Perhaps IoT businesses are just IT businesses. So why the fixation on hardware? Hardware has long been considered the cornerstone of the IoT industry, but we're increasingly recognizing that success hinges not on the physical devices themselves but on the software that drives them. This shift mirrors patterns we've seen in the world of cloud computing, where software is being written without considering the hardware it runs on. Let's dive into this transformation.<h3>Commoditization: Scaling Through Common Hardware</h3>At the heart of the transformation from hardware-centric to software-focused solutions is commoditization. Jason Kridner, founder of <a href="http://beagleboard.org/" target="_blank">BeagleBoard.org</a>, aptly stated:"If you make a reprogrammable machine, then that machine can be used in a whole lot of other applications. We can pretty much put everything into that box. Then you drive the hardware to commodity because that’s what scales."By commoditizing hardware, companies can focus their resources on software development, where true differentiation occurs. Kridner emphasizes that it's the software engineers who ultimately "win" because “that’s where you put your differentiation”. Hardware becomes the scalable foundation upon which innovative software solutions are built.The saying: “hardware is hard” still resonates with many in the industry. Hardware development is associated with complexities, high investments, lengthy development cycles and significant security risks. By treating hardware as a common, reliable platform, businesses may find ways to work around these obstacles and focus on their core competencies.<h3>Software Defined Everything</h3>Hardware lacks flexibility, until it doesn’t. The move from hardware commoditization to software-defined solutions feels like a natural progression. “Software gets abstracted from the hardware layer”, as Gregory Roth from <a href="https://monogoto.io/" target="_blank">Monogoto</a> explains, “leading to more flexible solutions”.This abstraction enables agility and it liberates companies from the complexities of hardware-specific constraints. Francisco Xavier Garcia (<a href="https://www.picmg.org/" target="_blank">PICMG</a>) reinforces this point:"You want a modular solution that is quickly scalable and quickly deployable because you want to respond swiftly to the rapidly changing industry. When you have embedded software and hardware tightly coupled, your flexibility is limited, your adaptability to respond to threats is limited. Extensibility is needed, where you can extend your capability and grow what your system can do. Software gives you all that ability."By decoupling software from hardware, businesses enhance their ability to innovate, adapt, and scale, which is essential in a market characterized by constant change.<h3>Adopting Cloud Practices: Stealing from the Best</h3>The IoT industry's shift mirrors the evolution seen in cloud computing over the past decade. In 2011, Adam Wiggins, one of Heroku’s founders, published the <a href="https://12factor.net/" target="_blank">Twelve-Factor App</a>, outlining twelve principles of modern application development. While many of these principles seem obvious today, they weren’t at the time, highlighting how far we've come in embracing common cloud practices.Brandon Satrom from <a href="https://blues.com/" target="_blank">Blues</a>, drew inspiration from these principles to create a similar set for the connectivity landscape called the <a href="https://12factor.io/" target="_blank">Twelve Factor Thing</a>. This initiative aims to enhance developer efficiency and product reliability, increasing market trust in connected devices that adhere to these principles. Satrom doesn’t want to impose these 12 factors to anyone, but merely initiate the conversation on how to grow the market in terms of efficiency, security and reliability.Consider the common practices such as automated testing, DevOps, continuous integration, you name it. By applying these practices to the world of IoT, reliance on specific hardware components is significantly reduced – a lesson learned during recent chip shortages that exposed the vulnerabilities of tying software too closely to specific hardware.Eventually, there are a couple of things – or principles – we should be able to agree upon, already now. Think of the way we should approach <a href="https://12factor.io/factors/firmware/" target="_blank">firmware as a collection of dynamic assets</a> which change frequently through standardized firmware update processes. The necessity of testing, the 12 Factor Thing specifically mentions&nbsp;<a href="https://12factor.io/factors/hardware-in-the-loop/" target="_blank">Hardware-in-the-Loop test</a> (HIL). And, one of Satrom’s Things he’s most passionate about: <a href="https://12factor.io/factors/transport/" target="_blank">Harmonization of Radio Access Technologies</a> “you don’t want developers having to learn everything from soup to nuts” if they add an additional radio technology. This will “slow them down and increase the likelihood of failures”.<h3>Conclusion</h3>The IoT industry's future lies in embracing the fact that software people ultimately take all the credits. Hardware gets commoditized to better suit software-defined solutions, minimizing the learning curve for developers who can build their solutions using familiar interfaces and APIs, regardless of the underlying hardware.This approach not only streamlines development and deployment but also empowers businesses to swiftly respond to market changes and emerging threats. By focusing on software, companies can enhance flexibility, scalability, and adaptability. It's not just about building better machines; it's about redefining how we manage those machines and how we, as an industry, can better serve the needs of a constantly changing world.<blockquote>This article is based on the discussion that took place at <a href="https://iotstars.com/" target="_blank">IoT Stars</a>, held during Embedded World North America on October 8, 2024. Thanks to Brandon Satrom (Blues), Jason Kridner (<a href="http://BeagleBoard.org" target="_blank">BeagleBoard.org</a>), Francisco Xavier Garcia (PICMG) and Gregory Roth (Monogoto) for sharing their take on Software Defined IoT. To join future discussions on the developments in IoT, visit <a href="https://iotstars.com/" target="_blank">iotstars.com</a>.</blockquote>

In a world where software drives innovation, why are we still fixated on hardware in IoT?

The Reprogrammable Machine: The rise of software-defined IoT

Embedded World 2024, held at Nuremberg Messe from April 8 to 11, was a brilliant event for all engineers, designers, and researchers involved in embedded systems.&nbsp;With almost a thousand exhibitors and a huge schedule of related events, Embedded World has really lived up to its reputation of showcasing the industry's cutting-edge.&nbsp;Wevolver team members took the chance to meet engineers from all over the world. Here are some of the highlights.&nbsp;<h2>OKdo: LLMs at the edge.&nbsp;</h2>At Embedded World 2024, <a href="https://www.wevolver.com/profile/okdo" target="_blank" rel="noopener noreferrer">OKdo</a> - a key player in single-board computing and IoT solutions - &nbsp;showcased its significant advancements in industrial computing technologies. One of the highlights was the display from <a href="https://www.liquid.ai/" target="_blank" rel="noopener noreferrer">Liquid AI</a>. Liquid AI is developing a new generation of AI foundation models built from first principles - going beyond generative pre-trained Transformers (GPTs). Liquid neural net technologies unlock efficiency, robustness, explainability, and state-of-the-art performance. Generative AI on the edge (for exampl,e on a Raspberry Pi 5 or Rock 5a) becomes possible.&nbsp;<img src="https://images.wevolver.com/eyJidWNrZXQiOiJ3ZXZvbHZlci1wcm9qZWN0LWltYWdlcyIsImtleSI6ImZyb2FsYS8xNzEzOTUxMTA5ODUwLU9LIHN0YW5kLmpwZyIsImVkaXRzIjp7InJlc2l6ZSI6eyJ3aWR0aCI6OTUwLCJmaXQiOiJjb3ZlciJ9fX0=" style="width: 100%px;" class="fr-fic fr-dib">The OKdo Embedded World stand. Credit: OKdo.<h2>Enabling cutting-edge PCB design</h2>Altium, a PCB design software company, &nbsp;introduced its new enterprise-grade system, Electronics Lifecycle Management (ELM), a system for enterprise-level hardware design and development, which was a featured presentation in the Exhibit Hall. The booth experience also featured live demos of solutions, including Altium Designer, Altium 365, Nexar API, and Octopart. Altium is transforming electronic development by allowing anyone in the electronics industry to connect with team members and partners and with solutions and data to work more productively.&nbsp;“Altium is uniquely positioned to reshape the future of electronics development and production. We were extremely excited to share our vision of a transformed industry ― which is fast becoming a reality ― with the world’s leaders in embedded technology at Embedded World 2024.” – VP Enterprise Marketing, Altium, Naomi Hamels&nbsp;See Altium in action below:&nbsp;<iframe width="640" height="360" src="https://www.youtube.com/embed/aZeWMmDGFKg?&amp;wmode=opaque&amp;rel=0&amp;enablejsapi=1&amp;origin=https://www.wevolver.com" frameborder="0" allowfullscreen="" class="fr-draggable" data-lf-form-tracking-inspected-kn9eq4roawkarlvp="true" data-lf-yt-playback-inspected-kn9eq4roawkarlvp="true" data-lf-vimeo-playback-inspected-kn9eq4roawkarlvp="true" data-gtm-yt-inspected-9="true" id="253684619" title="Altium at Embedded World 2024" data-gtm-yt-inspected-23="true">&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;</iframe><h2>Exploring NXP's advancements in motor control and TSN communication&nbsp;</h2>At Embedded World 2024, NXP showcased advancements in smart edge technologies aimed at improving device automation and connectivity. They introduced new technologies and products for various markets, including automotive, industrial, and smart home sectors. Demonstrations highlighted NXP's solutions for software-defined vehicles, smart home interoperability, and industrial automation. The event also featured expert-led tours and demonstrations exploring AI, edge computing, and more.<iframe width="640" height="360" src="https://www.youtube.com/embed/fFeo0q6AdWA?&amp;feature=youtu.be&amp;wmode=opaque&amp;rel=0&amp;enablejsapi=1&amp;origin=https://www.wevolver.com" frameborder="0" allowfullscreen="" class="fr-draggable" data-lf-form-tracking-inspected-kn9eq4roawkarlvp="true" data-lf-yt-playback-inspected-kn9eq4roawkarlvp="true" data-lf-vimeo-playback-inspected-kn9eq4roawkarlvp="true" data-gtm-yt-inspected-9="true" id="597216751" title="Wevolver at Embedded World: NXP Semiconductors" data-gtm-yt-inspected-23="true">&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;</iframe><h2>The latest in automotive local dimming control with Synaptics</h2>AT EW 2024, Synaptics launched its <a href="https://www.synaptics.com/products/embedded-processors" target="_blank" rel="noopener noreferrer">Astra platform</a> of AI-native embedded processors designed to enhance the capabilities of Internet of Things (IoT) devices. This platform focuses on providing multi-modal, AI-enabled microprocessors (MPUs) and microcontrollers (MCUs) that are optimized for power efficiency and high performance. Astra supports a unified software experience across various operating systems and includes an adaptive AI framework, which facilitates the integration of AI into devices. It also offers robust connectivity solutions, such as Wi-Fi and Bluetooth, and is designed to help developers easily incorporate AI features into their products.During our visit, we also got to see their SmartBridge technology live. This advanced automotive display processor is a local dimming IC that supports large, high-contrast, high-resolution automotive LCDs while consuming lower overall power and cost. SmartBridge can handle up to 30-inch displays and 6k resolutions. It leverages a proprietary algorithm to analyze incoming images and control the brightness of each LED in the array.&nbsp;<iframe width="640" height="360" src="https://www.youtube.com/embed/xGelGpfUJFw?&amp;wmode=opaque&amp;enablejsapi=1&amp;origin=https://www.wevolver.com" frameborder="0" allowfullscreen="" class="fr-draggable" data-lf-form-tracking-inspected-kn9eq4roawkarlvp="true" data-lf-yt-playback-inspected-kn9eq4roawkarlvp="true" data-lf-vimeo-playback-inspected-kn9eq4roawkarlvp="true" data-gtm-yt-inspected-9="true" id="506257900" title="Wevolver at Embedded World: Synaptics Incorporated" data-gtm-yt-inspected-23="true">&nbsp;&nbsp;&nbsp;&nbsp;</iframe><h2>Code reading with MVTec Software</h2>At the Embedded World 2024, <a href="http://www.mvtec.com" target="_blank" rel="noopener noreferrer">MVTec Software</a> showcased its extensive portfolio of machine vision software, prominently featuring their products HALCON and MERLIC. These products are designed to integrate seamlessly with various hardware configurations, supporting a broad spectrum of operating systems, AI accelerators, CPUs, and embedded boards. The company emphasized the adaptability of their software to different industrial needs, making it suitable for smart cameras, sensors, and other embedded vision applications.MVTec's presentation revolved around the theme "Embedded Vision = YOUR Hardware + OUR Software," underscoring the customizable nature of their solutions to enhance hardware capabilities with sophisticated vision software. This approach not only facilitates enhanced performance and compatibility across different machine vision standards like GigE Vision, GenICam, and USB3 Vision but also promotes quicker time-to-market for products developed using their systems. The technologies highlighted include advanced features for code reading, optical character recognition (Deep OCR), and deep learning applications such as anomaly detection.<iframe width="640" height="360" src="https://www.youtube.com/embed/CHitadg5uKg?&amp;wmode=opaque&amp;rel=0&amp;enablejsapi=1&amp;origin=https://www.wevolver.com" frameborder="0" allowfullscreen="" class="fr-draggable" data-lf-form-tracking-inspected-kn9eq4roawkarlvp="true" data-lf-yt-playback-inspected-kn9eq4roawkarlvp="true" data-lf-vimeo-playback-inspected-kn9eq4roawkarlvp="true" data-gtm-yt-inspected-9="true" id="37302711" title="Wevolver at Embedded World: MVTec Software Demo" data-gtm-yt-inspected-23="true">&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;</iframe><h2>The 2024 State of Edge AI Report</h2>At Embedded World we were thrilled to share the 2024 State of Edge AI Report with you. &nbsp;This report represents months of hard work, research, and dedication to exploring the dynamic world of Edge AI applications across industries.<h3>Report Highlights:</h3>#1: Industry-specific insights– Each industry is covered by a dedicated chapter that provides industry-specific insights, descriptions, and examples. ,#2: Real-life case studies – Get comprehensive insights from featured sections of real-world case studies.&nbsp;#3: A look into the future - The report also provides a peek into the exciting generative AI technology and its convergence with edge computing before delving into the challenges still hindering Edge AI today.We hope you download the report below!&nbsp;<iframe width="540" height="665" src="https://c8056756.sibforms.com/serve/MUIFAIvLRE9FbYyICZVAlYl6oaSYA1y_4gbHWS82SbbrrHstSB6VMs2aj57Tvg0e16L_UZVprsukgYH2yB927lpe-0uPg3ZEAhJiDPRQb2_oMEFQLYiEcG6zXHV6DwzxZPaJo25g3EzCRsYZ5z3ey0dkbIZE_M-AQ1kRN3i2_dq-vjEv7FQC29p_-Vu8tD5vbyaerZZF9IYqTQ8D" frameborder="0" scrolling="auto" allowfullscreen="" style="display: block;margin-left: auto;margin-right: auto;max-width: 100%;" class="fr-draggable" data-lf-form-tracking-inspected-kn9eq4roawkarlvp="true" data-lf-yt-playback-inspected-kn9eq4roawkarlvp="true" data-lf-vimeo-playback-inspected-kn9eq4roawkarlvp="true" data-gtm-yt-inspected-9="true" data-gtm-yt-inspected-23="true">&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;</iframe>

Wevolver was present at the recent Embedded World event in Nuremberg to launch the 2024 Edge AI report. 


