAZINE: (ISSN 1094-6969) (IIMMF9) is published bimonthly by The Institute of Electrical and Electronics Engineers, Inc. Headquarters: 3 Park Avenue, 17th Floor, New York, NY 10016-5997 +1 212 419 7900. Responsibility for the contents rests upon the authors and not upon the IEEE, the Society, or its members. Individual copies: IEEE members $20.00 (first copy only), nonmembers $25.00 per copy. Subscriptions: $6.00 per member per year (included in Society fee) for each member of the IEEE Instrumentation and Measurement Society. Nonmember subscription prices available on request. Copyright and Reprint Permissions: Abstracting is permitted with credit to the source. Libraries are permitted to photocopy beyond the limits of U.S. Copyright Law for private use of patrons: 1) those post-1977 articles that carry a code at the bottom of the first page, provided the per-copy fee indicated in the code is paid through the Copyright Clearance Center, 222 Rosewood Drive, Danvers, MA 01923 USA; 2) pre-1978 articles without fee. For other copying, reprint, or republication permission, write Copyrights and Permissions Department, IEEE Service Center, 445 Hoes Lane, Piscataway, NJ 08854 USA. Copyright © 2015 by the Institute of Electrical and Electronics Engineers, Inc. All rights reserved. Postmaster: Send address changes to IEEE Instrumentation & Measurement Magazine, IEEE, 445 Hoes Lane, P.O. Box 1331, Piscataway, NJ 08855-1331 USA. Canadian GST #125634188

Printed in the U.S.A.

April 2018

IEEE Instrumentation & Measurement Magazine 1

editorial I&M around the World


ravelling around the world‌ it definitely opens up your eyes and broadens your mind! And that's why each April issue of our magazine is dedicated to a different region. This allows all regions to show to the rest of the world their work in the field of instrumentation and measurement. What kind of I&M research is going on in that particular region? What are the difficulties? Where do they put the focus? In this April 2018 issue of our magazine we travel to‌ Africa!


This time we have two guest editors: Prof. Aime LayEkuakille and Dr. Mohamed Khalil. They both have strong connections with the scientific community in Africa and are experts in the field. As such they are the perfect ones to show us what is going on in the field of instrumentation and measurement in Africa. I would like to take the opportunity to thank them both for the valuable time they spent on this issue. Let's start our journey to Africa! Groetjes,

IEEE Instrumentation & Measurement Magazine

April 2018

guesteditorial AimĂŠ Lay-Ekuakille, University of Salento, Italy Mohamed Khalil, Doble Engineering, United Kingdom

I&M in Ultrasound Technology


e are pleased to introduce this section regarding applications of ultrasound in measurements. Two papers are considered important and included in the section. Ultrasound remains one of the most well-known radiations used for active and passive detection. It is often used for distance measurement and obstacle ranging. The aforementioned papers deal with acoustic sensing on mobile robot and invasive grinding, respectively. Acoustic sensing is used in many automotive applications because of its simplicity. It is an asset for radar systems including ground penetrating radar. The paper aims at allowing the mobile robot to follow line in an

April 2018

autonomous way employing an array of sensors instructed by a fuzzy logic system. The second paper is related to the use of ultrasound as guide for minimally invasive grinding in order to clear blood clots. Blood clots can be processed by means of medicine but it takes a long time; especially for urgent and concerned cases, a grinding by means of ultrasound could be necessary. In this case ultrasound is used to guide surgeons during their intervention by trying to help in clear blood clots but also by detecting indirectly the type of physiological material. We do hope the readers will enjoy both papers even with two different applications.

IEEE Instrumentation & Measurement Magazine 3

Advanced Acoustic Sensing System on a Mobile Robot: Design, Construction and Measurements Patrick Kapita Mvemba, Simon Kidiamboko Guwa Gua Band, Aimé Lay-Ekuakille, and Nicola Ivan Giannoccaro


n this paper, we describe the construction and charcaterization of a low-cost ultrasonic sensing system for obstacle avoidance on a mobile robot. In some locations and in certain conditions where electronic components are not readily available and not affordable, it is more advantageous to design an obstacle sensing system with a single sensor (e.g., HC-SR04) to reduce the construction cost. We decided to use only one sensor, mounted on a servo on the front of a mobile robot that scans and detects obstacles within the interval from 15° to 165° according to our design, to allow the estimation of the distance of currently detected obstacles with the help of a fuzzy rules set. The embedded fuzzy algorithm will select what obstacles should be avoided to perform collision-free navigation. A microcontroller with an Arduino bootloader was used to perform calculations and control the sensor (HC-SR04) and actuator (SG90 mini gear). The robot has two independent wheels, driven by geared PM dc motors, via the H driver L928N. The ultrasonic sensing system accuracy can be improved by considering ambient temperature in sound speed computation.

Ultrasound in Robotics Ultrasound-based signals are widely used in robotics to perform distance measurements and/or objects detection. Applied ultrasound technology has been developing in recent decades as the electronic technology develops, especially in high-power semiconductor devices. The application of ultrasonic systems has become increasingly widespread [1] in areas such as: ultrasonic measurement of distance, depth and thickness; ultrasonic testing; ultrasound imaging; ultrasonic machining, such as polishing and drilling; ultrasonic cleaning; and ultrasonic welding. Distance measurement using ultrasonic principles is as follows: the ultrasonic transmitter emits an ultrasonic wave in one direction and starts timing when it is launched. Ultrasound signals spread in the air and return immediately when they encounter an obstacle on the way. At last, the ultrasonic receiver stops timing when it receives the reflected wave. As ultrasonic spread velocity is 340 m/s in the air, based on the timer record t, we can calculate the distance s between the obstacle and transmitter, namely: 4


340 t (1) 2

which is the so-called time difference distance measurement principle. The principle of ultrasonic distance measurement uses the already-known air spreading velocity, measuring the time between emission and after receiving the reflection when it encounters obstacle, and then calculates the distance between the transmitter and the obstacle according to the time and the velocity. Thus, the principle of ultrasonic distance measurement is the same as with radar. Distance formula is expressed as:

l = v · th (2)

where l stands for the measured distance, v is the ultrasonic spreading velocity in air, and th represents time (th is half the time value from transmitting to receiving) [2]. In the design, the sensor emits pulses (40 kHz) when the microcontroller sends a command to the sensor trigger pin (Fig. 1). The sensor measures the amount of time it takes for the 40 kHz signal to come back and then outputs a variable-width pulse proportional to the time measured. The microcontroller excites the trigger pin for this given interval spanning 15° to 165° of the servomotor positions by taking into account the variable-width pulse for each of the degrees within the interval to determine how far away the obstacles are. A piece of fuzzy logic rules, implemented in the ATMEGA328P [3] fuzzy inference system, allows the controller to be able to send back measurement results in linguistic terms. The proposed application that uses only one acoustic sensor is a particular case of beamforming [4] in obstacle detection especially in automotive and vehicle navigation. Many on-board systems use acoustic propagation, such as radar, to detect obstacles and other cars. When they are low-cost, they use simple acoustic radar and cannot release an image of the obstacle or barrier [5]. The main measurement issue in these systems involves sensor array processing. Arrays play a basic role in radar, radio astronomy, sonar, communications, directions-searching,

IEEE Instrumentation & Measurement Magazine 1094-6969/18/$25.00©2018IEEE

April 2018

Fig. 1. Sensing system's block diagram.

seismology, medical diagnosis and care, and measurements. The implementation of arrays of sensors to achieve certain performance criteria involves trade-offs among the array geometry, the number of sensors, signal-to-noise and signalto-interference ratios, as well as a number of factors. From a signal processing perspective, the main objective is to detect the signal arriving from a particular look direction and cancel out any interfering signals and noise. A beamforming processor can achieve this by electrically steering the Smart Antenna, rather than mechanically guiding it as done in years gone by. The advantages of beamforming are improvements in communication quality, throughput and efficiency, especially in fields such as radar, sonar, seismology and wireless communications. A sensor array receives incoming signal information. A beamformer processes the spatial samples collected to provide the required spatial filtering [6], [7]. The sensor array may be arranged in a number of different configurations, two of which are the Uniform Linear Array (ULA) and Uniform Circular Array (UCA) in 2D space. The beamformer linearly combines the spatially-sampled time series from each sensor to obtain a scalar output time series. Beamformers are grouped into three different classes: fixed, optimum and adaptive. Fixed beamformers are analogous to bandpass filters, and they strive to pass signals spatially from a desired look direction and suppress all other signals arriving from all other angles.

The variable k is initiated with the integer value of 15 that corresponds to the first servo position, (the sensor is mounted on the servo on the front of the robot), and then the value of k is incremented into a loop whenever the estimation of the distance is computed and returned by the function: calculate Distance(). In that way the controller continuously updates data from the mobile robot surrounding space. Fig. 2 shows the activation of the fuzzy variable (middle) related to the front obstacle detection. The obstacle covers the interval between 69° and 88° according to the servo position. It is also shown that it is detected at the distance of an order of 16.2 cm. The formula relating the speed of sound, distance, and time traveled is as follows:

Apart from the beamforming approach, based on fixed array, that allows a scanning of the interested area, the design of the proposed architecture is based on an acoustic sensing system mounted on an ad hoc robot for mobile scanning. Fig. 1 shows its block diagram in which the microcontroller successively reads for all the servo positions (from 15° to 165°) and pulses from the ultrasonic sensor by exciting its trigger pin, in a loop cycle as follows:

speed =

distance (3) time

Rearranging this formula, we get the formula we will use to calculate the distance:

Design and Architecture

April 2018

for(int k=15;k<=165;k++) {  servoMotor.write(k);  delay(30);   distance = calculateDistance(); }

distance = speed ⋅ time (4)

The time variable will be provided by the ultrasonic range finder as the time it takes the ultrasonic pulse to leave the sensor, bounce off the object, and return to the sensor [8]. We actually divide this time in half since we only need to measure the distance to the object, one way. The speed variable is the speed at which sound travels through air. The speed of sound in air changes with temperature and humidity. Therefore, to accurately calculate distance, we will need to consider the local

IEEE Instrumentation & Measurement Magazine 5

Fig. 2. (a) Front obstacle detection, and (b) useful train of signals after obstacle targeting (the sensor moved from 15° to 165°).

ambient temperature and humidity. The formula for the speed of sound in air that factors in temperature and humidity is:

C = 331.4 + ( 0.606 ⋅ T ) + ( 0.0124 ⋅ H ) (5)

where C is the speed of sound in meters per second (m/s), the constant 331.4 is the speed of sound (in m/s) at 0 °C and 0% humidity. T is the temperature in °C and H (in %) is the humidity (relative humidity), ultrasonic ranging module HC - SR04 [9] provides 2 to 400 cm non-contact measurement function, and the ranging accuracy can reach 3 mm. The modules include ultrasonic transmitters, receivers and control circuits. The basic principle of work is as follows: ◗◗ uses an IO trigger for at least 10 μs high level signal, ◗◗ the module automatically sends an 8 cycle burst of ultrasound at 40 kHz and detects whether there is a pulse signal back ◗◗ if the signal returns back, through high level, the time of high output IO duration is the time from sending ultrasonic signals to their return.

Test distance =

high level time ⋅ sound speed (6) 2

(without a temperature sensor) depends on the temperature. But the orange plot (with a temperature sensor) is quite steady, and the greatest variation at 30 cm is 0.03 cm. With a two-way range measurement, the deviation will be 14.4 mm at a range of 2 m which is not a lot but more than the wavelength which is 9 mm at 40 kHz. According to (5), at 27 °C, the speed of sound is in the order of 347.7 m/s, and this implies that the obstacle is at 31.3 cm. This estimated distance drops down to 30.0 cm if the default sound velocity (344.8 m/s) is used.

Fuzzy Rules Fuzzy logic, unlike classical logic, is tolerant to imprecision, uncertainty and partial truth. In the context of mobile robot control, a fuzzy logic based system has the advantage that it allows intuitive nature of sensor-based navigation and can easily transform linguistic information into control signals [11]. The basic structure of a fuzzy logic controller consists of three conceptual components: fuzzification of the input–output variables; a rule base that contains a set of fuzzy rules; and a reasoning mechanism that performs the inference procedure on the rules and given facts to derive a reasonable output.

Temperature Compensation for Sensor Range Accuracy The accuracy can be improved by compensating for the variation of speed of sound with temperature. Sound moves through air at a speed dependent on the ambient temperature according to the following equation:

C = 331.3 + 0.606 ⋅ T (7)

where C is in m/s and T is in °C. The formula is good up to at least +/-30 °C. In this study we neglected the dependence of humidity because it is so small. Fig. 3 shows the accuracy at different temperatures [10]. As we can see, the blue plot variation 6

Fig. 3. Distances measured at different temperatures.

IEEE Instrumentation & Measurement Magazine

April 2018

Fig. 4. (a) Sensor range, and (b) fuzzy variables, the input of the sensor is defined as: near, middle and far.

The fuzzy logic system is widely used in many industrial fields [12]. In this paper, fuzzy rules allow the robot to avoid obstacles and navigate without hitting any of them, and the controller uses sensor signals as input to control the speed of motors. Fig. 4 shows that distances measured are within [0, 30] cm and defined as: near, middle and far with respect to fuzzy variables. The sensor is attached to the servo and can move between 15° to 165°, and the interval is divided into three areas: Left, Front and Right. The inputs are distance values measured from the obstacle to the ultrasonic sensor on the robot, and the output variables are velocities of the left and right wheels, where each of motors is being controlled by the PWM output of the microcontroller. The controller has three inputs and two outputs. Inputs are distances between the robot and the obstacle and the outputs are the motor speeds. For every input, we have defined three membership functions (far, medium and near) and for every output there are three membership functions (Reverse, Stop and Forward), and thus, there are 216 (2 3 ⋅ 33) possible rules [13]. The minimum number of necessary rules is three, but of course, the obtained behavior is very primitive. Another important factor is the computational time, and this was one of the main reasons that led us to simplify the fuzzy

Table 1 – Fuzzy rules descriptions Left_ Motor

Right_ Motor












































April 2018

algorithm. To reduce the number of the rule we considered only two membership functions for the input ultrasonic sensor (far, near) at different servo positions. Analyzing the rules and after some testing experiments we decided to take only eight rules. These fuzzy control rules for avoiding collision with any obstacles are given in Table 1 and Table 2. Since the ATMEGA328P increments the servo position from 15° to 165° by ones, the myservo.write and the delay libraries which we used were executed 150 times. We can consider the


IEEE Instrumentation & Measurement Magazine 7

Fig. 5. a) Sensor mounted on the mobile robot, and b) current consumption within 35 minutes.

writing to the servo to be instantaneous, while the delay takes 10 milliseconds. Overall, 150 executions each taking 15 milliseconds took a total of 150 ⋅ 0.010 seconds, or 1.5 seconds. Thus, the execution time was about 1.5 seconds.


References [1] HC-SR04 User Guide, Nov. 20, 2017, [Online]. Available: http://

The overall system is designed around an ATMEGA328P where ultrasonic and temperature sensors are connected to inputs, and servo and dc motors are actuators connected to microcontroller outputs. The ultrasonic sensor is mounted on the servo on the front of the robot, and motors are controlled by H driver L928N. All of the components, the servo, the microcontroller, the temperature sensor and the ultrasonic sensor are mounted on the 2WD Motors Smart Car chassis [14] (Fig. 5). The system power, including the two dc motors, is supplied by 4 AA batteries. When the dc motors are disconnected, the consumption drops down to 8mA, and if we enable the two motors and the H-bridge, the consumption goes up to 260 mA. [2] S. T. Maruthuru, “Automatic material separating conveyor,” Int. J. Innovative Science, Engineering & Technology, vol. 3, no. 12, Dec, 2016. [3] TMEGA328 Datasheet, Atmel, Nov. 20, 2017, [Online]. Available: datasheet.pdf. [4] A. Lay-Ekuakille, G. Vendramin, and A. Trotta, “Acoustic sensing for safety automotive applications,” in Proc. 2nd Int. Conf. on Sensing Technology, 2007. [5] A. Lay-Ekuakille, P. Vergallo, D. Saracino, and A. Trotta, “Optimizing and post processing of a smart beamformer for


obstacle retrieval,” IEEE Sensors Journal, vol. 12, no. 5, pp. 1294-

In this paper we have described the construction of a low-cost ultrasonic sensing architecture for obstacle avoidance. We have presented an ultrasonic sensing system for a mobile robot where at a given servo position (15° to 165°), the fuzzy logic controller translates ultrasonic sensor measurements directly to actuator actions. The algorithm is implemented on an ATMEGA328P. The robot has two independent wheels, which are driven by geared PM dc motors. Control of the motors is accomplished by the H bridge driver L928N, and the ultrasonic sensor is mounted on the servo in front of the robot and connected to the controller analog input port. By considering the ambient temperature in sound speed calculations, we have improved the sensor accuracy. The microcontroller is programmed in C language, and fuzzy rules are tested to perform collision-free navigation toward any given goal and corridor following. The experimental results have shown that the proposed sensor architecture provides an efficient and flexible solution for a light autonomous mobile robot. The presented 8

approach can be also improved in a combination with a genetic algorithm and can give reliability to the detection technique using ultrasonic sensors.

