This article presents a comprehensive technical analysis of CPLD vs FPGA, focusing on their architectural distinctions, performance metrics, design flows, and implementation methodologies. 

CPLD vs FPGA: A Comprehensive Technical Analysis and Implementation Guide

<p id="isPasted">In wireless product development, what happens behind the scenes when selecting, designing, and integrating an antenna into a product? What is the journey of a PCB antenna, or more specifically, a PCB-mounted embedded antenna?</p><p>Designing and integrating an antenna may seem complex, but it can be broken down into manageable steps. It begins with defining the product’s requirements, which drive the antenna specifications. Next, you select and design the appropriate antenna type. Then, verify the design’s performance through simulations. After building and testing prototypes, manufacturing begins.</p><p>Each step demands careful attention. This article offers a clear, step-by-step guide to the PCB antenna design process, explaining each stage and providing practical tips to help you successfully integrate the antenna into your wireless product.</p><h2><p>Step 1: Requirements Engineering</p></h2><p>The first step in designing an antenna for wireless solutions is to identify and document the product's essential needs. This process is known as Requirements Engineering (RE).</p><p>RE focuses on understanding what needs to be designed, not how. To effectively define these requirements, the design team must conduct a thorough problem analysis. This analysis should address several critical questions to ascertain the operational and environmental conditions under which the product will function. These include:</p><ul><li>What types of communication protocols will the product support (e.g., WiFi, cellular, GPS)?</li><li>What operating frequency and bandwidth are required?</li><li>What are the coverage, gain, and cost expectations?</li><li>Will the product operate indoors, outdoors, or under specific environmental conditions such as extreme temperatures or high humidity?</li><li>How much space can be afforded for antenna clearance on the PCB?</li></ul><p>The product requirements define the necessary antenna specifications, which help narrow down the possible antenna technologies. When proposing antenna solutions, include a detailed assessment of the costs, timeline, and, if possible, recommendations for placement and orientation.</p><p>Document all antenna and product requirements in a way that supports testing. These requirements should guide the design process and serve as a checklist during testing to ensure the design meets the goals.</p><h2><p>Step 2: Design</p></h2><p>The design stage starts by using the requirements defined in the RE phase to select the appropriate antenna type and design its specifications to meet those needs.</p><p>The antenna must be designed to cover the required frequency and range. Its key parameters, such as efficiency and directionality, should align with the product’s needs. The polarization of both the transmitting and receiving antennas must match to maximize power transfer. The Voltage Standing Wave Ratio (VSWR) should be kept low and the return loss high.</p><p>When attaching the antenna to a PCB, match the antenna's impedance with the characteristic impedance of the feed line connecting it to either a transmitter or a receiver. Impedance mismatches can significantly reduce the RF link budget, lowering the antenna's efficiency and range.</p><p>Remember, an antenna is a sensitive component. Its performance is greatly influenced by its placement, surrounding components, the ground plane, and the enclosure or casing.</p><p><img src="https://s3-us-west-1.amazonaws.com/wevolver-project-images/froala%2F1729781924459-1729781924459.png" alt="Antenna design pcb" class="fr-fic fr-dii"></p><ul><li><strong>Antenna Placement:</strong> Antenna placement on the PCB is crucial. Datasheets often suggest various placement options. Choose one of these for optimal performance. If you need to place the antenna elsewhere, you'll need to validate the placement through careful study.</li><li><strong>Nearby Components:</strong> Placing metallic objects near an antenna can interfere with its performance. Metallic objects reflect the antenna’s radiated energy back to the antenna. It’s best to avoid this, or at least account for these components when designing and tuning the antenna. Finalize the device's mechanical layout before tuning the antenna, as changes later on can disrupt the tuning.</li><li><strong>PCB Size and Ground Clearance:</strong> PCB size is crucial for ground-dependent antennas because the board itself affects radiation. Follow ground clearance guidelines in datasheets to avoid poor performance due to detuning or reduced efficiency.</li><li><strong>Casing:</strong> Materials used to manufacture the device casing can be critical for the antenna’s performance. This is even more critical for small antennas. If the dielectric material properties are available, it is better to choose a material with low permittivity and dielectric losses that are as low as possible. Such materials are virtually transparent to the antenna’s radiation.