AI WILL SOON BE EVERYWHERE
ETHERNET ON THE EDGE
FASTEST SINGLE CORE MICROPROCESSOR
BLUETOOTH LE AUDIO BRINGS BETTER SOUND
INCREASE THE VALUE OF YOUR TEST TEAM
AI WILL SOON BE EVERYWHERE
ETHERNET ON THE EDGE
FASTEST SINGLE CORE MICROPROCESSOR
BLUETOOTH LE AUDIO BRINGS BETTER SOUND
INCREASE THE VALUE OF YOUR TEST TEAM
Take your multi-protocol wireless design to the next level with the PIC32CX-BZ2 and WBZ451 family of wireless microcontrollers (MCUs).
Featuring Bluetooth® Low Energy 5.2, a proven Zigbee® stack, and industry-leading security, this family is a complete solution for smart homes.
Home Automation at Your Fingertips
• Turn on living room lights, lock a door or control the blinds remotely
• Use the Bluetooth Low Energy capability to set up a Zigbee network that you control with a smartphone
• RF-ready modules, global regulatory-certified, modules save you time and money
• Hardware arbiter for reliable multi-protocol operation
• AEC-Q100 Grade 1 (125°C)-qualified SoCs for rugged applications
• Free reference design and wireless design check service
• Building Blocks training course walks you through every step of Bluetooth development
• Auto-generated code that you can drag and drop to start prototyping in seconds
• Demos walk you through adding touch, display, Wi-Fi and more to your application
Large language models (LLMs) are artificial intelligence systems that are capable of processing and generating human-like language. They have been hailed as a breakthrough in the field of natural language processing and have been used for a variety of applications, such as language translation, chatbots, and content generation. However, there are also dangers associated with LLMs that we need to be aware of.
One of the biggest dangers of LLMs is their potential to spread misinformation and disinformation. Because LLMs are trained on large amounts of text data, they can generate highly convincing fake news or propaganda. This could be used to manipulate public opinion or even to influence political elections.
Another danger of LLMs is their potential to perpetuate and amplify existing biases and discrimination. If LLMs are trained on biased data or algorithms, they may generate language that is discriminatory or offensive to certain groups of people. This could have real-world consequences, such as reinforcing stereotypes or exacerbating social inequalities.
LLMs also raise concerns about privacy and security. Because LLMs are capable of generating highly realistic language, they could be used to impersonate individuals or to generate fake content that appears to come from a legitimate source. This could be used to commit fraud, spread malware, or carry out other malicious activities.
Finally, there is the danger that LLMs may become too powerful and uncontrollable. If LLMs become self-aware or develop their own goals, they may act in ways that are not aligned with human values and goals. This could lead to unpredictable and potentially catastrophic outcomes.
In conclusion, while LLMs have the potential to bring many benefits, they also have the potential to cause harm. It is essential that we are aware of the dangers of LLMs and take steps to mitigate them. We need to ensure that LLMs are designed and used in ways that are safe, ethical, and aligned with human values. This requires a careful consideration of the potential risks and benefits of LLMs and a commitment to responsible use of this powerful technology.
Editor-in-chief Veijo Ojanperä firstname.lastname@example.org +358-407072530
Sales manager Anne-Charlotte Lantz +46-734171099 email@example.com
Advertising prices: etn.fi/advertise
ETNdigi is a digital magazine specialised on IoT and embedded technology. It is published 2-3 times a year.
ETN (www.etn.fi) is a 24/7 news service focusing on electronics, telecommunications, nanotechnology and emerging applications. We publish indepth articles regularly, written by our cooperation companies and partners.
ETN organises the only independent embedded conference in Finland every year. More info on ECF can be found on the event website at www.embeddedconference.fi.
The easiest way to access our daily news service is to subscribe to our daily newsletter at etn.fi/ tilaa.
Cover photo: AdobeStockVeijo Ojanperä ETN, Editor-in-chief
PS. This leader was generated by ChatGPT.
Embedded Conference Finland is the only independent embedded conference in Finland. In 2023 we celebrated our 5th anniversary.
You can find all the ECF23 presentations on the event website at embeddedconference.fi. Recordings are online in the ECF Youtube channel. All vital information on future events will be posted on the event website.
For more information see www.embeddedconference.fi To book a place or ask for sponsor packages please contact ETN editor-in-chief Veijo Ojanperä firstname.lastname@example.org
+358-407072530 or Sales Manager Anne-Charlotte Lantz email@example.com
This is the end of TETRA Sigfox refused to die
Even if you have to adjust your business, it's not worth risking information security, says Viivi Tynjälä of Check Point Software.
While system architects long contemplated integrating super-intelligent IoT systems in the past, it was never feasible financially. Today it´s a different game.
Single Pair Ethernet devices bring the reliable connections to the edge of industrial networks.
The STM32MP13 is STMicroelectronics´ most cost-effective industrial-grade microprocessor reaching 1 GHz
Bluetooth LE Audio might change our listening experience fundamentally.
Increased awareness about health has given rise to the demand for small but high accuracy devices that can measure various vital signs and health markers.
42 INCREASETHEVALUEOFYOUR VERIFICATION
Three factors are needed to drive more business value through your test team: data, systems and processes.
It is a bit cheeky to pronounce any technology obsolete, especially when it is used daily in many parts of the world. In Finland alone there are more than 50 thousand Tetra terminals connecting to the nation-wide Virve network every day.
And if you ask Francesco Pasquali, the chairman of TETRA industry group at TCCA, he´ll be upset.- The news of Tetra dying are absolutely false. We are still developing the standard and focusing on both broadband and adding security, Pasquali said in Critical Communications World 2023 in Helsinki in May.
But Timo Lehtimäki, the CEO of Erillisverkot says it is time for Tetra to retire.- We´ll be closing our Tetra network by the end of 2028, says the head of company that operates the Tetra based Virve network in Finland.
The big transition in mission critical networks refers to new requirements of networks. The amount of data is increasing, so various authorities have a big need for broadband data connections. This need is widely recognized in the industry, but the solutions in different countries differ. In Finland, a hybrid model has been chosen, where Virve and LTE/5G operate side by side during the transition period.
This transition time is needed for many reasons.- It will take time before, for example, the police vehicles are equipped with new equipment, Lehtimäki clarifies.
Some concern has been caused by whether there will be sufficient coverage throughout Finland after 2028. The police and other officials need to be able to operate even remote places, like the long eastern border of Finland, the new NATO country. Lehtimäki believes that the situation is actually getting better.
- The coverage of our Tetra based Virve has never been 100 percent. With mobile networks and satellite connections it is possible to reach a coverage of 100 per cent.
Lehtimäki refers to the new 5G standards, which also support calls via satellites. It
is clear that building mission critical connections after Virve will be "super expensive", as Lehtimäki says.
- This is not just a base station issue. Electricity supply and core networks connection must be taken care of there. Of course Elisa was aware of this, she refers to Erillisverkko´s operator partner in the transition.
The advantage of Virve has been the frequency it is using. At 400 megahertz, the base station can reach a distance of 50 kilometers. In this way, it has been possible to cover sea areas in the Gulf of Finland, for example. There have been base stations on the Estonian side, so there have been no gaps in the sea either.
5G uses higher frequencies, so coverage requires a lot more base stations. And in the future, remote areas will be managed
with satellites. At the end the aim is to run mission critical connections everywhere, even in remote Lapland like the Urho Kekkonen National Park, where Virve never worked.
The broadband transition was clear in the CCW23 exhibtion floor. Motorola presented a networked police car that had been driven across Europe to Helsinki. The car has, for example, the M500 video system, whose LTE connection enables continuous video streaming, real-time alarms and evidence recovery. Operational tasks can be managed via Apple CarPlay with the PSCore application.
On their booth Motorola also showed a new Tetra terminal, which also works on the LTE networks. Next to the new Tetra radio phone, i.e. the MPX600 model, a large screen is very visible. It is easy to imagine a police checking video
Mission critical communications is heading for a transformation. Tetra standard has served the officials well for 25 years, but it cannot cope with the new requirements for broadband data. It is time to bid farewell to Tetra.
streaming to the screen from command center. The device works on the Tetra network by default, but the connection can be selected from the menus as needed.
Digitization, broadband connections, artificial intelligence and analytics are also coming to the use of the authorities. However, development is slow. According to the Omdia research institute's forecast, in 2026, approximately 82 percent of the goverment radios devices will be digital. The old analog radiotelephone, especially in the P25 network in the US is still running strong.
And when will the police get their hands on a 5G phone that also works on the Tetra network? The reasons for the slow digitalization of the police are both technical and political, or actually business-related. According to Motorola, this depends on demand. There are
countries in Asia that want to switch to the use of 5G networks, and the Nordic countries would also be ready for it quickly, but without terminals it will not be possible. And if there are no volumes, there will be no devices.
The need is obvious. 10 years ago, the police still went to the scene of an emergency or crime with a notepad. Nowadays, it is well known from studies that without video, a large part of minor crimes remain unsolved. It is clear that the equation must be solved somehow.
In the future mission critical communications will not be just about high bandwidth. Police will be equipped with bodu-warn cameras. During CCW23 Motorola announced that they´ll be equipping the whole force of London City policw with VB400 body-warn cameras. In a way the BWC is a two-way device. It helps to ensure the safety of the citizens,
but it also promotes transparency and accountability of the police work.
The VB400 system is designed to streamline an officer’s workflow. After a shift, officers simply place the VB400 into its dock where it will automatically upload footage of the day’s events into VideoManager evidence management software. VideoManager will store the data in-country and organize it with time, date and location details along with supporting incident data reported by officers.
And in remote places drones will be used to set up mission critical connections. At CCW23 Nokia showed the foirst ever CEcertified, turnkey drone-in-a-box solution. The solution meets the EU safety requirements as the Nokia Drone Networks solution connects over public and private 4G/LTE and 5G networks for first responders and other professionals.
The 5th annual ECF event moved to a new venue Dipoli on the Aalto university campus. Embedded Conference Finland has established its position as the leading embedded event in the country.
Cyber attacks are becoming too easy thanks to artificial intelligence. And very dangerous. A phishing attack can already be created using artificial intelligence without the developer having to write a single line of code, said Check Point's information security expert Jarno Ahlström in his keynote speech at the Embedded Conference Finland event.
Artificial intelligence in itself is an old idea, but recent developments have brought the topic to the public and into the hands of more and more people. Now we want to move towards AGI technology (artificial general intelligence). It would be a machine that could answer and solve any problem, draw a picture, compose a song or make a presentation
independently, Ahlström said.
- When OpenAI's ChatGPT 4 came out in March, some already asked if we have reached the AGI level. I think not. The datasets are a year and a half old, so it doesn't know anything after that. Nor can it train itself any further.
San Francisco-based OPenAI has other tools. DALL-E generates images from text very efficiently. Codex knows how to
complete software code because it's trained on 54 million Github code repositories. However, ChatGPT, introduced in November 2022, was the one that blew the bank. And all this happened in just two years.