Embedded World 2024: Trends, Tech, and Trailblazers

MATT Framework is the core software application that was designed to allow users to calibrate, prototype, develop testing scenarios, and manually operate the <a href="https://www.adaptarobotics.com/matt/" target="_blank">MATT robot</a>, making it as easy to interact with as possible. It functions like an application server with a dedicated user interface for seamless robot control and operation.<img src="https://images.wevolver.com/eyJidWNrZXQiOiJ3ZXZvbHZlci1wcm9qZWN0LWltYWdlcyIsImtleSI6ImZyb2FsYS8xNzA4NTU2OTc4Nzg1LU1BVFQtcHJvZHVjdC1saW5lLXVwLTEwMjR4NjI3LmpwZWciLCJlZGl0cyI6eyJyZXNpemUiOnsid2lkdGgiOjk1MCwiZml0IjoiY292ZXIifX19" style="width: 100%px;" class="fr-fic fr-dib">MATT Robot product line-upMATT Framework, as a software product, has expanded over time to offer multiple APIs that accommodate various programming languages, reducing user-side integration efforts.In an earlier version of the Framework, which was already experiencing effortless adoption among MATT Robot users, the Python API was introduced. Looking towards further improving the functionality of the software, the next objective was adding a Python Read-Eval-Print Loop (REPL). Up until recently, carrying out such a task proved to be technologically challenging, but the difficulty diluted with the advancement of available and used methods and knowledge.<h2>Python runs in the browser</h2>To see the set intention being implemented into MATT Framework, multiple methods have been explored in order to find the most appropriate and more efficient one for the current situation. In the end, adding Python to the software was carried out through Pyodide.Pyodide is a Python distribution designed specifically for the browser environment, offering seamless integration of JavaScript and Python through a robust foreign function interface (FFI). It leverages the power of WebAssembly, enabling the execution of Python code with full support for the standard library and the ability to load additional packages from PyPI, the Python Package Index.<img src="https://images.wevolver.com/eyJidWNrZXQiOiJ3ZXZvbHZlci1wcm9qZWN0LWltYWdlcyIsImtleSI6ImZyb2FsYS8xNzA4NTU3MDA1NTk2LVB5c2NyaXB0LWlzLWEtcG9wdWxhci1wcm9qZWN0LXVzaW5nLXB5b2RpdGRlLTEwMjR4NjU4LmpwZWciLCJlZGl0cyI6eyJyZXNpemUiOnsid2lkdGgiOjk1MCwiZml0IjoiY292ZXIifX19" style="width: 100%px;" class="fr-fic fr-dib">PyScript is a popular project using Pyodide — pyscript.netWhen it comes to browser-based Python implementations, Pyodide stands out as a comprehensive and feature-rich solution. With the arrival of WebAssembly, Pyodide takes advantage of this technology to deliver a highly capable Python environment while ensuring compatibility across different browsers. Unlike other alternatives such as Transcrypt, which require a pre-compilation step to transpile the Python code into JavaScript, Pyodide offers a more direct and streamlined approach.Moreover, Pyodide excels in its ability to directly load pure Python packages from PyPI. This expands its functionality and allows developers to leverage a wide range of existing Python libraries, enhancing the capabilities and versatility of their browser-based applications. In comparison, alternatives like Brython may not provide the same level of support or seamless integration with PyPI.<h2>Enhancing MATT Console</h2>To make the utilization of MATT robots during testing as smooth as possible, the integrated MATT Framework software was designed to contain a console section that allows the robot operator to easily interact with their test cases. Aiming to make automation as accessible as possible, the necessary code for performing test operations is generated by the console. Considering that MATT is a testing robot built to interact with devices having touchscreens and buttons, the most common operations consist of taps, swipes, identifying icons, and saving pictures.MATT’s console was initially built to support JSON API. As the number of teams utilizing the device testing robot grew, the necessity of updating the console by integrating Python API was met, as it was getting increasingly difficult for Python users to operate MATT.<img src="https://images.wevolver.com/eyJidWNrZXQiOiJ3ZXZvbHZlci1wcm9qZWN0LWltYWdlcyIsImtleSI6ImZyb2FsYS8xNzA4NTU3MDMyMTEwLUxhcHRvcC1jb250cm9sLXRlc3Rpbmctcm9ib3Qtc2NhbGVkICgxKS5qcGciLCJlZGl0cyI6eyJyZXNpemUiOnsid2lkdGgiOjk1MCwiZml0IjoiY292ZXIifX19" style="width: 100%px;" class="fr-fic fr-dib">Laptop control of testing robot — Adapta Robotics on UnsplashThe initial intention of the update (named ServicePack 2 software update) was only to incorporate Python code generation support in order to facilitate robot operation. Soon, the objective evolved into making the best of both worlds by implementing bidirectional code translation that supports both the Python and JSON APIs. This meant that anyone who is using the JSON API can now take advantage of this feature to convert the test cases to Python code and vice-versa.Translating API&nbsp;callsTo achieve the above mentioned, during the technical development of the update it has been made use of the already implementation of the Python API as a wrap around the existent JSON command protocol and the API’s significant surface overlap.For example, a Python step that tells the system to tap a specific location on a device looks like this:<img src="https://images.wevolver.com/eyJidWNrZXQiOiJ3ZXZvbHZlci1wcm9qZWN0LWltYWdlcyIsImtleSI6ImZyb2FsYS8xNzA4NTU3MTA4MjE2LTE3MDg1NTcxMDgyMTYucG5nIiwiZWRpdHMiOnsicmVzaXplIjp7IndpZHRoIjo5NTAsImZpdCI6ImNvdmVyIn19fQ==" class="fr-fic fr-dib">While its corresponding JSON command is:<img src="https://images.wevolver.com/eyJidWNrZXQiOiJ3ZXZvbHZlci1wcm9qZWN0LWltYWdlcyIsImtleSI6ImZyb2FsYS8xNzA4NTU3MDk3ODAzLTE3MDg1NTcwOTc4MDMucG5nIiwiZWRpdHMiOnsicmVzaXplIjp7IndpZHRoIjo5NTAsImZpdCI6ImNvdmVyIn19fQ==" class="fr-fic fr-dib">By patching some of the methods in the Python API, specifically those for sending commands over a socket connection, the generation the JSON commands were made possible simply by calling the Python method inside the browser using Pyodide. The one-way conversion was not just possible, but easier to implement than initially expected.<img src="https://images.wevolver.com/eyJidWNrZXQiOiJ3ZXZvbHZlci1wcm9qZWN0LWltYWdlcyIsImtleSI6ImZyb2FsYS8xNzA4NTU3MTg2MzI1LWRvY3VtZW50YXRpb24tcG9ydGFsLTEwMjR4NTc0LmpwZWciLCJlZGl0cyI6eyJyZXNpemUiOnsid2lkdGgiOjk1MCwiZml0IjoiY292ZXIifX19" style="width: 100%px;" class="fr-fic fr-dib">Documentation portal for the MATT, the device testing robot – docs.mattrobot.ai<h2 id="isPasted">JSON to Python&nbsp;</h2>Next, the focus was on completing the development of the reverse transform, enabling the conversion of JSON commands into Python calls. To accomplish this, a converter class, specifically designed to accept JSON commands as input and generate corresponding string representations of the Python calls, was created.<h2>Limitations</h2>In mathematics, a transform is deemed reversible when the sequence A =&gt; B =&gt; A1 results in identical values for both A and A1. In the present case, however, the code translation process is only partially reversible, because when handling iterative loops, they become flattened into individual operations.<img src="https://s3-us-west-1.amazonaws.com/wevolver-project-images/froala%2F1708557239611-limitations-when-translating-API.gif" style="width: 100%px;" class="fr-fic fr-dib">Limitations when translating API calls<h2 id="isPasted">Refactoring to React</h2>An important step that was achieved during the implementation of the developed MATT Framework software update was rewriting the console feature using React. This came as a natural choice since the component was separable from the main project while also being very heavy in the business logic department.By utilizing <a href="https://prismjs.com/" target="_blank">PrismJS</a>&nbsp;and its capabilities, enhancing the component with syntax highlighting was not only easy but also necessary (contributing to the overall scope of delivering the robots’ operator an accessible, easy-to-use interaction). The end of the process presented itself with numerous files inside the console editor project. For this reason, it has been compiled into a separate module using WebPack with the result of the compilation being imported into the main project.<h2>Conclusion</h2>Comprising all added updates and new functionalities, the picture featured below shows the improved user interface emphasizing the append mode, syntax highlighting, and instant API to API translation.Ultimately, the ability to execute Python code in the browser through Pyodide has proved to be a game-changer for the utilized web-based user interface. This development decision not only reduced the workload significantly, but it allowed many more features to be shipped through the release updates that ultimately closely follow the MATT Robot product line roadmap.The above-described walkthrough serves as a great example of how library choices can impact the product’s roadmap and why thorough technology research is crucial.<img src="https://s3-us-west-1.amazonaws.com/wevolver-project-images/froala%2F1708557270884-Improved-console-1024x572.jpeg" style="width: 100%px;" class="fr-fic fr-dib">The improved console of the MATT Framework software

MATT Framework is the core software application that was designed to allow users to calibrate, prototype, develop testing scenarios, and manually operate the MATT robot, making it as easy to interact with as possible.

How adding Python control to MATT Framework improved the functionality of Our Robots

As the world of business continues to shift and evolve, one of the most important factors for success is asset tracking. It involves keeping track of valuable company assets to optimize their usage, prevent loss, and ensure that they are properly maintained. This is important for businesses of all sizes, as it can help to increase efficiency, reduce costs, and improve asset maintenance. Asset tracking is especially critical in today’s rapidly evolving business landscape, where globalization and digitalization have made it more important than ever to have visibility and control over assets. By implementing an asset tracking system, businesses can gain a competitive edge and ensure that they are able to make the most of their resources.<h2>Methods of Asset Tracking: how to use Manual vs Advanced</h2>There are several methods for tracking assets, including manual methods like paper records and spreadsheets, as well as more advanced methods like asset tracking software and RFID (radio-frequency identification) technology or&nbsp;<a href="https://maxlab.io/blogs/2023/smart-parking-lot/" target="_blank">machine vision&nbsp;</a>. The right method will depend on the type and number of assets being tracked, the level of detail required, and the budget available for tracking technology.In today’s increasingly digital world, asset tracking is more important than ever. As businesses expand and become more globalized, it’s essential to have a way to track assets across multiple sites and countries. Asset tracking provides businesses with visibility into their assets around the globe, allowing them to make informed decisions about their operational costs, inventory needs, and more.<h2>Popular Asset Tracking Technologies: Barcode Scanning, RFID, and GPS Tracking</h2><img src="https://s3-us-west-1.amazonaws.com/wevolver-project-images/froala%2F1706004197876-barcode-tracking.webp" style="width: 300px;" class="fr-fic fr-dib">Barcode scanning is a widely used asset tracking technology that has been around since the 1970s. It involves attaching a barcode, which has encoded information about the asset, such as its serial number or other identifying information, to the asset. A barcode scanner is then used to read the barcode and access the asset’s information. Barcode scanners are widely available and relatively inexpensive, making them a popular choice for many businesses.One of the main advantages of barcode scanning is its simplicity. Barcode scanners are easy to use, and many industries are already familiar with the technology. This makes it easy for businesses to implement barcode scanning into their existing asset tracking systems. In addition, barcodes are small and can be attached to assets in a variety of ways, making them easy to use in a wide range of applications.<img src="https://s3-us-west-1.amazonaws.com/wevolver-project-images/froala%2F1706004217534-RFID.webp" style="width: 300px;" class="fr-fic fr-dib"><a href="https://en.wikipedia.org/wiki/Radio-frequency_identification" target="_blank" rel="noopener" id="isPasted">Radio Frequency Identification (RFID)</a> is another popular asset tracking technology. RFID tags are attached to assets, and an RFID reader is used to read the tags from a distance. RFID tags contain more information than barcodes, such as the item’s location, date of manufacture, and warranty information. RFID tags are also more durable than barcodes, making them better suited for long-term tracking applications. The downside to RFID tags is that they require a more expensive reader than barcodes and can be more difficult to install on assets.<img src="https://s3-us-west-1.amazonaws.com/wevolver-project-images/froala%2F1706004233873-asset-car-tracking.webp" style="width: 300px;" class="fr-fic fr-dib">The most advanced asset tracking technology is Global Positioning System (GPS) tracking. GPS tracking involves attaching a GPS-enabled device to an asset and using satellite signals to track its location in real time. GPS tracking is used in many industries, such as transportation, agriculture, logistics, and security. GPS tracking can provide valuable insights into asset movement and can help businesses improve their efficiency by providing timely updates on their assets’ locations. The biggest downside to GPS tracking is that it requires an expensive device as well as an ongoing subscription fee for access to satellite signals.<h2>Conclusion: The Benefits of Implementing Asset Tracking Solutions for Businesses</h2>There are many benefits to using asset tracking in your business. By keeping track of your assets, you can ensure that you have adequate inventory on hand and that you are utilizing your resources in the most efficient manner possible. Asset tracking also helps you keep an accurate record of your financials and ensure that your assets are secure.Finally, asset tracking can help you make better decisions about future investments. By analyzing the data gathered through asset tracking, you can determine which assets are performing well and which need to be replaced or upgraded. This allows you to allocate resources more effectively and make informed decisions about future investments.

As the world of business continues to shift and evolve, one of the most important factors for success is asset tracking. It involves keeping track of valuable company assets to optimize their usage, prevent loss, and ensure that they are properly maintained.