</li></ul><p>For more in-depth guidance during antenna design, refer to our previous white paper covering six essential considerations for PCB antenna design.</p><h2><p>Step 3: RF Simulation</p></h2><p>To shorten and ensure reliable design cycles for PCB antennas, designs must be verified early on. However, predicting antenna performance is challenging, as solving Maxwell's equations requires complex algorithms. Electromagnetic simulators can help predict performance and optimize antenna designs before building physical prototypes.</p><p>Remember, however, these simulators are tools, not replacements for the designer. They can predict performance based on the designs you give them, but you need a strong understanding of antenna theory to use them effectively.</p><p>Choosing the right simulation method is also important. Popular algorithms include the Method of Moments (MoM), Finite Element Method (FEM), Finite Difference Time Domain (FDTD), and Finite Integration Technique (FIT). Each has specific advantages and disadvantages summarized below:</p><table><thead><tr><th scope="col">Algorithm</th><th scope="col">Advantages</th><th scope="col">Limitations</th></tr></thead><tbody><tr><td>Method of Moments (MoM)</td><td><ul><li>Analyzes complex layered structures effectively.</li><li>Simulates designs with many ports efficiently, without added time costs.</li></ul></td><td><ul><li>Cannot handle general 3D structures, limiting its use to planar designs.</li><li>Demands large storage and increases computational time sharply with the size of the problem.</li></ul></td></tr><tr><td>Finite Element (FEM)</td><td><ul><li>Can analyze arbitrarily shaped 3D structures, not just layered setups, offering an advantage over the MoM.</li><li>Simulates designs with many ports efficiently, without added time costs.</li><li>Offers the most flexible approach to electromagnetic analysis.</li><li>Serves as a frequency-domain solver, making it ideal for problems with periodicity like phased arrays, frequency-selective structures, and band gap antennas.</li></ul></td><td><ul><li>Simulating complex antenna structures uses significant computer memory.</li></ul></td></tr><tr><td>Finite Difference Time Domain (FDTD)</td><td><ul><li>Can analyze 3D structures.</li><li>It has an advantage over the FEM method in that it does not require solving a matrix, allowing it to handle very large problems with relatively small amounts of computer memory.</li><li>Highly suitable for parallel processing, which allows it to leverage modern GPUs to speed up simulations.</li><li>Simpler to implement a basic FDTD solver compared to MoM and FEM solvers.</li><li>Well-suited for simulating specific antenna types such as horn antennas, waveguides, microstrip antennas, conformal antennas, and small planar array antennas.</li></ul></td><td><ul><li>Requires running a separate simulation for each port in a design, making it less efficient for designs with many ports.</li><li>Not optimal for certain types of antennas like wire antennas, aperture antennas, reflector antennas, and those with electrically large structures.</li></ul></td></tr><tr><td>Finite integration technique (FIT)</td><td><ul><li>Effective for analyzing antennas with inhomogeneous materials, wideband issues, and ultra-wideband (UWB) applications.</li><li>FIT underpins many commercial simulation tools because it can handle the entire spectrum of electromagnetics, from DC to high-frequency and optical applications.</li></ul></td><td><ul><li>Not well-suited for simulating wire antennas.</li></ul></td></tr></tbody></table><h3><p>Choosing the Right Simulation Method</p></h3><p>Selecting the appropriate EM(Electromagnetic) analysis tool depends on factors like ease of model creation, integration with circuit simulation tools, and the level of EM expertise required.</p><p>For planar structures, such as PCB interconnects and planar antennas, the MoM is most effective. For fully 3D structures like connectors, packages, and 3D antennas, FEM or FDTD methods are better.</p><p>For complex 3D structures with many ports, FEM is ideal. FDTD is best for large structures with fewer ports, such as antenna placements on vehicles or aircraft, and for evaluating antenna performance near detailed human body models.</p><h2><p>Step 4: Performance Tests and Optimization</p></h2><p>After optimizing the designs, prototypes are built for testing. These tests take place in an anechoic chamber designed to absorb all sound waves. Key measurements include return loss, efficiency, maximum gain, Effective Isotropic Radiated Power (EIRP), Total Radiated Power (TRP), Total Isotropic Sensitivity (TIS), and impedance. These measurements fully assess the antenna’s performance. If the results show any issues, the design can be adjusted.</p><p>Testing falls into two categories: passive and active.