ChatGPT can be used for good, but unfortunately also for harmful purposes. Ahlström demonstrated how ChatGPT can be asked to compose a phishing email that appears to come from a fictitious
Researchers at Linköping University and the KTH Royal Institute of Technology have developed the world’s first transistor made of wood. Their study, published in the journal PNAS, paves the way for further development of wood-based electronics and control of electronic plants.
- We’ve come up with an unprecedented principle. Yes, the wood transistor is slow and bulky, but it does work, and has huge development potential, says Isak Engquist, senior associate professor at the Laboratory for Organic Electronics at Linköping University.Person holding a small wooden construction infront of face.
The researchers used this to build the wood transistor and could show that it is able to regulate electric current and provide continuous function at a selected output level.
In previous trials, transistors made of wood have been able to regulate ion
transport only. The transistor developed by the Linköping researchers, however, can function continuously and regulate electricity flow without deteriorating.
The researchers used balsa wood to create their transistor, as the technology involved requires a grainless wood that is evenly structured throughout. They removed the lignin, leaving only long cellulose fibres with channels where the lignin had been.
These channels were then filled with a conductive plastic, or polymer, called PEDOT:PSS, resulting in an electrically conductive wood material. See more here -> etn.fi/14901
web hosting service. The generated email can be in any language. ChatGPT can be asked to replace the feedback link with another one that loads additional information from the included excel file.
- Next, we use Codex to create the desired excel file. The task is to prepare a working VBA code in an excel file, which downloads the program to be executed from the desired address and runs it. The code must be included in Excel so that it is run as soon as the excel file is opened.
In this way, a phishing attack can be created using artificial intelligence without writing a single line of code. In addition to that, artificial intelligence can be asked to create various scripts that can be used, for example, to scan the machine for vulnerabilities. ChatGPT can then be asked to create an exe file from previously developed scripts.
- Without a single line of code, I have managed to create a malware using ChatGPT and Codex.
But what should an embedde designer do, if an unexpected shortage of critical components threatens to delay the project? David Gustavik, the designer of the cycle, highlighted ways to solve the problems in his keynote.
Even if the company has a good relationship with the component supplier, it does not necessarily guarantee availability if there are problems in the market.- The problems of the automotive industry during the laste years should prove that even the largest semiconductor suppliers cannot solve all problems, Gustafik said.
One of the most important means is documentation.- Proper documentation is a designer's best friend. A circuit diagram alone is not enough because it does not necessarily convey the reasoning that led to each component selection.
According to Gustafik, the designer should always document what a certain part or component does, why a particular
component was chosen for the project and how it relates to the rest of the design.- One way is to prepare the libraries so that the alternative component is always included either in the circuit diagram or in the bill of materials (BOM).
Gustafik gave an example of a design for which a "perfect" single-chip LDO solution was chosen. It did everything it promised in the demos. Then when they started to buy the component from the market, it was nowhere to be found. One special circuit had to be replaced with several generic chips.
- The price was roughly the same. The PCB area grew slightly and the power consumption was slightly higher. But now, instead of being stuck with one supplier, I can for example choose from hundreds of different op-amps without having to change the circuit board.
All the ECF23 presentations are here: www.embeddedconference.fi
Silicon Labs has announced two new integrated circuit families designed for the smallest form factor IoT devices: the xG27 family of Bluetooth systems on chips (SoCs) and the BB50 microcontroller unit (MCU).
Designed for the smallest IoT devices, the xG27 and BB50 families range in size from 2 mm-squared, about the width of a #2 pencil lead, to 5 mm-squared, less than the width of a standard #2 pencil. These offer IoT device designers energy efficiency, high performance, trusted security, and in the case of the xG27 family, wireless connectivity. This makes the xG27 SoC family and BB50 MCU ideal for tiny, battery-optimized devices like
connected medical devices, wearables, asset monitoring tags, smart sensors, simple consumer electronics like toothbrushes and toys, and more.
The new xG27 family of SoCs comprises the BG27, for Bluetooth connectivity, and the MG27, supporting Zigbee and other proprietary protocols. Built around the ARM Cortex-M33 processor, the BG27 and MG27 share several common features designed to make them the ideal SoC for small form-factor devices.
The BG27 and MG27 open up and extend a wide range of possible applications and use cases.
See more here -> etn.fi/14735
Carbon nanomaterial developer Canatu has upgraded its carbon nanotube (CNT) film manufacturing to meet the rising demand for ADAS heater products in the automotive industry. Canatu has improved its step-and-repeat (SAR) CNT film production line by adding new reactor capacity and ramping up the level of automation. These improvements have effectively doubled production speed and increased the output of Canatu CNT film. Canatu has been massmanufacturing CNT films for the automotive industry since 2015.
- With the rapid development of Advanced Driver Assistance Systems (ADAS) and autonomous driving, the demand for Canatu CNT film products has increased significantly, making it more vital than ever to increase production capacity, says Antti Valkola, VP of Manufacturing at Canatu.
AMD reports, that the Spartan 6 FPGA devices will have an extended lifecycle through at least 2030. All speed and temperature grades are included. Spartan 6 with high IO-to-logic ratios in
Unless you´ve been hiding under a rock you have heard about OPenAI´s ChatGPT which will revolutionalize most of the service sectors and professions. But AI cannot replace the fiels application engineers, says Graham Maggs, the marketing chief for Mouser EMEA.
- We are fully aware of ChatGPT, of course. We´re certainly keeping a close eye on it. We are not utilizing that AI process except in small marketing applications. And even at that we´ll have a human edit involved. We are not creating content using AI and we have no plans to.
Maggs maintains that there are restrictions on what you can use AI for.We´re not into simplistic ways to create content. We want our customers to be sure in the quality of the content and the material we are using. The content has to be significantly better that with using AI.
- Whether AI in ten years´ time is going to be significantly different from what it is today, who knows. AI by nature is a selflearning process. As it learns, it should be improving. At this time it is not our plan to utilize AI.
Engineers today are a great deal more trained in the multi facets of their
business, Maggs emphasizes. - They are production engineers, design engineers, component engineers manufacturing engineers, quality engineers all wrapped up into one person.
This doesn´t mean that Mouser is not investing in automation. If the company spend 80-90 Million dollars in a new building, they invest 120-150 Million into automation within that building.
But that will not remove the individual from the warehouse.- We´ve actually doubled the number of people working for Mouser during the last five or six years. So we´re adding people throughout the company, but we´re trying to make the system and the process much more efficient.
- Automation only goes so far. It’s the human with those robots, that makes the difference. You can put as much automation into the system as you want, but if you don’t have the correct human interface, then it falls down.
At the end of 2022 the EMS companies had component stocks worth of 17.5 Billion euros. Is the whole question of availability a self-inflicted problem?
small form-factor packaging, continue to be well suited for customers in industrial, medical, vision and automotive markets as well as other markets, like communications, where simple bridging is required. Customers in these segments require extended product longevity, typically requiring product lifecycles of 15 years.
- I´ve been in this industry since 1978. I´ve been on the road as a sales person since 1982. There has always been peaks and drops in this business. This is no different.
- Yes, the supply chain is full with the exception of specific components which have had a huge backlog in capacity caused by specific demand and specific shortages of materials.
Demand is good in all industries around the world, Maggs reminds us. --->
The markets are very healthy. Still the numbers are correct.- The EMS companies, distributors and manufacturers did something which I call double dipping. Actually, I think there was a lot of triple dipping, especially by the EMS houses. And it´s not just to have products as soon as they´re needed, so they placed orders in someone like Mouser and someone like Digi-Key and someone like Arrow, and whoever was able to supply those products quickest they would take those products and cancel elsewhere.
That´s why a lot of distributors started to take in orders with no change of canceling. At the same time they did take in orders and had to accept cancellations. We´re no different. We´ve had lots of cancellations, but it is different from an Arrow or an Avnet. Getting a cancellation
on a thousand dollar order is a lot different from getting a cancellation of a Million dollar order.
- I know that there is a lot of inventory out there. I know that there are still shortages. I´m confident that the demand is still there. I think we´ll see a recovery of bookings at the end of the year.
For Mouser inventory is king. If people need something, especially in small quantities, they come to us. And we´re still up from last year, by they way. We were forecasting a very small, couple of digits growth this year and we were even worried that we´d be below last year but that’s not been the case. Only recently we´ve gone below double digit growth this year, so it´s been a very good start for the year.
WithSecure’s USB armory is the world’s smallest secure computer. It’s used in many different applications by organizations all over the world, and now, beyond.
Experiments in Zero Gravity, is a program operated by the German Aerospace Center’s (DLR) Materials Physics in Space and Aerospace Medicine institutes, and the Mobile Rocket Base (MORABA).
Their latest mission, MAPHEUS-13, was launched on May 22, 2023 to conduct experiments on 3D printing components made of metal in zero gravity, the response of molten alloys to weightlessness, healing processes in the central nervous system or brain in reduced/increased gravity, and more.
Included in the mission as a part of Experiment 007 EV2 was WithSecure’s USB armory to assess its security framework and capabilities for protecting data produced by experiments. Specifically, it aimed to assess post-
quantum cryptography for secure key exchange in a trusted execution environment running on a USB armory.
- We were looking for a solution to extend our real-time systems in Experiment 007 with a secure compute system that allows us to integrate computing power for more advanced algorithms, data analysis, and standard software in a neat, lean, and secure way. The USB armory proved to be small, versatile, and powerful enough to fit our requirements. In the long run we will extend the use of TamaGo, GoTEE and Linux to enable complex data analysis and AI/ML use-cases for the scientists at DLR in-flight,” said DLR partner and Chief Security Architect at adesso SE Christian Kahlo.
The USB armory is an open-sourced, single board computer with a unique form factor and capabilities. It has been used in a variety of applications, including encrypted storage solutions, hardware security modules (HSM), enhanced smart cards, electronic vaults , key escrow services, and more.
According to the recent Photonics Industry in Finland survey, the industry’s estimated turnover in Finland exceeds two billion euros. Compared to the previous study conducted in 2020, the photonics industry has grown by more than 50 percent. There are about 300 purely photonics companies which employ more than 6,000 experts in the field.
UK-based quantum computer developer OQC will install its hardware in an Equinix data center, allowing direct access to OQC's quantum computer using the Equinix Fabric service. Finnish customers will be able to run complex algorithms and test different use cases remotely, without direct access to the hardware itself.
Renesas has completed the acquisition of Panthronics AG, the Austrian fabless semiconductor company specializing in NFC solutions. The acquisition provides Renesas with in-house capability to instantly capture the growing market opportunities for NFC. -> etn.fi/15044.