1299, 2012. [6] B. D. Van Veen and K. M. Buckley, “Beamforming: a versatile approach to spatial filtering,” IEEE ASSP Mag., vol. 5, no. 2, pp. 4 -24, 1998. [7] A. Lay-Ekuakille, G. Vendramin, and A. Trotta, “Implementation and characterization of novel acoustic imaging through beamformers for automotive applications,” S.C. Mukhopadhyay and G. S. Gupta Lectures Notes in Electrical Engineering Series, “Smart Sensors and Sensing Technology,” Springer-Verlag, 2008. [8] “Project tutorial for Arduino compatible products,” Nov. 20, 2017, [Online]. Available: [9] Ultrasonic Ranging Module HC- SR04, Nov. 20, 2017, [Online]. Available: [10] “Improve ultrasonic range sensor accuracy,” Nov. 20, 2017, [Online]. Available:

IEEE Instrumentation & Measurement Magazine

April 2018

[11] G. C. D. Sousa, “Fuzzy logic applications to power electronics and drives- an overview,” in Proc. IEEE Conf. Industrial Electronics,

research deals with artificial intelligence, measurements and instrumentation.

Control, and Instrum., 1995. [12] K. Hirota, Ed., Industrial Applications of Fuzzy Technology. Tokyo, Japan: Springer, 1993. [13] C. G. Rusu, “Fuzzy based obstacle avoidance system for autonomous mobile robot,” in Proc. IEEE Int. Conf. Automation Quality and Testing Robotics (AQTR), 2010. [14] Dual Full-bridge Driver L298N,, Apr. 25, 2016, [Online]. Available: Robotics/L298_H_Bridge.pdf.

Patrick Kapita Mvemba ( received the Master's degree in electronics engineering from the ISTA University, D. R. Congo in 2008 where he has been appointed as Assistant Professor. He is currently a Ph.D. student at the University of Siena, Italy, working in the field of measurements and instrumentation. Simon Kidiamboko Guwa Gua Band received the Master's degree in electronics engineering from the ISTA University, D. R. Congo in 1980 and the Ph.D. degree from the University of Ancona, Italy in 2009. He is currently Professor of Artificial Intelligence at ISTA University. His main

April 2018

Aimé Lay-Ekuakille (SM'12) is the Director of the Instrumentation and Measurement Lab at University of Salento, Italy and teaches measurement and instrumentation courses. He received the Master's degree in electronics engineering from the University of Bari, Italy, the Master's degree in clinical engineering from the University of L'Aquila, Italy and the Ph.D. degree in electronics engineering from the Polytechnic of Bari, Italy. He chairs the IEEE IMS TC34 “Nanotechnology in Instrumentation and Measurement,” is part of IEEE Nanotechnology Council AdCom and serves as chairman of IMEKO TC19 “Environmental Measurements.” His main research areas are environmental, industrial and biomedical instrumentation and measurement, including the use of nanotechnology, and renewable energy. Nicola Ivan Giannoccaro received his M.S. degree in electronics engineering and the Ph.D. degree in advanced production systems from the Politecnico of Bari, Italy in 1996 and 2000, respectively. He is Associate Professor with the Department of Innovation Engineering, University of Salento, Lecce, Italy. His research interests include mechatronic systems, control of mechanical systems, modal analysis and dynamical identification.

IEEE Instrumentation & Measurement Magazine 9

Ultrasound-Guided Minimally Invasive Grinding for Clearing Blood Clots: Promises and Challenges Dalia Mahdy, Ramez Reda, Nabila Hamdi, and Islam S. M. Khalil


echanical removal of blood clots is a promising approach towards the treatment of vascular diseases caused by pathological clot formation in the circulatory system. These clots can form and travel to deep seated regions in the circulatory system and result in significant problems as blood flow past the clot is obstructed. A microscopically small helical microrobot (Fig. 1a) offers great promise in the minimally-invasive removal of these clots. The simple design of the microrobot, which was originally presented at Tohoku University by Ishiyama et al., enables fabrication at micro and nano scales [1]. The incorporation of a magnetic material to the helical microrobots allows them to controllably navigate using an external source of magnetic field and feedback control [2]. The external source of magnetic field has been provided using a configuration of electromagnetic coils or rotating permanent magnets, while several research groups have relied on visual feedback to design feedback control systems. The greatest power of these microrobots in biomedical applications has emerged when clinical imaging modalities were incorporated to provide feedback and enable accurate in vivo tracking and control towards a desired position in three-dimensional space. Sylvain Martel's and Brad Nelson's groups have succeeded in achieving control of untethered microrobots, magnetotactic bacteria [3], and artificial bacterial flagella [4] using magnetic resonance imaging and fluorescence-based in vivo imaging, respectively. These ongoing in vivo attempts indicate that we will be able to use microrobots to achieve new minimally-invasive diagnosis and therapeutic procedures. For example, the microrobot that we are using consists of a helical body and a magnetic head (Fig. 1b), with a magnetization vector perpendicular to the long axis of its helical body, and can swim controllably in a vessel using a rotating magnetic field in milliTesla range. This level of control enables the microrobots to swim controllably and selectively target a blood clot, as shown in Fig. 1c and Fig. 1d. The formation of blood clots

starts with the activation and aggregation of platelets followed by the activation of blood coagulation factors. These factors generate strands of fibrin surrounding the platelets, and more platelets and blood cells are entrapped inside the fibrin network. The continuous interaction between the tip of the helical microrobot and the fibrin network of the clot has proven to be efficient in decreasing the size of the clot in vitro [5]. However, the translation of this concept into in vivo trials requires several challenges to be overcome. Among these challenges is the need to achieve real-time tracking and motion control of the helical microrobot using feedback provided by a clinical imaging system. In addition, the accurate estimation of the blood clot size during the interaction of the helical microrobot is also necessary to study the influence of the microrobot on the clot. Our in vitro experimental model is shown in Fig. 2. The system integrates several modules to control the motion of the helical microrobots, localize the microrobot using ultrasound feedback (Fig. 2a), and analyze the volume and composition of the clot during interaction with the microrobot. The first module is a permanent magnet-based robotic system (Fig. 2b) that consists of two rotating dipole fields. Single rotating dipole field can also be used to externally actuate the microrobot, however, a magnetic force will pull the microrobot along its lateral direction towards the inner wall of the vessel. This pulling magnetic force is decreased when the two synchronized rotating dipole fields are used and the microrobot is positioned in their common center, as shown in Fig. 2b. Rotation of the dipole fields enables the microrobot to swim through the medium inside a catheter segment that contains the blood clots and a flowing stream of phosphate buffered saline (PBS) (Fig. 2c). Blood clots with volume of 94.24 mm3 are prepared and inserted inside the catheter segment, and PBS is injected at a flow rate of 10 ml/hr against the direction of the swimming microrobot. The second module of the in vitro model provides imaging of the microrobot and the clot. An ultrasound system (HD 5 Diagnostic Ultrasound System, Philips

This work was supported by funds from the German University in Cairo and the DAAD-BMBF funding project. The authors also acknowledge funding from the Science and Technology Development Fund in Egypt (No. 23016). 10

IEEE Instrumentation & Measurement Magazine 1094-6969/18/$25.00Š2018IEEE

April 2018

Fig. 1. Clearing of blood clot is achieved using a helical microrobot and rotating magnetic field. (a) The microrobot consists of a magnetic head and helical tail. (b) The helical tail is attached to a cylindrical magnet with axial magnetization (red arrow). (c) The tip of the helical microrobot tears the three-dimensional fibrin network of a blood clot and enables blood cells to break free. (d) An externally actuated helical microrobot decreases the size of blood clot via mechanical grinding under the influence of rotating magnetic field. The straight and curved arrows indicate the direction of swimming and rotation, respectively.

and Neusoft Medical Systems, Amsterdam, The Netherlands) and two digital cameras are used to localize the microrobot and visualize the clot in each trial, respectively. The third module of the in vitro model provides quantitative analysis on the pre- and post-conditions of the blood clots and also provides physical information during interaction. The three modules of the in vitro model are utilized on approximately 1-hour old blood clots, corresponding to a clinically relevant time regarding the treatment's effectiveness of several conditions of vascular obstruction. The aim of their integration is threefold: first, to actuate the helical microrobot at a distance; second, to control the motion of the helical microrobot towards the clot under ultrasound guidance; and third, to analyze the pre- and post-conditions of the clot to verify the effectiveness of this minimally-invasive approach. Our approach begins by localizing the microrobot using ultrasound feedback (Fig. 2a). The transducer is fixed with respect to the catheter segment which is embedded into gelatin to achieve air-free coupling. Closed-loop motion control of the helical microrobot is implemented upon detecting its ultrasound feedback (Fig. 3). The position tracking error is April 2018

calculated based on the positions of the blood clot and the microrobot, and the angular velocity of the rotating dipole fields is adjusted such that the position error decreases asymptotically (Fig. 2d). This closed-loop control is followed by continuous grinding of the blood clot. Two cameras are used to simultaneously capture the morphology of the clot during each trial for off-line calculation of the clot volume. In addition, the PBS past the blood clot is collected and the number of blood cells is calculated. The volume of the clot is constructed using the visual feedback and provides overall information pertaining to the morphology of the clot, whereas the number of blood cells provides a measure of the influence of the helical microrobot on the composition of the blood clot. Fig. 2e shows experimental results conducted using 1-hour old blood clot samples. The samples are prepared using blood taken from healthy donors. This experiment is approved by Institutional Review Board, and donors' written informed consents are obtained. We use 4-mm-long helical microrobots with diameter of 364 Îźm. The swimming speed of these microrobots depends on the frequency of the rotating magnetic field. For instance, average swimming speeds of 10 mm/s and 26 mm/s

IEEE Instrumentation & Measurement Magazine 11

Fig. 2. Motion control of a helical microrobot towards a blood clot is achieved under ultrasound guidance. (a) The microrobot is localized and its position is feedback to a closed-loop control system. (b) A permanent magnet-based robotic system provides rotating magnetic field to actuate the microrobot. (c) The microrobot swims under the influence of the rotating magnetic fields towards the clot. (d) Motion control is achieved towards a reference position (black dashed line) that represents the position of the clot. (e) The microrobot grinds the clot and decreases its size with time. The volume ratio represents the current estimated volume of the clot to its original volume before grinding. (f) The count of the blood cell indicates that the helical microrobot tears the fibrin network and allows cells to break free from the clot.

are observed under the influence of rotating magnetic field of 5.5 mT at frequencies of 20 Hz and 45 Hz, respectively. However, under the influence of ultrasound feedback, the speed of the microrobot varies (proportional to the position error between the clot and the microrobot) and decreases as the microrobot approaches the clot. The average speed of the helical microrobot is 5.32 mm/s during the closed-loop control trials, and mechanical grinding of the clot is achieved upon contact. Ischemic tissue, deprived of blood because of the clot, will not survive. Therefore, the microrobot must achieve enough grinding and dissolution of the thrombus, and hence recanalization of the blocked artery, within a biologically meaningful time window which imposes a significant limitation on the grinding time [6]. Hence, the grinding time is limited to 40 minutes in all of our in vitro trials. This limitation can be overcome by controlling the removal rate of the clot using the frequency of the rotating magnetic field. The removal rate of the blood clots is proportional to the rotation frequency of the 12

microrobot. For instance, negligible removal is observed below actuation frequency of 30 Hz. Removal rate of -0.23 mm3/ min (n=6), -0.885 mm3/min (n=6), and -0.315 mm3/min (n=6) are measured under the influence of rotation frequency of 30 Hz, 35 Hz, and 40 Hz, respectively. Again, we observe a decrease in the removal rate above actuation frequency of 40 Hz, owing to the increased damping at relatively high frequencies and due to the step-out frequency of the microrobot, i.e., frequency above which the microrobot no longer aligns with the applied magnetic field [5]. Another method of assessment of our minimally-invasive technique is the analysis of the decomposition of the collected PBS past the grinding. We count the blood cells past the clot and also observe that their number is directly proportional to the actuation frequency of the microrobot. In the absence of interaction between the clot and the microrobot (Fig. 2f), the cell count is 162.75±81.25×104 cell/ml (n=6), whereas the count increases to 832.91±206.6×104 (n=6) cell/ml for grinding at frequency of 35 Hz.

IEEE Instrumentation & Measurement Magazine

April 2018

Fig. 3. Mechanical grinding of blood clots is achieved using two synchronized rotating dipole fields and ultrasound feedback. Position of the microrobot is determined using the ultrasound scans and is used to generate position error based on the position of the clot. The angular positions and velocities of the two rotating dipole fields are used to synchronize their motion to exert magnetic torque on the dipole of the microrobot and mitigate the magnetic force along its transverse direction. The red and green dashed curves and arrows indicate the original ultrasound waves and the reflected waves via the helical microrobot only, respectively.

The continuous interaction between the tip of the helical microrobot and the fibrin network of the clot has proven to be efficient in decreasing the size of the clot in vitro and allowing blood cells to break free from the fibrin network of the clot. The hope of translating this concept into in vivo trials one day implicates that we need to overcome several challenges. The first challenge is the optimal control of the microrobot in the real circulatory system while being exposed to the natural blood flow, which varies according to the type of blood vessel in question, April 2018

its cross sectional area, and the blood velocity in a specific region of the body. This challenge necessitates an optimization of the size of the microrobot as well as the respective mechanical properties in function of type and location of the clot. As real-time tracking of the helical microrobot in vivo is essential, we are using ultrasound feedback as a non-invasive and safe clinical imaging system. Another challenge is to find a suitable method to assess the mechanical grinding's effect on the blood clot in vivo, especially since the echogenic signal of fresh blood

IEEE Instrumentation & Measurement Magazine 13

clots is relatively weak. Would it be possible to accurately analyze the changes of a blood clot's size during rubbing? Or should we rely on other parameters rather than the size and the number of blood cells past the clot? In addition, as much as using biodegradable material sounds relieving, the fate of the microrobot in vivo, once the clot is removed, needs to be studied. It is also essential to mitigate the interaction of the microrobot with the normal vessel wall in vivo and inhibit the coagulation around the microrobot itself. Therefore, it may be possible to coat the microrobot using a material similar to those used in vascular stents and catheters. Another promising aspect of using microrobots is the possibility to use them as vehicles for in situ delivery of thrombolytic drugs in addition to the mechanical grinding; in this case the combination therapy could enhance the local effect on the blood clot while preventing the severe complications that can be related to the systemic administration of thrombolytic drugs. In this context, an in vivo comparative study between the pure mechanical grinding and the combination with in situ chemical lysis requires further investigation.

Acknowledgment The authors would like to thank Dr. A. M. R. Abuelatta from the Diagnostic and Interventional Radiology of the National Cancer Institute for assistance with the preparation of the in vitro model. They would also like to thank Mr. A. El Sharkawy and Ms. S. Hesham for assistance with the experimental work.

References [1] K. Ishiyama, M. Sendoh, and K. I. Arai, “Magnetic micromachines for medical applications,” J. Magnetism and Magnetic Materials,

microrobotic flagella,” Advanced Materials, vol. 27, no. 19, pp. 2981-2988, Apr. 2015. [5] I. S. M. Khalil, A. F. Tabak, K. Sadek, D. Mahdy, N. Hamdi, and M. Sitti, “Rubbing against blood clots using helical robots: modeling and in vitro experimental validation,” IEEE Robotics and Automation Letters, vol. 2, no. 2, pp. 927-934, Apr. 2017. [6] S.-I. Ando, “What does a perfect blood pressure meter look like from a clinician point of view?” IEEE Instrum. Meas. Mag., vol. 17, no. 3, pp. 15-20, Jun. 2014.