</p><p>Passive antenna testing evaluates the antenna alone, without the influence of the rest of the device. The antenna port is isolated from the RF front end and connected to a vector network analyzer (VNA) set to specific frequencies and amplitudes. The VNA measures data, which is then used to calculate efficiency, gain, directivity, and EIRP.</p><p>Active antenna testing measures the antenna as part of the entire system, including the RF front end. The main measurements are TRP and TIS.</p><p>TRP measures the total power radiated by the antenna when it is transmitting across a frequency band. TIS measures the average sensitivity of the receiving antenna system over a frequency band. TRP and TIS are key indicators of signal quality, offering a reliable gauge of the system’s overall performance.</p><h3><p>Practical Considerations During Testing</p></h3><p><img src="https://s3-us-west-1.amazonaws.com/wevolver-project-images/froala%2F1729781924565-1729781924565.png" alt="Antenna design considerations" class="fr-fic fr-dii"></p><ul><li>If possible, make the prototype from the same materials, using the same manufacturing processes and facilities as the final product. This ensures accurate testing.</li><li>Start critical measurements and prototype testing early. This speeds up the successful production of the final product.</li><li>Antennas must be customized to their specific environment and materials, as they may not perform as expected based on datasheets. Testing should simulate real-world conditions by considering the intended use cases and positions, such as near the head, body, or on different surfaces.</li><li>When planning tests, verify if the product must meet specific regulatory requirements. For instance, for cellular IoT devices, TRP and TIS are crucial, as failing to meet carrier standards for these parameters can prevent product approval.</li><li>Follow regulatory standards for electromagnetic emissions and safety. Include electromagnetic emissions compliance checks throughout the design and testing process. This ensures legal compliance and prevents interference with other devices.</li></ul><p>Antenna testing offers significant advantages. It allows designers to optimize the antenna’s design for better radio performance, considering factors like product size, typical use, manufacturing costs, and RF exposure. Even a small improvement, such as a 6 dB gain, can double coverage, while a 3 dB gain can cut the power amplifier’s energy consumption in half, leading to longer battery life. As a result, users experience extended range and enhanced signal quality.</p><h2><p>Step 5: Manufacturing</p></h2><p>The final step is manufacturing. Success in this stage depends on a stable design, a reliable production process, and effective end-of-line testing. Even the best designs are useless if they can't be produced efficiently and with high quality. To achieve this, consider the following when selecting and collaborating with a manufacturer:</p><ul><li><strong>Early Collaboration:</strong> Involve the manufacturer early in the design process. Close collaboration between design, manufacturing, and materials teams throughout all stages—from design and prototyping to production—ensures a smooth transition to manufacturing, saving time and costs.</li><li><strong>Quality Assurance:</strong> Choose a manufacturer that offers robust quality assurance, including innovative end-of-line testing methods. Ensure any overseas facilities comply with local legal standards and maintain strict confidentiality agreements.</li><li><strong>Timely Production:</strong> The manufacturer must meet your deadlines and pricing requirements without compromising on quality. Confirm their average lead time to ensure they will deliver on schedule and within budget.</li><li><strong>Fabrication Tolerances:</strong> Select a fabrication process capable of meeting tight tolerances. Use prototyping and testing to evaluate the manufacturer’s capabilities and limitations and how their fabrication tolerances impact antenna performance.</li></ul><h2><p>Conclusion</p></h2><p><img src="https://s3-us-west-1.amazonaws.com/wevolver-project-images/froala%2F1729781924596-1729781924595.png" alt="Red pcb design" class="fr-fic fr-dii"></p><p>Designing and manufacturing antennas is a complex process. It is crucial for the success of any wireless product. Choosing the right partner is essential. The design team must work closely together. They need to understand the product’s requirements. They must also follow regulations on frequency, power output, EMC, and testing.</p><p>The design should consider factors like PCB size, ground plane, casing, placement, and nearby components. Early design verification is important. Rapid prototyping helps speed up production. Ultimately, even the best designs are only valuable if they can be manufactured efficiently.</p><h4 id="isPasted">Ready to start your PCB Antenna prototype?</h4><p><a href="https://factory.macrofab.com/quick_quote?utm_term=electronic%20manufacturing%20systems&utm_campaign=ampliva-search-electronics-manufacturing&utm_source=bing&utm_medium=cpc" target="_blank">Get an Instant Quote Now</a></p><p><br></p>