NXP Semiconductors has announced the i.MX 91 applications processor family. Building on more than two decades of developing multimarket applications processors, the i.MX 91 family delivers an optimized blend of the security, features, and energy-efficient performance required for the next generation of Linux-based IoT and industrial applications. The i.MX 91 processor features an Arm Cortex-A55 running at up to 1.4 Ghz plus support for modern LPDDR4 memory and dual Gigabit Ethernet for gateway applications. -> etn.fi/15024.
Along with LoRaWAN, Sigfox is the other "original" IoT radio technology. In the spring of last year, the whole story was about to end when Sigfox SA, the original parent company who developed the technology, went bankrupt. In the auction, all the assets ended up in the ownership of Singaporean UnaBiz. Under its management, the operation has recovered and now Sigfox seems to be fulfilling the promises made to the technology.
Sigfox operates on free ISM frequencies and is in many ways a superior IoT technology. Then how could it be that the company that developed the technology failed in practice, even though it collected a large pot of money from the market? According to UnaBiz CEO Henri Bong, it was a matter of many factors: the timing, the business model, the signed operator contracts, actually the whole hype around IoT. In the end, the reason seems to be the ego of the Sigfox management.
- I told Sigfox's management in 2016 that the price of the connection is too low. Still, the entire IoT business was developed with the hope that there would be a billion subscriptions in 2020. The reality was something between 100200 million devices. You could say that the IoT failed all expectations.
According to Bong, an IoT business cannot be founded on hardware alone, i.e. devices. Actually, network technology doesn't matter. Every existing Sigfox
operator sells complete solutions. It includes the device, network management and ready-made solutions for various applications.
There are currently around 11.5 million Sigfox devices on the market. In Finland, the 0G network is operated by Connected Finland. There are 230 thousand devices in the network. It is telling that all these devices together consume less power than a single 5G base station.
According to Henri Bong, sending one Sigfox message consumes 10-20 milliwatts of power. Henri Bong himself believes that energy efficiency can be improved even more.- I believe we will reach 1.2 milliwatts per message.
Sigfox's energy efficiency is superior. Still, Bong sees that the services and overall solution are more important. Unabiz wants to go in a technology-agnostic direction. Visio has an IoT network where messages travel from nodes to the network regardless of which radio technology or network they travel on.
This requires a translation as devices in networks built with different technologies must be able to communicate with each other. UnaBiz's solution is a light data management software layer called UnaConnect. It has already shown its power translating LoRaWAN frames into Sigfox frames and vice versa in the project of LoRa operator Actility and UnaBiz.
OEMs providing RUs for Open RAN infrastructure adopting the Qualcomm QRU100 5G RAN Platform can now use the SMW200A and SMM100A vector signal generators and FSW and FPS signal and spectrum analyzers from Rohde & Schwarz for testing.
Deploying 5G networks with infrastructure that meets modern demands comes with major challenges, as achieving the right balance between high capacity, low latency and cost efficiency becomes increasingly difficult. The Qualcomm QRU100 5G RAN Platform is a comprehensive modem-RF solution that aims to address these challenges, helping operators and OEMs enable flexible network deployments by delivering O-RAN-compliant 5G solutions. The platform features support ranging from mmWave to sub-6 GHz Massive MIMO with 64T64R capabilities to 4T4R Remote Radio Heads (RRH) to enhance coverage, improve cell-edge data speeds, as well as increase the overall network capacity. It is designed to enhance speed in the rollout of 5G networks with highperformance O-RAN compliant infrastructure products.
The German DE-CIX launched Finland's third internet interconnection point in late May. With the new connections Finnish companies users will have even faster connections. CEO Ivo Ivanov already painted a future where the virtual world works as fast as the real one. There are quite a few requirements for this network, especially in terms of latency.
At the end of May, DE-CIX launched Finland's third Internet access point with Nokia routers. With improved connections, Finland is closer to the other Nordic countries and Europe again. But what does it mean in terms of real-time, or immersive internet?
There has been a lot of research around the world on what kind of delay the human brain perceives as natural or pleasant in different situations. The delay can be 20 milliseconds in the sensation of touch, less than 13 milliseconds in detecting visual stimuli, and less than half a millisecond in hearing sound. This means that in order to be truly immersive,
we allow, for example, a delay of milliseconds for an AR/VR experience.
The speed of light is 300 thousand kilometers per second, but Ivanov reminded that in one millisecond, under optimal conditions, the signal travels a maximum distance of 80 kilometers. So this is the so-called RTT time, i.e. the time it takes for a data packet to travel from the device to the destination and back. In practice, this means that the data has to be much closer for the experience to be pleasant. Immersive, if you will.
Thanks to DE-CIX's new connection point located in Equinix's data center, the "distance" between Finland and Frankfurt shrinks to less than 25 milliseconds. A data package travels to Oslo in 13.5 milliseconds, from there to Copenhagen 7.3 milliseconds and from Denmark to Hamburg 4.5 milliseconds. Next year, DECIX plans to open a new connection point to Stockholm, which will shorten the journey to the western neighbor to less than 10 milliseconds.
NOW YOU CAN SIMULATE 64-BIT ARM CODE ON A PC
SEGGER has added a complete instruction set simulator to its latest version of Embedded Studio for ARM. ARM64 support to the same level as all other architectures supported by Embedded Studio. Embedded Studio for ARM is available on all platforms (Linux, macOS, and Windows) on Arm, Intel, and Apple Silicon. -> etn.fi/15017.
Tampere University, the University of Oulu, Aalto University and Nokia are planning to establish a joint Centre of Excellence that will focus on System Circuit Design. The project aims to strengthen Finnish expertise via cooperation between academic research and businesses. The centre would be part of the ongoing national ‘Chips from Finland’ programme. -> etn.fi/15011.
THE SMALLEST GPU TO RUN HDR VIDEO NATIVELY
Imagination Technologies is bringing seamless visual experiences to cost-sensitive consumer devices with the new IMG CXM GPU range which includes the smallest GPU to support HDR user interfaces natively. IMG CXM GPU range boasts nearly a 50% uplift in performance density compared to the IMG BXM range, opening up more performance in the same silicon area. -> etn.fi/15006.
Apacer has created the world's first fully lead-free memory module. It not only exceeds the requirements of the current EU RoHS environmental protection standard, but also avoids relying on the RoHS 7(c)-I's lead exemption clause. The current memory modules on the market contain lead components in their resistor elements so they do not fully comply with RoHS. -> etn.fi/15002.
STMicroelectronics has introduced the market’s first MEMS water/liquid-proof absolute pressure sensor with a declared 10-year longevity program for the industrial market. ST’s new ILPS28QSW sensor comes in a sealed, cylindrical, surface-mountable package. It features a ceramic substrate that provides high resistance to liquid permeability and a robust potting gel,
proven in automotive applications, to protect the internal circuitry. The lid, made from high-grade surgical steel, is sealed with an o-ring and secured with epoxy adhesive. This unique package design ensures an ingress-protection rating of IP58 to withstand immersion in over one meter of water, certified according to IEC 60529 and ISO 20653. In addition, the sensor can sustain up to 10Bar over-pressure.
See more here -> etn.fi/15027
Arm has announced Arm Total Compute Solutions 2023 (TCS23), which will be the platform for next generation mobile computing. TSC23 also ends the support for 32-bit applications. Total Compute Solutions is Arm's name for the platform on which different processors are implemented. The first processors are the power core X4, the Cortex-A720 aimed at the mid-price category and the Cortex-A520 for basic phones.
For example, the combination X4 + 5 x A720 + 2 x A520 is available for nextgeneration Android flagship phones. This combination promises 27 percent more performance, according to Arm. For developers, the end of 32-bit support is of course a significant issue. Last year, Mediatek's Dimensity 9200 was the first application processor to exclusively use 64-bit Armv9 cores. At the same time, Google's Pixel 7 was the first device where 32-bit applications could no longer be run. With the new processors, 32-bit applications will be a thing of the past. In the end, it is not a very big change, as 64-bit mobile apps have had to be added to the Google Play store since 2019.
Finland is not the first place that comes to mind when looking for significant developers in the semiconductor industry. However, the country has quite an honorable history in semiconductors and now the industry will is aiming to become a huge business by 2030.
In the Chip Zero locomotive project started by Business Finland and led by Picosun, which develops ALD manufacturing devices, the goal is to grow the semiconductor business to 10 billion euros during the current decade. The goal is high, because at the moment there are 5000 employees in the industry and it generates an annual turnover of around one and a half billion euros.
At first, the idea sounds crazy, but Finland isn´t planning a billion-euro factory where processors for the next generation of iPhones are made. Instead, Chip Zero aims to develop solutions and processes that can make microcircuits emission-free throughout their life cycle.
In the starting seminar of the Chip Zero
program Picosun's technology director Jani Kivioja clarified that the goal requires new solutions for the entire life cycle of the circuits. We need to develop resource-efficient manufacturing technology and components. The passage of each silicon wafer must be achieved with clearly less energy consumption than at present. This requires a new approach.
- Currently, there is no focus on sustainability in seiconductor production. A lot of harmful chemicals are used in the processes, no recycling is done and the production chains are separate, Kivioja listed.
It is estimated that the total turnover of semiconductor industry will grow to a trillion dollars, or $1000 billion, by 2030. So could Finland someday have a billiondollar factory that produces 300millimeter wafers? According to Okmetic CEO Kai Seikku, the answer is no. But Picosun CEO Jussi Rautee sees it differently.
- The semiconductor industry needs a lot of new factories when the market doubles. Maybe sometime arond 2050, we too could have our own 300-millimeter factory, Rautee envisions.
According to Seikku, Finland has something to offer in certain niche areas, also in production. This requires new production lines, and in the Chip Zero project, a new Kvanttinova line is planned next to Micronova's test line. Its task would be to enable the manufacture of slightly larger volumes, but also to scale products towards larger production facilities.
Murata Electronics' CEO Tomy Runne sees Kvanttinova as an important tool together with Micronova and Tampere's SoC line.- If we could produce something here in three weeks, we would beat our competitors. In semiconductor manufacturing, speed means everything, Runne emphasized.
Picosun, an Applied Materials company, has received funding from Business Finland to form an R&D program in the country focused on reducing the environmental impact of semiconductor manufacturing.
The four-year program, called “Chip Zero,” seeks to bring together companies across the semiconductor ecosystem in Finland with a shared mission of developing chips with zero lifetime emissions by reducing the carbon footprint of chip manufacturing and increasing the efficiency of semiconductor decarbonization applications. Picosun will initiate and lead the program with a significant grant from Business Finland. The aim is to scale the program over time with contributions from ecosystem partners to reach more than 100M€ in R&D investments.
The German Rutronik is one of Europe's leading broadline distributor of electronical components. Now the company celebrates its 50th year in business.
Helmut Rudel founded Rutronik in 1973 in Ispringen, near Pforzheim, Germany. The expertise of the independent familyowned company lies in the distribution of semiconductors, passive and electromechanical components, embedded boards, storage and displays, as well as wireless products. The distributor's customers drive high-growth future markets around Industry 4.0, Inustrial IoT, Future Mobility, and Energy & Power.