Dalia Mahdy is currently a Research Associate with the Medical Micro and Nano Robotics of the German University in Cairo. She received her Bachelor's degree in computer engineering and is currently working toward the M.Sc. degree in computer engineering. Her research interests include design and development of minimally-invasive techniques for targeted drug delivery using microrobots. Ramez Reda is currently an Assistant Professor of Neurology at Ain Shams University. He held the position of Director of the Neurology Unit at Ain Shams University until 2012 and serves as a Consultant Neurologist at several major hospitals across Egypt, including Nozha International Hospital and Cleopatra Hospital. Nabila Hamdi is currently an Assistant Professor and the Head of the Molecular Pathology Unit of the German University in Cairo. Dr. Hamdi graduated from the Medical University in Sfax, Tunisia and spent the practical years of her medical studies in Munich, Germany. She received her Ph.D. degree in molecular genetics pathology in 2009 from the German University in Cairo.

vol. 242-245, no. 1, pp. 41-46, Apr. 2002. [2] B. J. Nelson, I. K. Kaliakatsos, and J. J. Abbott, “Microrobots for minimally invasive medicine,” Annual Review of Biomedical Eng., vol. 12, pp. 55-85, Apr. 2010. [3] O. Felfoul, M. Mohammadi, S. Taherkhani, D. de Lanauze, Y. Z. Xu, et al., “Magneto-aerotactic bacteria deliver drug containing nanoliposomes to tumour hypoxic regions,” Nature Nanotechnology, Aug. 2016. [4] A. Servant, F. Qiu, M. Mazza, K. Kostarelos, and B. J. Nelson, “Controlled in vivo swimming of a swarm of bacteria-like


Islam S. M. Khalil ( is currently an Assistant Professor with the German University in Cairo and the Director of the Medical Micro and Nano Robotics Laboratory. He received the Master's and Ph.D. degrees in mechatronics engineering from Sabanci University, Istanbul, Turkey. For two years, he has been a Postdoctoral Research Associate with the Robotics and Mechatronics Research Group and MIRA-Institute for Biomedical Technology and Technical Medicine, University of Twente, the Netherlands.

IEEE Instrumentation & Measurement Magazine

April 2018

lifegraduation after

Erik Timpson

Africa— Where Metrology Began


n this issue there are some wonderful advanced measures including acoustics. What are some simple measures and where did they begin? According to NCSL International, the history of measurement began in Africa, specifically with the Egyptian cubit—a fascinating story about having to bring your cubit in every full moon or die. Now that is an effective calibration program! My hope in mentioning this is not to fill you with fear. I hope that I fill you with curiosity about how to calibrate whatever instrument you are currently dependent on. I am focused on Africa for this article, and after much thought, I decided that best way to do this was to use the IEEEXplore platform to find other interesting papers coming from Africa. Then after considering we are all from Africa, depending on how far back one traces our DNA, I dismissed that idea as a distraction and started reading. I loved 25 papers total, but being forced to make this article a quick read, I'll focus on three. All with a measure focus including specifics of magnets, smartphones, and optical measures. O. O. Ogidi et al., from the University of Cape Town, did a wonderful job in their paper titled “Development of a Test Rig for Eccentricity Fault Studies on an Axial-Flux Permanent Magnet (AFRM) Wind Generator” [1]. Here I note with wind turbines being a great part of a mixed energy solution of the future, the authors do especially well in focusing on machine fault diagnosis. With great figures, equations, and simple explanations, I was entertained. Here's hoping their life after graduation is going well. Charl A. Opperman et al., from the University of Pretoria, did well in looking at the most readily available platform to human kind— the smartphone. In their paper titled, “Smartphones as a Platform for Advanced Measurement and Processing,” they conclude that a PC is still much faster than a phone for processing [2]. Yet the phone has a wide variety of

April 2018

sensors, and its processing power is increasing. I'm sure they are correct whey they say, “Instead of having to purchase specialized hardware and software, open source software…” is more simple. There is elegance in simplicity. Abdelrahman E. Afifi et al., from Cairo, titled their paper “Fiber Optical Coherence Domain Polarimetry for PM Fiber Measurements” [3]. Also with beautiful equations, figures, and measurements, they verified the theoretical model with experimentation, making for wonderful entertainment. I was reminded of George Box who said, “All models are wrong, some models are useful.” The authors devised a very useful model with a more environmentally robust system than before. In the end, I know we scholars are busy writing papers and reading in our areas of our expertise. Perhaps take a break and look for papers not on subject alone but from a certain geographic area or with a different search parameter. Consider effective calibrations for your measures and instruments. And have a wonderful life after graduation. Cheers, Erik

References [1] O. O. Ogidi, P. S. Barendse, and M. A. Khan, “Development of a test rig for eccentricity fault studies on an axial-fluxpermanent magnet (AFPM) wind generator,” in Proc. 2014 International Conference on Electrical Machines (ICEM), pp. 1562-1568, 2014. [2] C. A. Opperman and G. P. Hancke, “Smartphones as a platform for advanced measurement and processing,” in Proc. 2012 IEEE International Instrumentation and Measurement Technology Conference, pp. 703-706, 2012. [3] A. E. Abdelrahman, A. R. El-Damak, T. A. Ramadan, and M. H. Ahmed, “Fiber optical coherence domain polarimetry for PM fiber measurements,” J. Lightwave Tech., vol. 35, no. 16, pp. 35693576, 2017.

Dr. Timpson may be contacted at, and his bio is available at

IEEE Instrumentation & Measurement Magazine 15 1094-6969/18/$25.00©2018IEEE

Experimentally Driven Demystification of System Identification for Nonlinear Mechanical Systems Mark Vaes, Yves Rolain, Johan Pattyn and Gerd Vandersteen


he goal of this work is to develop a low cost hardware-based system identification (SI) demonstrator. It targets the students and practicing engineers with an illustration of the threads for model extraction caused by nonlinear distortion. The focus lies on hands-on training with a high return on effort. The training starts from real experiments that are performed on the demonstrator hardware, which is a real system that is introduced here. It can easily be used at home by the trainee without the need for external support or expensive equipment. The steep learning curve that often scares potential users of advanced SI methods is hereby flattened, and the practical applicability of these methods is demonstrated simultaneously. The joint use of practical, clear and simple teaching materials and hardware-based illustration can provide enough knowledge, understanding and confidence for the end users to apply the methods on industrial scale systems. The first test case presented here is a mechanical setup. It consists of a nonlinear mass-spring-damper system built using commonly available components for a total cost of less than 50 Euro ($US 60). When applying a signal with a low amplitude to the designed system, its behavior almost perfectly matches that of a Linear Time Invariant (LTI) device. This remains true whenever the amplitude of the input signal is kept very small. Increasing the amplitude gradually introduces weak and intuitive nonlinear distortion. This regime of operation is used to illustrate the use of the analysis and estimation methods that are presented in a framework that closely approximates the theoretical assumption of a periodic in, same period out (PISPO) system. These systems impose the period of a periodic input to the resulting output signals. Increasing the amplitude further challenges the trainees' intuition even more. A swept sine test for increasing and decreasing frequency shows that the system becomes bi-stable in the frequency band near the resonance peak. The peak then shifts in frequency when

the excitation amplitude is further increased. These phenomena are typical for a nonlinear (vibrating) system containing jumps. The main payback of this approach for the trainee is the knowledge that the well-known Frequency Response Function (FRF) measurement of the mechanical system can indeed be used as an enabler for the detection of nonlinearity and the nonlinear SI. The experiments show the need for the detection, quantification and qualification of nonlinear effects. Existing high return local linearization methods [1], [2] are introduced and applied directly to the system under test. The major advantage of these methods based on real experiments lies in the direct visual feedback that is provided to the user under a format that is familiar, hence directly usable and understandable. A first presentation of the setup to test groups already involved in SI seems to indicate that the approach can strongly stimulate, motivate, attract and help potential users both at the academic and industrial level.

System Identification Research System identification research (SI) has advanced at a fast pace over the last years. Given the current momentum, one can safely predict that extensive research will lead to further future growth. As of today, SI has become a reliable way to identify a parametric model for many real-world systems starting from experimental data. On the other hand, practitioners, both in academic institutions and in industry, often use linear time invariant (LTI) models to tune a design or to characterize a system, even if they know that the actual system is nonlinear. This is due to the steep learning curve for nonlinear system modeling and design, the large variety of approaches that impede an easy selection of the â&#x20AC;&#x153;bestâ&#x20AC;? method, and the lack of widespread nonlinear identification education. The payback for a nonlinearity assessment in SI is potentially very high, both for industrial and academic users. Being able to differentiate between noise and nonlinear distortion in

This work was supported in part by: the Fund for Scientific Research (FWO-Vlaanderen); the Flemish Government (Methusalem); the Belgian Government through the Interuniversity Poles of Attraction (IAP VII) Program; and the VUB (SRP-19). 16

IEEE Instrumentation & Measurement Magazine 1094-6969/18/$25.00Š2018IEEE

February 2018

experimental data and to measure the level of nonlinear distortion has many advantages. It leads to more accurate models, avoids over modeling, and results in a more profound understanding of the actual behavior of the system. To spread the concept of nonlinear system identification despite its inherent complexity, to convince practitioners, and to attract student attention, a set of systems is developed here. The idea is to teach the use of nonlinear SI tools with the help of hands-on exercises on real, yet inexpensive and easy to build systems. These exercises are performed on easy to build systems in different fields of engineering. The toy examples are conceived such that the trainee faces all of the challenges that can be expected for a real industrial system: the presence of measurement noise, process noise, model errors, and other imperfections are handled and assessed rather than being eliminated on purpose. One of the test case systems is a mechanical system seen in Fig. 1. Despite its simplicity, the system behaves as a mass-spring-damper system [3] where the nonlinear distortions become more prominent with increasing movement amplitude. A combination of the visual illustration provided by the vibration of the system and the hands-on exercises is expected to be a major advantage and an enabler for better understanding of a nonlinear systems' behavior. Before introducing the mechanical system, the idea behind this project is explained. A simple Frequency Response Function (FRF) measurement shows the influence of the nonlinear distortion on the behavior of a system. From there on, a user centric point of view is taken to show the first steps in the quantification and identification of the nonlinear systems.

Current State of the Art: Identification in an Industrial Context LTI identification methods [4] prove to be efficient when it comes to the characterization and modeling of linear systems. They are currently widely used in many fields of engineering both for design and prediction. If the system contains some nonlinear behavior, as is often the case, it can become tricky to use these methods and interpret their outcome properly as will be discussed. Whenever a system is mainly linear, and hence has a very weak nonlinear contribution, the methods can still be approximately valid [1], [2]. However, if the system is excited in an amplitude range where the nonlinear distortions become noteworthy, problems will arise. The methods will still deliver results, but they are no longer as usable as before or are sometimes simply erroneous. The validation of the model also becomes complicated. Some tests such as the whiteness test of residuals will validate the model. However, it is clear that the system does not belong to the model class; hence, the test should invalidate the model. This does not mean that the test is wrong, but rather that it is misused. The basic hypothesis of this test is the linearity of the system under test which is clearly questionable in this case. This clearly shows the relevance of the nonlinear detection which is proposed here. For practitioners who are not aware of these issues, this nonlinear distortion leads to a poor interpretation of the February 2018

Fig. 1. The mechanical mass spring damper test case system.

systems' behavior. Many practitioners nevertheless continue to rely on the LTI methods due to prior experience with previous identification projects and/or the lack of nonlinear behavior awareness. Such an example where the use of the LTI identification methods can fool the user is shown in the next section. Some simple measurements on a mechanical toy system illustrate the behavior of the nonlinear distortions and their influence on the degradation of the result.

A First Look at the Mechanical Test System The mechanical example system (Fig. 1) that is measured in this section has been created specifically to illustrate the nonlinear behavior of a system. The results of a simple FRF measurement are then obtained using classical LTI methods only. The practitioner then visualizes the influence of nonlinear distortions on the measured system response and eventually also on the identification process. The description of the specific properties that are incorporated in the system design is discussed when the system behavior is analyzed. The (classical) steps used here to measure the FRF are intended to show the trainee that regular methods can only be used in a proper context.

Setup of the Experiment A well-designed, informative experiment is a mandatory requirement for any identification process. The first step requires interaction with the practitioner to roughly guess the bandwidth and the handling capabilities of the system under test. The measurement hardware needed for the experiment can be selected on this input. Industry practitioners commonly have multiple measurement platforms that are readily available. Based on prior knowledge, the practitioner decides which hardware best suits the experiment. Academics and students normally have access to measurement equipment through the courses the student follows and through research setups. However, there is still the option of using the integrated soundcard of a PC to perform measurements. MATLAB ÂŽ and other software generally provide the tools needed to easily access the soundcard as if it were a measuring instrument. This way, the microphone and headset jack connection can be used

IEEE Instrumentation & Measurement Magazine 17

Measuring with the Soundcard One of the possible measurement setups that can be used to identify a system is the use of the built in sound card of a computer. The audio out connection is used as the signal generator and the stereo microphone connection is used to measure the input and output of the system (stereo offers two channels). Obviously, the practitioner needs to take into account that the maximum sampling frequency of such a soundcard is 44 kHz such that only audio band frequency systems can be measured. The soundcard has two major advantages. Firstly, it is a low cost alternative as it is always provided in computers and laptops. Secondly, if a laptop is used, it is possible to learn and perform system identification everywhere on the go. With good software tools, it is even possible to adapt the measuring variables like fs to the needs of the experiment. Fig. 2 shows the results of a measurement done on the vibrating system shown in Fig. 1. Note that the resonance frequency of that measurement is different from the resonance frequency of the other figures that are analyzed. This is because the springs were replaced with a rubber wire to test the influence of the spring properties. This shows that using different materials will create different setups that still can be measured in the same way. The major disadvantage of using the soundcard as a measurement “instrument” is that it is limited by the power of the signals. In the case of the vibrating setup, the output from the MEMS accelerometer had to be amplified to be able to measure the output with a decent SNR. This shows that although the use of this setup looks simple, it can become more complex to use in practical measurements of unknown systems. After all, it was never meant to be used as an instrument, and hence such issues were to be expected. Therefore, it may be better to invest some money to buy inexpensive measurement hardware that most instrumentation manufactures offer to students at a reduced price.

Fig. 2. Output of the mass spring damper system using the soundcard as measurement equipment. The system was excited with a random phase multisine.


as the input and output. This can be an advantage as it offers a more mobile setup that can also be used off-site. For the experiment conducted here, we use an ELVIS II engineering lab workstation by National Instruments. This device is also provided to the students for the measurement courses within the Electronical Engineer curriculum. Secondly, a sensible choice of the excitation signal is required to obtain qualitative information about the system. This boils down to the selection of the frequency band of interest as stated before and the input amplitude range before starting the experiment. Making a poor choice here will potentially result in a model that is not informative and therefore useless. For the system at hand, it is known that the frequency band of interest ranges from 15 to 35 Hz by design. The amplitude range used for the input signal is designed to cover the 10 mVRMS up to maximum value 350 mVRMS range. To extract the FRF, a class of excitation signals is selected next.

Design of the Input Signal The input signal is selected to comply with the experimental conditions defined above. Remember that at this moment, the behavior of the system is still unknown. To be able to easily extract an FRF and a non-parametric noise model for the measurements simultaneously, a periodic excitation signal is preferred here. Experiments can also be performed with noise excitations but are left out here for the sake of simplicity. For more details, see [1], [2]. In the setup proposed, a full multisine signal is used and the advantages of this signal in an identification context will be shown. More specifically a random phase multisine is used [1], [2]. The signal is called random because of the random choice of the phase  k. The reason behind this choice becomes clear later on when some identification algorithms are applied. This signal is mathematically defined as follows:

u(t ) =


N /2

A sin( kω t + ϕ ) (1) N k =1




where Ak ∈ is the user-defined real amplitude and  k is a user-selected random phase value distributed in [0, 2). This signal excites the system at the frequencies k 0 to cover the frequency band of interest. Note that the dc component is not excited. In the time domain, the signal looks like a periodic, noisy signal as is shown in Fig. 3a. The advantage of this forcing signal when compared to a random noise excitation lies in its periodicity. The trainee has full control over the amplitude spectrum, both in amplitude level and in frequency allocation. A single realization, which means that one selected set of phases  k is drawn, of the input signals spectrum is shown together with its time domain form (Fig. 3). Once the input forcing signal is chosen, it is time to start the measurement.