Asset Tracking: how to to Optimize Usage and Maintenance

We’ve heard the saying that modern cars are like computers on wheels. Well, vehicles are also like smartphones on wheels. Familiar and favorite apps from your phone are now available natively inside your ride. You can listen to streaming music, play a movie, map a route to the nearest ATM, and soon even order dinner or your favorite beverage on the way home or to work.We’re in the age of <a href="https://www.synopsys.com/blogs/chip-design/software-defined-vehicles-changes-auto-design.html" target="_blank">software-defined vehicles</a>, where automotive functions and features are driven more by software than by mechanical components. Over-the-air (OTA) updates allow carmakers to correct software bugs, deliver new capabilities (including apps!), and assess the vehicle’s operation. This way of constantly modifying the vehicle also presents opportunities for collaboration with third parties for new offerings—and new revenue streams.However, as with any electronic system, vehicles can be ripe for cyberattack if not properly protected. OTA updates, especially when they involve third parties, represent a potential area of vulnerability. In this article, I’ll highlight key security considerations of OTA updates and in-vehicle apps and the measures that should be taken to keep cars safe and secure.<img src="https://images.wevolver.com/eyJidWNrZXQiOiJ3ZXZvbHZlci1wcm9qZWN0LWltYWdlcyIsImtleSI6ImZyb2FsYS8xNjk3NjMxMDI3NDAzLW9wdGljYWwtc2NhdHRlcmluZy1hdXRvbW90aXZlLndlYnAiLCJlZGl0cyI6eyJyZXNpemUiOnsid2lkdGgiOjk1MCwiZml0IjoiY292ZXIifX19" style="width: 300px;" class="fr-fic fr-dib"><section id="isPasted"><h2>Automotive Software Is Driving Vehicle Evolution</h2>OTA updates are delivered over a cellular network, Wi-Fi, or other radio frequency- (RF-) based methods that provide software updates. This grouping of technologies offers a fast and convenient way for vehicle manufacturers to resolve issues and future-proof their cars, all without requiring car owners to visit the dealer. There isn’t currently a standardized way in the automotive industry to verify software updates. As such, it’s possible for an OEM to have multiple ways to confirm software updates for some of its components or to rely on a complex supply chain for the delivery of these updates. But this should change as the software-driven features become further ingrained in the industry.In the U.S., the National Highway Traffic Safety Administration (NHTSA) offers security guidance through its <a href="https://www.nhtsa.gov/sites/nhtsa.gov/files/2022-09/cybersecurity-best-practices-safety-modern-vehicles-2022-tag.pdf" target="_blank" data-lf-fd-inspected-kn9eq4roawkarlvp="true">Cybersecurity Best Practices for the Safety of Modern Vehicles</a> report. The report outlines steps for vehicle manufacturers to take to mitigate the risk of cyberattacks. NHTSA isn’t alone in this endeavor. The EU has issued<a href="https://unece.org/transport/documents/2021/03/standards/un-regulation-no-155-cyber-security-and-cyber-security" target="_blank">&nbsp;UN Regulation No. 155,</a> which has clear requirements that require OTA capabilities to update vehicles in order to sell in that market.Efforts generally begin with a risk assessment, outlining components in a vehicle that any application would interact with. But then, there are a lot of questions about risk and responsibilities. As an example, let’s consider a self-parking application that’s activated by a smartphone. The vehicle in question already has autonomous driving capabilities. However, to enable self-parking, the driver needs to install an app that is tied to a parking garage company and must be allowed to communicate with the vehicle. For this application to work, the carmaker needs to provide location and other vehicular data to the app so that the car can navigate safely through the garage to an available parking spot once prompted by the driver. Is the OEM responsible for monitoring and managing the app’s security and capability, or is it the supplier's responsibility to ensure that the app contains nothing malicious or compromised? Many of these questions will be resolved by legal departments, but proper cybersecurity hygiene is the primary focus for minimizing these types of events.In-vehicle apps such as the self-parking example represent a new revenue stream, and the opportunities are wide open for a host of new and exciting capabilities. Imagine having a car that knows your favorite morning coffee order—once you step inside for your commute, the car has already placed the order, handled the payment, and is navigating the fastest route to the café for order pickup. Or, what if different in-vehicle apps could interact with one another? Your car senses—thanks to an app plus a tracking tag on the bag—that you have clothes in the trunk for the dry cleaner. The dry-cleaning app communicates with the café app, directing the car on the most efficient route to drop off the dry cleaning and pick up your coffee.The apps themselves must have built-in protections against malicious attacks. When they’re ready to be updated, the OTA updates must flow securely from server to vehicle, with little risk that the data will be intercepted and compromised.These days, carmakers and app developers are collaborating closely, though a primary focus is getting new capabilities to market. At the consumer level, it may be several years before we start seeing these example scenarios come to fruition. Enterprise-level innovations may come sooner. <a href="https://www.thewrap.com/pizza-hut-new-driverless-delivery-vehicle-scares-black-mirror-fans/" target="_blank">Pizza Hut, for example,</a> is exploring a driverless concept vehicle equipped with a pizza oven, a move that could transform how food is ordered, prepared, and delivered.Regardless of when these new offerings are available, the time is ripe to start addressing the security considerations.&nbsp;</section><section><h2>Keeping Connected Vehicles Safe and Secure, Inside and Out</h2>The risk assessment is a solid starting point that requires the right people for the task. Also important is assigning areas of responsibility, as well as accounting for the steps and actions needed to mitigate security risks. Carmakers would be wise not to rely on third-party app developers to understand the security requirements of their vehicles. These requirements will vary widely across vehicle types and manufacturers. A secure set of application programming interfaces (APIs) for the app developers is a good safeguard and opens an opportunity to monitor and limit access. This way, if anything inappropriate happens with the app, the vehicle would know how to react and potentially protect itself. Another important element is establishing a protective layer for the vehicle subsystems, enabling a safe, secure way for third parties to interact with the vehicle to deliver value-added services.Key elements of an effective OTA software update security program include having a robust public key infrastructure (PKI), well-defined software verification and validation practices, and effective vehicle software deployment, storage, and activation. The automotive industry can also take some lessons from the mobile device industry – smartphone makers, for example, must apply robust security measures to ensure that their devices, the apps installed on them, and the data accessible by these apps remain safe and secure. Carmakers must do the same across all their product lines.A variety of security and monitoring technologies is available to support carmakers. For the vehicles themselves, automotive engineers can employ <a href="https://www.synopsys.com/automotive.html" target="_blank">electronic digital twin solutions</a>, which provide virtual environments to test silicon solutions before they are deployed. As we see more consolidation of in-vehicle compute platforms, traditional methods to test silicon for security become impractical given all the touch points in these complex environments. Electronic digital twins provide an answer. <a href="https://www.synopsys.com/solutions/silicon-lifecycle-management.html" target="_blank">Silicon lifecycle management (SLM)</a> sensors and monitors can track electronic component behavior over time, from in-design stages to the field. With SLM, automotive engineers have a way to evaluate how devices are performing under their workloads over the long lifespan of vehicles.On the software side, <a href="https://www.synopsys.com/software-integrity/security-testing/static-analysis-sast.html" target="_blank">static application security testing</a>, <a href="https://www.synopsys.com/software-integrity/security-testing/fuzz-testing.html" target="_blank">fuzz testing</a> to uncover security weaknesses, and <a href="https://www.synopsys.com/software-integrity/code-dx.html" target="_blank">application vulnerability correlation</a> to diagnose and manage software risk are among the solutions available to safeguard in-vehicle apps and OTA updates.OTA updates are an efficient and cost-effective way for carmakers to future-proof vehicles, provide updates, or keep cars out of the dealerships for recall fixes. As with any electronic component or software solution, these updates as well as the in-vehicle apps that enrich the driver experience can all be vulnerable to cyberattacks. Conducting a thorough risk assessment, assigning areas of security responsibility, and deploying both hardware and software security solutions can help keep vehicles safe and sound for the long haul.Read more about&nbsp;<a href="https://wevlv.co/synopsys3" target="_blank" rel="noopener noreferrer">Key Considerations for Securing In-Vehicle Apps</a>. Read more about Synopsys in previously published articles:<a href="https://www.wevolver.com/article/designing-energy-efficient-ai-accelerators-for-data-centers-and-the-intelligent-edge" target="_blank" rel="noopener noreferrer">Designing Energy-Efficient AI Accelerators for Data Centers and the Intelligent Edge</a><a href="https://www.wevolver.com/article/why-ai-requires-a-new-chip-architecture" target="_blank" rel="noopener noreferrer">Why AI Requires a New Chip Architecture</a></section>

Balancing Innovation with Safety in the Age of Smart Mobility



Key Considerations for Securing In-Vehicle Apps

<h2 id="isPasted">The days of monolithic binaries are over</h2>As the IoT market is slowly maturing, IoT systems are becoming increasingly complex and versatile. As a result, embedded devices are composed of distinct firmware components made by different parties working together, often running on multiple MCUs simultaneously.This multi-party involvement has implications for how firmware is developed and updated. Rather than creating single binaries which are built and flashed to devices at once, firmware is constructed from multiple components. Each of these components can be managed and updated separately, just like cloud applications, allowing you to do what you do best. Whether it’s developing modem firmware, application code, bootloader software, drivers for sensors, or interfaces with hardware secure modules.Let’s explore the opportunities of remotely managing multi-component firmware in IoT devices by using the LwM2M standard.This article is the part of a series about OTA software updates for IoT devices. Read the first article <a href="https://www.avsystem.com/blog/remote-firmware-updates-for-iot-devices/" rel="noopener" target="_blank">here</a>.<h2>Problems with full-firmware updates</h2>When dealing with IoT devices which are highly constrained due to their limited available bandwidth, power and memory size, the process for device firmware updates (DFU) has to be carefully crafted. Updating the firmware can be resource-intensive, certainly when the network connection is unstable, leading to a lengthy update process which requires the radio to be on and drains the device’s battery.As IoT devices become more complex, so do their firmware, leading to increasingly large update files. This not only increases the necessary time and required energy to perform an update, it also makes the update more prone to errors as the devices require a stable connection for a longer period of time.In contrast to single-binary firmware which are to be updated fully (as described in our <a href="https://www.avsystem.com/blog/remote-firmware-updates-for-iot-devices/" rel="noopener" target="_blank">latest article</a>), embedded firmware can be composed of separated code blocks which can be configured and updated independently. A component-based update process can significantly reduce the required resources for updating the device’s firmware.<h2 id="isPasted">Deconstructing embedded device firmware</h2>Firmware is usually stored in the device’s flash memory; non-volatile memory which persists after a reboot and can be electronically rewritten. To virtually isolate different firmware components, flash memory is split into multiple partitions. Different firmware components – such as the bootloader, application firmware or the tinyML model – reside in dedicated flash partitions. Consequently, one partition can be safely updated while the others remain untouchedEmbedded devices usually don’t contain just one MCU. Instead they often contain multiple processors each being responsible for a dedicated task. Next to the main MCU, the device may include one or multiple radio modems for cellular or Bluetooth communication. It may include smart sensors which include local processing to add more intelligence. For example, cameras which detect the number of cars, bikes or people passing by, or microphones which only detect specific sound events. All of these MCUs may contain flash partitions which can become updatable.<img src="https://images.wevolver.com/eyJidWNrZXQiOiJ3ZXZvbHZlci1wcm9qZWN0LWltYWdlcyIsImtleSI6ImZyb2FsYS8xNjk4MDc1NDgxMzg4LWJsb2dwb3N0LWZsYXNoLW11bHRpLTEuanBnIiwiZWRpdHMiOnsicmVzaXplIjp7IndpZHRoIjo5NTAsImZpdCI6ImNvdmVyIn19fQ==" style="width: 302px;" class="fr-fic fr-dib">A multi-MCU IoT device with partitioned flash memoryMaking flash partitions updatableIn order to make a firmware component updatable, a dedicated partition – called the update partition – needs to be created for each component. In essence, this means that each component consists of two partitions; one that runs the actual firmware, and one which is reserved to store new firmware images during firmware updates. In case flash size is restricted, update partitions can be created in external flash memory.<img src="https://images.wevolver.com/eyJidWNrZXQiOiJ3ZXZvbHZlci1wcm9qZWN0LWltYWdlcyIsImtleSI6ImZyb2FsYS8xNjk4MDc1NDk5MTc1LWJsb2dwb3N0LWZsYXNoLW11bHRpLXVwZGF0ZS1tY3UtMi5qcGciLCJlZGl0cyI6eyJyZXNpemUiOnsid2lkdGgiOjk1MCwiZml0IjoiY292ZXIifX19" style="width: 302px;" class="fr-fic fr-dib">A multi-MCU IoT device with partitioned flash memory containing update partitions<h2 id="isPasted">Multi-Component Firmware Updates</h2>Let’s explore what really happens deep down in the device’s flash memory when a firmware component is being updated.The process starts when new firmware is released and uploaded to the repository. Once uploaded, the LwM2M Server informs the device about the availability of new firmware, enabling the device to download the firmware image.DownloadingOnce the firmware update is initiated for a specific component, the device starts downloading the new firmware image into the dedicated update partition. After the download has completed, the application initiates a reboot. At boot, the bootloader validates the authenticity and integrity of the new firmware, once validated it initiates the swapping process.Swapping firmwareIn contrast to Linux-based systems which support “position-independent code”, microcontrollers usually can’t boot firmware directly from the update partition. As a result, the firmware residing in the update partition needs to be swapped with the original firmware. When using the open source MCUboot bootloader, the two images are swapped using the flash partition known as the scratch partition. The scratch partition acts as a buffer, allowing the bootloader to switch the new and existing firmware images between their respective partitions. Throughout the swapping process, the device goes offline and becomes unresponsive to server commands.In the swap process, the bootloader first copies a chunk of the new firmware from the update partition to the scratch partition. It then moves the existing firmware to the update partition. Finally, it moves the new firmware chunk from the scratch partition to the original partition. This process repeats until the entire new firmware image has been moved into the original partition, completing the swap.Running new firmwareAfter the swap has been completed, the bootloader initiates the new firmware from the newly flashed image. If initialized successfully, the device starts running the new firmware. The new firmware is expected to write a flag in the flash memory to indicate that the update was successful, which is done after performing several self-checks, e.g. initializing the hardware.RollbackIf no successful update flag is added to the flash memory, the bootloader assumes something went wrong and performs a rollback during the next boot. The rollback triggers the swap procedure to revert the initial firmware. Once the swap is complete, the bootloader initiates the old firmware, ensuring the device remains operational.<img src="https://images.wevolver.com/eyJidWNrZXQiOiJ3ZXZvbHZlci1wcm9qZWN0LWltYWdlcyIsImtleSI6ImZyb2FsYS8xNjk4MDc1NTYxNzk2LXJvbGxiYWNrLndlYnAiLCJlZGl0cyI6eyJyZXNpemUiOnsid2lkdGgiOjk1MCwiZml0IjoiY292ZXIifX19" style="width: 302px;" class="fr-fic fr-dib">DFU process for a distinct firmware component<h2 id="isPasted">Presenting the LwM2M Advanced Firmware Update Object</h2>In the <a href="https://www.avsystem.com/blog/remote-firmware-updates-for-iot-devices/" rel="noopener" target="_blank">previous article</a>, we’ve discussed the Firmware Update Object (/5) as defined in the <a href="https://technical.openmobilealliance.org/OMNA/LwM2M/LwM2MRegistry.html" rel="noopener" target="_blank">LwM2M Object and Resource Registry</a>. The creation of updatable firmware components can be accomplished through implementing the Advanced Firmware Updates Object (/33629). Object 33629 extends the functionality of the Firmware Update Object (/5) by enabling multiple Object Instances to represent the different components of device's firmware. The specific purpose of these components are not standardized and can vary depending on the implementation.Similar to Object 5, the Advanced Firmware Updates process comprises 4 States (idle, downloading, downloaded, updating) and no less than 13 Update Results which contain all the likely outcomes of the firmware update process; providing developers with information about the update status and results.Currently, developers are using Object 33629 as a custom LwM2M Object implementation, until it’s officially adopted into the <a href="https://technical.openmobilealliance.org/OMNA/LwM2M/LwM2MRegistry.html" rel="noopener" target="_blank">LwM2M Object and Resource Registry</a>. <img src="https://images.wevolver.com/eyJidWNrZXQiOiJ3ZXZvbHZlci1wcm9qZWN0LWltYWdlcyIsImtleSI6ImZyb2FsYS8xNjk4MDc1NTg4MTYzLTExLmpwZyIsImVkaXRzIjp7InJlc2l6ZSI6eyJ3aWR0aCI6OTUwLCJmaXQiOiJjb3ZlciJ9fX0=" style="width: 302px;" class="fr-fic fr-dib">View of the Firmware Update feature within&nbsp;<a href="https://www.avsystem.com/coiote-iot-device-management-platform/" rel="noopener" target="_blank" id="isPasted">Coiote IoT Device Management Platform</a><h2 id="isPasted">Conclusion</h2>The shift from monolithic firmware to a multi-component structure in embedded devices presents an opportunity to streamline and enhance the OTA device firmware update process. This approach offers flexibility in updating specific components independently, reduces resource usage, and provides a more efficient update process.As IoT devices become increasingly complex, moving towards multi-component FOTA updates is a logical and beneficial step. It facilitates the concurrent development of different firmware components by various parties which ensures the overall quality and functionality of IoT systems over time. As the IoT market continues to mature, this practice is likely to become the norm, making firmware updates more efficient, reliable, and user-friendly.