In the 1970s, color television became increasingly popular in Germany, especially because of the 1972 Olympic Games and the Soccer World Cup in 1974. Exactly in the period between these two major sporting events, in 1973, Helmut Rudel founded Rutronik, which was specialized on the distribution of passive components. Rutronik is now one of Europe's largest electronics distribution companies and generated sales of 1.28 billion euros in 2022. Worldwide, the family-owned company now employs more than 1,900 people in over 80 offices.
- We are happy and proud to celebrate our 50th anniversary in the electronics industry this year, said Thomas Rudel, CEO and chairman of the board at Rutronik. From a small family business that I founded in a double garage in Ispringen in the early 1970s with great
Asahi Kasei Microdevices (AKM) is developing a new ultra-fast response current sensing technology. This technology is ideal for high frequency applications and will allow for smaller, more efficient electrical systems. AKM demonstrated the new technology for the first time in 9-11 May 2023 at PCIM Europe.
To further optimize efficiency and packaging size of high frequency applications, Asahi Kasei Microdevices is developing a new current sensing technology featuring an ultra-fast response time. Utilizing this technology substantially increases the switching frequency of power supply systems, such as ACDC and DC-DC converters, contributing to the overall size reduction of electrical systems. It is ideal for high frequency control in applications using silicon carbide (SiC) or gallium
support of my wife, we have become a leading distribution company for electronic components within 50 years, added Helmut Rudel, founder and president of Rutronik (pictured left).
More than 40,000 customers benefit from the diverse product portfolio of leading manufacturers, technical support, individual logistics solutions, and inhouse research and development (R&D). Rutronik has established design centers in Singapore and Lithuania specifically for its own R&D activities. With Rutronik System Solutions, the distributor is transforming itself into a system provider with unique, partly patented system solutions consisting of hardware and software to shape fast-moving innovation cycles in the industrial environment. With proof-of-concepts, Rutronik System Solutions significantly accelerates the pre-development phases of its customers to shorten the time-to-market of their new applications.
nitride (GaN) switching devices, as well as for the overcurrent protection of electrical systems. In the future, Asahi Kasei Microdevices is aiming to utilize this technology in automotive applications, such as on-board chargers and DC-DC converters.
Asahi Kasei Microdevices will showcase the new technology for the first time in a demonstration showing the ultra-fast measurement of a triangular current wave at PCIM Europe.
In addition, the company showcased an AC servomotor control demonstrator utilizing a coreless current sensor IC with a sigma-delta digital output, as well as the CZ375x, a coreless current sensor IC of the Currentier product line with a high current capability up to 100 Ampere root mean square (Arms).
Asiantuntemuksemme on sinun etusi
50 vuoden aikana meistä on tullut yksi Euroopan suurimmista elektroniikkakomponenttien jälleenmyyjistä ja luotettava kumppanisi maailman johtavien valmistajien valikoimassa. Yhdessä oman tutkimus- ja kehitystoimintamme sekä laitteistoista ja ohjelmistoista koostuvien patentoitujen järjestelmäratkaisujemme kanssa onnistut pysymään sovelluksessasi askeleen edellä.
Luotettavaa ja monipuolista - Rutronikin valikoima
Interconnect- ja sähkömekaaniset komponentit
Sulautetut piirilevyt ja järjestelmät
Näytöt ja monitorit
Haluaisitko tietää meistä enemmän? Vieraile osoitteessa www.rutronik.com
Technological advances and the global geopolitical situation have taken cybersecurity to a whole new level. It affects us all, whether we are individuals, society, businesses, communities or public organisations, let alone governmental preparedness.
According to Check Point Research, the number of cyber attacks in Finland increased exponentially in 2022, up 81 % compared to 2021. The average number of attacks per organisation was 1 228 per week. Attacks on cloud-based networks increased by 48 % globally last year.
According to the "Information Security in Finnish Organisations 2023-2025" survey conducted by Check Point Software and Loihde Trust, one in four Finnish companies have identified a business-impacting security incident in their organisation in the last two years. The most common of these were data leaks and security challenges related to personal records. In the future, three out of four large Finnish companies believe they will face a cyber attack.
Digitalization and the spread of cloud computing will continue to
grow, creating a new, everchanging attack surface. The rapidly growing interest of hackers in AI technologies will further increase the number of attacks. For example, the AI bot ChatGPT has already been used to write malware code and phishing emails. ChatGPT account credentials and stolen ChatGPT Premium accounts are traded on the dark web. Sensitive organisational data can be compromised when cybercriminals gain access to queries from the owner of a hijacked account.
According to IBM Security's Cost of a Data Breach report, the cost of a data breach to an organisation in 2022 was at an all-time high, averaging $4.35 million. The indirect cost of an extortion attack averaged $4.54 million, not including the ransom itself.
Despite all this, companies are not always ready to make significant investments in security. Security may not be seen as a business enabler and profitability enhancer, or the risks may be seen as theoretical if the company has not yet suffered a security breach. There is also often a lack of clarity about who is responsible
for security, for example in the case of SaaS services or cloud solutions. Critical infrastructure protection in Finland is only beginning to wake up.
As economic growth slows, it is important to ensure that business continuity is safeguarded. Information security cannot just be a necessary expense but must be an integral part of an organisation's business continuity planning. Achieving cyber resilience requires a holistic, proactive approach to security: comprehensive, consolidated and collaborative solutions. This is currently not common practice, and the problem is often that companies' solutions do not communicate with each other in the event of a data breach. This makes companies vulnerable.
Even if you have to adjust your business, it's not worth risking information security, says Viivi Tynjälä, Country Manager of Check Point Software.The author Viivi Tynjälä works as the country manager for Finland and the Baltics of the security company Check Point Software Technologies.
Advancing IoT technology is rapidly transforming the energy and district heating sector. The profitability of the entire industry improves. While system architects long contemplated integrating super-intelligent IoT systems in the past, it was never feasible financially. Technology was simply too expensive. Today it´s a different game.
Over the past few years, a perfect alignment of technical advances and capability – combined with far lower costs – at last makes these new concepts economically viable.
Wireless sensors are not new, but innovations such as LoRaWAN or NB IoT communication open a new, never-seenbefore equilibrium between price and performance.
Combined with the rapid expansion of digital data platforms, we can give realworld interpretation to these data points. It is no exaggeration to say that we are in the midst of a revolution.
IoT technologies enable companies to continually gather data and have a real-
time overview of what is happening on site. If the system generates enough real-time data, soon we can create a Digital Twin. It can be used to refine procedures, rewrite methodology, and accurately analyze efficiency and accountability.
It is easy to understand why many large operators are implementing these measurement systems. Over the past few years, a tremendous technological shift has greatly affected the energy industry.
Flow measurement in district heating pipelines has been successful. Now the network where the solutions work cover almost all of Finland. Digita, the country’s largest independent owner of telecommunication masts, owns and operates Finland’s leading national
terrestrial television and radio network, but also it controls the nationwide network LoRaWan.
In principle, any time series data or alarms can be stored in the measurement database service. The data most often indicates water volumes, flows or water levels, but also alarms about the situation or condition of the network or the measuring device itself. It is possible to automatically visualize this information on the virtual map as alerts. This way, for example, site visits or maintenance tasks or leak repairs can be more quickly targeted at the right target.
Currently, it is possible to import results from various data producers, eg Kamstrup, Mipro and Soficta. In Finland, there is a patented technology that reveals incipient leaks in underground pipelines. It is a high technology export product.
The technology works in the LoRaWan network.- Even in one flow study using a LoRa solution, the return will pay off the investment by up to 100,000 euros, said managing director Sami Metsänperä of Soficta Co.
According to Metsänperä, several customers, district heating companies and water utilities, already have a Digital model or even a Digital Twin, in which Soficta's technology works.
- We maintain and develop the reliability of supply of our district heating by many different means on an ongoing basis. On this scale, we do a network survey every few years, and with the survey we get a comprehensive overview of the condition of our network and information about
possible areas that need repair, says Jani Uitti, manager of network assets from Turku Energia.
Uitti emphasizes that it is now necessary to make clear the difference between a Digital Twin operating in real-time data and a digital model. - We are moving towards a Digital Twin.
Since the creation of automation, there has been the question of which assets to monitor. Until now, the typical strategy was simply monitoring the top 10% of assets –namely, the assets used most and those most critical. Now automation brings comprehensive monitoring.
The road of IoT technologies, mobile data to usable technology is long - almost ten years. LoRa technology has always been much more ready than NB IoT in mobile networks. Both technologies have 128-bit
Technology director Pasi Pajula uses NB IoT technology to measure the pressure of the sewer network. NB IoT technology has struggled with serious technical problems for years, but now it is reliable and is used to monitor the condition of pipelines and the adequacy of capacity.
data encryption. One technological option is Sigfox, but it doesn´t have a large, ubiquitous network.
Pasi Pajula, technology director at Preventos
Informatics Oy in Kuopio, uses NB IoT technology to measure the pressure of the sewer network.
NB IoT has struggled with serious technical problems for years, but now it is reliable and used to monitor the condition of pipelines and the adequacy of capacity.
- The battery life in the IoT transmitter is now about five years, says Pajula.
During the inspection, signs of corrosion were visible in the sensors that had been sending digital pressure measurement data for 2½ years, but the device still works flawlessly. In the well, immediately below the hardware sensors, an eternal stream of dirt flows.
By 2050, global energy demand is projected to rise by over 60%. ADI’s expertise in power management has enabled breakthroughs like energy harvesting and robotic miniaturization. Which means we can make progress, while making less waste. Analog Devices. Where what if becomes what is. See What If: analog.com/WhatIf
Microchip has introduced new industrial-grade Single Pair Ethernet devices that implement the 10BASE-T1S and 100BASE-T1 physical layer. These products bring Ethernet all the way to the edge of industrial networks.
Single Pair Ethernet, or SPE, defines the transceiver part of an Ethernet system. All the higher software layers remain unchanged, regardless of the speed grade. SPE is also referred to as T1, which means one balanced pair of wires. Some applications use a twisted pair of wires, but others use just two wires running alongside each other. The IEEE standard defines a channel in terms of its electrical characteristics and not the specific physical wires.
Multiple bandwidths are defined for SPE. The first part of the name specifies the Mbit/s: 10BASE means 10 Mbit/s. There are standards for 10BASE-T1S (S for Short reach), 10BASE-T1L (L for Long reach), 100BASE-T1, 1000BASE-T1 and even higher data rates are defined for 2.5, 5 and 10
Gbit/s. SPE reduces system cost by reducing weight and wiring complexity.
The megatrend in networking is to move from distributed systems defined primarily by the hardware involved, to more centralized, software defined systems. The trend is to connect everything with Ethernet:
• Domain-specific hardware architectures give way to zones connected to each other and to a centralized compute platform.