Measuring the Frequency Response Function The FRF of the system is measured first for a low RMS value of the input signal (A = 10 mVRMS). The amplitude Ak of the exciting multisine is then set higher. Note that multiple periods

IEEE Instrumentation & Measurement Magazine

February 2018

Fig. 4. Result of 3 FRF measurements with increasing amplitude at the input. Blue: A = 10 mVRMS; Red: A = 150 mVRMS; Black: A = 350 mVRMS; Green diamonds: FRF at A = 350 mVRMS averaged over 4 periods. Fig. 3. Example of a random phase multisine with excited frequencies f = 10 â&#x2C6;&#x2019; 200Hz, sampling frequency fs = 1000Hz, # points N = 2000 and VRMS = 1. a) multisine in the time domain. b) multisine in the frequency domain. Note that the noise level at â&#x2C6;&#x2019;300dB is due to numerical precision problems (15 digits) in MATLAB.

have been measured and only the steady-state is analyzed. This is applicable for all of the measurements in this paper. The FRF of the system is represented by the blue line in Fig. 4. The mechanical system behaves as a noisy LTI massspring-damper system with a resonance peak around 25 Hz. Classical LTI identification methods can be used to identify and parameterize the system [4]. For many practitioners, this very measurement is even considered to be sufficient to start the parametric identification process. In the LTI framework, the system is behaving linearly by assumption. Therefore, one expects that the FRF remains unchanged under an increasing excitation amplitude. Repeating the experiment using a scaled up version of the original excitation signal, with A = 150mVRMS (red) first and eventually A = 350mVRMS (black), shows a serious problem. The measured FRF represented by the black line in Fig. 4 shifts in frequency and becomes very noisy. This comes as a surprise, because one intuitively expects the SNR to increase for a growing excitation amplitude. This leaves the trainee with seemingly contradictory results, calling for further analysis and deeper understanding.

Measurement and Analysis of the Noise Behavior The paradoxical result obtained above will very probably convince LTI-skilled users that the poor quality of the FRF measurement is due either to the presence of a measurement error or to the presence of an unexpected and sudden drop of the SNR of the spectral measurement at the input and/or output of the system. The measurement could be attributed to a bad measurement setup resulting from bad connections, for example. However, the current measurement can easily be shown not to suffer from these problems. The amplitude of the measured input spectrum is close enough to its expected counterpart. Performing a sanity check on the setup readily shows February 2018

that no bad connections or bad cables come into the picture either. This shows the presence of yet another problem. When looking at the SNR of the input and output spectra, the SNR of the output spectrum decreases very fast with an increasing input amplitude. The practitioners' intuition readily attributes the problems to this drop, and the classical reaction is to average the measurement over successive periods. This allows the random character of the measurement errors to be easily removed. One of the advantages of a periodic excitation signal is that both the signal and the noise standard deviation (STD) can be recovered during this process. Whenever the perturbation is noisy and independent of the input signal, averaging of the FRF should result in a smoother FRF. This is because the random noise will be averaged out to zero while the signal part is deterministic and is not averaged out. The same experiment will also measure the level of the noise. In practice, averaging the measured FRF over multiple successive periods does not smooth the result, as seen in Fig. 4 in which the Green diamond points are exactly on top of the black line. As a conclusion, averaging did not influence the strange behavior of the FRF measurement. The reason for its presence is still not revealed. The black line FRF in Fig. 4 is indeed a valid measurement. The only option left is that the noisy result is due to an effect that is highly correlated to the input signal. Hence, the FRF becomes dependent on the input signal and this boils down to a nonlinear distortion. The measurements show that this distortion becomes more prominent when the amplitude is increased. The trainee is then invited to find out if and how a nonlinear behavior may enter the test system.

Root Cause of the Nonlinear Distortion Finding the root cause of the nonlinear distortion in a system can be a challenge. To allow for an intuitive analysis of the system, we simplify the setup to reduce it to its original concept (Fig. 5). The system consists of only three components: a Vshaped frame, a spring and a mass. This system is actuated by a shaker that displaces the rigid frame which moves the springs

IEEE Instrumentation & Measurement Magazine 19

Fig. 5. First realization of the concept using an elastic band, wooden mass and shaker.

and eventually results in a vertical movement of the mass. The materials used for this system can vary depending on what is available in the workplace. In our case, birch wood is used for the V-frame and a rubber band acts as the spring. A Brüel & Kjær accelerometer and shaker are used as output and input, respectively, to remove all doubts about the sensor and the actuator behavior. Note that the accelerometer is not shown in Fig. 5. Initially, this setup was used as a purely visual experiment. After the visual experiments, an accelerometer replaced the mass to analyze the behavior in a quantitative way. The advantage of this simplified system is that we know by design that the nonlinear distortion is created by the rubber band as this system only has three components. To reinforce our intuition, quick analysis shows that the V-frame is stiff enough to directly transfer the energy to the springs. The mass does not change over time either, and therefore, one can safely conclude that the origin of the nonlinear distortion lies within the springs.

Seeking the Origin of the Nonlinear Distortions The measurements performed earlier (Fig. 3) show that increasing the amplitude, hence the power of the excitation, results in an increased “noise” distortion at the system output. This “noise” is shown to be systematically linked (or heavily correlated) to the input signal. This conclusion can be drawn as the shape of the distortion for our test system does not change when the measurement is repeated with the same input signal. Hence, there is a clue to the assumption that this systematic effect results from the presence of a nonlinear distortion in the system. For further analysis, the trainee is pushed to conduct design based reasoning to detect the origin of the nonlinearity using a simplified white box model of the setup. We start from the simplified concept introduced in the previous section. A vertical force F is applied by the shaker to the V-frame. The frame acts as a rigid body and completely transfers the force to the flexible spring that is assumed to have a stiffness k. Eventually, the stiff mass m will move vertically. The system can be abstracted by a mass-spring-damper model. 20

A first reasoning uses a linear spring and damper model on a dynamical system at one frequency. Hence, the static relation between the applied force F and the displacement X of the mass attached to the springs is linear as shown in the red trace on Fig. 6. Most practitioners always rely on the assumption that the linear model of the behavior is valid for a “small” range excitations amplitudes. The “small” signal hypothesis is then obeyed which results in a linear relation between F and X. However, when the span of the variation of the Force F increases further, the linear models used in the abstraction will become questionable. In the white box case, it is sensible to question the linearity of the spring model first: an increasing input power F increases the displacement X of the mass and hence, stretches out the spring more. Logically, we assume that the origin of the nonlinear distortion lies in the nonlinearity of the spring and put it to the test.

Nonlinearity Test: Swept Sine Up-Down Test A measurement that empowers a better understanding of the behavior of this type of nonlinear mass-spring-damper system is a swept sine measurement. Using a sine with a linearly varying sweep with a very low amplitude (power) causes the system to behave like a LTI resonant system, as could be expected. Increasing the power causes a deformation of the FRF first and the occurrence of a big jump in the FRF when the sweep excites the system around the resonance frequency. In this case, a sweep up and a sweep down test result in a different behavior. It is then possible using these measurements to determine whether the system contains a hardening spring or a softening spring by looking at the dependence of the jump frequency on the excitation power. If the jump occurs at a frequency lower than the linear resonance frequency, the nonlinear system is softening. If the jump occurs at a higher frequency, the system behaves as a hardening spring. More information can be found in [5].

Fig. 6. Spring characteristic of a mass-spring-damper system with a hardening spring. Applying a high Force F results in nonlinear distortions.

IEEE Instrumentation & Measurement Magazine

February 2018

Fig. 7. Result of the swept sine in the time domain. a) sweep up.b) sweep down.

Measurement Results The simplified system is measured with a swept sine test. A quick look at a low and a high amplitude result reveals the systems' behavior. The results of the sweep up and sweep down are shown in Fig. 7 and Fig. 8., in which the system is bi-stable at the resonance frequency and is behaving as a hardening spring. This phenomena can occur in systems that can be modeled by duffing oscilator [6]. When applying the same swept sine, with a small amplitude, the jump effect is still present, but it becomes very small. This shows that the system is behaving non-linearily, even at low amplitudes.

Conclusion about the Experimental Test The results of the swept sine test show that the system is nonlinear, even for very small signals. They show that this demonstrator, although the conceptual model is simple, is more complex than originally predicted by the white box model. Even the the low amplitude displacement is already large enough to create a nonlinear distortion in the spring. Another problem is that the movement of the mass is not happening in one dimension as expected. As the movement is not constrained in the vertical plane, any mass imbalance will result in a horizontal and rotational movement, too. This shows that even the theoretical analysis has to be handled with care, as the hypothesis spanning it may not be sufficient to understand the full behavior of the system. In the next section, an improved version of the demonstrator is presented that deals with the rotational and horizontal mass movement due to imbalance while keeping the system low-cost enough to remain affordable to build.

An Improved Low-Cost Demonstrator To constrain the mass movement to be vertical even when the mass imbalance is present, the system needs to be modified. Two arms are added perpendicular to the ones of the simple February 2018

concept to provide a stabilization of the mass movement in the vertical plane. To obtain a low cost system, a remaining challenge is to replace the actuator and the sensor of the system by a low cost device. The reason behind this is that our initial goal is to offer a simple to build and inexpensive mechanical setup. The shaker and accelerometer from Brüel & Kjær that are present in the initial proof-of-concept design are replaced by low-cost alternatives that are widely available in general stores. The shaker is replaced by a broadband loudspeaker as the working principle is similar even if the quality of the loudspeaker is less. The accelerometer is also replaced by a cheaper, less accurate alternative, a MEMS accelerometer test board. The result is a low-cost demonstrator (50 Euro; $US 60) usable for nonlinear system identification (Fig. 1). The mechanical system also needs to be stable and repeatable besides being inexpensive. To test its robustness and stability, FRF measurements were repeated over time. The first results are shown in Fig. 4. The low-cost system was designed to mimic the behavior of the second prototype system as closely as possible. Hence, it will also contain both a linear and a nonlinear behavior. This provides the trainee with a broader spectrum of possibilities when compared to case of a purely nonlinear system.

Power Dependent FRF of a Linearizable System As the previous analysis showed that the “noisy” behavior of the FRF is highly correlated to the input signal, the trainee is asked to repeat the experiment with a similar but different excitation signals. The idea is to keep the power and the band of excitation of all the signals the same except for the phases. In this case, as the previous multisine measurements were taken using a random phase multisine, different sets of randomly chosen phases should be used that are drawn from a uniform distribution between 0 and 2π. A signal containing one set of random phases is called a realization of the signal. The trainees' attention is hereby put on the hypotheses and conditions

Fig. 8. Spectrum of the swept sine in the frequency domain. Blue: sweep up; Red: sweep down. Around the resonance peak, the system is bi-stable due to the nonlinear distortions. Note that the small jumps in the frequency spectrum are due to the transients caused by the fast sweeping.

IEEE Instrumentation & Measurement Magazine 21

Fig. 9. Result of the FRF for two different realizations of the random phase multisine excitation.

that are required to allow combinations of these experiments. We know now that the behavior of the nonlinearity depends on the power spectrum of the input signal, and only signals sharing the same power spectrum are to be combined. The trainee will then measure the system response using more than one phase realization of the same random phase multisine, which means that several experiments with different input signals are performed. The result of the FRF measurement for two different realizations of the random phase multisine is shown in Fig. 9. Each realization clearly results in a different “noisy” FRF measurement. The result shows that the nonlinear distortion for each realization of the input signal depends on the phase spectrum of that input signal. This is certainly not a disadvantage, but turns out to be a major advantage for the measurement method. Rather than averaging the measurement spectra over multiple, successive periods of the excitation signal, it is now possible to start averaging the spectra over multiple realizations of the excitation. The “noise” from the different realizations turns out to be random “nonlinear noise” that does not change with time but only with a change of the phase spectrum of the signal. This way of working [1], [2] eases the understanding of the advantage of averaging the measurement noise between the spectra associated to successive periods of the same input signal, and the averaging and calculating of the variance of the nonlinear distortion noise that will only vary if the phase spectrum of the input signal is changed. The idea behind this “smart” averaging is shown in Fig. 10. When the trainee repeats the same set of experiments for a different power spectrum, it immediately becomes very clear that the measured response depends on the power spectrum. Taking the mean value of the FRF over all realizations yields an approximated FRF that is called the best linear approximation [1], [2]. This method does not only measure the “best” linearized frequency response in mean square sense, but calculating the variance of the spectra to their mean values of the excited lines also gives the level of the signal to nonlinear distortion ratio. This information helps one to get an idea about the deviation between the system behavior and the behavior of a linear time-invariant behavior. More information 22

Fig. 10. The robust method. The BLA can be found through averaging over multiple realizations. Also the level of noise and the level of nonlinear distortions can be found as the nonlinear distortions are assumed to act as noise (from [1], used with permission).

about this method, the so-called robust method, can be found in [1] and [2].

Finding the Best Linear Approximation Two sets of 50 different realizations of the input multisine excite the system with A = 10mVRMS and A = 350mVRMS for all 50 signals. Four periods/realizations are measured. Instead of averaging the system response over multiple periods, it is now averaged over multiple realizations. The measured best linear approximation (BLA) of the two experiments is shown in Fig. 11 (black lines). Also, the level of the noise and the level of the stochastic nonlinear distortion are shown. The FRF is indeed much smoother when comparing with the result of a measurement with a single realization as shown in Fig. 9. This average curve represents the BLA of the system for that specific power spectrum, such that:

  GBLA arg min   Y  GU  G



Fig. 11. Result of the BLA with 50 realizations (black) with A = 10mVRMS (full lines) and (A = 350mVRMS) (dashed lines). Green: Level of the disturbing noise. Red: Level of the stochastic nonlinear distortions + noise.

IEEE Instrumentation & Measurement Magazine

February 2018

Note that the level of the stochastic noise due to nonlinear distortion is higher than the level of the noise when looking at the results of the measurement using a high input power. This would not be the case if the system would behave as an LTI system. This knowledge allows the user to assess how big the nonlinear distortion is when compared to the measured noise. Based on this information, the user can decide whether using the LTI identification methods and ignoring the nonlinear distortion is a possible way to obtain a simple model that remains valid enough to describe the system adequately for the application or if he has to change its identification methods. When looking back at the first measurement results (Fig. 3), it is now clear that the level of the noise due to the nonlinear distortion increases with an increasing input power. This can be further confirmed by using the robust method with the different power levels. The results also show two other phenomena that would not appear if the system was behaving as a LTI system: the resonance frequency shifts to the left with an increasing power and the system becomes dampened.

Analysis of the Dynamics When looking at the evolution of the BLA for an increasing amplitude, it can be seen that both the resonance peak and the damping change. The frequency is shifting to the lower values, pointing in the direction of a softening spring behavior. The resonance amplitudes drop more with increasing amplitude of the input signal. These changes in the dynamics of the system are due to the nonlinear distortion. For a user in the industry, this information can be important as these shifts can lead to dangerous situations. A shift in resonance peak can lead to excessively high vibrations in mechanical systems that can even cause it to fail. When comparing the measured softening effect with the hardening model explained earlier when the spring goes outside its linear region, it looks like the white box model is wrong. The result of the analytical approach shows a hardening effect, rather than the experimentally obtained softening effect. The reason for this discrepancy is due to the moderate power level used for the excitation signal, which results in a dominating friction distortion. When the sine sweep is used, as shown previously, the system behaves as a hardening spring as the jump occurs after the resonance when using a sweep up. The difference lies in the complete power level that is inserted in the system at a single frequency by the swept sine, while the power is distributed evenly in the multisine case. This causes a higher displacement of the mass in the former case. The softening effect measured and shown in Fig. 11 is due to some other (unknown) nonlinear distortions which could be friction in the spring connection. This contradiction shows the trainee that even for a simple system, the analytical approach does not always provide enough and/or reliable information to identify the system. This is also the case with real industrial systems: a user often encounters unexpected behavior during a measurement. This February 2018

Fig. 12. Input signal for the Fast method.

shows the high importance of prior knowledge to be able to perform identification. Through this method, called the robust method, it is already possible to retrieve a lot of information about the system, but by adapting the input signal even more, it is possible to qualify the nonlinear distortion. This means that it is possible to see if the contribution of the distortions is due to even nonlinear distortions (e.g., x2) or odd nonlinear distortions (e.g., x3) or a combination of both [7].

Qualification of Nonlinear Distortions It is known that even nonlinear distortions (e.g., x2) mainly increase the noise level of the FRF measurement while the odd nonlinear distortions (e.g., x3) increase the nonlinear noise level and can change the dynamics of the system [1], [2]. In the present case, the softening/hardening effect is the result of an odd distortion. This means that knowing the nature of the nonlinear distortions (odd or even) is very important when identifying a system. Additionally, this information helps to improve the model of the system as it helps to assess the structure of the nonlinear distortion. To be able to qualify the system using one single experiment, the input signal's frequency grid is engineered in a different way.

Input Signal The multisine used earlier excites every spectral line within the band of interest with the same amplitude. To be able to detect and separate the odd and the even nonlinear distortions, the input spectrum is adapted (Fig. 12). The difference is that the even frequencies and some of the odd frequencies are left out. Within a block of consecutive odd frequencies (in this case four spectral lines is a block), one random odd frequency is slected to be left out. Using this signal allows one to detect the linear, the odd and the even nonlinear distortions seperately. Detailed information on how to create such a signal can be found in [1], [2]. Note that the dc component is not excited. The measurement results at the output of the system are shown in the next section.