As the IoT market is slowly maturing, IoT systems are becoming increasingly complex and versatile. As a result, embedded devices are composed of distinct firmware components made by different parties working together, often running on multiple MCUs simultaneously.

Long live multi-component FOTA updates

Low-power computer vision provides a new opportunity to gain a practical understanding of the world through data collection and vision in remote areas.


Leveraging low-power embedded computer vision to solve urgent sustainability and safety challenges

The Challenges Ahead of Edge AI and How to Solve Them.

2023 Edge AI Technology Report. Chapter IX: Challenges of Edge AI

A Deep Dive Into Edge AI in Energy-Constrained Environments


2023 Edge AI Technology Report. Chapter V: TinyML

<h1 id="isPasted">2023 Edge AI Technology Report. Chapter IV: Hardware &amp; Software Selection</h1>Edge AI, empowered by the recent advancements in Artificial Intelligence, is driving significant shifts in today&#39;s technology landscape. By enabling computation near the data source, Edge AI enhances responsiveness, boosts security and privacy, promotes scalability, enables distributed computing, and improves cost efficiency.Wevolver has partnered with industry experts, researchers, and tech providers to create a detailed report on the current state of Edge AI. This document covers its technical aspects, applications, challenges, and future trends. It merges practical and technical insights from industry professionals, helping readers understand and navigate the evolving Edge AI landscape.<img src="https://images.wevolver.com/eyJidWNrZXQiOiJ3ZXZvbHZlci1wcm9qZWN0LWltYWdlcyIsImtleSI6ImZyb2FsYS8xNjg4NDg0MjA0ODUyLUVkZ2UgQUkgRG93bmxvYWQgYmFubmVyICgyKS5wbmciLCJlZGl0cyI6eyJyZXNpemUiOnsid2lkdGgiOjk1MCwiZml0IjoiY292ZXIifX19" style="width: 300px;" class="fr-fic fr-dib"><h2 dir="ltr"><iframe width="540" height="335" src="https://c8056756.sibforms.com/serve/MUIFABFML2jk7knWjihwfmaWcFVV6GvM3WWUP6qryWvjht4DWOrppyQ7-VLwYu9wbjkQCCMxaWS9vmneqio8zYAai1OKHh0-OA61OuFt5xdYgRxECqvgNTAnjjob8456fOSNTOOCe8YLBqs4fiLh0kTJSsBgWCDDvnFUMZSuafNiO9Cbhh5H7fBa4doE8a4CzpRXWMOtAOLUtRmq" frameborder="0" scrolling="auto" allowfullscreen="" style="display: block;margin-left: auto;margin-right: auto;max-width: 100%;" class="fr-draggable" data-lf-form-tracking-inspected-kn9eq4roawkarlvp="true" data-lf-yt-playback-inspected-kn9eq4roawkarlvp="true" data-lf-vimeo-playback-inspected-kn9eq4roawkarlvp="true"></iframe></h2><hr><h2 dir="ltr">Introduction</h2>When selecting and implementing Edge AI systems, it&#39;s important to consider the limitations of edge devices and their compatibility with other hardware and software. Edge devices like sensors and smartphones often have limited processing power, storage, and battery life. These limitations can negatively affect the performance of Edge AI systems, as they may require significant resources to perform complex tasks, such as image recognition and natural language processing. To overcome these limitations, AI models must be optimized to work with limited resources, and specialized hardware such as field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), and GPUs can accelerate computation at the edge.<img src="https://images.wevolver.com/eyJidWNrZXQiOiJ3ZXZvbHZlci1wcm9qZWN0LWltYWdlcyIsImtleSI6ImZyb2FsYS8xNjg3ODg3MjI2OTA1LW1pY3JvY2hpcC5wbmciLCJlZGl0cyI6eyJyZXNpemUiOnsid2lkdGgiOjk1MCwiZml0IjoiY292ZXIifX19" style="width: 300px;" class="fr-fic fr-dib">A field-programmable gate array (FPGA). Image credit: Microchip.Edge AI systems are designed to operate for extended periods, so the hardware selected should be capable of supporting future upgrades. It should be designed to accommodate new technologies and features that may be required in the future. Additionally, compatibility is a critical consideration when selecting software for Edge AI systems. Because Edge AI systems often require integration with existing infrastructure and hardware, it&#39;s important to select software that is compatible with a wide range of devices and platforms. This includes both hardware compatibility, such as compatibility with different types of sensors and processors, and software compatibility, such as compatibility with different operating systems and programming languages.<h3>Hardware Considerations</h3>Ensuring that the chosen hardware meets the requirements of Edge AI systems involves considering various factors. One key consideration is processing power, as Edge AI applications require hardware capable of processing data quickly and accurately. The hardware&#39;s computing power and clock speed play crucial roles in achieving real-time processing, which is essential in Edge AI systems. Another critical consideration is the amount and type of memory available in the hardware. Edge AI applications necessitate sufficient memory to store and process large volumes of data, with a focus on fast and real-time processing capabilities. There is a race to have the highest number of TOPS (tera-operations per second) in the hardware industry, but it&#39;s not a one-to-one comparison.<blockquote>&ldquo;The memory system handling, DMA configuration, software compression of weights, intelligence, and decompression all matter significantly. Focusing only on the box numbers is similar to focusing solely on the megapixel or gigapixel of a phone&#39;s camera without considering other factors. This tunnel vision is an error for the industry to make.&rdquo;&nbsp; - Rahul Venkatram, Senior Product Manager, Machine Learning, Arm</blockquote>Power consumption is another vital aspect, particularly for devices with limited power resources like IoT devices. Hardware optimization for low power consumption becomes crucial to ensure prolonged device operation without frequent battery replacements or recharging. Techniques such as utilizing low-power chips, hardware accelerators, and intelligent power management systems contribute to achieving low power consumption in Edge AI systems. In a conversation with Henrik Flodell, Director of Product Marketing at Alif Semiconductor, he emphasized that &ldquo;reducing power consumption is a key parameter users need to consider when they pick suitable devices for edge systems that will use ML.&rdquo;&nbsp;The Autonomous Intelligent Power Management (aiPM&trade;) technology from Alif is a vital solution for power consumption. It helps with selectively powering on the sections of an SoC that are needed when they are needed, and it powers them off when they are not, based on instantaneous processing demand and use case. It makes use of dual-processing regions within the MCU to speed up machine learning. The always-on, high-efficiency region continuously senses the environment, while the high-performance region wakes up when needed so it can execute heavy workloads and then return to sleep. This approach allows powerful devices to behave like low-power MCUs when necessary, which enables longer performance of smart IoT devices on smaller batteries.<blockquote>&ldquo;Systems that have to rely on CPU-bound processing tend to be power hungry, as the inference operations would take longer to finish, and the controller would run at 100% for much longer periods of time. Well-designed modern platforms offload the ML processing on accelerators designed for such workloads and can, therefore, finish faster and go to sleep sooner.&rdquo;&nbsp; - Henrik Flodell, Direct of Product Marketing, Alif Semiconductor.</blockquote>Seamless connectivity is also essential for Edge AI systems, requiring hardware that enables smooth data exchange with other devices and cloud-based platforms. The hardware should offer various connectivity options, including Wi-Fi, Bluetooth, and cellular networks, facilitating efficient communication and data processing with other devices. This connectivity ensures that data is processed and analyzed quickly and efficiently, contributing to the overall performance of the Edge AI system.<img src="https://images.wevolver.com/eyJidWNrZXQiOiJ3ZXZvbHZlci1wcm9qZWN0LWltYWdlcyIsImtleSI6ImZyb2FsYS8xNjg3ODg3MjkzMjM2LWFyZHVpbm8ucG5nIiwiZWRpdHMiOnsicmVzaXplIjp7IndpZHRoIjo5NTAsImZpdCI6ImNvdmVyIn19fQ==" style="width: 300px;" class="fr-fic fr-dib">The Edge Control from Arduino is a remote monitoring and control solution optimized for outdoor environments that enables users to collect real-time data from smart sensors and leverage AI at the edge. Image credit: Arduino.<h3>Software Considerations</h3>When it comes to Edge AI systems, selecting software with specific characteristics is crucial for optimal operation. Analyzing various factors ensures that the chosen software meets the requirements of the Edge AI system. Compatibility stands as a significant consideration in software selection. The software should run efficiently on the hardware, enabling real-time data processing. Moreover, compatibility with other software components used in the system, such as operating systems, libraries, and frameworks, ensures seamless integration and functionality.Scalability is another vital aspect to consider in Edge AI system design. Given that Edge AI systems often handle large amounts of data, the software must be capable of handling the real-time processing and analysis demands associated with such data. Scalable software guarantees the system can accommodate increasing data volumes, processing requirements, and user requests without compromising performance.The accuracy of the software employed in Edge AI systems holds great importance. These systems heavily rely on accurate data analysis and processing to provide meaningful insights and support decision-making. Therefore, the software must exhibit high accuracy and reliability in analyzing and processing data.Interpretability, or the software&#39;s ability to explain its results comprehensibly, plays a crucial role in Edge AI system design. Interpretability allows users to understand the system&#39;s decision-making process and provides insights into data analysis. This aspect becomes particularly significant in applications where the decisions made by the Edge AI system have substantial implications, such as healthcare and finance. The software used in Edge AI systems should prioritize interpretability, presenting results in a clear and understandable manner.<h3>Integration Considerations</h3>Integrating Edge AI with other systems is essential for seamless and efficient operations. As Edge AI systems generate vast amounts of data that require processing, analysis, and sharing with other systems, selecting communication protocols becomes imperative. Different protocols are used by Edge AI systems to communicate with other systems, and it is crucial to choose a protocol that is compatible with the integrated systems.Several commonly used protocols in Edge AI system integration include MQTT, CoAP, and HTTP. MQTT is a lightweight protocol widely utilized in IoT systems due to its low overhead and power consumption. CoAP, on the other hand, is designed specifically for resource-constrained networks, making it suitable for Edge AI systems. HTTP, a widely used protocol in web-based applications, is also suitable for Edge AI systems that communicate with web-based applications. The choice of communication protocol should align with the requirements of the Edge AI system and the systems being integrated.<img src="https://images.wevolver.com/eyJidWNrZXQiOiJ3ZXZvbHZlci1wcm9qZWN0LWltYWdlcyIsImtleSI6ImZyb2FsYS8xNjg3ODg3MzU4NTUyLUlNQUdFUyAyLTA0LmpwZyIsImVkaXRzIjp7InJlc2l6ZSI6eyJ3aWR0aCI6OTUwLCJmaXQiOiJjb3ZlciJ9fX0=" style="width: 300px;" class="fr-fic fr-dib">Comparison of protocols in terms of power consumption and resource requirement.In addition to communication protocols, data formats are another critical consideration when integrating Edge AI with other systems. Edge AI systems generate data in various formats, and it is vital to ensure compatibility between the data generated by the Edge AI system and the systems it interacts with. Data formats must be aligned and properly transformed or translated to facilitate seamless data exchange and interoperability between the systems.Global technology provider, Arm, has approached integration from a holistic thought process about where and how to deploy processing. David Maidment, Senior Director, Secure Device Ecosystem at Arm, explained their approach as follows: &ldquo;We&#39;ve landscaped our approach in two directions. The first is scaling down traditional cloud-native workloads, such as microservers. The second direction is taking IoT and embedded devices and making them connected, updateable, and secure. We&#39;re seeing these two trends converge, with higher-end, higher-power devices coming down in cost and power, and lower-end, lower-power devices being pushed up in performance capabilities.&rdquo;This results in two distinct software ecosystems. The first is the traditional cloud software ecosystem, with portable workloads that can run on servers with offload and acceleration. The second is a more traditional Linux box that was initially designed for a single function, like a router or gateway but is now being developed as a connected general-purpose box with capabilities that can change during its lifetime through new applications and AI models.<h3>Security Considerations</h3>Ensuring security is a critical factor when integrating Edge AI with other systems. Since Edge AI systems generate vast amounts of data, including sensitive information, protecting it from unauthorized access becomes imperative. Secure communication protocols such as SSL/TLS and SSH should be employed to establish secure channels for data transmission between the Edge AI system and other systems. These protocols encrypt the data, preventing eavesdropping and tampering during transit.To reinforce security, access controls and authentication mechanisms must be implemented. This ensures that only authorized users can access the data generated by the Edge AI system, safeguarding it from unauthorized individuals. Secure software development practices, such as regular code reviews and vulnerability assessments, should also be followed to identify and mitigate potential security risks in the integrated systems.Encryption is pivotal in securing Edge AI systems and should not be overlooked. By converting sensitive data into an encoded form, encryption ensures that even if the data is compromised, it remains unintelligible to unauthorized entities. Given that Edge AI systems often process sensitive data like personal health information, financial records, and valuable intellectual property, encrypting this data is crucial in preventing unauthorized access, theft, or manipulation.<img src="https://images.wevolver.com/eyJidWNrZXQiOiJ3ZXZvbHZlci1wcm9qZWN0LWltYWdlcyIsImtleSI6ImZyb2FsYS8xNjg3ODg3NDU1ODEyLTA1IGltYWdlLTAxLmpwZyIsImVkaXRzIjp7InJlc2l6ZSI6eyJ3aWR0aCI6OTUwLCJmaXQiOiJjb3ZlciJ9fX0=" style="width: 300px;" class="fr-fic fr-dib">Key usage in a cryptographic system of an IoT device and attack paths. Gain a broader perspective on Edge AI by revisiting our&nbsp;<a href="https://www.wevolver.com/article/edge-ai-a-comprehensive-guide-to-its-engineering-principles-and-applications" rel="noopener noreferrer" target="_blank">comprehensive guide</a>.<hr><h1 dir="ltr" id="isPasted">The 2023 Edge AI Technology Report</h1>The guide to understanding the state of the art in hardware &amp; software in Edge AI.Click through to read each of the report&#39;s chapters.&nbsp;<h2 id="isPasted"><a target="_blank">Report Introduction</a></h2><h3><a href="https://www.wevolver.com/article/2023-edge-ai-technology-report-chapter-i-overview-of-industries-and-application-use-cases" rel="noopener noreferrer" target="_blank">Chapter I: Overview of Industries and Application Use Cases</a></h3><h3 dir="ltr"><a href="https://www.wevolver.com/article/2023-edge-ai-technology-report-chapter-ii-advantages-of-edge-ai" rel="noopener noreferrer" target="_blank">Chapter II: Advantages of Edge AI</a></h3><h3><a href="https://www.wevolver.com/article/2023-edge-ai-technology-report-chapter-iii-edge-ai-platforms" rel="noopener noreferrer" target="_blank">Chapter III: Edge AI Platforms</a></h3><h3>Chapter IV: Hardware and Software Selection</h3><h3><a href="https://www.wevolver.com/article/2023-edge-ai-technology-report-chapter-v-tinyml" rel="noopener noreferrer" target="_blank">Chapter V: Tiny ML</a></h3><h3><a href="https://www.wevolver.com/article/2023-edge-ai-technology-report-chapter-vi-edge-ai-algorithms" rel="noopener noreferrer" target="_blank">Chapter VI: Edge AI Algorithms</a></h3><h3><a href="https://www.wevolver.com/article/2023-edge-ai-technology-report-chapter-vii-sensing-modalities" rel="noopener noreferrer" target="_blank">Chapter VII: Sensing Modalities</a></h3><h3><a href="https://www.wevolver.com/article/2023-edge-ai-technology-report-chapter-viii-case-studies" rel="noopener noreferrer" target="_blank">Chapter VIII: Case Studies</a></h3><h3 dir="ltr"><a href="https://www.wevolver.com/article/2023-edge-ai-technology-report-chapter-ix-challenges-of-edge-ai" rel="noopener noreferrer" target="_blank">Chapter IX: Challenges of Edge AI</a></h3><h3><a href="https://www.wevolver.com/article/2023-edge-ai-technology-report-chapter-x-future-of-edge-ai" rel="noopener noreferrer" target="_blank">Chapter X: The Future of Edge AI and Conclusion</a></h3><iframe width="540" height="335" src="https://c8056756.sibforms.com/serve/MUIFABFML2jk7knWjihwfmaWcFVV6GvM3WWUP6qryWvjht4DWOrppyQ7-VLwYu9wbjkQCCMxaWS9vmneqio8zYAai1OKHh0-OA61OuFt5xdYgRxECqvgNTAnjjob8456fOSNTOOCe8YLBqs4fiLh0kTJSsBgWCDDvnFUMZSuafNiO9Cbhh5H7fBa4doE8a4CzpRXWMOtAOLUtRmq" frameborder="0" scrolling="auto" allowfullscreen="" style="display: block;margin-left: auto;margin-right: auto;max-width: 100%;" class="fr-draggable" data-lf-form-tracking-inspected-kn9eq4roawkarlvp="true" data-lf-yt-playback-inspected-kn9eq4roawkarlvp="true" data-lf-vimeo-playback-inspected-kn9eq4roawkarlvp="true"></iframe> <hr><h2>The report sponsors.</h2><img src="https://images.wevolver.com/eyJidWNrZXQiOiJ3ZXZvbHZlci1wcm9qZWN0LWltYWdlcyIsImtleSI6ImZyb2FsYS8xNjg3ODg3NTQ5OTc0LTExICgxKS5wbmciLCJlZGl0cyI6eyJyZXNpemUiOnsid2lkdGgiOjk1MCwiZml0IjoiY292ZXIifX19" style="width: 300px;" class="fr-fic fr-dib">