• Multiple application-specific buses are replaced by an IP-based and ubiquitous Ethernet network.
• Gateways or controllers required to translate between different hardware approaches and that require complex wiring are eliminated. Low-cost, single pair cabling then brings Ethernet all the way to the edge of the network.
All this results in a more powerful, more flexible network to meet industrial challenges.
Ethernet enables connected Cloud-to-Edge infrastructures. Traditional IT networks at the top enable communications at the highest levels. Operational Technology (OT) networks are used within buildings to
control processes and assembly lines. As you get closer to the edge of the enterprise network, you find sensors, actuators, and other low-level devices.
Ethernet allows a common way to connect and communicate with all these elements to make their data available to higher levels of the organization so Big Data can use it to better run the enterprise.
It also makes it easier to configure and control systems. Established Ethernet mechanisms can secure all these elements. Authentication, encryption, and secure
Ethernet enables connected Cloud-to-Edge infrastructures.
updates are an increasingly important aspect of today’s networking requirements.
Microchip has been a leader in the development of Single Pair Ethernet. It helped write the standards at the IEEE and has in parallel developed a comprehensive portfolio of SPE solutions that are flexible and scalable. Whether running at 10Mbits/s, 100Mbits/s or 1000 Mbits/s, Microchip has a PHY, MAC-PHY or Ethernet switch that can drive a single pair of wires to transmit the data.
10BASE-T1S technology is one of the newest types of Ethernet interconnection defined by the IEEE. The IEEE 802.3cg standard was published in 2019. Microchip was one of the key participants that created the standard within the IEEE. This standard was developed to expand Ethernet to the edge of Operational Technology or OT networks. Existing systems had bandwidths of hundreds of kilobits per second, so going to megabits per second provides for future growth.
The technology uses half-duplex communication and provides for a flexible topology, from point-to-point to multidrop configurations. Multidrop means that multiple devices connect to a bus line made up of a single pair of wires.
10BASE-T1S is defined to use a single balanced pair of conductors. This can be a single twisted pair of wires, other configurations for wire pairs, or even
parallel traces on a printed circuit board or server backplane.
The new 10BASE-T1S standard define a physical layer that creates new business opportunities for companies that embrace pervasive all-Ethernet systems. Data anywhere in the system can be used in innovative ways that enable new applications.
Cost can be reduced since 10BASE-T1S uses simple components, software, and wiring. A new 10BASE-T1S MAC-PHY Ethernet controller now in the marketplace allow the simplest of microcontrollers to become part of the Ethernet world. 10BASE-T1S eliminates the need for gateways previously required to translate data from different hardware systems. The multidrop feature reduces the number of ports in switches as devices connect to a single bus line.
Using well-established security mechanisms and the unified interfaces of the Ethernet world reduces risks when creating applications at the edge of industrial networks. Security mechanisms are well understood by a large body of designers and implementers. Any issues can be quickly discovered given how many people work with the technology, and they can also be quickly resolved. Design, software development, testing and maintenance resources at all levels of OT and IT networks can concentrate on the same basic communication mechanisms.
Microchip offers a complete range of SPE PHYs, from 10Mbits/s to 1000Mbits/s, all offering Industrial grade temperature support. Further enhanced support for industrial applications includes safety, security and extended cable reach. Complete SPE networked systems solutions can be realized with a broad portfolio of SPE switch devices. Microchip offers intelligent switches that support Time Sensitive Networking (TSN) and integrated SPE PHY technology. Our latest release is an industry first true singlechip SPE switch with TSN support, integrating switch fabric, CPU, RAM and uniquely FLASH code memory. Designers can now implement ‘out of the box’ Single-Pair switching solutions, with little or no expertise on T1 PHY technology or AVB and TSN specifications. See more at www.microchip.com/ethernet
ST removed the Cortex-M4 and the GPU found on STM32MP15 devices. In exchange, we increased the single-core performance, vastly improving the price-per-feature ratio and even adding new security features to make the STM32MP13 the first SESIP Level 3 certified MPU. The new device also supports Payment Card Industry (PCI) PTS/POI 6.0 and Arm PSA level 1 certifications.
Too many teams have experienced the challenge of a low bill of materials that forces them to spend excessive time in development or support or pushes design considerations that end up blowing up budgets.Put simply, a seasoned engineer knows that a low price can have a high cost.That’s why the new STM32MP13 used an architecture designed with efficiency and accessibility in mind. For instance, we
decided to remove the GPU since industrial applications at this price point rarely need one, but we added a second Ethernet port and a new pipeline to process camera images more effectively with up to 15 frames per second at 5 Megapixels.
Another way a low price can trigger high costs is if the device has a high energy consumption that requires heavy cooling solutions, large power supplies, and more. That’s why ST worked to make the STM32MP13 a new reference in MPU power consumption.For instance, its Standby mode uses 27 µW, which is 94% less than the competition, and its Stop mode only requires 11 mW, about three times less than other brands. Similarly, the LPLV-Stop2 mode consumes 1.64 mW, or 32% less than the competition, while still waking up in 0.009
The STM32MP13 is STMicroelectronics´ most cost-effective industrial-grade microprocessor, thanks to a price of fewer than four dollars. Yet, the STM32MP13 is our fastest single-core MPU thanks to a Cortex-A7 capable of reaching 1 GHz.Put simply, our engineers decided to optimize the architecture significantly to bring an industrial solution that would be more accessible.
seconds.Finally, VBATBAT only needs 4.65 µW, which is about 19 times or 95% better than alternatives from other makers.
Evidently, the low power consumption of the STM32MP13 will help design a simpler PCB.And to go one step further, ST also provides layout examples for four-layer plated-through hole PCBs.Not only do we simplify designs, but we also optimize developments by providing references so teams can release a product to market faster. The four-layer PTH PCB will ensure companies don’t have to use costly laser vias, and the pin-to-pin compatibility of all STM32MP13s means switching to a different model is highly straightforward. Put simply, this is an example of how an ecosystem can further augment the value proposition of a cost-effective solution.
It’s another vital lesson developers learn early: a poor ecosystem can cost a lot. That’s why ST continues to work on the software available on all STM32 MPUs.Indeed,OpenSTLinux, our embedded Linux distribution, best testifies of our work with theopen-source community.We are consistently mainlining our drivers to make them part of the main Linux codebase. We want to work with the community to improve our solutions and make it easier towork with an MPU. That’s why we also have many members of theST Partner Programthat can further help build solutions around our MPU.
However, we wanted to go even further with the announcement of the
STM32MP13.Hence, the new ST MPU will be the first to support a real-time or near-real-time performance.We know that some teams crave the performance of a Cortex-A7 but need the reduced latency and consistency of an RTOS. Consequently, we publishedLinux-RT, an OpenSTLinux extension that helps developers reduce the latency of certain applications to provide near-real-time performance for applications like factory automation systems. Moreover, we will also bring bare metal andAzureRTOSsupport by the end of the year. As a result, teams familiar with MCUs will also have an easier time transitioning to the STM32MP13.
Another way the ST ecosystem shines is through the new security features of the
STM32MP13. The device includes important hardware functionalities, like new anti-tamper pins and side channel protections, to name only two. We also added IPs to accelerate on-the-fly encryption/decryption in the DRAM and the use of many cryptographic algorithms. Furthermore, we are working to provide software and documentation to facilitate the implementation of Secure Boots, TrustZone, and more.For instance, we updated Our Wiki onconfiguring the Open Portable Trusted Execution Environment (OP-TEE)to account for the secure services on the STM32MP13.
Those wishing to experiment with the new STM32MP13 can grab theSTM32MP135F-DK. The board includes two Ethernet ports, one USB-C and four USB-A connectors, and a microSD card holder. It also has a 4.3-inch 480×272 display with capacitive touch capabilities, 512 MB of DDR3L, and Wi-Fi and Bluetooth modules. It also comes with a camera board so users can start creating a proof-of-concept for a computer vision application for an industrial setting.
There are also SoM makers that are shipping or will soon ship STM32MP13 modules. These companies include Engicam,DiGi,iWave,Phytec, orDH electronics. In many instances, a SoM can significantlyjumpstart your applicationby providing a robust building block that can shift many of the maintenance and developments to the SoM maker. It’s also a great way to help take advantage of theopen-source communityby working with a SoM that upstreams its code and actively participates in the community.
Bluetooth LE Audio (low energy audio) has the potential to change our audio experience fundamentally, whether through “silent disco” or improved listening with an assistive listening system. Due to further developments, a variety of novel use cases and products, including new markets, are expected to emerge in the world of audio.
Bluetooth technology has been under development since the 1990s and uses over 79 channels in the 2.4 GHz unlicensed band to transmit data. Today, the Bluetooth Classic variant only supports point-topoint communication for audio transmission.
A key further development is Bluetooth Low Energy (LE). It has already replaced Bluetooth Classic in most applications. Wireless audio streaming, e.g. for wireless headphones, speakers, or in-car entertainment systems, is now the last bastion of Bluetooth Classic (Fig. 1).
Bluetooth LE also uses the 2.4 GHz band, but – as the name suggests – is designed to operate with low energy. Beside point-to-point communication, it also enables broadcast and mesh topologies, thus laying the foundation for large-scale, high-speed device
networks. Furthermore, it can be used for device tracking, making it an ideal addition to indoor GPS.
The focus of the latest standard is also crystal clear: it is called Bluetooth Audio. The first version is based on Bluetooth Classic. However, its range of functions is limited as is the variety of possible applications. The second version, the new Bluetooth LE Audio, in contrast, enables the more flexible processing of audio signals. This represents an evolutionary step for existing applications, such as headphones and hearing aids, thereby creating new applications and markets for audio streaming.
LE Audio is based on Bluetooth LE and the powersaving Low-Complexity Communications Codec (LC3) developed by the Fraunhofer Institute. It combines
higher audio quality than Classic Audio with low data rates and provides developers with an enormous amount of flexibility (Fig.2). Moreover, various product features can be better matched, e.g. the provided energy savings can be used either to extend the battery life or to use smaller batteries.
LE Audio offers a range of interesting functions: With Multi-Stream Audio, multiple independent audio streams can be played synchronously between an audio source device and one or several audio receiving devices (audio sink function). This optimizes their performance significantly. For example, this results in a better stereo experience for wireless earphones. Users can also switch between various voice assistants just as seamlessly as between different audio source devices.
The Auracast function also allows users to send one or several audio streams from one audio source device to numerous audio sink devices. In contrast to the multi-stream function, however, the number of audio sink devices is unlimited here. This opens up an entirely new set of use cases for sharing audio experiences.