Result at the Output To explain the idea used to separate even and odd nonlinear distortion contributions, a sine wave excitation is used. When

IEEE Instrumentation & Measurement Magazine 23

Fig. 14. Result of the Fast Method on the mechanical system. Black: result of the FRF. Red o: odd nonlinear distortions. Blue +: even nonlinear distortions. Green: noise level. Fig. 13. An overview of the linear, odd and even contributions when using an odd random phase multisine as input.

a sine wave with frequency f is applied to a linear system, a spectral response appears at the same frequency f at the system output but with a potentially different amplitude and phase. This is not the case for a nonlinear system: using a sine wave as an input signal results in a sum of harmonically related sine waves, called harmonics, at the output (Fig. 13). This knowledge can be extended for multisines. When the nonlinear distortions are even (e.g., x2, x4), the even nonlinearities applied to the odd lines present in the excitation result in spectral energy that is present at even frequencies only. Whenever the even bins are not excited in the multisine, they can act as a detection for even nonlinear contributions. If the nonlinearity of the system is odd (e.g., x3, x5), the odd summation of odd frequency bins results in spectral energy present at odd frequencies. As the odd frequencies are also carrying the linear contribution of the output, randomly selected odd frequencies are left out from the input spectrum. As these lines are not excited, they cannot carry any contribution from a linear behavior and therefore only contain the odd spectral contributions. The result of the linear and nonlinear contributions when applying an odd random phase multisine to the test system is shown in Fig. 13. This method using a single input signal is called the fast method. The signal is now applied on the mechanical system with an input power of A = 350 mVRMS. The results are shown in Fig. 14. In Fig. 14, the system is shown to have odd and even nonlinear distortions at a level that is significantly higher than the level of the noise. This shows that although the system might seem simple and straightforward, the measured complexity is higher than what is originally predicted by the white box models. The advantage of this method is the fast qualification of the system as only multiple periods of one realization of the input signals are needed. Gathering knowledge about the system will help to identify the system more efficiently. The disadvantage of this method is the lower frequency resolution due to the absence of the odd and even detection lines in the input signal. 24

It is possible to combine the robust and the fast method to get most information out of the system [1], [2].

Conclusions Using LTI identification methods works well until nonlinear distortion start to kick in. In many real life cases, systems behave only approximately as LTI systems and are essentially nonlinear systems. Neglecting the resulting distortions can lead to dangerous situations. It is up to the practitioner to measure as much information as is possible about the system before making the selection of an appropriate identification method. In general, only a few practitioners know how to deal with nonlinear distortions and tend to ignore the danger. To address this problem, a set of systems is being developed to help introduce practitioners to the knowledge necessary to gather information about the system they are trying to identify and to deal with this information accordingly. They allow hands-on exercises to minimize the gap between learning the concepts that allow one to deal with nonlinear systems and dealing with real life systems. Using intuitive approaches allows practitioners to see how the nonlinear distortions are behaving and what their effect is on the system behavior. Once the practitioner better understands how the methods work, direct use of the methods on their own real life system is empowered. Making it low cost and portable makes sure that the practitioner has the freedom of learning when and where he/ she wants.

References [1] R. Pintelon and J. Schoukens, System Identification, A Frequency Domain Approach, New York, NY, USA: IEEE Press, 2012. [2] J. Schoukens, R. Pintelon and Y. Rolain, Mastering System Identification in 100 Exercises, New York, NY, USA: IEEE Press, 2012. [3] D. J. Inman, Engineering Vibration, 2nd ed., Upper Saddle River, NJ, USA: Prentice-Hall, 2001. [4] L. Ljung, System Identification: Theory for the User 2nd ed., Upper Saddle River, NJ, USA: Prentice-Hall, 1999.

IEEE Instrumentation & Measurement Magazine

February 2018

[5] C. Touze and M. Amabili, “Nonlinear normal modes for damped geometrically nonlinear systems: application to reduced-order

applied digital signal processing and parameter estimation / system identification.

modelling of harmonically forced structures,” J. Sound and Vibration, vol. 298, no. 4-5, pp. 958-981, 2006. [6] G. Duffing, Erzwungene Schwingungen bei Veränderlicher Eigenfrequenz, Braunschweig, Germany: F. Vieweg u. Sohn, 1918. [7] R. Pintelon and J. Schoukens, Frequency Response Measurements in the Presence of Nonlinear Distortions, Hoboken, NJ, USA: WileyIEEE Press, pp. 73-118, 2012.

Mark Vaes graduated as an Industrial Engineer in Electromechanics from the Erasmushogeschool Brussel, Belgium. In February 2013, he joined the ELEC department as a Ph.D. student. His main interests are in the field of system identification of linear and nonlinear systems. Yves Rolain received the E.E. (Burgerlijk Ingenieur) degree in July 1984, the degree of computer sciences in 1986, and the Ph.D. degree in applied sciences in 1993, all from the Vrije Universiteit Brussel (VUB), Brussels, Belgium. He is currently a Research Professor at the VUB in the Department ELEC. His main interests are microwave measurements and modelling,

February 2018

Johan Pattyn, joined the navy to become a Radar Technician after two years of engineering studies. Since December 2009, he has been with the Department ELEC of the Vrije Universiteit Brussel (full time tenure). He is responsible for rapid PCB prototyping and is also involved in the maintenance and repair of the instruments and circuits for the student labs. Gerd Vandersteen received his electrical engineering degree from the Vrije Universiteit Brussel (VUB), Brussels, Belgium in 1991. In 1997, Dr. Vandersteen received a Ph.D. degree in electrical engineering from the Vrije Universiteit Brussel/ELEC. During his postdoc he worked at the micro electronics research center IMEC as a Principal Scientist in the Wireless Group with the focus on modeling, measurement, and simulation of electronic circuits in state-of-the-art silicon technologies. This research was in collaboration with the Vrije Universiteit Brussels. Since 2008, he has been a Professor at the Vrije Universiteit Brussels/ELEC with interests in measuring, modeling, and analysis of complex linear and nonlinear systems.

IEEE Instrumentation & Measurement Magazine 25

basicmetrology Richard Davis

The Last Measurement of the Speed of Light


easuring the speed of light, and ultimately using this speed to define the meter in the International System of Units (SI), has much to teach us about basic metrology. These lessons will gain new relevance later this year if the revised SI gains final approval, as expected. The history culminating in the last measurement of the speed of light is reviewed in this issue’s column. Galileo is generally credited as the first to attempt a measurement of the speed of light, c. This was in 1638. He and an assistant were each equipped with a lantern whose cover could be quickly removed. Galileo also was equipped with some kind of clock. The measurement began when he uncovered his lantern and started timing. As soon as the assistant saw the light from Galileo's lantern, he uncovered his own lantern. As soon as he saw the light from his assistant's lantern, Galileo read the change in time Δt on his clock. By standing as far as possible from each other at a known distance D, the speed of light would thus be measured to be c = 2D/Δt. Galileo found that Δt was essentially zero and concluded that, even if c is finite, it is too fast to be measured using this method. The first measurements that could be interpreted as measuring a finite value for c were astronomical observations by Ole Roemer at the Paris Observatory (1675) and James Bradley in England (1728). They studied different celestial phenomena, but their observations could each be used to infer the value of the speed of light expressed in the units of their times. In retrospect, Roemer's data could have been used to calculate a value of c to within 20% of the accepted value; Bradley's result was even closer. These remarkable astronomical observations and their interpretation, though interesting, will not be discussed further.

Fizeau Improves on Galileo Instead, let's return to Galileo's experiment. By the mid-nineteenth century, advances in technology rejuvenated Galileo's basic idea of how to conduct a measurement of c, as Hippolyte Fizeau reported in 1849. In essence, Fizeau replaced Galileo's lantern with an intense light source. The lantern cover was replaced by a rotating notched wheel containing 720 teeth. The wheel could be made to rotate at different rates, which could be measured, thereby replacing Galileo's clock. Galileo's assistant was replaced by a mirror positioned at a known distance D 26

in a neighboring town, about 8 km away. The optics were carefully adjusted so that the observer saw a point of light, which looked something like a star, when the returning beam passed through a notch, but the star was eclipsed when the returning beam encountered a tooth. The first eclipse occurred at a rotation rate of 12.6 Hz, the re-appearance of the star occurred at twice this rate, the second eclipse at three times the rate, and so on. Fizeau was able to calculate the value of c from this data, and his result agreed with the value previously obtained by astronomers. Other experimenters soon followed, using improved designs with which they achieved significantly greater accuracy; but Fizeau did it first. Several animations of his experiment can be found on YouTube.

Bergstrand Improves on Fizeau's Method by Using Electronics Let's now move forward by a century, when the Fizeau experiment became the inspiration for a measurement of c reported in 1949 by Erik Bergstrand of the Geophysical Survey Office in Stockholm [1]. Bergstrand was interested in surveying and in commercializing a new type of surveying instrument which he called “the geodimeter.” His geodimeter was basically a Fizeau apparatus modernized by replacing both the spinning notched wheel and the observer with electronics. Bergstrand used a fast switch (Kerr cell) to create light pulses and an electronic detector (photomultiplier tube) to detect them. The Kerr cell and phototube were both activated by a crystal-controlled oscillator whose frequency, about 8 MHz, determined whether the phototube would register the flash for a given distance 2D or not. The oscillator took the place of Fizeau's toothed wheel. Bergstrand begins the report of his apparatus [1] with this simple statement: “Fizeau's principle of determination of the velocity of light can be used for the measurement of lengths.” What's important here is the synergy between length measurements and knowledge of an accurate value of c: Bergstrand measured the value of c in meters per second to verify the operation of his geodimeter. To do this, he had to know the distance D, of the order of kilometers, measured in terms of the international meter held at the International Bureau of Weights and Measures (BIPM) [2]. He also needed to apply a correction for the index of refraction of air, so that his experimental observations could be corrected to vacuum conditions. The value

IEEE Instrumentation & Measurement Magazine 1094-6969/18/$25.00©2018IEEE

April 2018

of c that he obtained was sufficiently accurate to calibrate similar instruments, which could then replace conventional methods for measuring geodetic lengths. Bergstrand's determinations of c were competitive with other methods [3] until K. D. Froome at the National Physical Laboratory (NPL) in Teddington, England, UK perfected a completely different way to measure c.

Froome Eclipses the Geodimeter Froome measured the speed of electromagnetic (EM) radiation at microwave frequencies, corrected to vacuum conditions. EM radiation includes the well-known spectrum from radio waves to gamma rays. As we all know, visible light is a part of this spectrum. Froome's final result [3], [4] was carried out with microwaves of frequency f around 72 GHz, which means that the wavelength  of the radiation was about 4 mm. The formula c = f  is exact at conditions which Froome could approximate to sufficient accuracy in his laboratory. In fact, that is the basic formula that Froome used to measure c. His microwave source had a calibrated frequency and his length measurements were made in terms of the wavelength of the red emission of a cadmium lamp (whose wavelength was traceable to the international meter, thanks to the efforts of Michelson and others a half-century earlier [2]). The length measurement was the greatest contributor to Froome's uncertainty budget; the uncertainty of the air index of the refraction was half as much, and the uncertainty of the microwave frequency was 1/20 as much. (We will see that this feature—relative uncertainty of frequency measurements much smaller than relative uncertainty of length measurements—is a theme which is repeated in the last measurement of c.) Froome's final result, c = (299 792 500 ± 100) m/s, is consistent with the last measurement of the speed of light, but 100 times less accurate.

The Team at NIST/Boulder Makes the Last Measurement of the Speed of Light What happened within the next 20 years to reduce the uncertainty of the measured value of c by two orders of magnitude? Several things: the meter was redefined in 1960 in terms of the wavelength Kr of a specified emission line of a krypton-86 lamp. The lamp is operated at the triple point of nitrogen (temperature about -210 °C), where gas, liquid, and solid phases co-exist in equilibrium. Without giving all the significant figures, Kr ∼ 600 nm, which we see as orange. Next, the unit of time, the second, was redefined in 1967 in terms of a microwave frequency obtained from an electronic transition of a cesium-133 atom. Cesium atoms are the heart of a device commonly referred to as the cesium atomic clock. The clock oscillates at a stable frequency fCs which, since 1967, has been internationally agreed to have the exact value 9 192 631 770 Hz. This frequency defines the second. April 2018

But the most disruptive new technology by far was the development of frequency-stabilized lasers [5]. These provide a source of coherent light at a single frequency. By 1972, a team at the National Bureau of Standards (now the National Institute of Standards and Technology, NIST) in Boulder, Colorado, USA had reported the last measurement of the speed of light. They accomplished this by first measuring the infrared frequency fCH4 of a methane-stabilized He-Ne laser in terms of fCs. The large frequency ratio fCH4/fCs ∼ 104 was a challenge. The ratio was measured to high accuracy by using a frequency chain involving five different stabilized lasers and five different microwave sources, with various clever techniques used as necessary to synthesize the frequencies required in a chain connecting fCs to fCH4 [6]. The authors of [6] attributed their ability to measure fCH4 with respect to the cesium frequency to the overall 100-fold improvement in their determination of c. Of course, the final step was still the determination of  CH4, about 3.39 μm, in terms of  Kr. John Hall, one of the team members, recalled years later: We did collectively find a good way to measure the speed of light. Of course that was relative to the then-existing length standard, which was some incoherent radiation from a krypton discharge lamp, operated at the triple point of nitrogen… The experimenters often needed to quickly respond when the optical path would be momentarily blocked by frozen nitrogen ice floating through the light beam [5]. Why was this the last measurement of the speed of light? Surely as technology improved, measurement of laser frequencies beyond the infrared into the optical range would become possible, further increasing the accuracy. And surely better techniques could be found to measure optical wave-lengths in terms of  Kr. While the methane frequency had already been measured to 6 parts in 1010 and lower uncertainties were already in the pipeline, the wavelength of the methane laser could only be measured to 3.5 parts in 109, with no prospects for further improvement. This was because the shape of the orange emission line of krypton is an envelope of wavelengths which is not perfectly symmetric about its maximum, making any significant improvement in basic length metrology problematic. It was clear that the meter definition based on krypton involved a technology that had already become obsolete after only 12 years of service. The NIST/Boulder team summed up the situation succinctly (see Text Box).

The Present Definition of the Meter The solution adopted by the world community of scientists was the second alternative presented in the Text Box, meaning that there could be no new determination of the speed of light. The exact value of c, officially sanctioned in 1983, is: 299 792 458 m/s, thereby redefining the SI meter. The same value had been

IEEE Instrumentation & Measurement Magazine 27

basicmetrology continued A new value of the speed of light with [improved] accuracy should be achievable if the standard of length were redefined. Alternatively, one can consider defining the meter as a specified fraction of the distance light travels in one second in vacuum (that is, one can define the speed of light). With this definition, the wavelength of stabilized lasers would be known to the same accuracy with which their frequencies can be measured [6].

light travels in vacuum during a specified fraction of a second. Change the fraction and you change the length of the meter— but not the speed of light! Later this year, the Planck constant, the elementary electrical charge, the Boltzmann constant, and the Avogadro constant will likely be “fixed” (that is, will get the speed-of-light treatment) to redefine the kilogram, the ampere, the kelvin and the mole. More about this when the time comes.

References [1] E. Bergstrand, “Velocity of light and measurement of distances by

recommended by a committee of international experts in 1973 and it was soon adopted by astronomers, although the value did not enter the SI officially until a decade later.

[2] R. Davis, “The story of invar,” IEEE Instrum. Meas. Mag., vol.

The Meter Definition has been a Success

[3] A. G. McNish, “The speed of light,” IRE Trans. Instrum., vol. I-11,

A lot has happened to confirm the wisdom of redefining the meter by fixing the value of the speed of light. The definition still serves us well despite 35 years of exponential growth in the science and technologies that rely on this definition [5]. The frequencies of many stabilized lasers continue to be measured with increased accuracy, and each frequency measurement automatically implies a wave-length known to the same accuracy. For example, the recommended experimental value of fCH4 is now some 200 times more accurate than was reported in 1972, and the wavelength  CH4 is consequently known to the same relative uncertainty through the relation  CH4 = c / fCH4, as envisioned in the Text Box. The fixed value for c also facilitates the use of time-of-flight measurements of EM radiation to measure large distances. A common rule of thumb in global positioning is “a nanosecond is 30 cm.”

Final Thoughts And yet…the idea that we humans can define the value of such an important constant as c can seem presumptuous. It is not. We have devised a system of units which includes the meter as the unit of length, and c was once measured in this system of units. However, the meter is now defined as the distance


high-frequency light signaling,” Nature, vol. 163, p. 338, 1949. 20, pp. 27-29, 2017, [Online]. Available: IEEExplore: https://doi. org/10.1109/MIM.2017.8036694. pp. 138-148, 1962. [Online]. Available: IEEExplore: https://doi. org/10.1109/IRE-I.1962.5006618. [4] K. D. Froome, “Determination of the velocity of short electromagnetic waves by interferometry,” Proc. Roy. Soc. A, vol. A213, pp. 123-141, 1952. [5] John L. Hall, “Remembering Isfahan,” Physics Today, vol. 68, p. 44, 2015, [Online]. Available: PT.3.2784. [6] K. M. Evenson et al., “Speed of light from direct frequency and wavelength measurements of the methane-stabilized laser,” Phys. Rev. Lett., vol. 29, pp. 1346-1349, 1972.