What to Consider When Selecting Your Edge AI Hardware and Software.

2023 Edge AI Technology Report. Chapter IV: Hardware & Software Selection

If you&rsquo;re developing an IoT device, odds are that you want it to do some valuable computations to solve an important problem. Maybe you want to <a href="https://www.mistywest.com/posts/5-reasons-to-use-an-iot-framework/" target="_blank">deploy sensors in remote locations</a>, develop a device that can perform data analytics to monitor a renewable energy source, or build a medical device that can use computer vision to detect early signs of an illness.Whatever you&rsquo;re building, at some point you may start to wonder: should your device perform these important computations in the cloud or at the edge? Choosing between computing on the cloud or on the edge is a decision that can impact things like your device&rsquo;s cost or efficiency &ndash; and no one wants to make the wrong decision initially and spend time and money later down the line to pivot to the correct one.&nbsp;We asked some of our experienced engineers what they think someone needs to consider when choosing between computing in the cloud or on the edge.&nbsp;<h2>What is cloud computing?</h2>The &ldquo;cloud&rdquo; refers to the collection of servers that can be accessed over the internet &ndash; popular <a href="https://www.mistywest.com/posts/azure-vs-aws-which-iot-framework-is-right-for-you/" target="_blank">cloud providers include Amazon Web Services, Microsoft Azure</a>, and Google Cloud.&nbsp;These servers can provide on-demand computing resources to store and process data. You can think of the cloud as a centralized location for your files and programs, and you can connect any device to the cloud to access them. Services like Dropbox or Google Drive are some of the many cloud-based services out there.&nbsp;<img data-fr-image-pasted="true" src="https://www.mistywest.com/wp-content/uploads/2022/05/MistyWest_Edge_Vs_Cloud-03.png" alt="" width="100%" height="auto" srcset="https://www.mistywest.com/wp-content/uploads/2022/05/MistyWest_Edge_Vs_Cloud-03.png 1360w, https://www.mistywest.com/wp-content/uploads/2022/05/MistyWest_Edge_Vs_Cloud-03-300x155.png 300w, https://www.mistywest.com/wp-content/uploads/2022/05/MistyWest_Edge_Vs_Cloud-03-1024x528.png 1024w, https://www.mistywest.com/wp-content/uploads/2022/05/MistyWest_Edge_Vs_Cloud-03-768x396.png 768w" sizes="(max-width: 1360px) 100vw, 1360px" class="fr-fic fr-dii">&nbsp;Cloud computing describes the idea of performing computations in the cloud. These computations can include data analysis and visualization, computer vision and machine learning. An example of cloud computing in action is when your average smart home speaker sends your audio input to the cloud where it is interpreted by algorithms and sends back a response. &nbsp;&nbsp;&nbsp;<h2>What is edge computing?</h2><a href="https://blogs.nvidia.com/blog/2022/01/05/difference-between-cloud-and-edge-computing/#:~:text=The%20main%20difference%20between%20cloud,deploy%20them%20at%20the%20edge" target="_blank">The edge describes the &ldquo;edge&rdquo; of the network.&nbsp;</a>It includes devices that are an entry/exit point to the cloud, but are not part of the cloud itself. For example, a server in a data centre is part of the cloud; your smartphone and router that connects to that server is part of the edge.&nbsp;&nbsp;<img data-fr-image-pasted="true" src="https://www.mistywest.com/wp-content/uploads/2022/05/MistyWest_Edge_Vs_Cloud-02.png" alt="" width="100%" height="auto" srcset="https://www.mistywest.com/wp-content/uploads/2022/05/MistyWest_Edge_Vs_Cloud-02.png 1360w, https://www.mistywest.com/wp-content/uploads/2022/05/MistyWest_Edge_Vs_Cloud-02-300x155.png 300w, https://www.mistywest.com/wp-content/uploads/2022/05/MistyWest_Edge_Vs_Cloud-02-1024x528.png 1024w, https://www.mistywest.com/wp-content/uploads/2022/05/MistyWest_Edge_Vs_Cloud-02-768x396.png 768w" sizes="(max-width: 1360px) 100vw, 1360px" class="fr-fic fr-dii">&nbsp;Edge computing describes the idea of performing computations on the edge. This way, the processing is done closer to or at the location where the data is collected or acted upon.An example of an edge computing process is object detection on an autonomous vehicle. The vehicle processes data from its sensors and uses the results to avoid obstacles. Unlike your smart home speaker, the data it collects is processed locally, rather than sending it to the cloud.&nbsp;&nbsp;<h1>Key Considerations</h1>There are a couple of key questions to consider when choosing between the edge and cloud computing.&nbsp;<h2>What is the quality of your device&rsquo;s network?</h2>Performing computations on the cloud can work well when you have high-bandwidth, low-latency, and stable connection to the internet, as you will be needing to send your data back and forth between cloud servers and your device. If your device is intended to be used, for instance, in a home or office with a good internet connection, this back and forth can be done relatively seamlessly.&nbsp;In most cases, if the computation is done on the edge, it won&rsquo;t be affected by poor or lost internet connection in a remote location; the processing can continue since it is not computed in the cloud. You wouldn&rsquo;t want your vehicle&rsquo;s object detection to stop working while on a long road trip; that&rsquo;s one of the reasons why autonomous vehicles frequently perform computations like object detection on the edge.&nbsp;&nbsp;<h2>How quick and how often does your data need to be processed?</h2>Edge computing can be ideal in cases where your customer needs response times from your device to be faster than what can be achieved with a decent network connection, such as monitoring vital components of a system. The latency from travel-time between the device and the cloud can be reduced or eliminated completely. As a result, the data can be processed right away. If the data processing itself is quick, you could achieve real-time responses from your device.&nbsp;Cloud computing is beneficial when device use is intermittent. Smart homes devices are a good example of this again, where running computations in the cloud lets you share the same computing resources between multiple customers. This reduces costs by avoiding the need to provision your device with upgraded hardware to run the data processing.&nbsp;&nbsp;<img data-fr-image-pasted="true" src="https://www.mistywest.com/wp-content/uploads/2022/05/MistyWest_Edge_Vs_Cloud-05.png" alt="" width="100%" height="auto" srcset="https://www.mistywest.com/wp-content/uploads/2022/05/MistyWest_Edge_Vs_Cloud-05.png 1360w, https://www.mistywest.com/wp-content/uploads/2022/05/MistyWest_Edge_Vs_Cloud-05-300x155.png 300w, https://www.mistywest.com/wp-content/uploads/2022/05/MistyWest_Edge_Vs_Cloud-05-1024x528.png 1024w, https://www.mistywest.com/wp-content/uploads/2022/05/MistyWest_Edge_Vs_Cloud-05-768x396.png 768w" sizes="(max-width: 1360px) 100vw, 1360px" class="fr-fic fr-dii">&nbsp;<h2>What part of your data is important to you?</h2>Computing on the edge is useful if you only care about the result of your data after it has been processed. You can send only what is important to store in the long-term in the cloud, and doing so will allow you to reduce the cost of storing and processing data in the cloud. For example, if you are creating a traffic surveillance device that needs to report levels of congestion on a road, you could preprocess the videos on the edge&ndash;Instead of running hours of raw video in the cloud&ndash;and only send images or clips of the traffic when traffic is present.It&rsquo;s possible you need to keep the data to <a href="https://www.mistywest.com/posts/improving-dataset-creation-for-machine-learning/" target="_blank">build out your machine learning dataset</a>, or you plan on analyzing the raw data in other ways in the future. If you are already sending your raw data to the cloud, it may be ideal to perform calculations in the cloud as well.&nbsp;<h2>What are your device&rsquo;s power and size limitations?</h2>If you expect your device will be restricted in power and size, given that it has a good network connection, sending the computing work to be done on the cloud will allow your device to remain small and low-power. Google Home and Amazon Alexa, for example, will capture the audio and send it to the cloud for processing, allowing complex computations to be run on the audio that would not be possible to run on the small computers inside the devices themselves.&nbsp;<img data-fr-image-pasted="true" src="https://www.mistywest.com/wp-content/uploads/2022/05/MistyWest_Edge_Vs_Cloud-04.png" alt="" width="100%" height="auto" srcset="https://www.mistywest.com/wp-content/uploads/2022/05/MistyWest_Edge_Vs_Cloud-04.png 1360w, https://www.mistywest.com/wp-content/uploads/2022/05/MistyWest_Edge_Vs_Cloud-04-300x155.png 300w, https://www.mistywest.com/wp-content/uploads/2022/05/MistyWest_Edge_Vs_Cloud-04-1024x528.png 1024w, https://www.mistywest.com/wp-content/uploads/2022/05/MistyWest_Edge_Vs_Cloud-04-768x396.png 768w" sizes="(max-width: 1360px) 100vw, 1360px" class="fr-fic fr-dii">&nbsp;<h2>Is your data processing model your Intellectual Property?</h2>If you are making a consumer device and the method you are using to process data is part of your Intellectual Property (IP), you may need to consider how you plan to protect it. Putting your IP on your device without a robust security plan can leave it vulnerable to hacks. If you don&rsquo;t have the knowledge or resources to secure your IP on the edge, it may be best to leave it on the cloud, which already has security measures in place.&nbsp;<a data-rel="lightbox-image-0" data-rl_caption="" data-rl_title="" href="https://www.mistywest.com/wp-content/uploads/2022/05/MistyWest-Cloud-Vs-Edge-FlowChart-01.jpg" target="_blank" title=""><img data-fr-image-pasted="true" src="https://www.mistywest.com/wp-content/uploads/2022/05/MistyWest-Cloud-Vs-Edge-FlowChart-01.jpg" alt="" width="100%" height="auto" srcset="https://www.mistywest.com/wp-content/uploads/2022/05/MistyWest-Cloud-Vs-Edge-FlowChart-01.jpg 1801w, https://www.mistywest.com/wp-content/uploads/2022/05/MistyWest-Cloud-Vs-Edge-FlowChart-01-275x300.jpg 275w, https://www.mistywest.com/wp-content/uploads/2022/05/MistyWest-Cloud-Vs-Edge-FlowChart-01-938x1024.jpg 938w, https://www.mistywest.com/wp-content/uploads/2022/05/MistyWest-Cloud-Vs-Edge-FlowChart-01-768x839.jpg 768w, https://www.mistywest.com/wp-content/uploads/2022/05/MistyWest-Cloud-Vs-Edge-FlowChart-01-1406x1536.jpg 1406w" sizes="(max-width: 1801px) 100vw, 1801px" class="fr-fic fr-dii"></a>&nbsp;<h2>Some final considerations for choosing between edge and cloud computing</h2>There are quite a few things to consider when choosing between computing on the edge or in the cloud. In complex problems, you may benefit from using a combination of both, by leaving some parts of your processing on the edge and the rest on the cloud.We created the above flow chart to help guide you through which path you should take, but there may always be more to consider for your device, depending on how complex your use case is.&nbsp; This article was originally written by Amanda Luu, former Product Design Engineer Coop at MistyWest,&nbsp;<a href="https://www.mistywest.com/posts/edge-vs-cloud-computing-which-solution-is-better-for-your-connected-device/" rel="noopener noreferrer" target="_blank">for the MistyWest blog</a>.