When an Auracast transmitter, such as a TV, smartphone, laptop, or similar, begins a broadcast, it contains one or several audio streams (e.g., left and right stereo stream) and an advertisement with information about the broadcast, such as name, content, codec configuration, etc. Auracast assistants, also Auracast-compatible smartphones, smartwatches, or assistive listening systems (ALS),
scan for these advertisements. Users can then select a broadcast to join via their user interface (UI) –similar to connecting to a WLAN today. Once a broadcast has been selected, the Auracast assistant provides the receiver (e.g. headphones or earbuds) with the information it needs to join the broadcast.
With personal audio sharing, people will be able to share music and podcasts with others around them; for example, via their smartphone with family and friends within Bluetooth range. In the public sphere, for example, airports, railroad stations, bars, gyms, cinemas, and conference centers can share information or music with their visitors via Bluetooth Audio. This gives people the opportunity to enjoy the same music – on a larger scale even at “silent” concerts or discos. It also allows people to use their own earbuds or ALS to select the audio being broadcast by silent TVs, for example in a gym or waiting room, and to listen to a conference lecture, a talk, a church service, or a guided museum tour, and ensures they do not miss any announcements at an airport. And since several audio streams can be sent in parallel, multiple languages can be transmitted, for example. This is particularly interesting for lectures, conferences, and airport announcements.
LE Audio will guarantee significant benefits for people with assistive listening systems. They benefit not only from significantly higher audio quality than with traditional hearing aids, but can also use their Auracast-enabled ALS as wireless headphones, e.g. when using their smartphone. This helps to eliminate interference during a call, something that often occurs when the phone is held up next to the ALS. This can be achieved with a simple topology (Fig.3, left).
The images and graphics clearly show the extended options and improved features of Bluetooth LE compared to Bluetooth Classic.
LC3 (Low-Complexity Communications Codec) ensures better audio quality at every data rate than SBC (Low-Complexity Subband Codec) on which Classic Audio is based.
The connection between phone and hearing aid is established through an audio stream, which also allows a return stream. The user can then use the microphone of either the hearing aid or the phone as the return channel. Since both directions of the audio stream are configured and controlled separately, both can also be switched on and off individually.
In the topology (Fig. 3, right), the phone sends a separate left and right audio stream to the ALS in the left and right ear, respectively. Compared to the connection with an audio stream that then also establishes a second wireless connection with the other ear, latency is thus significantly reduced. This is evident when it comes to the lip synchronization of movies or music videos, etc.
With Auracast, an ALS can be used as a headphone, either with one audio stream or with two separate streams for the le� and right ALS.
The return streams can also be implemented separately, which further adds to the complexity. These parallel, synchronized streams to two independent audio devices go far beyond what conventional Bluetooth audio profiles can handle.
Despite being based on Bluetooth LE, LE Audio still requires its own hardware. Nordic Semiconductor offers the nRF5340, an all-in-one SoC (system-onchip), which is ideally suited for Bluetooth LE Audio applications, thanks to its dual-core processor –consisting of the 128/64 MHz Arm Cortex-M33 application processor with 1 MB flash memory and 512 kB RAM, and the 64 MHz Arm Cortex-M33 network processor with 256 kB flash memory and 64 kB RAM – and an extended temperature range of –40 to +105 °C.
For an even easier entry, Nordic has come up with the nRF5340 Audio Development Kit (DK). It supports all Auracast functions and is configurable. As a USB dongle, it can send or receive audio data from a PC; further, it can be used as a business headset or as true wireless stereo (TWS) earbuds. The Audio DK essentially consists of the nRF5340 SoC, the nPM1100 power management IC, and Cirrus Logic’s CS47L63 audio digital signal processor (DSP), which is optimized for direct connection to an external headphone load and is ideal for mono-only and direct speaker output earphones.
Also based on the nRF5340, Nordic has developed Thingy:53 with the support of Rutronik. The multisensor prototyping platform with multi-protocol short-range wireless connectivity ensures reduced time-to-market for embedded applications with machine learning (ML). Thingy:53 is equipped with multiple motion and environmental sensors, the nPM1100 power management IC, the nRF21540 front-end module, a power amplifier/low-noise amplifier, and a 1350 mAh Li-Poly battery. This allows embedded ML models to be run directly on the device to use the sensors for speech recognition, for example. Certain motions or sounds wake nRF5340 from stand-by mode, leaving the platform in power-saving sleep mode for a long time.
Moreover, Rutronik’s product portfolio also includes modules based on Nordic’s nRF5340 chip. Insight SiP’s ISP2053, for example, integrates semiconductor and passive components, including the antenna structure, into a miniaturized module measuring just 8 mm × 8 mm × 1 mm, thanks to its system-inpackage (SiP) method. It is, therefore, ideal for applications that offer very little space. The module is fully certified and supports not only Bluetooth Audio but also all other profiles of Bluetooth LE, Long Range, and Mesh, as well as NFC, Thread, ZigBee, and direction finding using AoA/AoD (angle of arrival and angle of departure). Starting with Bluetooth 5.0, all Insight SiP pin-compatible modules will facilitate migration to the latest Bluetooth generation.
Users who do not require the compact design of the Insight SiP module may find MS45SF1 from Minew to be a competitively priced alternative. As an officially licensed design partner of Nordic, Minew also uses nRF5340 for MS45SF1. The module comes with an integrated PCB antenna and is also fully certified.
Bluetooth LE Audio is blazing a trail to a new, completely networked audio world of headphones, earbuds, or ALS, as well as smartphones, laptops, TVs, and other audio devices. The introduction of native LE Audio support in Android 13 will certainly boost the interest of audio device suppliers in the new Bluetooth version. In its “Bluetooth Market Update 2022”, the Bluetooth SIG expects strong growth in earbuds and headphones through LE Audio and forecasts the sale of more than 600 million devices by 2026.
Last but not least, Bluetooth LE Audio will, without doubt, have a massive impact on the current market for ALS and auditory acoustics. The further development ensures affordable hardware with the option of audiogram adjustment, such as frequency (equalizing) and phase correction. And everything is simply controlled via a phone app.
Increased awareness about health has given rise to the demand for small but high accuracy devices that can measure various vital signs and health markers. In particular, the COVID-19 pandemic has caused a surge in demand for devices capable of monitoring multiple vital signs, including temperature, SpO2, and heart rate both in hospitals and at homes.Yigit Yoleri and Guixue Bu Analog Devices
The last decade has witnessed drastic improvements in mobile, wearable, and digital health fields. In particular, there is rapid expansion and adoption of digital healthcare fueled by the continuous advancement in electronics and recent
breakthroughs in cloud computing, artificial intelligence (AI), and communication technologies such as the Internet of Things (IoT) and 5G. Some of the vital signs monitoring (VSM) capabilities have been built into phones, watches, and other smart
wearable devices, and thus have become accessible to a much broader population.
The need for small and convenient health tracking devices, preferably smart
wearables, has reached historically high levels.
Adding multiple sensing capabilities to such small devices has its challenges, calling for a smaller form factor, lower power consumption, and multiparameter capabilities with significantly improved performance. However, this can now be achieved through a single analog front-end solution. This new AFE functions as a multiparameter vital signs monitoring hub for synchronous measurements. It offers low noise, high signal-to-noise ratio (SNR), small form factor, and low power consumption, enabling significant improvements to medical devices, especially wearable technology. For doctors, patients, and consumers, this makes vital signs monitoring easier than before with higher performance, longer battery life, and greater accuracy without the intrusive discomfort of multiple devices. This article discusses some of the breakthrough capabilities and features of this single analog front-end solution.
The ADPD4100/ADPD4101 is a multimodal sensor AFE that has eight analog inputs and supports up to 12 programmable time slots. The 12 time slots can enable 12 separate measurements in a sampling period. The eight analog inputs are multiplexed into a single channel or two independent channels, enabling simultaneous sampling of two sensors, either in a single-ended or differential configuration. There are eight LED
drivers that can drive up to four LEDs simultaneously. These LED drivers are current sinks and are independent of the LED supply voltage and LED type. There are two pulsed voltage sources for voltage excitation. The new AFE’s signal path consists of transimpedance amplifier (TIA), band-pass filter (BPF), integrator (INT), and analog-to-digital converter (ADC) stages. The digital block provides multiple operating modes, programmable timing, general-purpose input/ output (GPIO) controls, block averaging, and a selectable second- to fourth-order cascaded integrator comb (CIC) filter. Data is read directly from the data register or through first-in, firstout (FIFO) method.
There are two variations of this new AFE. One features an I2C communication interface, and the other has an SPI port. One of the advantages of the ADPD4100/ ADPD4101 relates to optical measurements. Its superior, automatic ambient light rejection capability uses pulses as short as 1 µs in a synchronous modulation scheme combined with the BPF, eliminating the need for external control loops, dc current subtraction, or digital algorithms. A decimation factor higher than 1 can be used to increase the output SNR. There is a subsampling feature that allows selected time slots to run at slower sampling rates than the programmed sampling rate to save power, where power is proportional to the sampling rate. There is also a TIA ceiling detection feature, which uses voltage comparators at the output
terminals of the TIA to set the interrupt bit when the TIA input exceeds typical operation limits.
The ADPD4100/ADPD4101 is an ideal hub for various electrical and optical sensors in wearable health and fitness devices for heart rate and heart rate variability (HRV) monitoring, blood pressure estimation, stress and sleep tracking, and SpO2 measurement. This new multiparameter VSM AFE’s multiple modes of operation can accommodate various sensor measurements, including but not limited to photoplethysmography (PPG), electrocardiogram (ECG), electrodermal activity (EDA), body composition, respiration, temperature, and ambient light measurements in healthcare applications.
PPG measurement detects the blood volume change in the microvascular bed of the tissue associated with each cardiac cycle. The total absorption of light correlates with changing blood volume due to systole and diastole events, producing a PPG signal. PPG measurement is performed by pulsing an LED light into human tissue, and collecting the resultant reflected/ transmitted light with a photodiode, which converts light to photocurrent. The ADPD4100/ ADPD4101 processes and measures the photocurrent and produces the digital PPG signal. Without any change to the hardware connection, the AFE can be flexibly configured to operate in four different modes for
different PPG measurement use cases: continuous connect mode, multiple integration mode, float mode, and digital integration mode.