Richard Davis ( joined the International Bureau of Weights and Measures (BIPM) in 1990 following eighteen years at NIST- Gaithersburg, Maryland. He began at NIST as a post-doctoral fellow in the electrical standards group. Later, he had technical responsibility for dissemination of the unit of mass from the United States’ national prototype of the kilogram. At the BIPM, he worked in the mass department until he retired in 2010 as department head. He continues as a consultant to the BIPM Director. Richard is a Life Member of IEEE and a Fellow of the American Physical society.

IEEE Instrumentation & Measurement Magazine

April 2018

A New Low Cost Power Line Communication Solution for Smart Grid Monitoring and Management Giovanni Artale, Antonio Cataliotti, Valentina Cosentino, Dario Di Cara, Riccardo Fiorelli, Salvatore Guaiana, Nicola Panzavecchia, and Giovanni Tinè


odern smart grids require the improvement of measurement and communication infrastructures of distribution networks, at both medium voltage (MV) and low voltage (LV) levels. Distributed sensing and measurement systems are needed to provide all necessary data for grid monitoring, control and management, as well as for the implementation of a number of smart functionalities, such as remote control of distributed generators (DGs), real time analysis of power flows, automatic meter reading (AMR), demand side management (DSM), grid automation and so on [1]-[6]. Acquired network data include typical electrical network quantities and status variables (such as powers, voltages, currents, switches status, DGs power production, and remote commands) and also environmental and other parameters (temperatures, security or safety warning signals, etc.). Such data must be exchanged between the different players of the smart grid (distribution system operators, active users, prosumers). Thus, a fundamental smart grids requirement is the development of a capillary communication system, which must be reliable, cost-effective and easy to be installed in both MV and LV distribution networks. Nowadays, the most common solutions make use of wireless, GSM or power line communication (PLC) systems. This last option is a very suitable solution, whose main advantages can be summarized in their low installation and service costs, intrinsic security from cyber-attacks, and direct and complete control by distribution system operators (DSO). In fact, power lines are already located in the territory and they are owned by the DSO, thus communication provider costs are avoided and potential intruders would encounter difficulties in accessing the network. PLC has been widely implemented to support different smart applications in LV networks, such as AMR or DSM. Its employment is also envisaged by some recent standards for remote control of DGs connected to distribution grids (for example, CEI 0-21 in Italy or VDE-AR-N 4105 in Germany). While the PLC use in LV distribution networks is generally

consolidated, its application at MV level has posed some issues concerning the behavior of electrical network components (MV cables, power transformers, overhead lines) at relevant PLC frequencies and the signal transmission effectiveness over the MV grid. As an example, some of the authors' research activities in the field have been focused on measurement issues and procedures for the characterization and modeling of the PLC channel [7], with a particular focus on the frequency range reserved by the CENELEC EN 50065-1 for PLC signals transmission, i.e., 50-148 kHz. Another important issue for PLC exploitation in MV distribution grids is the signal coupling, which commonly requires the installation of dedicated MV couplers in both primary and secondary substations. To allow PLC signal transmission (with adequate bandwidth and different digital modulation techniques) and signal circuit insulation from the power grid, MV couplers must have low impedance in the desired PLC frequency band and high impedance at power system frequencies. Commercial couplers installation in the whole distribution grid is expensive and difficult; in fact, apart from equipment and manpower costs, further overhead costs are due to temporary service interruption, since the coupler installation requires the substation disconnection from the grid and it is not easy inside the existing air insulated and, even more, gas insulated MV switchboards. To overcome these problems, the authors have patented an innovative PLC coupler solution [8], which is based on the exploitation of capacitive dividers of voltage detecting systems (VDS). Such devices are normally installed in MV switchgears (from 1 kV to 52 kV) to indicate the mains voltage presence and ensure worker safety during operations on MV switchgears [9]. The developed solution includes a proper interface circuit between the PLC transceiver and the VDS socket to transmit/ receive the PLC signal to/from the MV network. The proposed solution allows simplifying the coupling system installation and reducing both equipment and manpower costs, because

The solution presented in this paper received the 2016 Best Application in I&M Award, announced at the 2017 IEEE International Instrumentation and Measurement Technology Conference (I2MTC 2017). April 2018

IEEE Instrumentation & Measurement Magazine 29 1094-6969/18/$25.00Š2018IEEE

such a solution does not require modifications of existing MV switchgears. To allow transmitting/receiving the PLC signal through the VDS socket, an interface card is connected between the PLC transceiver and the socket itself (otherwise, the PLC signal would be almost totally short circuited to earth). A schematic circuit of the proposed solution is represented in Fig. 3 [8], in the case of line-to-earth PLC signal injection (i.e., the signal is injected between one cable core and its earth-connected shield).

Interface Card Design and Related Measurement Issues

Fig. 1. Voltage detecting system with portable indicator on a MV switchboard in a secondary substation.

the electronic interface board is less expensive than a commercial MV coupler (which is not needed anymore) and no service interruption and MV switchboard modifications are required for its installation (the interface board can be directly connected to the existing VDS socket). The proposed PLC coupling solution and some experimental on-field tests are described in the following sections.

Basic Idea of the New PLC Coupling Solution Basically, a VDS is made by a MV capacitive divider, which provides an LV signal to a voltage presence detector; usually such a detector is an external plug-in device, which is connected to the dedicated MV switchboard socket by means of a two-phase plug and hosts an audible and/or visible indicator for voltage presence signalling (such as a blinking light shown in Fig. 1). From a circuital viewpoint, the capacitive divider MV terminal is connected to the MV bus-bars; a voltage limiting device, a measuring circuit component (usually a capacitance), and a short circuiting device are connected between the capacitive divider LV terminal and the earth (Fig. 2). The whole circuit can be represented by the series of the MV divider capacitance and the equivalent LV circuit capacitance, which also includes the stray capacitance; the equivalent LV circuit capacitance is higher than the MV divider one. Thus, the measured voltage at the VDS socket terminals is proportional to the mains voltage; the voltage reduction ratio is proportional to the capacitances ratio. The patented solution consists in replacing the voltage detector with a new external plug-in device, through which the PLC signal can be injected or received; thus, the adoption of 30

The interface card was designed starting from a simulation study aimed at characterizing the behavior of each electrical element of the MV network in the PLC frequency range. Models of MV cables, MV/LV transformers and VDS system were developed and experimentally verified. The developed models were used to design the PLC signal coupling solution through the VDS capacitive divider [7]. Finally, the designed solution was prototyped and experimentally tested, by means of both laboratory and on-field measurement campaigns. The developed interface card (Fig. 3) consists of a transmission (Tx) and a reception (Rx) circuit, both having a variable inductance (L) in parallel to the VDS socket line-to-ground connection and an amplifier. The function of the inductance is to obtain a resonant circuit between such impedance and the VDS capacitance at the PLC signal center frequency; this creates a high-impedance line-to-earth path for the signal, thus enabling it to transmit/receive the PLC signal to/from the MV network. The amplifierâ&#x20AC;&#x2122;s function is to increase the PLC signal level in both Tx and Rx modes (before transmitting and after receiving the signal, respectively). Furthermore, in Tx mode, the maximum signal transfer to the MV network is obtained through the impedance matching circuit, which allows adapting the transmitting output impedance to that of the MV capacitive divider. On the other hand, in Rx mode, a pass-band

Fig. 2. Schematic circuit of the Voltage Detecting System. [9]

IEEE Instrumentation & Measurement Magazine

April 2018

Fig. 3. Schematic circuit of the interface card: reception (Rx) circuit and transmission (Tx) circuit. [8]

filter is used to reduce the noise on the PLC signal. The desired operation mode (Tx or Rx) is selected through two switches, which are controlled by a PLC transceiver digital output [9]. In order to obtain repeatable and accurate results reducing the influence of stray parameters and noise, measurement procedures were developed to set the various interface card parameters. In fact, when an interface card is connected to an MV switchboard, the equivalent capacitance is a priori unknown. Furthermore, stray capacitances of the installation site are unpredictable. The measurement procedure allows tuning the inductance L, the impedance matching circuit and filter parameters, and setting the best values for both Tx and Rx circuits. A frequency response is firstly measured sweeping a sinusoidal signal with 50 kHz of SPAN around the center frequency. To ensure high input impedance, the measurements are performed with a digital oscilloscope (Fig. 4).

Fig. 4. Transmission station assembled inside a secondary substation. An interface card prototype is connected to the VDS socket panel. April 2018

It measures the received signal in the time domain and performs its FFT spectrum with a frequency resolution of 300 Hz. The inductances, the impedance matching circuit and the filter parameters are tuned to achieve the target bandwidth; for example, a 6 dB bandwidth of 15 kHz is required for an efficient transmission of an nPSK modulated symbol with a symbol rate of 9600 baud/s (Fig. 5). After, a transmission test helps to perform the final fine adjustments. It should be noted that the measurement procedures allow finding the target settings in the desired PLC frequency range even in the presence of the mains voltage, thus avoiding any service interruption during the interface card installation.

On-Field Experimental Tests The performances of the VDS coupler prototype were verified in different lines of the MV distribution networks on the

Fig. 5. MV PLC transmission channel frequency response during an on-field sweep test around the center frequency of 86 kHz.

IEEE Instrumentation & Measurement Magazine 31

Table 1–Fastest achievable bitrate for different MV lines [10] Operating condition of substations under test Substation #1

Intermediate Substation

Substation #2

Bit rate (bit/s)



by pass

by pass



by pass

by pass

by pass




by pass




Line length (km)


islands of Ustica and Favignana (small islands in the Mediterranean Sea). The on-field tests were performed in the presence of both mains voltage (i.e., 20 kV, 50 Hz), AMR PLC signals and noise. Different network topologies were considered including MV lines connecting by-pass, nodal and terminal substations and exploring also the case of possible intermediate substations. In the test campaign presented here, an nPSK modem was used with a maximum symbol rate of 9600 baud/s. By means of the mentioned measurement procedure, the interface card was tuned at its first connection to each VDS system of the different secondary substations. As expected, different values of variable inductance were found in different secondary substations. However, an efficient signal transmission was obtained for all of the tested topologies and for different center frequencies, with a transmission data rate up to 19.2 kbit/s. Some of the on-field experimental results for bidirectional communication are summarized in Table 1, where the fastest achievable bitrate is reported, corresponding to a minimum success rate of 98% (percentage of received information bit packets, with respect to the transmitted ones) [10]. In the table, the different network configurations used for the tests are specified, in terms of lengths of the MV lines connecting the substations under test and their operating conditions (by-pass, nodal or terminal). As an example, one of the MV lines tested is shown in Fig. 6, which connects two nodal substations of the Favignana MV network.



Modulation QPSK QPSK coded QPSK BPSK coded

Conclusions The obtained results show that the proposed PLC coupler solution can be suitable for its usage on MV distribution networks. In this framework, tests are in course for some potential smart grid applications, i.e., smart metering and remote control of distributed generators (DGs); the preliminary results have confirmed the feasibility of the proposed solution for exploiting PLC communication between DSO and distributed measurement systems over both MV and LV grids. Further ongoing research activities concern the coupling solution integration with the basic VDS functionality at power system frequency (i.e., voltage detection) and its communication performances improvement. As regards this last aspect, in order to improve data rate and transmission bandwidth, the performances of the proposed PLC solution are under study in the case of OFDM signals transmission.

References [1] C. Muscas, M. Pau, P. A. Pegoraro, S. Sulis, F. Ponci and A. Monti, “Multiarea distribution system state estimation, “ IEEE Trans. Instrum. Meas., vol. 64, no. 5, pp. 1140-1148, May 2015. [2] G. Rietveld, J.-P. Braun, R. Martin, et al.,"Measurement infrastructure to support the reliable operation of smart electrical grids,” IEEE Trans. Instrum. Meas., vol. 64, no. 6, pp. 1355-1363, Jun. 2015. [3] A. Cataliotti, V. Cosentino, D. Di Cara, S. Guaiana, N. Panzavecchia, and G. Tinè, “A new solution for low-voltage distributed generation interface protection system,” IEEE Trans. Instrum. Meas., vol. 64, no. 8, pp. 2086-2095, Aug. 2015. [4] A. Cataliotti, V. Cosentino, D. Di Cara and G. Tinè, “LV measurement device placement for load flow analysis in MV smart grids,” IEEE Trans. Instrum. Meas., vol. 65, no. 5, pp. 9991006, May 2016. [5] G. Artale, A. Cataliotti, V. Cosentino, D. Di Cara, S. Guaiana, S. Nuccio, N. Panzavecchia, and G. Tinè, “Smart interface devices for distributed generation in smart grids: the case of islanding,” IEEE Sensors J., vol. 17, no. 23, pp. 7803-7811, Dec. 2017. [6] D. Della Giustina, P. Ferrari, A. Flammini, S. Rinaldi, and E. Sisinni, “Automation of distribution grids with IEC 61850: a first approach using broadband power line communication,” IEEE Trans. Instrum. Meas., vol. 62, no. 9, pp. 2372–2383, Sep. 2013. [7] A. Cataliotti, V. Cosentino, D. Di Cara, and G. Tinè,

Fig. 6. Favignana MV line connecting two nodal substations named “4 Vanelle” and “Torregrossa.” A third nodal substation, named “S. Francesco,” is connected in the middle of the MV line [10]. 32

“Measurement issues for the characterization of medium voltage grids communications,” IEEE Trans. Instrum. Meas., vol. 62, no. 8, pp. 2185–2196, Aug. 2013.

IEEE Instrumentation & Measurement Magazine

April 2018

[8] R. Fiorelli, A. Cataliotti, D. Di Cara, and G. Tinè, “Coupling circuit for power line communications, “Assignee: STMicroelectronics s.r.l., Patent US08896393 B2, Grant Date: 25/11/2014. Priority date: 22/12/2010, [Online]. Available: patents/US8896393. [9] "Live working - Voltage detectors - Part 1: Capacitive type to be used for voltages exceeding 1 kV a.c.,” IEC 61243-1, 2009,

of Pa­ler­mo, in 2005 and 2009, respectively. Currently he is a Researcher at the Institute of Intelligent Systems for Automation (ISSIA) of the National Research Council of Italy (CNR), Pa­ler­mo, Italy. His current research interests include power quality measurements, current transducers characterization in nonsinusoidal condition, power line communications and smart grids.

[Online]. Available: [10] G. Artale, A. Cataliotti, V. Cosentino, D. Di Cara, R. Fiorelli, S. Guiana, and G. Tinè, “A new low cost coupling system for power line communication on medium voltage smart grids,” IEEE Trans. Smart Grid., vol. PP, no. 9, Nov. 2016.

Giovanni Artale ( received the M.S. degree in electronics engineering and the Ph.D. degree in electronics and telecommunications engineering from University of Pa­ler­mo, Pa­ler­mo, Italy, in 2010 and 2014, respectively. Currently, he is a Research Scholar at DEIM, University of Pa­ ler­mo. His current research interests include low frequency harmonic analysis algorithms, power line communications, and smart grids. Antonio Cataliotti (M' 01) ( received the M.S. degree in electrical engineering from University of Pa­ler­mo, in 1992 and the Ph.D. degree in electrical engineering in 1998. Since 2005 he has been an Associate Professor in electrical and electronic measurements at DEIM, University of Pa­ler­mo. His current research interests include power quality measurements, power line communications and smart grids. Valentina Cosentino ( received the M.S. and Ph.D. degrees in electrical engineering from University of Pa­ler­mo, in 2001 and 2005, respectively. Currently she is Assistant Professor in electrical and electronic measurements at DEIM, University of Pa­ler­mo. Her current research interests include energy and power quality measurements, detection of disturbances sources in power systems, virtual instrumentation and smart grids. Dario Di Cara (M' 16) ( received the M.S. and Ph.D. degrees in electrical engineering from University

April 2018

Riccardo Fiorelli ( received his bachelor's degree in electronics engineering from the Politecnico di Milano University, Italy in 2003. In the same year, Riccar­do joined STMicroelectronics. Since 2005, he has worked as a Senior Application Engineer for Industrial Products. He is especially an expert in the Power Line Communication system and application field. Salvatore Guaiana ( received the M.S. degree in electronics and fotonics engineering from University of Pa­ler­mo in 2012. Currently he is a Ph.D. student at DEIM. His current research interests include power line communications, development of microcontroller systems and communications protocols for smart grids applications. Nicola Panzavecchia ( received the M.S. degree in computer science and the master's degree in domotics and building automation from University of Pa­ ler­mo, in 2007 and 2011, respectively. Since 2007, he has worked as Consultant, Programmer and Web System Administrator. Currently he is a Research Fellow at ISSIA-CNR, Pa­ler­mo. His current research interests include software development for energy management and embedded systems, smart grids, power line communications. Giovanni Tinè (M' 04) ( received the M.S. degree in electronics engineering and the Ph.D. degree in electronics, computer science, and telecommunications engineering from University of Pa­ler­m o, in 1990 and 1994, respectively. Currently, he is a Researcher with ISSIA-CNR, Pa­ler­mo. His current research interests include electromagnetic compatibility, power-line communications and smart grids.