Choosing between computing on the cloud or on the edge is a decision that can impact things like your device’s cost or efficiency. Here’s how you can ensure you make the right decision early on in your product development roadmap.

Edge Vs. Cloud Computing: Which Solution Is Better For Your Connected Device?

IoT Stars joined Embedded World 2023 to bring together top IoT professionals and discuss the latest products, services, and market trends. From innovative start-ups and scale-ups to established players, the event highlighted the need for collaboration in tackling the challenges of the rapidly growing IoT market.Ray Ozzie said back at IoT Stars 2022, &quot;We&#39;re all in this together.&quot; Whether competitors or partners, collaboration is essential in simplifying technology and educating customers. We&rsquo;re building on each other&rsquo;s tools and features. Eventually, it&#39;s all a Tetris game.Google, IBM and Ericsson leaving the IoT space led some to question the viability of IoT economics. Brandon Satrom from Blues Wireless countered &ldquo;the economics didn&rsquo;t work for them, but they do work for us&rdquo;. IoT simply takes longer because it&rsquo;s not general cloud compute. Marcin Nagy from AVSystem added that the hyperscalers aren&#39;t necessarily leaving the space, they want to keep as much data as possible. &ldquo;We synergize with them by building robust solutions&rdquo;.One of the most significant challenges still facing the market is the chip shortage, as Benjamin Cab&eacute; from the Zephyr Project pointed out. Building future-proof IoT means building firmware which can easily be ported to another hardware platform. &ldquo;You don&rsquo;t want to tie too much into a particular hardware system, because you never know if the chip you&rsquo;re using all of a sudden won&rsquo;t be available anymore&rdquo;.<img src="https://images.wevolver.com/eyJidWNrZXQiOiJ3ZXZvbHZlci1wcm9qZWN0LWltYWdlcyIsImtleSI6ImZyb2FsYS8xNjc5NDk2NzA5NTMxLU1CMjMwMzE0X2lvdHN0YXJzXzAzMjMtbG93cmVzLmpwZyIsImVkaXRzIjp7InJlc2l6ZSI6eyJ3aWR0aCI6OTUwLCJmaXQiOiJjb3ZlciJ9fX0=" style="width: 766px;" class="fr-fic fr-dib">Benjamin Cab&eacute;, ZephyrStephen Berard, CTO of Nubix cried: &ldquo;The cloud gets all the cool features like Kubernetes, CI/CD and automated testing. Embedded folks need this too!&rdquo;. With embedded tools like Zephyr and Nubix, we slowly start seeing these cool cloud features being incorporated into embedded development.<img src="https://s3-us-west-1.amazonaws.com/wevolver-project-images/froala%2F1679496747482-MB230314_iotstars_0450-lowres.jpg" style="width: 766px;" class="fr-fic fr-dib">Stephen Berard, NubixSimplification in IoT is both the key challenge, and the key catalyst for IoT adoption, according to Ken Yu, CEO of RAKwireless. Most use cases come from very specific industrial scenarios and the biggest problems here is to provide a plug and play solution, &ldquo;like turning a light switch&rdquo;. Brandon Satrom seconded Ken&rsquo;s comment by bringing tribute to unsexy use cases, &ldquo;those are the ones that can get deployed and bring value faster. Just report one thing like temperature or movement, get that live and see the value it brings.&rdquo;<img src="https://s3-us-west-1.amazonaws.com/wevolver-project-images/froala%2F1679496894670-MB230314_iotstars_0937-lowres.jpg" style="width: 766px;" class="fr-fic fr-dib">Brandon Satrom, Marcin Nagy, Ken Yu, Dora Terjek, Brandon Lewis (left to right)However, implementation remains tricky due to fragmentation, interoperability, and poor APIs, as noted by Marcin. To solve some of these key issues and drive scale, open-source solutions like Zephyr and open standard like LwM2M are necessary, Dora Terjek from Soracom highlighted. Fortunately, the collaborative, open-source movement in the hardware space is gaining momentum. The open-source RISC-V is becoming a real competitor to Arm for chip designs, and Zephyr is becoming the go-to operating system for embedded IoT devices, leveraged by companies like Nubix, Blues Wireless, and AVSystem.Let&rsquo;s celebrate hyperscalers leaving, the rise of open-source in IoT, and the deployment of unsexy use cases! Together, this might be the perfect cocktail for innovative IoT companies to start scaling up in 2023.

Let’s celebrate hyperscalers leaving, the rise of open-source in IoT, and the deployment of unsexy use cases! Together, this might be the perfect cocktail for innovative IoT companies to start scaling up in 2023.

IoT Stars brings tribute to unsexy use cases

During Mobile World Congress 2023 in Barcelona, IoT Stars gathered the biggest players in IoT to discuss how to bring the pieces together and start making an impact that will shape the industry. <a href="https://iotstars.com" target="_blank">IoT Stars</a>, a community of IoT professionals focused on sharing knowledge and experiences, welcomed entrepreneurs, developers, engineers, investors and press from the IoT ecosystem.For the eighth annual installment of the event, IoT Stars brought together some of the top experts to discuss how to make a lasting impact.<h2 dir="ltr">Bringing all the pieces together</h2>The Internet of Things (IoT) industry is a dynamic and rapidly evolving landscape, with fast-falling pieces that need to be positioned intelligently in order to realize IoT use cases that generate value.<img src="https://images.wevolver.com/eyJidWNrZXQiOiJ3ZXZvbHZlci1wcm9qZWN0LWltYWdlcyIsImtleSI6ImZyb2FsYS8xNjc3NzgyOTQzMDAyLV9BNzQ4NjYwLmpwZyIsImVkaXRzIjp7InJlc2l6ZSI6eyJ3aWR0aCI6OTUwLCJmaXQiOiJjb3ZlciJ9fX0=" style="width: 766px;" class="fr-fic fr-dib">IoT Stars presenting IoTetrisThe pioneering work has been done as all individual IoT components are available as of today. Cellular connectivity is provided by innovative MVNOs like Soracom and IoT Creators. Or connectivity can be realized by building private networks using existing hardware from companies like RAKwireless and MultiTech in combination with software tools from Monogoto. Easy-to-implement antennas from Ignion and off-the-shelf hardware with advanced sensing capabilities from Kallipr are within our reach.<h2 dir="ltr">Making an impact in the IoT playfield</h2>A panel of experts discussed how we&rsquo;re making an impact as an industry and many agree that our biggest impacts lie in our customers&rsquo; success. <a href="https://www.linkedin.com/in/gerhard-loots-298bb93/?originalSubdomain=au" target="_blank">Gerhard Loots</a> from <a href="https://kallipr.com/" rel="noopener noreferrer" target="_blank">Kallipr</a> shared a water utility business case that saved 250 million liters of water, and a rail track company that increased efficiency by $25 million in six months. &ldquo;That&#39;s not about our win. It&#39;s about their win&rdquo;, he concluded.<img src="https://images.wevolver.com/eyJidWNrZXQiOiJ3ZXZvbHZlci1wcm9qZWN0LWltYWdlcyIsImtleSI6ImZyb2FsYS8xNjc3NzgzNDEyNzM4LV9BNzQ4ODczLmpwZyIsImVkaXRzIjp7InJlc2l6ZSI6eyJ3aWR0aCI6OTUwLCJmaXQiOiJjb3ZlciJ9fX0=" style="width: 766px;" class="fr-fic fr-dib">Left to right: Jaap Groot, Daniel Quant, Gerhard Loots, Nicolas Lesconnec, Christine Perey<a href="https://www.linkedin.com/in/christineperey/?originalSubdomain=ch" target="_blank">Christine Perey</a>, who moderated the panel discussion, believes that sharing our successes is what will have the biggest impact on the future. &ldquo;We don&rsquo;t want competitors to know about the advantage that we&rsquo;ve gained through the adoption of new technology. We have great success stories, but can&#39;t talk about them. We can&#39;t give data and yet the data is the thing that&#39;s going to really convince the next customer. So how do we address that?&rdquo;<h2 dir="ltr">There is impact, but why don&rsquo;t we see it?</h2>The IoT impact and use cases are clearly there &mdash; and there are many more examples &mdash; but they&rsquo;re just not always being shared. <a href="https://www.linkedin.com/in/nicolaslesconnec/?originalSubdomain=fr" target="_blank">Nicolas Lesconnec</a> from <a href="https://www.soracom.io" target="_blank">Soracom</a> believes one reason for this is that proven use cases are often not specifically IoT success stories. &ldquo;IoT is just one of the Tetris blocks. I love to speak about the IoT success story, but from the customer&rsquo;s point of view, IoT is just one tiny part of it.&rdquo;IoT antenna supplier&nbsp;<a href="https://ignion.io" style="background-color: transparent;" target="_blank">Ignion</a>&nbsp;is providing solutions that simplify hardware development and pave the way for the creation of maximum value. &ldquo;How we measure impact is simply by success, by the value that we can bring, and by what customers tell us,&rdquo; says&nbsp;<a href="https://www.linkedin.com/in/grootjaap/" style="background-color: transparent;" target="_blank">Jaap Groot</a>. &ldquo;At the end of the day, 30 million+ antennas for products that are communicating today over networks... for us, that&#39;s impact. Nobody wants to talk about it: &lsquo;It&#39;s my big secret how I designed my device. It&#39;s so special that I don&#39;t want anybody to have this.&rsquo; We have to talk about success stories.&rdquo;<a href="https://www.linkedin.com/in/danquant/" target="_blank">Daniel Quant</a> from <a href="https://www.multitech.com" target="_blank">Multitech</a>, which fulfilled a hand sanitizing solution of 6 million units last year, warned that &ldquo;when somebody in a segment does something disruptive, it doesn&#39;t take long before other companies follow.&rdquo;<h2 dir="ltr">Use cases of the future</h2>In order to show the innovative use cases that are shaping the future, IoT Stars invited leading companies to showcase real products that are on the market today.<img src="https://images.wevolver.com/eyJidWNrZXQiOiJ3ZXZvbHZlci1wcm9qZWN0LWltYWdlcyIsImtleSI6ImZyb2FsYS8xNjc3NzgzNjk2OTcwLV9BNzQ4NzA1LmpwZyIsImVkaXRzIjp7InJlc2l6ZSI6eyJ3aWR0aCI6OTUwLCJmaXQiOiJjb3ZlciJ9fX0=" style="width: 766px;" class="fr-fic fr-dib">Laurens Slats (left) in conversation with Itamar Kunik from Monogoto<a href="https://monogoto.io" target="_blank">Monogoto</a> set out to simplify connectivity, recognizing that innovation lies in simplifying processes for developers. <a href="https://www.linkedin.com/in/itamar-kunik-0142031/?originalSubdomain=il" target="_blank">Itamar Kunik</a> called it &lsquo;the network for developers, by developers&rsquo; during a demonstration of a hybrid LTE solution the organization created using <a href="https://www.rakwireless.com/en-us" target="_blank">RAKwireless</a>&rsquo; 5G small cell. <a href="https://www.linkedin.com/in/jmarcelino/?originalSubdomain=uk" target="_blank">Jose Marcelino</a> from RAKwireless added that private network solutions empower developers and organizations with a firmer grip on their own infrastructure.&nbsp;<img src="https://images.wevolver.com/eyJidWNrZXQiOiJ3ZXZvbHZlci1wcm9qZWN0LWltYWdlcyIsImtleSI6ImZyb2FsYS8xNjc3NzgzNzM5NDAwLV9BNzQ4ODA4LmpwZyIsImVkaXRzIjp7InJlc2l6ZSI6eyJ3aWR0aCI6OTUwLCJmaXQiOiJjb3ZlciJ9fX0=" style="width: 766px;" class="fr-fic fr-dib">Jose Marcelino (left) from RAKwireless presenting the 5G small cell, with Marc Pous (right)&nbsp;Successful and practical use cases are already available, enabled by the building blocks created by IoT partners that are leading the way in innovation. So&nbsp;isn&rsquo;t it time we stopped keeping our successes to ourselves? Instead, what if we recognized a collective responsibility to share our impacts, in order to develop the IoT market?<h2 dir="ltr">Creating maximum value for a better future</h2><a href="https://www.linkedin.com/in/leonard-lee-nextcurve/" target="_blank">Leonard Lee</a> held a fireside chat with <a href="https://iotcreators.com" target="_blank">IoT Creators&#39;</a><a href="https://www.linkedin.com/in/afzalmangal/?originalSubdomain=nl" target="_blank">&nbsp;Afzal Mangal</a>, who called on the IoT industry to think beyond impact, and instead on how we can create maximum value. &ldquo;We usually think of IoT impact in terms of business impact,&rdquo; said Mangal. &ldquo;But is this the impact that we imagined 10 or 20 years ago? I think most of us signed up for IoT to discover maximum values beyond monetization, costs, and profits.&rdquo;&nbsp;<img src="https://images.wevolver.com/eyJidWNrZXQiOiJ3ZXZvbHZlci1wcm9qZWN0LWltYWdlcyIsImtleSI6ImZyb2FsYS8xNjc3NzgzMTc5MDc0LV9BNzQ4NzcwLmpwZyIsImVkaXRzIjp7InJlc2l6ZSI6eyJ3aWR0aCI6OTUwLCJmaXQiOiJjb3ZlciJ9fX0=" style="width: 766px;" class="fr-fic fr-dib">Leonard Lee (left) in conversation with Afzal MangalOne example took place just a few weeks ago: Two main highways between the Hague and Utrecht were closed because of a bridge on the verge of collapse. Thanks to a network of sensors placed on the bridge&rsquo;s structure connected to IoT Creators&rsquo; cellular network, authorities could monitor cracks forming over time and identify whether structural faults justified its closure. With large amounts of pedestrian foot traffic, trains, and vehicles crossing the bridge each day, Mangal believes &ldquo;this is maximum value creation with IoT because it goes beyond saving costs.&rdquo;There is undoubtedly great potential for maximum value still to be discovered. &quot;There are a lot of things possible today, because there is a lot that we can do with IoT,&quot; concludes Mangal. Could it be that the biggest impact IoT can have, lies beyond economic factors?<h2 dir="ltr">Realizing value beyond economic factors</h2>We believe the biggest impact comes from close collaboration within the industry. The right partners can provide you with the all essential building blocks to realize solutions that solve real problems. IoT Stars created in 2023 the IoTetris as an example of how IoT companies can work together to assemble the essential blocks to build the IoT stack that leads to maximum value.<a href="https://iotbuilder.iotstars.com" target="_blank">IoTetris</a> is already available to play, so go ahead and give it a go. And to continue the discussion, be sure not to miss our&nbsp;<a href="https://iotstars.com" target="_blank">next IoT Stars networking event</a>. We&rsquo;re rounding up the key players from the world of embedded development on March 14th during Embedded World in Nuremberg, and it&rsquo;s sure to be just as insightful.