Making your prototype from the same materials using the same manufacturing processes and facilities as your final product ensures accurate testing and better final results. This article also outlines other best practices. 

PCB Antenna Design: A Step-by-Step Guide

I3C is becoming a strong alternative to I2C. Learn about the main differences, benefits, and uses of these two important communication protocols. Explore the world of inter-integrated circuit communication and see which protocol is best for your next project.

I3C vs I2C: Unraveling the Battle of Communication Protocols

<p dir="ltr" id="isPasted">The logistics industry today is grappling with multi-layered challenges, prominently among which is sustainability. Traditional methods of transporting goods, particularly perishable goods, rely heavily on energy-intensive refrigerated cargo trailers, resulting in substantial carbon emissions and operational costs. Researchers at <a href="https://www.tecnalia.com/en/" target="_blank" rel="noopener noreferrer">Tecnalia</a> have found a solution that can disrupt this space and overcome this longstanding challenge: their approach is to harness the potential of low-power IoT technology.&nbsp;</p><p dir="ltr"><img src="https://images.wevolver.com/eyJidWNrZXQiOiJ3ZXZvbHZlci1wcm9qZWN0LWltYWdlcyIsImtleSI6ImZyb2FsYS8xNzIwNTQwODA0NTM1LTE3MjA1NDA4MDQ1MzUucG5nIiwiZWRpdHMiOnsicmVzaXplIjp7IndpZHRoIjo5NTAsImZpdCI6ImNvdmVyIn19fQ==" width="602" height="401" class="fr-fic fr-dii"></p><p dir="ltr">In 2023, Mouser Electronics, together with Nordic Semiconductor, Soracom, and Crowd Supply, initiated the <a href="https://www.wevolver.com/article/connect-for-good-low-power-wireless-sustainability-challenge" target="_blank" rel="noopener noreferrer">Connect for Good Challenge</a> with the goal of enabling innovation for social and environmental good. It rewarded the most innovative low-power wireless solutions aiming to encourage efforts towards achieving the UN’s Sustainable Development Goals (SDGs). The winner was a Tecnalia project led by researcher Iván Arakistain, a technologist and researcher at Tecnalia. It focused on sustainable logistics infrastructure, providing a solution for monitoring and detecting anomalies.&nbsp;</p><p dir="ltr">Central to Arakistain's project is the seamless integration of advanced technology and embedded artificial intelligence (AI) to optimize the transport of perishable goods. Through sophisticated monitoring systems and anomaly detection algorithms, the solution ensures the integrity of the cold chain without the reliance on energy-intensive refrigeration. The incorporation of the <a href="https://www.wevolver.com/specs/nrf9160-development-kit" target="_blank">nRF9160 Development Kit</a>, renowned for its low energy consumption and robust cellular connectivity, underpinned the project's success.&nbsp;</p><h2 dir="ltr">Leveraging Embedded AI</h2><p dir="ltr">The Tecnalia solution revolves around a smart device connected to the Internet with multiple sensors and GPS capabilities. These features allow for effective monitoring and tracking of items. The core aspect of this solution lies in integrating AI directly into the device. This embedded AI is miniaturized and used within numerous everyday objects. By leveraging this technology, the project enhances the handling of perishable goods, ensuring that their required temperature levels are maintained throughout transportation. Additionally, the device incorporates a specialized sensor akin to an "electronic nose," which monitors, detects, and alerts users to the presence of harmful organic compounds. This functionality serves to identify spoiled food items, offering an added layer of quality control and safety.&nbsp;</p><p dir="ltr">This innovative approach sets a new standard for eco-friendly transport in the logistics industry, with far-reaching implications for global sustainability initiatives. Moreover, the project's emphasis on operational efficiency underscores its potential to drive substantial economic benefits for many stakeholders.&nbsp;</p><p dir="ltr"><img src="https://images.wevolver.com/eyJidWNrZXQiOiJ3ZXZvbHZlci1wcm9qZWN0LWltYWdlcyIsImtleSI6ImZyb2FsYS8xNzIwNTQwODAzOTUxLTE3MjA1NDA4MDM5NTEucG5nIiwiZWRpdHMiOnsicmVzaXplIjp7IndpZHRoIjo5NTAsImZpdCI6ImNvdmVyIn19fQ==" width="602" height="344" class="fr-fic fr-dii"></p><h2 dir="ltr">Navigating Challenges, Embracing Opportunities</h2><p dir="ltr">As Arakistain's project gains traction, it is expected to encounter both challenges and opportunities on the path to widespread adoption. Convincing stakeholders to embrace innovative solutions necessitates not only technological prowess but also strategic communication and collaboration. However, initiatives like the <a href="https://www.wevolver.com/article/connect-for-good-low-power-wireless-sustainability-challenge" target="_blank">Connect for Good Challenge</a> and the <a href="https://www.wevolver.com/article/connect-for-good-challenge-winners-logistics-infrastructure-monitoring-smart-bees-and-bioreactors" target="_blank">innovative projects</a> it sparked serve as catalysts for change, fostering innovation and collaboration in pursuit of a more sustainable future.