IEEE Instrumentation & Measurement Magazine 33

futureI&M trends in

Irina Florea

The Need for Standardization in Instrumentation and Measurement


hen we talk about measurements, we think of using instrumentation and applying measuring methods and procedures. But do we ever consider what lies beneath all of this? I mean, why is it necessary to measure temperature, pressure, length and so many other things by means of comparison with a standard unit? The answer is that there are standards that must be applied in order to be able to compare your results to someone else's. So, we reach the issue of standardization. Everything that we do in the field of instrumentation and measurement represents the result of applying standards. Standards refer to the quality of the process, to the environmental conditions that must be taken into account, and also to other factors that can influence each step of the process under discussion. When we talk about the quality of a process, we often refer to the International Organizations for Standardization (ISO) standards. These standards were created and are still under development, with the purpose of allowing people to “speak” the same language. As an example, let us take the specifications of a product/equipment/system that help us purchase the right one, optimal for our application. Meanwhile, specifications prove that there is still innovation in design of products/ equipment/systems, since the specifications tend to solve more and more problems. And still, how do we know that we have the right specifications for our needs? Our needs are usually adapted to certain standards (for example, we need a certain accuracy for measuring temperature in an explosive environment, we need certain quality standards to connect any equipment in an electrical network, or we need certain standards to fulfill to drive a car on public roads). Consequently, specifications are designed in order to ensure that the product/equipment/system meets the standard criteria. Due to the continuous development of technical fields, there is growing interest for standardization. So we cannot talk yet of the maximum development in creating standards.


Relating to the measurement area, for ensuring the optimal accuracy and reliability, there is always a need for standards to be adapted to the new technologies. First of all, to understand the importance of standards, we have to keep in mind the idea that the first ISO standard was released in 1951 and it concerned “standard reference temperature for industrial length measurements.” From that moment until present, almost 22,000 international standards have been published. So, there is a proof of the importance of standardization not only in technical areas, but in all of our everyday lives. Beside ISO standards, there are also other organizations that have published standards and guidelines, international or national, that need to be fulfilled in order to obtain a compliant product/equipment/system. Some of them are IEC, ITU, and ETSI, and there are many more. Regarding IEC standards, those are designed for electrical, electronic and related technologies. ITU standards are developed for communications at all levels. These standards refer to any communication technology: radio, air, cable communication. For instance, GSM technology is designed to fulfill ITU and ETSI standards. ETSI is the organization that refers to the area of information and communication technologies (ICT). Also, when we need to transfer measured data from a device mounted far away from the place where we are, we use remote measurements and data that are sent over a communication protocol. The protocol derives from standards applied. With regard to the ISO standards, it is important to mention that their form is under the surveillance of the International Laboratory Accreditation Cooperation (ILAC) and the International Accreditation Forum (IAF). The ILAC offers accreditation of metrological activities, while the IAF offers accreditation of certification activities. The process of standardization aims at obtaining optimal results with minimum effort. These results are not correlated only with the technical area but with everything around us, from the quality of water, to air that we breathe and to food. So, coming back to our subject of standardization in instrumentation and measurements, this is a matter in which any new technology has to undergo development. Without any standards to compare, any innovative process will terminate with no results. Imagine what life without same standards would mean: no safety standards, no reliable traceability, our

IEEE Instrumentation & Measurement Magazine 1094-6969/18/$25.00©2018IEEE

April 2018

work would not be judged by the same standard, everybody could sell and buy anything (it would be just a matter of the lowest price and no quality standards), and no standard to evaluate one's capabilities. Chaos? Yes, that is the correct word to describe life without standards. We can thank the numerous standardization bodies, all over the world, focused internationally or nationally, for their work which is reflected in our everyday lives, in the continuous improvements that change the quality of every process or product/equipment/system. If you still are not convinced of the importance of standards, just keep in mind two things: why are there so many entities that deal with standardization, and when is the last time you have come across a mention of a standard when acquiring/developing/improving something?

Irina Florea ( earned her B.S. degree in 2010 at the University Politehnica of Bucharest in the Department of Electrical Engineering. From 2010 to 2012, she pursued her Master of Science degree in electrical engineering at the same university, studying sensors and transducers and successfully writing her dissertation on instrumentation and advanced measurement systems. She presented research on instrumentation and data acquisitions at the IEEEâ&#x20AC;&#x2122;s 1st Annual International Measurement University, hosted by the Instrumentation & Measurement Society in Sardagna, Trento, Italy in July 2008. She began her work in industry at Renault Technologie, Romania in 2012 as an Acoustic Engineer. In August 2016, she became a Lead Engineer for Advanced Driving Assistant Systems with the same company.

aprilcalendar For more information on the meetings, please go to the I&M Society Web site at I2MTC 2018 / May 14-17, 2018 IEEE International Instrumentation and Measurement Technology Conference Houston, TX, USA MEMEA 2018 / June 11-13, 2018 IEEE International Symposium on Medical Measurements & Applications Rome, Italy

AUTOTESTCON 2018 / September 17-20, 2018 AUTOTESTCON National Harbor, MD, USA ISPCS 2018 / September 30 - October 5, 2018 International IEEE Symposium on Precision Clock Synchronization for Measurement, Control and Communication Geneva, Switzerland

CIVEMSA 2018 / June 12-14, 2018 International Conference on Computational Intelligence and Virtual Environments for Measurement Systems and Applications Ottawa, Canada

April 2018

IEEE Instrumentation & Measurement Magazine 35

societynews Ruqiang Yan

TC-39 Measurements in Power Systems


echnical Committee 39 (TC-39) of the IEEE Instrumentation and Measurement Society was founded in 2009 to create a reference for many members of the society working in the field of definition and characterization of measurement systems, devices, components and methods for modern power grids. The current roster of the TC includes nineteen members from eight countries. Since the beginning, the main activity of TC-39 has been the organization and sponsorship of the IEEE International Workshop on Applied Measurements for Power Systems (AMPS), whose first seven editions were held in Aachen, Germany yearly from 2010 to 2016, while the eighth edition was in Liverpool, UK September 20-22, 2017. The workshop provides a forum where qualified people can deeply discuss the different aspects related to instrumentation and measurement in power system applications and bring up critical opinions and innovative solutions to the challenges facing the new generation of measurement equipment. Over the years, AMPS, which also offers the opportunity for TC-39 to act as a bridge between academia and practitioners, has continuously grown in terms of both numbers of contributions and attendees and gained a high reputation in the scientific


community. Furthermore, the Transactions on Instrumentation and Measurement hosts, in a dedicated Special Section, the technically extended versions of some of the papers presented at AMPS, selected after a careful peer review process. TC-39 also organizes special sessions in larger conferences, such as the Special Session on “Measurements for Emerging Power Systems” at I2MTC 2016, held May 2326, 2016 in Taipei, Taiwan, and the Special Session on “New Power Quality Measurement Issues in Modern Power Systems” proposed for I2MTC 2018, to be held next year in Houston, Texas, USA. The other main activity of TC-39 is represented by standards development. In particular, in the last years, a joint project with TC-38 of IEC (Instrument Transformers) has been promoted to define a new Dual Logo Standard (IEEE-IEC). The Project Authorization Request P61869-105 “Recommended Practice for Uncertainty Evaluation in the Calibration of Instrument Transformers” was approved in May 2017 by the IEEE Standards Association. The kick-off meeting of the Joint Working Group, which also includes several IEEE members from other societies, was held November 7-8, 2017 in Bologna, Italy.

IEEE Instrumentation & Measurement Magazine

April 2018


IEEE I&M Society Awards Call for Nominations


he IEEE Instrumentation & Measurement Society (IMS) is soliciting nominations for its society and other Awards. To view the full detailed listing of each award please visit our Awards page on the IMS website: Nominations are due on varying dates, so please carefully refer to each specific award listing. Nominators should utilize the forms associated with each award description found on the website. For more information, please contact the Society Awards Chair, Reza Zoughi:

Outstanding Young Engineer Award

Award Nominations due by 1 August 2018

Distinguished Service Award

J. Barry Oakes Award Prize: $3,000 USD which may be used to attend a technical workshop or I2MTC or AUTOTESTCON; Registration at I2MTC or AUTOTESTCON for year in which lecture is presented; Plaque designating the individual as the recipient of the IEEE J. Barry Oakes Advancement Award. Eligibility: 35 years of age or younger at the time of the nomination. Other qualifications of the nominee include one or more of the following: Nominee actively engaged in engineering work in the field of I Nominee may hold a position in academia, government, or industry. Qualifications include one or more of the following: Demonstrated contributions to I&M science and engineering; potential leadership/project management skills; potential to serve as role model for other engineers. Nominees must exhibit actions that reflect positively on and enhance the reputation of the IMS The IEEE J. Barry Oakes Advancement Award will be used to provide a question and answer lecture during the annual I2MTC or AUTOTEST. Exceptionally, and upon motivated request by the recipient, the presentation will be given to another event fully sponsored by the IEEE IMS. April 2018

Prize: The Award consists of $2,000 and a plaque. Also, up to $1,000 will be paid to the recipient for transportation to the place of the presentation. Eligibility: The I&M Outstanding Young Engineer Award recognizes an outstanding young IMS member who has distinguished him/herself through achievements, which are technical, of exemplary service to the IMS, or a combination of both early in his/her career. The nominee must not have reached their 39th birthday and must be an IMS member at the time of nomination.

Prize: The Award consists of $2,000 and a plaque. Also, up to $1,000 will be paid to the recipient for transportation to the place of the presentation. Eligibility: The IMS Distinguished Service Award is presented each year to an individual who has given outstanding service to the IMS and to the I&M profession. All nominees must be, or have been, members of the IMS. Secondary considerations are service to the IEEE, IRE or AIEE, service to the engineering profession in general, technical accomplishments and outstanding technical leadership.

Technical Award Prize: The Award consists of $2,000 and a certificate. Also, up to $1,000 will be paid to the recipient for transportation to the place of the presentation. Eligibility: Any person with demonstrative and substantive achievement in the field of I&M may be nominated for the IMS Award. Membership in the IEEE is NOT a prerequisite. The IMS Society Technical Award is given to an individual or group of individuals for outstanding contribution or leadership in advancing instrumentation design or measurement technique.

IEEE Instrumentation & Measurement Magazine 37

announcementâ&#x20AC;&#x192;continued Career Excellence Award Prize: The prize is $5,000 and a plaque. In addition, the recipient may be reimbursed for travel expenses, not exceeding $1,000, to attend the ceremony during which the award is presented. Eligibility: A lifetime career in the field of instrumentation and measurement. The I&M Career Excellence Award is awarded to recognize a lifetime career of meritorious achievement and outstanding technical contribution by an individual in the field of instrumentation and measurement.

Outstanding Chapter Award Prize: $1,000 USD and a certificate. Pre-requisite: a minimum of two L31 forms have to be submitted to IEEE database for the application year and the previous one. This award will be given to the best chapter in a given calendar year based on activity.

Award Nominations due by 1 October 2018 Best Application in Instrumentation & Measurement Prize: $500 USD and a certificate. Candidate must be a Member (or higher-grade Member) or a Student Member of the IEEE and of the IMS at the time of accepting the Award. The purpose of the award is to recognize an individual whose idea applies measurement concepts or instrumentation technology in a novel way to benefit society. The application must be a working solution to an engineering need or problem.

Award Nominations due by 31 December 2018 Outstanding Technical Committee Award Prize: $1,500 USD to the TC and a certificate. All active technical committees of the IMS are eligible for this award. This award is given annually to the best technical committee of the IMS, one that best participates in I&M activities in an innovative way and delivers impact for the society.


IEEE Instrumentation & Measurement Magazine

April 2018

newproducts Robert Goldberg

Please send all “New Products” information to: Robert M. Goldberg 1360 Clifton Ave. PMB 336 Clifton, NJ 07012 USA e-mail:

Portable Single-Channel Wireless Oscilloscope Probe Operates to 30 MHz The IkaScope ultra-portable, 30 MHz, wireless oscilloscope probe weighs less than 60 grams and is designed to fit in your hand. With a WiFi connection to a smartphone, tablet or a computer, signals can be displayed and measurements can be easily made. Patented ProbeClick ® technology transforms the probe tip into an intelligent part of the oscilloscope. The probe detects when pressure is applied or released, allowing measurements to be automatically started and stopped without the need of a traditional RUN/ STOP button. The IkaScope is utilized as a wireless oscilloscope probe that displays measured signals and is configured via remote screen (tablet, smartphone or computer). Technically speaking, the lkaScope WS200 integrates a full-featured analog front end, a digital processing stage and a WiFi module. First, measured signals are conditioned, then converted to digital values via a high speed 8-bit ADC. Finally, a digital processing stage streams the captured signal frames over a high speed WiFi link. IkaScope incorporates an intelligent probe tip (called ProbeClick®) that detects when pressure is applied. This allows important power saving since all power consuming circuitry is in stand-by when a measurement is not in progress. That same intelligent probe tip is used to turn the IkaScope ON, which then turns OFF after 10 minutes of non-activity. During that period IkaScope keeps WiFi connection active for optimal responsiveness when quick measurements need to be made. April 2018

IkaScope may connect as well to your office or home WiFi network or it may create its own WiFi access point to which another display device may connect. IkaScope relies on a free application available on their website for display, analysis and configuration. IkaScope is highly portable. Therefore, applications that require quick, on-the-field measurements are a good match for this technology. Some examples are: ◗◗ Onsite maintenance and inspection ◗◗ Isolated measurements ◗◗ Design troubleshooting ◗◗ Education and research ◗◗ Manufacturing industry ◗◗ Aerospace industry ◗◗ Automotive industry For more detailed information, please visit www.ikalogic. com.

5G Network Emulation Solutions are Now 5G NR (New Radio) Ready Keysight Technologies has announced that their 5G network emulation solutions are now ready for 5G NR and will continue to support the new 3GPP NR standards. Keysight claims to be the first to launch network emulation solutions (5G Protocol R&D Toolset and 5G RF DVT Toolset) that enable device and chipset manufacturers to prototype and develop 5GTF chipsets and devices. Keysight's 5G network emulation solutions enable the device ecosystem to simplify workflows, share insights, and speed time to market. The workflow based solution portfolio uses a common and scalable software platform across a unified 5G development toolset that streamlines 5G modem and device R&D workflow. Keysight offers its users the ability to address 5G NR challenges and will support advanced channel bandwidth, beamforming, 8CC aggregation and multi-gigabit end-to-end IP data rates. Highlights: ◗◗ Keysight's 5G New Radio (NR) network emulation solutions allow users to prototype and develop 5G NR chipsets and devices

IEEE Instrumentation & Measurement Magazine 39 1094-6969/18/$25.00©2018IEEE

newproducts continued ◗◗ Keysight's solutions allow users to verify advanced 5G NR features (e.g., beamforming) across global spectrum requirements ◗◗ Keysight's solutions allow users to validate new 5G NR waveforms and complex 5G numerology with test automation tools More information about the 5G RF DVT toolset is available at A video demonstration of Keysight's 5G network emulation solutions is available on YouTube.