During Mobile World Congress 2023 in Barcelona, the biggest players in IoT gathered

IoT takeaways from MWC 2023

This article is part of the series on Getting Started with LwM2M. If you&rsquo;re new to LwM2M, make sure to start by reading the previous article: <a href="https://www.wevolver.com/article/whats-this-lwm2m-standard-and-why-should-you-care" target="_blank">What&#39;s this LwM2M standard, and why should you care?</a>&nbsp;In today&rsquo;s market, IoT devices come in different forms and shapes. Each device has a unique communication stack which comes with a particular data packet structure and a thoughtfully designed device management layer.Now imagine the struggles of an energy distributor who deployed various brands of smart meters over the years to measure the use of electricity in thousands of buildings. Each meter comes with a different data structure which needs to be formatted. Specific processes need to be implemented to remotely manage the different devices, and particular mechanisms need to be put in place to update the devices&rsquo; firmware.When devices use the LwM2M standard, IoT integrators can be assured that each device reports its data in a similar fashion. The configurations and firmware can be updated using similar server commands independent of the hardware, firmware version or sensors used. In this article, we will explore the three fundamental components of LwM2M that ensure interoperability: the standardized data format, LwM2M architecture and LwM2M operations.<h2 dir="ltr" id="isPasted">LwM2M Data Format - Objects &amp; Resources</h2>LwM2M is an application-layer communication protocol developed by the <a href="https://omaspecworks.org/" target="_blank">OMA SpecWorks</a> to simplify messaging and device management of IoT devices. The protocol defines communication semantics, device management mechanisms, over-the-air updates (FOTA), telemetry data collection and is well suited for resource-constrained devices.To realize data interoperability, the LwM2M standard uses so-called IPSO Smart Objects to represent configurations, functionalities and sensors. When structuring the IPSO Smart Objects in a specific way, a language is created which both the LwM2M Client and the LwM2M Server can comprehend. This language contains Objects, Object Instances and Resources.<img src="https://images.wevolver.com/eyJidWNrZXQiOiJ3ZXZvbHZlci1wcm9qZWN0LWltYWdlcyIsImtleSI6ImZyb2FsYS8xNjc3NzU0OTgzNTM5LVNtYXJ0IE9iamVjdHMyLmpwZWciLCJlZGl0cyI6eyJyZXNpemUiOnsid2lkdGgiOjk1MCwiZml0IjoiY292ZXIifX19" style="width: 766px;" class="fr-fic fr-dib"><h3>Objects</h3>Objects represent a physical component (a sensor, LED or actuator) or a logical entity (such as the preferred radio technology, Firmware Update configuration or security implementation). All objects are defined by an object ID.The LwM2M standard defines three&nbsp;mandatory objects&nbsp;which are considered as essential for the device to connect and communicate with a LwM2M server. The mandatory objects are:<ul><li dir="ltr">Object ID 0: LwM2M Security - contains confidential information about connections to the LwM2M Servers configured on the Client.</li><li dir="ltr">Object ID 1: LwM2M Server - contains non-confidential information about connections to the LwM2M Servers configured on the Client.</li><li dir="ltr">Object ID 3: Device - contains basic information about the device, such as e.g. serial number and firmware version.</li></ul>Next to the mandatory objects, additional objects can be implemented. Examples of non-mandatory objects are:<ul><li dir="ltr">Object ID 4: Connectivity Monitoring - provides information about the device&#39;s connectivity status, such as the signal strength and network type.</li><li dir="ltr">Object ID 5: Firmware Update - allows for remote firmware updates to be performed.</li><li dir="ltr">Object ID 6: Location - provides information about the device&#39;s geographic location, such as its latitude and longitude.</li><li dir="ltr">Object ID 11: LwM2M APN Connection Profile - enables a device to connect to an APN.</li><li dir="ltr">Object ID 3303: Temperature - reports temperature measurements.</li><li dir="ltr">Object ID 3313: Accelerometer&nbsp;- represents a 1-3 axis accelerometer.</li></ul><h3 dir="ltr">Object Instances</h3>Some objects may have multiple&nbsp;object instances. Examples of such objects include the object that manages connections to LwM2M Servers as multiple LwM2M Servers can be configured, or devices which contain multiple SIMs with different APN profiles. When devices contain multiple identical sensors, object instances are used to distinguish the different sensors.<h3 dir="ltr">Resources</h3>Resources represent specific properties of an object which can be read, written or executed. Each Resource is identified by a unique resource ID.For example, a temperature object has the resources current temperature, unit of measurement, and a timestamp. Similarly, the device object has resources like manufacturer name, serial number and firmware version. The location object has the resources latitude, longitude and altitude. All details of the objects and resources are defined in the <a href="https://technical.openmobilealliance.org/OMNA/LwM2M/LwM2MRegistry.html" target="_blank">OMA Registry</a>. There is also a possibility to define your own custom object if a specific feature or sensor is not listed in the OMA Registry.Each resource contains an identifier, a type (float, string, int, opaque) and allowed Server Operations (such as READ, WRITE or EXECUTE).&nbsp;Examples The Device object is represented with the identifier&nbsp;3, the resources include:<div align="left" dir="ltr"><table><tbody><tr><td>Resource ID</td><td>Resource name</td><td>Server Operation</td><td>Type</td></tr><tr><td>0</td><td>Manufacturer</td><td>Read</td><td>String</td></tr><tr><td>1</td><td>Model Number</td><td>Read</td><td>String</td></tr><tr><td>3</td><td>Firmware Version</td><td>Read</td><td>String</td></tr><tr><td>4</td><td>Reboot</td><td>Executable</td><td>-</td></tr><tr><td>13</td><td>Current Time</td><td>Writeable</td><td>Time</td></tr></tbody></table></div>The Temperature object is represented with the identifier 3303, the resources include:<div align="left" dir="ltr"><table><tbody><tr><td>Resource ID</td><td>Resource name</td><td>Server Operation</td><td>Type</td></tr><tr><td>5700</td><td>Sensor Value</td><td>Read</td><td>Float</td></tr><tr><td>5701</td><td>Sensor Units</td><td>Read</td><td>String</td></tr><tr><td>5518</td><td>Timestamp</td><td>Read</td><td>Time</td></tr></tbody></table></div>When sending data, each message is formatted as:<pre dir="ltr">&lt;object id&gt;/&lt;object instance id&gt;/&lt;resource id&gt; = &lt;data&gt;</pre>To send the latest temperature data, it could look something like:&nbsp;<pre dir="ltr">3303/0/5700 = 23.5</pre>To optimize bandwidth, the data is usually encoded to a space-efficient binary payload. Although the data can be sent with plain JSON, more common is to use CBOR or SenML to reduce the payload size.<img src="https://images.wevolver.com/eyJidWNrZXQiOiJ3ZXZvbHZlci1wcm9qZWN0LWltYWdlcyIsImtleSI6ImZyb2FsYS8xNjc3NzUzMTI0NDk4LUxvY2F0aW9uX29iamVjdC5qcGVnIiwiZWRpdHMiOnsicmVzaXplIjp7IndpZHRoIjo5NTAsImZpdCI6ImNvdmVyIn19fQ==" style="width: 766px;" class="fr-fic fr-dib">Sending Latitude and Longitude using the LwM2M data formatResource InstancesIn some cases, resources have multiple resource instances. For example, the Device object 3 includes the&nbsp;resource 6: Available Power Sources. The different instances describe the different power sources: 0: DC power 1: Internal Battery 2: External Battery 3: Fuel Cell 4: Power over Ethernet 5: USB 6: AC (Mains) power 7: SolarTo describe the availability of solar power, the message format becomes:&nbsp;<pre dir="ltr">3/0/6/7</pre><h2 dir="ltr" id="isPasted">LwM2M Architecture - Client &amp; Server</h2>The LwM2M architecture comprises the LwM2M Client and the LwM2M Server. The communication works independent of the (wireless) technology used and can run on top of both IP and non-IP networks. Popular IP technologies include WiFi and cellular. Non-IP includes Bluetooth, LoRaWAN or Wirepas. Non-IP networks can be addressed using IEEE 802.15.4 (6LoWPAN) networks. For example by running Thread over Bluetooth.&nbsp;The LwM2M standard uses the CoAP protocol to let devices easily communicate with the cloud. Think of CoAP as the HTTP protocol, but designed for resource-constrained devices. CoAP is often compared to MQTT as both protocols are commonly used in IoT applications. A key difference is that CoAP uses UDP by default while MQTT relies on TCP. The small overhead of UDP is often seen as an advantage when compared to TCP. Although CoAP can be used with TCP, UDP is often the preferred choice when efficient use of network resources is a priority.<img src="https://images.wevolver.com/eyJidWNrZXQiOiJ3ZXZvbHZlci1wcm9qZWN0LWltYWdlcyIsImtleSI6ImZyb2FsYS8xNjc3NzUzMjc1ODExLUx3TTJNX0FyY2hpdGVjdHVyZS5qcGVnIiwiZWRpdHMiOnsicmVzaXplIjp7IndpZHRoIjo5NTAsImZpdCI6ImNvdmVyIn19fQ==" style="width: 768px;" class="fr-fic fr-dib">Popular LwM2M Clients include Anjay (maintained by AVSystem) and Zephyr LwM2M Client (maintained by the Zephyr Project). The LwM2M Server provided by AVSystem is called Coiote and can be used to process device data, configurations and firmware updates.The LwM2M Client runs on the end device and communicates with the LwM2M server using a secure connection. The Client informs the Server about the supported objects and resources during its first connection as well as during periodic status updates. Throughout the lifetime of the device, the client sends notification to the server with telemetry data formatted as defined by the LwM2M standard. All data is captured and stored by the LwM2M Server where it can be further processed or forwarded to specific databases of platforms such as Azure IoT Hub, or AWS IoT Core.<h2 dir="ltr" id="isPasted">LwM2M Operations</h2>The third core feature of the LwM2M protocol are operations that can be initiated from both the Client and the Server. Client Operations are used to inform the Server about telemetry data or the state of devices. Server Operations are used to read or write data, or to remotely update device configurations or firmware over time.Two interfaces can be distinguished: The Information Reporting Interface, and the Device Management Interface.<h3 dir="ltr" id="isPasted">Information Reporting Interface</h3><img src="https://images.wevolver.com/eyJidWNrZXQiOiJ3ZXZvbHZlci1wcm9qZWN0LWltYWdlcyIsImtleSI6ImZyb2FsYS8xNjc3NzUzMzc0NDk1LUluZm9ybWF0aW9uIFJlcG9ydGluZyBJbnRlcmZhY2UuanBlZyIsImVkaXRzIjp7InJlc2l6ZSI6eyJ3aWR0aCI6OTUwLCJmaXQiOiJjb3ZlciJ9fX0=" style="width: 766px;" class="fr-fic fr-dib">The LwM2M Client can inform the Server about telemetry data or state changes in real-time, initiated by the Client. The Server can also define the device&rsquo;s behavior by requesting the device to send data at periodic intervals, or when values exceed a certain threshold (e.g. when temperature rises above 30 degrees). This can be accomplished using the operations: OBSERVE, SEND and NOTIFY.Server operation<ul><li dir="ltr">OBSERVE: When the Observe operation is invoked, the Client will start sending NOTIFY messages to the Server about its data or state at configurable intervals. The OBSERVE operation can also be canceled when sending CANCEL OBSERVE.</li></ul>Client Operations<ul><li dir="ltr">SEND: The SEND message is used by the Client to send data to the server without explicit request. Depending on the application firmware it can be used to report new measurements, or inform the server about a change in telemetry data or state.</li><li dir="ltr">NOTIFY: In response to the OBSERVE operation initiated from the server, the client sends NOTIFY operations with data. Usually, the server provides a bandwidth in which the client needs to send its data. E.g. At least every 1 hour, but no more than once every 15 minutes.</li></ul><h3 dir="ltr" id="isPasted">Device Management Interface</h3><img src="https://images.wevolver.com/eyJidWNrZXQiOiJ3ZXZvbHZlci1wcm9qZWN0LWltYWdlcyIsImtleSI6ImZyb2FsYS8xNjc3NzUzNDEwMTYyLURldmljZSBNYW5hZ2VtZW50IEludGVyZmFjZS5qcGVnIiwiZWRpdHMiOnsicmVzaXplIjp7IndpZHRoIjo5NTAsImZpdCI6ImNvdmVyIn19fQ==" style="width: 766px;" class="fr-fic fr-dib">The Device Management Interface allows the server to remotely connect with the LwM2M Client and manage its configuration, software, security etc. The interface comes with a set of standardized operations, all of which are initiated by the Server.<ul><li dir="ltr">DISCOVER: Retrieving the list of objects and resources supported by the Client.</li><li dir="ltr">READ: Retrieving the current value of a specific resource or an object as a whole.</li><li dir="ltr">WRITE: Modifying the value of a specific resource, or an object as a whole.</li><li dir="ltr">EXECUTE: Invoking an action or operation on a resource. For example instructing a device to reset, reboot or upgrade its firmware.</li><li dir="ltr">CREATE: Creating a new object instance.</li><li dir="ltr">DELETE: Deleting an object instance.</li></ul><h2 dir="ltr">Conclusions</h2>The IoT landscape is fragmented and it&#39;s easy to get lost when digging through the wireless technologies, messaging protocols and data formats. LwM2M promises to simplify the connected world through standardization and interoperability.This is accomplished as a result of the IPSO Smart Objects with predefined identifiers, allowing for the creation of a digital representation of the device. The standardized data format allows any LwM2M Server to easily interpret the data. The client-server architecture specifies how data flows between the end devices and the web. Lastly, the pre-defined operations enable devices to send data and be reconfigured over time.To get started with LwM2M, follow AVSystem&rsquo;s <a href="https://iotdevzone.avsystem.com/docs/quickstart/" target="_blank">Quick Start</a> guide, and build your first end to end LwM2M application within 20 minutes.<h3 dir="ltr" id="isPasted"><a href="https://iotdevzone.avsystem.com/docs/quickstart/" target="_blank">Start Building &rarr;</a></h3>