</p><p dir="ltr">In envisioning the future of sustainable logistics, the future course from Arakistain’s point of view team is clear: a shift to electrified transport is a necessity. The logistics industry will continue to evolve, driven by innovation and collaboration, and with a focus on sustainability and climate action, the prospects for a more sustainable future can grow ever brighter. Arakistain's project explicitly addresses the seventh, eleventh, and thirteenth SDGs, which target accessibility to affordable and clean energy, sustainable cities, and climate action. The success of this project stands as a testament to our ability to innovate and effect change, paving the way for more sustainable logistics and inspiring generations to come to embrace the challenge of building a better world.&nbsp;</p><p dir="ltr">The path ahead for this solution is laden with both promise and complexity. While the technology demonstrates its efficacy in controlled environments, scaling it across diverse regions and logistics networks presents unique challenges. The logistics industry's heterogeneity and reliance on established practices necessitate careful navigation of regulatory frameworks, infrastructure limitations, and stakeholder engagement.&nbsp;</p><p dir="ltr">The Tecnalia team must strike a delicate balance between technological innovation and practical implementation. This involves not only refining the technical aspects of the solution but also addressing logistical, regulatory, and cultural barriers to adoption. Collaborating with industry stakeholders, policymakers, and regulatory bodies becomes imperative to ensure the seamless integration of the technology into existing logistics networks.&nbsp;</p><p dir="ltr"><img src="https://images.wevolver.com/eyJidWNrZXQiOiJ3ZXZvbHZlci1wcm9qZWN0LWltYWdlcyIsImtleSI6ImZyb2FsYS8xNzIwNTQwODA0MjM0LTE3MjA1NDA4MDQyMzQucG5nIiwiZWRpdHMiOnsicmVzaXplIjp7IndpZHRoIjo5NTAsImZpdCI6ImNvdmVyIn19fQ==" width="602" height="401" class="fr-fic fr-dii"></p><h2 dir="ltr">Economic Implications, Market Dynamics, &amp; New Applications</h2><p dir="ltr">Beyond its environmental benefits, this project carries significant economic implications for logistics companies and end-users alike. By reducing reliance on energy-intensive refrigeration and optimizing transportation routes, the solution offers substantial cost savings for stakeholders. In fact, by minimizing the transport of refrigerated cargo, the proposed solution promises a 10% reduction in cost per kilometer and no less than 64 tons of annual carbon emissions per truck. Adopting such eco-friendly transport methods would also enhance brand reputation and market competitiveness, positioning companies as leaders in sustainability.&nbsp;</p><p dir="ltr">This Tecnalia project opens doors to new market opportunities and revenue streams. As consumers&nbsp;<iframe id="66cd9e99c4d4f52dca0829c9" width="250" class="specs-iframe" height="440" src="https://wevolver.com/iframe/specs/infineon-technologies-xensiv-pas-co2-5v-sensor2go-kit?iframe=true" frameborder="0" scrolling="no" data-lf-form-tracking-inspected-kn9eq4roawkarlvp="true" data-lf-yt-playback-inspected-kn9eq4roawkarlvp="true" data-lf-vimeo-playback-inspected-kn9eq4roawkarlvp="true"></iframe> increasingly prioritize sustainability, companies that embrace eco-friendly practices stand to gain a competitive edge. The technology not only addresses current market demands but also anticipates future trends, positioning stakeholders for long-term success in an evolving industry that is awaiting change.&nbsp;</p><p dir="ltr">Furthermore, while this project focuses on revolutionizing cold chain logistics, its impact extends far beyond this niche. The integration of low-power IoT solutions and advanced monitoring systems paves the way for innovation across diverse industries and applications. From structural health monitoring to audio detection and classification, the technology holds promise for addressing a wide range of societal challenges.&nbsp;</p><p dir="ltr">The Tecnalia team continues to explore novel applications of their technology, leveraging embedded AI and sensor networks to tackle pressing issues in fields such as infrastructure, healthcare, and environmental monitoring. By harnessing the power of data and connectivity, the project exemplifies the transformative potential of technology in addressing complex global challenges. It represents more than just a technological innovation; it embodies a vision for a more sustainable and interconnected world. By leveraging cutting-edge technology and collaborative partnerships, Arakistain and his team pave the way for a future where logistics is not only efficient but also environmentally responsible.&nbsp;</p><p dir="ltr">For more information, check out our <a href="https://www.wevolver.com/article/towards-sustainable-logistics-infrastructure-an-interview-with-tecnalia" target="_blank">interview with Iván Arakistain</a> about his winning project.</p>