Extended Sampling Oscilloscope Range P i c o Te c h n o l o g y h a s added three 15 GHz models including a 25 GHz model to its professional, portable and low-cost PicoScope 9300 Series of Sampling Oscilloscopes. The new 15 GHz models replace the preceding 9200 Series 12 GHz models, adding significantly upgraded specifications at lower prices, with the result that all Pico Sampling Oscilloscopes now operate under the PicoSample 3 software. These instruments combine Pico's cost-effective sampling technology with the convenience of USB and LAN control ports. The 9301-15 provides the benefits of two channels at 15 GHz bandwidth and prescaled trigger to 14 GHz. It delivers market-leading 16-bit sampling rate of 1 MS/s in support of fast-update eye diagrams, persisted traces, histogramming and statistical analysis. Equivalent sampling rate tops out at 15 TS/s—that is a time resolution of just 64 fs—along with an unusually long maximum trace length for sampling oscilloscopes of up to 32 kilosamples. With a 15 GHz sampling bandwidth, Pico's entry-level Sampling Oscilloscope aligns with today's popular gigabit data rates. 15 GHz bandwidth will support third harmonic characterization of serial data out to 10 Gb/s and fifth harmonic out to 6 Gb/s. Full touch screen control, menus that configure to the application at hand, comprehensive PRBS pattern lock, and eye-line step and scan, all add up to a powerful, low-cost instrument available for visualization, measurement and characterization of high-speed serial data. The third of the new 15 GHz models, the 9311-15 addresses single-ended Time Domain Transmission and Time Domain Reflection measurements. It is a significant upgrade to the predecessor 9211 in cable, component, backplane and PCB impedance and transmission characterizations and network analysis. In this model, system transition time (65 ps) halves distance resolution, and adjustable pulse width extends 40

reflected fault detection range from around 4 mm typically out to 400 meters (1350 ft). At 20 GHz, the 9311-20 continues to support fully differential and de-skewable TDR/TDT capability, and all 9300 models can be paired with the PG900 standalone fast pulse generators to achieve similar TDR/TDT capability. At 25 GHz, Pico has created the model 9302-25 to add 11.3 Gb/s clock recovery to the higher bandwidth models. Full details of all nine models at bandwidths of 15 GHz, 20 GHz and 25 GHz can be found at rf-products.

Release of ATEasy 10 with New Features and Performance Marvin Test Solutions, Inc. has announced the latest version of ATEasy®, its evolutionary test software suite, first released in 1991 and currently deployed in test and measurement applications worldwide. ATEasy provides test engineers with all of the necessary tools to efficiently develop, debug, document, maintain, and execute test applications. Building on nearly 3 decades of hardware, software and system design expertise, MTS has continually invested in ATEasy with the goal to provide an ATE software product that is easy to use and maintain, offers unrivaled long-term supportability, and meets the demanding test requirements of today's complex systems. The latest version, ATEasy 10, delivers faster run-times (up to 10x faster in benchmarked tests), new integrated user collaboration tools such as ATEasy Merge. Net Controls support the ability to embed ATEasy run-time executables, a test log template that provides the ability to analyze test results, and backward compatibility with all previous versions. Developed from the ground up as ATE software for test and measurement applications, ATEasy is the only commercially-available test development / test executive software with an integrated hardware abstraction layer (HAL) and full simulation capabilities. Preserving test program development investment, ATEasy offers backward source and execution compatibility with all previous versions. ATEasy is ideal for companies with large and/or complex automated test applications, as well as system builders and/or integrators of multiple systems who require the flexibility, customizability and complete set of tools included with ATEasy. Find more information at

IEEE Instrumentation & Measurement Magazine

April 2018

Vector Signal Generator Achieves Extremely High Pulse Rates in PDW (Pulse Descriptor Word) Streaming Applications Rohde & Schwarz introduces a new realtime control interface software option for simulated radar scenarios based on pulse descriptor word (PDW) streaming. The radar scenario simulator streams the PDWs to the R&S SMW200A vector signal generator directly via LAN. Equipped with the new software option, the vector signal generator processes these signals to simulate highly agile and dense radar signal environments. The R&S SMW200 acts as an agile signal source that generates the highest pulse rates with superior RF performance. It supports both classical pulsed signals and any I/Q modulated signals. This solution is especially well-suited for extremely long-duration tests of radar receivers. To cope with today's demanding radar signal simulations, ultralong playtimes are needed in order to simulate realistic radar environments. Pulse sequences are calculated pulse by pulse and streamed as PDWs to an RF signal source. This avoids long calculation times and saves memory space in the signal generator. The R&S SMW200A is able to generate extremely high pulse rates (up to 1 Mpulse/s), as required for simulating dense signal scenarios and complex radar environments. Users can connect their PDW-based radar scenario simulators directly to the vector signal generator via LAN. The R&S SMW-K503 software option allows easy, fast and cost-effective integration of the R&S SMW200A as a signal source in state-ofthe-art radar simulation environments. With its extremely high processing speed of up to 1 MPDW/s, the R&S SMW200A enables testing of radar receivers at extremely high pulse rates. Customers additionally benefit from the signal generator's excellent RF performance. With an optional integrated second signal path, frequently needed additional interfering signals, such as adjacent communications signals, can be implemented quickly and easily in the single-box solution. The R&S SMW200A with two independent paths is the ideal solution for testing DUTs with several channels or generating radar signals in two different frequency bands. Each path can receive PDWs independently via LAN and output them on the same frequency or different frequencies. Multiple channels can be coupled phase-coherent for simulating different angles of arrival (AoA). For more information, visit: ad/press/smw200a. April 2018

Next Generation of 16-Bit Digitizers Spectrum Instrumentation announces the first products of a completely new designed digitizer card series. It consists of the new platform-board M2p, which will be the PCIe base for all upcoming products for the coming years. The other element is the new 59xx module, which will be available in many variations. M2p platform and 59xx module form 13 different new digitizer cards with lots of options regarding speed and channels. The M2p.59xx series is initially available with three different speed grades of 20 MS/s, 40 MS/s and 80 MS/s and from one to eight channels per card. Based on Spectrum's unique modular design philosophy, 13 different models can be ordered, ensuring a perfect match against the required specifications of customers. More models will be released in 2018. Although the size of the product has been reduced into a half-length PCIe card by the design team, it still offers even more features than predecessor families. Each channel has a separate ADC and a fully individual programmable input amplifier with ranges between Âą200 mV and Âą10 V, programmable input offset for unipolar measurements, programmable input termination of 50 Ohms and 1 Mega Ohms and an integrated calibration circuit. Models are available with up to 8 single-ended and up to 4 differential channels. The reduced card length of 167 mm allows the 16-bit digitizers to fit into much smaller PC systems than beforeâ&#x20AC;&#x201D;ideal for compact OEM solutions. Up to 16 cards in one system can be synchronized using Spectrum's proven star-hub technology. That allows systems to be created with up to 128 channels, all sharing a common clock and trigger, in one single chassis. For synchronization with external equipment, clock and trigger inputs and outputs are standard. For additional flexibility, 4 individually programmable connectors are available directly on the front-plate that offer additional trigger inputs, status outputs, synchronous digital input lines, asynchronous I/O or a reference clock input for an integrated time stamping unit. All units include a base version of Spectrum's SBench 6 software for first tests and simple measurement tasks. More information about Spectrum can be found at www.

Inertial-Grade MEMS Capacitive Accelerometers Silicon Designs, Inc. has announced the availability of its Model 1525 Series, a family of commercial and inertial-grade

IEEE Instrumentation & Measurement Magazine 41

newproducts continued MEMS capacitive acce le ro me t e r s, o ffe ri n g low-noise performance. Design of the Model 1525 Series incorporates Silicon Designs' own high-performance MEMS variable capacitive sense element, along with a ±4.0 V differential analog output stage, internal temperature sensor and integral sense amplifier—all housed within a miniature, nitrogen damped, hermetically sealed, surface mounted J-lead LCC-20 ceramic package. The 1525 Series features low-power (+5 VDC, 5 mA) operation, excellent in-run bias stability, and zero cross-coupling. Five unique full-scale ranges, of ±2 g, ±5 g, ±10 g, ±25 g, and ±50 g are currently in production. Each MEMS accelerometer offers reliable performance over a standard operating temperature range of -40 °C to +85 °C. Units are also relatively insensitive to wide temperature changes and gradients. Each device is marked with a serial number on its top and bottom surfaces for traceability. A calibration test sheet is supplied with each unit, showing measured bias, scale factor, linearity, operating current, and frequency response. Carefully regulated manufacturing processes ensure that each sensor is made to be virtually identical, allowing users to swap out parts in the same G range with few-to-no testing modifications, further saving time and resources. This provides test engineers with a quick plug-and-play solution for almost any application, with total trust in sensor accuracy when used within published specifications. As the OEM of its own MEMS capacitive accelerometer chips and modules, Silicon Designs further provides full in-house customization capabilities to customer standards. The Silicon Designs Model 1525 Series tactical grade MEMS inertial accelerometer family is ideal for zero-to-medium frequency instrumentation applications that require high-repeatability, low noise, and maximum stability, including tactical guidance systems, navigation and control systems (GN&C), AHRS, unmanned aerial vehicles (UAVs), unmanned ground vehicles (UGVs), remotely operated vehicles (ROVs), robotic controllers, flight control systems, and marine and landbased navigational systems. For additional information on the Model 1525 Series or other MEMS capacitive sensing technologies offered by Silicon Designs, visit

Broadband 26GHz SMT Module for 5G Plextek RFI has announced the development of a multi-chip module (MCM) to cover the recently-designated European ‘Pioneer Band’ for millimeter-wave (mmWave) 5G around 26 GHz. The development of the Front-End Module (FEM) was carried out in collaboration with Filtronic Broadband. 42

The band 24.25 to 27.5 GHz was designated last year by the EU Radio Spectrum Policy Group (RSPG) as the preferred band, or ‘Pioneer Band,’ for mmWave 5G. The FEM comprises a GaAs low-noise amplifier (LNA), power amplifier (PA) and transmit/receive switch housed in a custom laminate surface-mount (SMT) package measuring 10 mm x 10 mm. The receive path gain is 20 dB across the full band, with a noise figure of 3.5 dB. Transmit path gain is 19 dB, and the output referred third order intermodulation (IP3) is +36 dBm. Low-loss RF filtering has been integrated into the package structure, with a band-pass filter after the LNA and a harmonic rejection filter after the PA. Insertion loss figures are 0.7 dB for the band-pass filter and 0.2 dB for the harmonic rejection filter. More information about Plextek RFI's work on developing 5G MMIC and components can be found on the Plextek RFI website at and also on Plextek's You Tube channel.

Faster Large-scale Laser Scanning H e x a g o n M a n u f a c t u ring Intelligence has announced the Leica Absolute Scanner LAS-XL, a new ultra-large scale portable laser scanner. Designed for industries and applications where both speed of measurement and metrology-level accuracy are indispensable, the expanded measurement field and point acquisition rate of the LAS-XL means large parts and surfaces can be fully digitized in far less time than ever before. Leica claims this product to be capable of a larger scanning field than any other laser scanner on the market. Operating on the same flying-dot scanning technology as the Leica Absolute Scanner LAS, the LAS-XL benefits from a scan-line width of up to 600 millimeters (24 inches) and a measurement stand-off distance of up to a full meter. The extreme flexibility this delivers makes the LAS-XL as ideal for mapping large blade surfaces as it is for digitizing aircraft and rail carriage interiors. Accurate to within just 150 microns, the LAS-XL is suitable for the wide range of applications for which increased measurement speed is extremely valuable. For more information, visit

IEEE Instrumentation & Measurement Magazine

April 2018

GaN Technology for Linear & Compressed Amplifier Circuits Cree's CGHV40200PP is a unique, gallium nitride (GaN) high electron mobility transistor (HEMT). The CGHV40200PP, operating from a 50 V rail, offers a general purpose, broadband solution to a variety of RF and microwave applications. GaN HEMTs offer high efficiency, high gain and wide bandwidth capabilities making the CGHV40200PP ideal for linear and compressed amplifier circuits. The transistor is available in a 4-lead flange package. Typical applications include 2-way radio, broadband amplifiers, radar amplifiers, and test instrumentation. It can also be used to develop Class A, AB, Linear amplifiers suitable for OFDM, W-CDMA, EDGE, CDMA waveforms. Features: ◗◗ Up to 2.7 GHz operation ◗◗ 21 dB small signal gain at 1.8 GHz ◗◗ 250 W typical Psat (Saturated Output Power) ◗◗ 50 V operation For more information, visit

New Direct-drive Linear Motor Stages Deliver Sub-nanometer Resolution New from PI, these ultra-precise motor stages are now available with 0 . 2 nanometer reso lu tion linear encoders, ideal for high-end alignment, scanning and automation applications, in fields such as photonics, biotechnology and laser optics. PI has released a new precision feedback option for its linear stages equipped with direct-drive ironless 3-phase motors. Two types of position feedback systems are available: absolute-measuring encoders providing 2 nanometer resolution and incremental encoders providing 0.2 nanometer resolution with effective 0.5 nanometer minimum incremental motion at the stage platform. Ironless linear motors are used when high dynamics needs to be combined with extremely smooth motion. They are ideal

April 2018

for applications where extremely constant velocity is required, such as in optics inspection, metrology, photonics, interferometry, and semiconductor test equipment. The frictionless, zero-wear motor drives are also popular in fast automation applications where reliability and maximum up-time are mission-critical. For more information, please visit

Low Noise, 2-Stage Bypass Amplifier for Tower Mounting RFMW has announced the design and support for the Qorvo QPL9065 Low Noise Amplifier (LNA). This ultra-low noise amplifier is specified with a 0.5 dB noise figure at 1.95 GHz. Designed with two amplification stages and internal, 2nd stage bypass switch, gain is selectable at 17.5 dB or 37.5 dB. Operational bandwidth is 450 to 3800 MHz. DC power comes from a single positive supply of 3.3 to 5 V and control is via 1.8V CMOS TTL logic without external circuitry. Applications include Tower Mounted Amplifiers, base station receivers and repeaters. The QPL9065 is offered in a 3.5x3.5 mm package. Samples are available for qualified applications through RFMW, Ltd. To learn more about RFMW, visit their website at Robert Goldberg ( has over 35 years’ experience with over 25 years in management of the design and development of hardware and software for a broad range of military electronic products involving digital, RF/Microwave, electro-optical and electromechanical systems. He is retired from ITT Aerospace Communications Division in Clifton, NJ, where he was responsible for Sensor Communication programs utilizing the application of sensor radios developed by ITT as a result of work with DARPA on the Small Unit Operations Situation Awareness System (SUOSAS). Prior to joining ITT, he held positions in systems test and systems engineering with Northrop Grumman in programs related to RF and IR electronic warfare systems. He is a Fellow of the IEEE and is currently chairman of the Fellows Evaluation Committee of the IEEE Instrumentation and Measurement Society.

IEEE Instrumentation & Measurement Magazine 43

comingin June 2018

advertisingindex The Advertisers Index contained in this issue is compiled as a service to our readers and advertisers; the publisher is not liable for errors or omissions, although every effort is made to ensure its accuracy. Be sure to let our advertisers know you found them through IEEE Instrumentation & Measurement Magazine.

I&M around the World: Africa Reach over 3,600 readers in the June 2018 issue of the IEEE Instrumentation & Measurement Magazine




ATI Pickering


Phone 1 408 749 7000 1 541 471 0700

For advertising rates for IEEE Instrumentation & Measurement Magazine, contact Onkar Sandal, Allen Press, 1-800-627-0932 x 218, or

2018 IEEE Instrumentation & Measurement Society 2018 Officers President - Max Cortner, Executive Vice President - Salvatore Baglio, Vice President of Finance - Dario Petri, Vice President of Conferences - Chi-Hung Hwang, Vice President of Publications - Mark Yeary, Vice President of Membership - Sergio Rapuano, Vice President of Technical Committees and Standards - Ruqiang Yan, Vice President of Education - Kristen M. Donnell, Treasurer - Juan Manuel Ramirez Cortes, Senior Past President - Reza Zoughi, Junior Past President - Ruth A. Dyer,

Administrative Committee (AdCom) 2015–2018 Salvatore Baglio, Zheng Liu, Dario Petri, Juan Manuel Ramirez Cortés,

2017–2020 Alessandro Ferrero, Helena Geirinhas Ramos, Sergio Rapuano, Mark Yeary,

2016–2019 Octavia A. Dobre, Kristen M. Donnell, Christophe Dubois, Chi Hung Hwang,

2018–2021 Sebastian Yuri C. Catunda, Marco Parvis, Gourab Sen Gupta, Gaozhi (George) Xiao,

Other AdCom Members EIC for IEEE Transactions on Instrumentation and Measurement – Shervin Shirmohammadi, EIC for IEEE Instrumentation & Measurement Magazine – Wendy Van Moer, Graduate Student Representative, Andrea Angioni, Undergraduate Student Representative - Michael Nacy, IEEE Young Professionals Program Representative - Melanie Ooi, I&M Society Executive Assistant - Judy Scharmann, Region 1-6 Liaison - Lee Barford, Region 7 Liaison - Branislav Djokic, Region 8 Liaison - Ferdinanda Ponci, Region 9 Liaison - Jorge Fernandez Daher, Region 10 Liaison - Ruqiang Yan, Chapter Chairs Liaison - Luca De Vito,


IEEE Instrumentation & Measurement Magazine

April 2018

April 2018  

I&M around the World: Africa

April 2018  

I&M around the World: Africa