Exploring the three fundamental aspects of IoT interoperability: a standardized data format, architecture and client-server operations.

IoT Device Management using LwM2M

The Internet of Things promised us a new era. One with ubiquitous connectivity which would lead to an abundance of data, allowing us to make smarter decisions. Today, reality caught up with this promise. Wireless networks are widely available, connectivity is relatively cheap to use and off the shelf hardware is available. However, instead of focusing on gaining meaningful insights from rich data, we&rsquo;re trapped in never-ending discussions about what wireless technology to use, which communication protocol to utilize or what security standard to implement.This often results in companies implementing different devices from different vendors, each using a different wireless technology and a different messaging protocol and data format. Once the data ends up in the cloud, it&rsquo;s unstructured and hard to make sense of. Device management is challenging as different devices adhere to different standards which need to be addressed differently, and the firmware update process needs to be redesigned for each device. Managing a heterogeneous fleet of devices and integrating the data into one platform is a heavy burden.<h2>Finding our way out of a labyrinth of wireless technologies&nbsp;</h2>As of 2017, a group of engineers promised an easy way out of this labyrinth of wireless technologies, messaging protocols and data formats. They proposed a simplified communication and device management standard to be implemented industry-wide. A new application-layer protocol for resource-constrained devices was launched called Lightweight Machine to Machine,&nbsp;or&nbsp;LwM2M. This novel protocol dictates a standardized way of formatting application data, and orchestrates device management and firmware updates mechanisms which should forever end the need for reinventing the wheel.<h2>LwM2M Basics</h2><a href="https://www.omaspecworks.org/what-is-oma-specworks/iot/lightweight-m2m-lwm2m/" target="_blank">LwM2M</a> is an application-layer communication protocol developed by the <a href="https://omaspecworks.org/" target="_blank">OMA SpecWorks</a> to simplify messaging and device management of IoT devices.&nbsp;LwM2M is the response to the ever-growing need for faster, easier and more power-efficient communication. The standardized message semantics enable embedded developers to focus on functionality rather than reinventing communication protocols. The efficient data format makes this standard well suited for resource-constrained IoT devices. Ideal for microcontrollers using expensive cellular connections or running low-power applications which require devices to optimize their communication.LwM2M simplifies the design and development of IoT solutions by standardizing the data format, and taking care of the complicated processes of secure device authentication. The standard also defines the process of sensor data collection, connectivity monitoring and firmware updates.<h2>LwM2M Architecture</h2><img src="https://s3-us-west-1.amazonaws.com/wevolver-project-images/froala%2F1671637030635-architecture.jpeg" style="width: 300px;" class="fr-fic fr-dib">The architecture of a LwM2M application comprises 3 components: the LwM2M client running on the end device, the LwM2M Server which manages the device, its data and firmware in the cloud, and the LwM2M Bootstrap Server a cloud service to authenticate and provision the LwM2M Client.<h3>LwM2M Client</h3>The LwM2M Client communicates with the LwM2M Server and LwM2M Bootstrap Server. It ensures a secure (i.e., authenticated and encrypted) connection with these servers, and sends the data in the right format as dictated by the standard.Several open source clients are available and ready to be used with your hardware platform of choice. Including:<ul><li>Anjay - maintained by AVSystem <a href="https://avsystem.github.io/Anjay-doc/index.html" target="_blank">link</a></li><li>Zephyr LwM2M Client - maintained by the Zephyr Project <a href="https://docs.zephyrproject.org/latest/connectivity/networking/api/lwm2m.html" target="_blank">link</a></li><li>Leshan&nbsp; - maintained by the Eclipse Foundation &nbsp;<a href="https://www.eclipse.org/leshan/" target="_blank">link</a></li><li>Wakaama - maintained by the Eclipse Foundation <a href="https://www.eclipse.org/wakaama/" target="_blank">link</a></li><li>IOWA - maintained by Ioterop (limited open source version available) <a href="https://github.com/IOTEROP/IOWA" target="_blank">link</a></li></ul><h3>LwM2M Server</h3>The LwM2M Server maintains the connections with the clients. Servers often come with a user interface allowing the end user to monitor the devices, send commands to one or multiple devices and schedule firmware updates.&nbsp;Specific operations, comparable to RESTful API methods, are defined in the LwM2M protocol. Operations for device management sent by the LwM2M server are DISCOVER, CREATE, READ, WRITE, DELETE and EXECUTE. The operations to manage sensor data collection are OBSERVE and CANCEL OBSERVATION. On the client side, a device sends data using the NOTIFY and SEND operations.From the LwM2M Server, data integrations to third party platforms can be initiated. As the data integrations are not defined in the LwM2M protocol, LwM2M Servers implement their data connector differently. Depending on the server implementation, APIs can be used to retrieve sensor data. Some pre-built connectors may be used to send data to popular IoT platforms such as AWS IoT Core or Azure IoT Hub.&nbsp;Popular LwM2M Servers are:<ul><li>Coiote IoT Device Management - from AVSystem <a href="https://www.avsystem.com/products/coiote-iot-device-management-platform/" target="_blank">link</a></li><li>ALASKA - from Ioterop <a href="https://ioterop.com/alaska" target="_blank">link</a></li><li>Leshan LwM2M server - from Eclipse (Open Source, evaluation purposes only) <a href="https://github.com/eclipse/leshan#test-server-sandbox" target="_blank">link</a></li><li>Thingsboard - <a href="https://thingsboard.io/" target="_blank">link</a></li><li>Cumulocity -&nbsp;from SoftwareAG <a href="https://www.softwareag.com/en_corporate/platform/iot/iot-analytics-platform.html" target="_blank">link</a></li></ul><h3>LwM2M Bootstrap Server</h3>The LwM2M Bootstrap Server is a cloud service which the LwM2M Client can connect to and often runs on the same instance as the LwM2M Server. LwM2M Clients usually connect to the Bootstrap Server during their very first boot-up, after an explicit command from the LwM2M Server, or after a factory reset.&nbsp;The Bootstrap Server provides the clients with credentials which are required to connect to a LwM2M Server. In addition, it initializes the data model and writes new configurations to the device. Once the device is provisioned using commands sent from the Bootstrap server, the Client is ready to start communicating with the LwM2M Server.The usage of the Bootstrap Server is optional, all necessary information can also be hard coded into the device. This is, however, not recommended as it poses complications with updating configurations over time such as renewing security credentials or migrating to different LwM2M servers.<h2>LwM2M Data model</h2>Each data message is sent using a data format which identifies standardized objects, object instances and resources, all defined by the OMA in the <a href="https://technical.openmobilealliance.org/OMNA/LwM2M/LwM2MRegistry.html" target="_blank">LwM2M Registry</a>.&nbsp;<img src="https://s3-us-west-1.amazonaws.com/wevolver-project-images/froala%2F1671637072647-objects-and-resources.jpeg" style="width: 300px;" class="fr-fic fr-dib">Devices contain different building blocks, each of these blocks is represented by an Object and identified by an Object ID. An object describes a particular part of a device. For example, the Firmware Update Objects is used to invoke and track status of the firmware update process. Objects can also describe the connectivity technology (e.g. cellular or WiFi), device information (serial number, manufacturer, firmware version), sensors (temperature, air quality) or peripherals (GPS, LEDs, buzzers). All Objects combined can be used to construct a digital twin; a virtual representation of the end device.Each Object may have different Instances, e.g. multiple temperature sensors embedded on one device.&nbsp;Object Instances have one or multiple Resources. For example, the &ldquo;device&rdquo; object has multiple resources such as the manufacturer, serial number and firmware version. The &ldquo;location&rdquo; object has the resources latitude and longitude.The data sent using the LwM2M standard follows the format<pre>&lt;object id&gt;/&lt;object instance id&gt;/&lt;resource id&gt; = &lt;data&gt;</pre>As an example, to send the temperature data of 23.5, the data format becomes:&nbsp;<pre>3303/0/5700 = 23.5</pre>The figure below illustrates this concept of object and resources.<img src="https://s3-us-west-1.amazonaws.com/wevolver-project-images/froala%2F1671637090093-data+format.jpeg" style="width: 300px;" class="fr-fic fr-dib"><h2>Getting started with LwM2M</h2>This article gave you an understanding of the need for the LwM2M standard. In my opinion, the best way to really learn how LwM2M works is through hands-on experience.AVSystem developed an open-source LwM2M Client called Anjay which can be used with different hardware platforms such as Nordic, Espressif, Raspberry Pi and STM32. AVSystem&rsquo;s LwM2M server called&nbsp;<a href="https://eu.iot.avsystem.cloud/" target="_blank">Coiote</a> has a free developers tier, allowing you to connect up to ten devices.Select your preferred hardware platform and get started with LwM2M:<ul><li>Nordic - <a href="https://iotdevzone.avsystem.com/docs/Anjay_integrations/Anjay_Nordic_integration/Thingy91/" target="_blank">link</a></li><li>Espressif - <a href="https://iotdevzone.avsystem.com/docs/Anjay_integrations/ESP32_integration/M5Stick/" target="_blank">link</a></li><li>STM32 - <a href="https://iotdevzone.avsystem.com/docs/Anjay_integrations/Anjay_ST_integration/Building_LwM2M_applications/Building_LwM2M_applications_with_I-CUBE-Anjay/" target="_blank">link</a></li><li>Raspberry Pi - <a href="https://iotdevzone.avsystem.com/docs/Anjay_integrations/Implementing_LwM2M_objects_on_RaspberryPi/" target="_blank">link</a></li></ul>Need help? Join the <a href="https://discord.avsystem.com/" target="_blank">AVSystem Discord</a>.<img src="https://s3-us-west-1.amazonaws.com/wevolver-project-images/froala%2F1671637107758-AVSystem+Discord.png" style="width: 300px;" class="fr-fic fr-dib">

Lightweight Machine to Machine, a simplified communication and device management standard for the Internet of Things.

What's this LwM2M standard, and why should you care?

PCB Arts 

Akenza-Fri Jun 30 2023 23:59:59 GMT+0200 (Central European Summer Time)

Akenza-Mon Jul 31 2023 23:59:59 GMT+0300 (Eastern European Summer Time)

Akenza-Thu Aug 31 2023 23:59:59 GMT+0300 (Eastern European Summer Time)

Akenza-Thu Aug 31 2023 23:59:59 GMT+0200 (Central European Summer Time)

Akenza-Sat Sep 30 2023 23:59:59 GMT+0300 (Eastern European Summer Time)

Mergeflow-Fri Jun 30 2023 23:59:59 GMT+0200 (Central European Summer Time)

Mergeflow-Mon Jul 31 2023 23:59:59 GMT+0300 (Eastern European Summer Time)

Mergeflow-Thu Aug 31 2023 23:59:59 GMT+0300 (Eastern European Summer Time)

Mergeflow-Thu Aug 31 2023 23:59:59 GMT+0200 (Central European Summer Time)

Mergeflow-Sat Sep 30 2023 23:59:59 GMT+0300 (Eastern European Summer Time)

Intemic-Fri Jun 30 2023 23:59:59 GMT+0200 (Central European Summer Time)

Intemic-Mon Jul 31 2023 23:59:59 GMT+0300 (Eastern European Summer Time)

Intemic-Thu Aug 31 2023 23:59:59 GMT+0300 (Eastern European Summer Time)

Intemic-Thu Aug 31 2023 23:59:59 GMT+0200 (Central European Summer Time)

Intemic-Sat Sep 30 2023 23:59:59 GMT+0300 (Eastern European Summer Time)

Canonical-Fri Jun 30 2023 23:59:59 GMT+0200 (Central European Summer Time)

Canonical-Mon Jul 31 2023 23:59:59 GMT+0300 (Eastern European Summer Time)

Canonical-Thu Aug 31 2023 23:59:59 GMT+0300 (Eastern European Summer Time)

Canonical-Thu Aug 31 2023 23:59:59 GMT+0200 (Central European Summer Time)

Canonical-Sat Sep 30 2023 23:59:59 GMT+0300 (Eastern European Summer Time)

embedded software

embedded software

Profiles