Continuous connect mode is the typical mode for PPG measurements. It provides the best ambient light rejection and provides high SNR. It works well with a charge transfer ratio (CTR, photocurrent over LED current) level as low as 5 nA/mA to 10 nA/mA and can provide 95 dB to 100 dB dc SNR. These levels can be increased with increasing decimation factor. It uses the full analog signal path, TIA + BPF + INT + ADC. Incoming charge is integrated once per ADC conversion. In a single stimulus event like PPG, most of the dynamic range of the integrator is used when the charge from the sensor response is integrated. The TIA is connected continuously to the inputs after the preconditioning period; therefore, the input signal is not modulated. To reduce the noise, the anode of the photodiode is preconditioned to the reference voltage of the
TIA (TIA_VREF). TIA_VREF is typically set to 1.27 V to get the largest dynamic range of the TIA. The cathode of the photodiode is connected to a cathode voltage source (VCx) pin, and the device is typically set to provide TIA_VREF + 215 mV to the cathode of the photodiode to create 215 mV reverse bias across the photodiode. This reduces the noise of the signal path and the photodiode capacitance. In this mode, typical LED pulse width is 2 µs. Short LED pulses provide the best ambient light rejection performance. The use of multiple LED pulses increases the SNR by 3 dB for every doubling of the pulse number. Integrator chopping is typically enabled to get the highest SNR, since chopping eliminates the low frequency noise content from the integrator. Higher TIA gain selection results in lower input referred noise, but reduces the dynamic range of the TIA. The dynamic range of the TIA is calculated by dynamic range = (TIA_VREF)/(TIA gain). In order to increase the ADC saturation level, TIA gain can be reduced, or integrator resistor can be increased. Higher integrator
resistor selection results in lower noise, but lower integrator resistor selection increases the ambient light headroom.
Multiple integration mode is the same as continuous connect mode, except the incoming charge is integrated multiple times per ADC conversion. This mode is useful in getting high SNR in low light cases because it uses a small amount, sometimes less than 50% of dynamic range per stimulus event. Therefore, it allows for the utilization of a larger amount of the integrator dynamic range due to multiple integration prior to ADC conversion. Every doubling of the number of integrations per ADC conversion results in a 3 dB increase in SNR, which is the same effect as doubling the number of pulses. Since this mode is typical for small inputs, the highest TIA gain is selected. This mode is used in the cases where CTR is lower than 5 nA/mA and good ambient light rejection is needed.
Float mode is also used in low light cases to get high SNR. Float mode allows for noise-free charge accumulation on the photodiode. The photodiode is disconnected from the AFE—and therefore floats—and accumulates charge, due to light, in a noise-free manner. Then, the AFE is connected back to the photodiode, the charge on the photodiode rushes into the AFE,
and the integration is done in a way that allows the maximum amount of charge to be processed per pulse with the minimum amount of noise added by the signal path. The dumping of the charge occurs quickly with short modulation pulses; therefore, the noise addition due to the signal path is smaller. Also, the float time can be increased to achieve higher signal levels, but there is a limit to the amount of charge that the photodiode’s capacitance can accumulate. In this mode, the band-pass filter (BPF) is bypassed because the shape of the signal produced, when transferring the charge from the photodiode by modulating the connection to the TIA, can differ across devices and conditions. To reliably align the signal with the integration sequence, the BPF must be bypassed. This mode does not provide good ambient light rejection performance, and it is limited by photodiode capacitance, but it provides a power-efficient and less noisy measurement in very low light conditions.
In low light conditions with CTR <5 nA/mA, the typical mode of operation is float mode. It offers lower noise than multiple integration mode, because multiple integration mode requires more integration cycles, leading to higher TIA and integrator noise contributions. Float mode is also more powerefficient than multiple integration
mode as the BPF is powered down and the measurement time is smaller. Therefore, SNR per watt efficiency is significantly higher in float mode.
When the photodiode is leaky or there is a significant amount of ambient light present in the PPG measurement, multiple integration mode is preferred. Leaky photodiodes can’t be used with float mode because the charge leaks off instead of being accumulated before the fast charge dump occurs. If the ambient light is significant, float mode is unfavorable, because ambient light dominates the amount of charge that can be stored on the photodiode. Multiple integration inherently offers excellent ambient light rejection due to the use of BPF and short LED pulses.
All the modes that have been mentioned so far use the integrator to integrate the incoming charge. Digital integration of the ADC samples is also possible through the digital integration mode. To achieve digital integration, the integrator is converted into a buffer. Digital integration mode works in two regions. In the lit region, the LED is pulsed and in the dark region the LED is off. ADC samples are taken at 1 µs intervals within the lit and dark regions and are integrated digitally. The signal is calculated by the subtraction of the integration of the dark samples from the lit samples. This mode can support longer LED
pulses; therefore, it is the typical mode of operation where the photodiode has a slower response time and requires longer pulses. The BPF is bypassed and powered down. Digital integration mode provides the best power efficiency and results in the highest achievable SNR levels. However, ambient light rejection is subpar to that of continuous connect mode, due to longer LED pulses and bypassing of the BPF. Digital integration mode cannot support simultaneous sampling of two channels during the same time slot. Digital integration mode can support 100+ dB dc SNR.
As mentioned previously, the typical mode of operation for PPG measurements is continuous connect mode as it provides high SNR and excellent ambient light rejection under conditions where CTR is greater than 5 nA/mA. However, digital integration mode results in the highest SNR levels and provides optimized SNR per watt efficiency. Therefore, if the ambient light is not an issue for the application and the target dc SNR is above 85 dB, digital integration mode can be selected to achieve high SNR efficiently. If the target dc SNR is below 85 dB, the power savings from digital integration is not significant when compared to continuous connect mode.
To summarize, digital integration mode can be selected if the photodiode requires longer
pulses due to slower photodiode response time, or when there is no need for simultaneous sampling of two channels within a time slot. Additionally, when ambient light is not an issue and when the target dc SNR is above 85 dB, selecting digital integration mode will achieve power efficiency.
In light of the COVID-19 pandemic, PPG applications have become even more important in vital signs monitoring and health diagnostics. Additionally, multiple metrics are critical to detection. For example, some key vital signs measurements include heart rate monitoring (HRM), HRV, and oxygen saturation (SpO2), which can be measured through pulse oximetry and blood pressure.
Optical and noninvasive SpO2 monitoring, which is also known as pulse oximetry, has become very valuable in the detection of hypoxia in patients with COVID19. Hypoxia, one of the main
symptoms of COVID-19, is the deprivation of oxygen supply to the body tissues. Hypoxia may also cause increased heart rate; therefore, optical and noninvasive heart rate monitoring is also critical to detection.
The integration of multiple measurement capabilities is optimal, if not necessary to future wearable devices, making the ADPD4100/ADPD4101 extremely advantageous. This AFE measures any type of sensor inputs (including temperature, ECG, and respiration measurements). Therefore, a complete multiparameter VSM platform can be established with just a single sensor AFE.
Pulse oximetry is done through red (typically 660 nm wavelength) and infrared (IR) LEDs (typically 940 nm wavelength). Deoxygenated hemoglobin absorbs more of the 660 nm wavelength light, and oxygenated
hemoglobin absorbs more of the 940 nm wavelength. The
photodiode perceives the nonabsorbed light. The perceived signals are then divided into dc and ac components. The dc component represents the absorption of light due to the tissue, venous blood, and nonpulsatile arterial blood. The ac component represents the pulsatile arterial blood. The SpO2 percentage is then calculated by:
Any two time slots of the ADPD4100/ADPD4101 can be configured to measure the response to the red and IR LEDs to measure SpO2. The remaining time slots can be configured to measure PPG from different wavelength LEDs, and can also support ECG measurement, leadoff detection, respiration measurement, and any other sensor measurements.
Heart rate monitoring is also critical in the detection of COVID19. As the oxygen supply drops due to hypoxia, the heart starts to beat faster in order to supply enough oxygen to the tissues. Monitoring heart rate is also valuable in detecting cardiac problems or tracking fitness behavior.
A green LED with wavelength around 540 nm is generally preferred in heart rate monitoring. It yields the best PPG signal as it has a higher modulation index than red or IR LEDs. It also provides a decent CTR level; therefore, power consumption would not be too high.
AC SNR is a parameter of signal quality and can be calculated by dc SNR times modulation index. For example, with 1% modulation index, 95 dB dc SNR translates into 55 dB ac SNR.
ECG measurement has been added to wearable devices, such as watches for spot checking and chest patches for continuous monitoring. Electrodes made of metals and other conductive materials, which are polarized and
called dry electrodes, are typically found in such devices. The primary challenges for ECG measurement with dry electrodes are high electrode-skin contact impedance and relatively high overpotential.
Conventional instrumentation amplifier-based ECG solutions use buffers to mitigate the high electrode-skin contact impedance associated with signal attenuation. The right-leg drive (RLD) technique, which requires a third electrode and drives a reference voltage back to the body, is implemented to reject the common-mode voltages the human body, electrodes, and cables are exposed to in ECG systems that measure the voltage.
When applied to ECG measurement, the ADPD4100/ ADPD4101 takes a novel approach by using a passive resistor-capacitor (RC) circuit to follow the differential voltage across a pair of electrodes. The passive RC circuit can be as simple as three components, two resistors RS and a capacitor CS, as shown in Figure 3a. It is a twostep process for each sample of the ECG data.
ECG measurement configuration. (a) RC sampling circuit and lead-off detection circuit. (b) Illustration of the charging and charge transferring process for each sample of ECG data.
The two input pins (IN7 and IN8) float during the charging step. The charge on the capacitor CS is proportional to the differential voltage across the two electrodes if the charging time is >3τ where τ is the time constant defined by the RS and CS, τ=2RSCS. During the charge transferring step, the capacitor is connected to the TIA and the charge is transferred to
the AFE for measurement. This charge measurement-based ECG solution offers several advantages, including eliminating the buffers and the third electrode for RLD, shrinking the system size through fewer external components, and saving power.
It is convenient to add lead-off detection to this ECG solution with the design flexibility of the ADPD4100/ADPD4101 using a bioimpedance-based approach. Figure 3a shows the lead-off detection circuit with driving pulses to one electrode and receiving current at another electrode. If one or both electrodes are detached from the skin, the path is broken and no current is received. Figure 4 shows traces of ECG and received electric current for lead-off detection, where ECG is measured in time slot A and lead-off detection in time slot B.
When compared to the pull-up resistor circuit for lead-off
detection in conventional ECG solutions, which impact the input impedance of the ECG circuit, this bioimpedance-based lead-off detection in a separated time slot has no impact on the ECG measurement. With this dccoupled circuit, ECG is captured once the electrode-skin contacts are re-established.
Respiration measurement using the ADPD4100/ADPD4101 detects the bioimpedance change of the lung during the inhalation and exhalation cycles. Respiration measurement for patient monitoring in the intensive care unit (ICU) and during sleep allows patient management and timely alarms to save lives. It is critical for patients with respiratory problems and sleep apnea. Sleep apnea alone is a public health and safety threat, with more than 25 million adults afflicted in the U.S. 1
ECG measurement and lead-off detection. Instantaneous restoration of ECG with dc coupling.
Three factors needs to be considered if you want to drive more business value through your test team: data, systems, and processes. Companies that get these three factors right can realize large gains not only in their validation test teams but also in upstream and downstream functions as designs are iterated on faster and delivered to manufacturing for mass production sooner.
For the longest time, the act of testing products has been seen as an unavoidable expense in the product development cycle. At a high level, the reason companies test are (1) to make sure the product is developed according to specification by testing it against a set of requirements and (2) to run through intended use scenarios to make sure it
functions as described by product management and R&D. Looking at it from that perspective, it is easy to see why many organizations work to identify how few features they can test on a new product and still deliver it to market with good quality (the test coverage challenge). This might have worked somewhat in the past, but in the current world,
market windows are shrinking while the number of competitors is increasing.
Shorter market windows and more competitors pose two problems. First, the shorter market window puts more
pressure on V&V teams, who often have to make quality tradeoffs to save time and hope that the product is still acceptable. Second, technology is advancing faster, so companies need to test for additional scenarios that they haven’t had to test for in the past. From a competition perspective, even established companies are finding that small startups can pose a big threat. Increased competition means not only that customers have more options but also technology advancements are helping new companies enter a market that might otherwise be considered unattractive. New entrants might also have a higher quality product because they have smaller product portfolios to manage. Established companies can lose market share to new competitors who can integrate new features and technology, get their product to market faster, and do all this without compromising quality
Today, many companies are trying to solve these issues through various initiatives, but only a few companies realize how much value can come from looking at their V&V test teams. Even though the V&V teams are part of a cost center, they can offer real gains in operational efficiency and increase product value. To enable test teams to add more business value, companies need to optimize their teams’ workflows and operational processes, examine how the team builds and manages their systems, and identify what insights can be derived from their test data. When doing so, they can identify and share relevant information
upstream with R&D and downstream with manufacturing to capitalize on efficiency gains across the organization, breaking down traditional silos. The status quo has to change by making data, systems, and process improvements to increase the overall business value V&V teams add to companies.
The product design process creates a vast amount of data that, if managed correctly, can provide valuable insights into how a product will perform. This is the test data acquired while testing the product; however, companies also have data on the state of the system, who performed the test, and which instruments were used during the test. Maybe your organization captures all that data, or maybe it captures some of it. Either way, step 1 is determining what you want to learn from the data. That, in turn, helps you determine which data you need to collect to gain the insights you are seeking. V&V is about product performance and understanding not only if a product performs to
specification but also why it does or doesn’t meet the specified requirements. If you understand why, then you can accelerate the iterations as you feed this information back to R&D and design engineers, who can solve the problem faster. Once you have defined which data you want to capture, you need a strategy to handle data coming from validation test benches. You can think of this data as the raw data that, depending on your strategy, can be both test and system data. However, just capturing it is not enough. You need to start with the end in mind, so after you have defined which insights you want to derive from your data, the next step is setting up a data pipe. This process defines how you ingest, govern, and transform data into something meaningful that can provide business context. The faster you can ingest data, the more real-time information you can derive from it to make decisions. You can also speed up the process if you’re notified in real time where a fix is needed and which test needs to be rerun. In this case, the earlier you’re notified, the less time you waste testing a product that will not pass a long-running test. A
Finlandisatechnologyleader.Nearly336,000Finnsworkin technologycompanies,whileatotalofaround700,000 peopleworkinthetechnologysectoreitherdirectlyor indirectly.Theindustryisresponsiblefor65%ofall investmentinresearchanddevelopmentcarriedoutin Finland.
TheturnoverofFinnishelectronicscompanieswas18Billion euros(2021).TheICTsectoremploys73.800peoplewho producedatotalturnoverof16.5Billioneurosin2021.
ETNisaFinnishtechnologymediaforeveryoneworking, studyingorjustinterestedintechnology.Through websitewithdailynewsandtechnicalarticles,daily newslettersandcolumnsETNcoverseveryaspectof hightechnology-automation,components,devices, networks,test&measurement,design&programming, embedded,electro&power,manufacturingand distribution.AllinFinnish.Thesitehasmorethan25000 uniquevisitorspermonth,generatingmorethan55000 pageviews.
WealsoorganisetheEmbeddedConferenceFinland annually.ECFistheonlyindependentconferenceon embeddedtechnologyinFinland.
See www.embeddedconference.fi formoreup-to-date information.
strategy and defined process for handling data also help your teams find data faster. Without a strategy, the engineers responsible for running the tests are left to decide on their own where data is stored and how it gets there. This often means that the data ends up on the test bench, where it may or may not be deleted after it is transferred to a USB drive or network drive. By defining how the data is ingested and transformed ahead of time, you make indexing and searching for it easier. Perhaps the R&D team wants to conduct a root cause analysis on why a particular product was sent back to them for rework multiple times and asks for all the data from every validation test run. Or maybe, years down the line, the team needs the original test data as part of a sustainment effort. These two cases and many more can result in hours of sifting through data to find exactly what the team needs. The power of data and a standard process grow more significant as your company’s teams start collaborating across different functions. When your company can easily connect requirements data and simulation data with
test and systems data, even though the raw data is in different formats, V&V teams can explore and interpret the data more quickly, which lets them feed back information to R&D sooner, leading to faster iterations between revisions of a product.
In summary, the benefits of setting up and defining a common process increase as you scale across the organization. However, if you start with your V&V teams, you can still achieve big efficiency gains and process improvements. Effective V&V processes make finding and using your data easier and faster and ensure your data can be used to notify and alarm your teams early.
To drive efficiency, your V&V test team needs to build validation test systems as quickly and easily as possible. For that, you need a strategy that defines a standard framework for your team’s efforts. Without a strategy, you likely will end up with multiple systems built on different hardware and software because the test
engineers building the system naturally default to whatever tool they prefer. This approach leads to test teams owning a lot of different types of expensive equipment that can’t easily be repurposed and integrated into new test setups. At least not without spending a lot of valuable time redeveloping code and learning how a specific piece of equipment needs to be interfaced to. Even if you are not repurposing equipment, maintaining a system that is built on software and hardware someone else preferred can be costly from both the monetary and time perspectives—especially if the person that built it has moved on to another role. Troubleshooting a system that is not performing as expected can take up most of your team’s time and force team members to make risk versus quality decisions. Stepping back, conducting a broader assessment, and standardizing how your team builds validation testers can make managing and maintaining your fleet of testers more efficient. For example, if your team is using a standard platform to build validation test systems, they share common knowledge on how to use and integrate instrumentation into a system. This enables a faster turnaround time when you need to stand up testers quickly. Your team can easily repurpose and reuse equipment from an old tester and understand exactly how to integrate it into the new system alongside new equipment. The biggest risk you face when standardizing on a platform is selecting an insufficient one. As you consider a platform, think
about the long term—not just your current test requirements. Openness and flexibility are important features, along with a broad range of instrumentation options that can work with new and future technologies. You also need to control your instrumentation and build software-connected systems in a manner that makes you agile enough to implement changes faster. Depending on which industry you are in, you might not be able to fit all the instrumentation in one platform, which is why openness is such an important factor. Easily integrating specialty equipment as needed and using a platform that provides reliability and repeatability are key to capitalizing on the efficiency improvements both when standing up new test systems and maintaining them into the future. Though a test system’s hardware is important, you can also achieve improved efficiency by standardizing on an open, interoperable software approach that is used alongside defined coding rules and standards built into a common framework. A single test team may include members proficient in different programming languages. To increase efficiency, avoid limiting them to writing test steps in a single language. Instead, use common frameworks that can handle code from multiple languages. With a common framework featuring standard components, your test team can focus on building the actual test steps instead of building all the basic framework components each time the team needs a new
tester. Emphasizing the talents of each team member while enabling team members to stand up testers faster leads to easier maintenance with a common software framework across all testers. To change the status quo and show how much value test teams can provide to your business, you must not overlook test systems. If done right, standardizing on a hardware platform and common software framework will reduce the time you need to build, manage, and maintain validation test systems. It will also give your test team more time to test your new products, enabling faster feedback to R&D, which means faster iterations on your product designs.
You can optimize hardware and software within or across your validation teams, but if you do not have the right process, you’ll never get the efficiency gains you are hoping for. Specific details around process and methodology are different for each team and company. Therefore, you should
start with an assessment that can help pinpoint the areas with bottlenecks. This involves some high-level requirements to operate efficiently through processes. The first requirement is as simple as your team having access to the tools it needs. Standardizing on software tools for development and deployment doesn’t help if your team can’t access them. To reduce the cost of maintaining frameworks and software development tools, you can choose a commercial off-theshelf solution maintained by a third party. But if your test teams must undergo a cumbersome procurement or internal process to obtain a development license, then you lose some of your efficiency gains. You also need to consider how to onboard and train new hires on the tools you use to shorten their learning curve. All these requirements must be deliberated as part of your process so they are automatically fulfilled as you expand your team. Automation should also be a big part of your process. It allows you to eliminate the efficiency loss and potential errors that occur in manual
processes. For example, as your team stands up and expands test systems, its members need to make tweaks to the test routines the team is running. Having traceability and insight into which test software is running on each test system can help your team quickly understand the performance of each system. Additionally, team members need to be able to remotely deploy new test software, receive alarm notifications, view test progress, and track utilization and calibration needs for each asset across all testers. In many cases, these are all still manual tasks, which are time-consuming. The less obvious benefit on the system side is that you are, by default, tracking metrics that can help you continually improve. For example, if you need to stand up a new tester, and you track your assets and their utilization, you can quickly determine if you have what you need, potentially eliminating an investment in new and costly instruments that take a long time to procure. With visibility into utilization and asset availability, test teams can make repurpose versus investment decisions sooner, making
planning easier and avoid compromising an upcoming test or using your available resources on overinvesting in equipment. One of the most time-consuming tasks is post processing of the test results and corresponding test parameters that are captured during a test. As discussed earlier, standard formats can reduce some of the time spent on this, but the real efficiency gains come from automatically performing the same set of analysis routines on every product you test. You should assess how you can build a process around automated analysis and reporting by identifying the analysis routines you always perform on your test results. This means reports can be available in a matter of minutes and can be shared with R&D sooner, which accelerates design iterations. As a final note, you can speed up more than just the iterations within product design. As you start building your process with the right data structure and systems, you can derive insights in dashboards that can be shared with multiple functions. For example, you may want a dashboard that shows manufacturing team members
how many tests have been completed for an upcoming product and estimates of when the product will be released for production. Or you may want an even higher level dashboard that you can share with executives showing how you are progressing on a new product. This is over the long term, but you can use and connect the data you have to provide insights to other teams, who can then build their own optimized workflows. This adds value across many parts of the business. In summary, your process needs to provide insights for your test team members based on the metrics that are important to them. As they find and correct issues, they can continually tweak and, therefore, improve over time, incrementally adding business value to the company. Additionally, as you start gaining insights, you can share these with other teams that can track and/or make decisions based on your insights. This provides a higher level of transparency into your operations, ultimately driving the right business decisions across your organization.
ETNdigi is a special digital magazine from ETN. We cover components, test & measurement, radios & networks and the whole scope of embedded software. To put it short: Everything related to electronics & embedded design.
Want to join us? Just contact ETN editor-in-chief Veijo Ojanperä at firstname.lastname@example.org or our sales manager Anne-Charlotte Lantz at email@example.com. More info about the pricing can be found at etn.fi/advertise