The logistics industry today is grappling with multi-layered challenges, prominently among which is sustainability.

Pioneering Greener Logistics: The Impact of Low-Power IoT Solutions

A re-development of classical CAN by Bosch in 2012, CAN FD is the protocol that operates all moving machines like cars, planes, robots, etc. This guide will unfold its foundational concepts, recent advancements, practical applications, and the challenges, for engineering professionals.

CAN FD: Revolutionizing Automotive and Industrial Communications

The Serial Peripheral Interface (SPI) protocol is a critical communication standard for high-speed data exchange in embedded systems. This article discusses its architecture, advancements, and practical applications in engineering to help engineers optimize and overcome design challenges. 

SPI Protocol: Revolutionizing Data Communication in Embedded Systems

Explore the fundamentals of Gray Code, its practical uses in digital systems, and the latest advancements driving innovation in technology and data encoding.

Gray Code: A Comprehensive Guide for Engineering Professionals

Pose detection technology leverages advanced machine learning algorithms to interpret human movements in real-time, enabling seamless, intuitive device control through simple gestures.

Enhancing Smart Homes with Pose Detection Using Neuromorphic Computing and Edge AI

<h1 dir="ltr" id="isPasted">Introduction</h1><p dir="ltr">In an era where rapid technological advancements continuously reshape how industries operate, edge artificial intelligence (AI) has emerged as a critical driver of innovation and efficiency. Edge AI brings computation and data storage closer to the locations where it's needed, accelerating response times while reducing bandwidth use. This technology is particularly transformative in industrial and commercial applications, where real-time data processing can significantly enhance operational efficiency and decision-making.</p><p dir="ltr">Enter NVIDIA Train Adapt Optimize (TAO) toolkit—an open-source suite designed to simplify and accelerate the creation of high-quality, customized, and enterprise-ready AI models. This toolkit democratizes access to state-of-the-art AI capabilities and streamlines the development process, enabling rapid deployment across a range of devices from cloud infrastructures to edge units.&nbsp;</p><p dir="ltr">When integrated with Edge Impulse, a leading development platform for edge AI, NVIDIA TAO becomes part of a powerful ecosystem that enhances the capabilities of developers and engineers to bring AI models to production faster. NVIDIA TAO and Edge Impulse represent a formidable combination, poised to accelerate the adoption of edge AI technologies across various sectors by offering tools that reduce complexity and enhance productivity.</p><p dir="ltr">This article explores how the NVIDIA TAO toolkit, along with Edge Impulse, facilitates the development and deployment of AI models on edge devices, illustrating the tangible benefits and real-world applications of this integration. By harnessing these advanced tools, developers can speed up the AI development lifecycle and achieve greater scalability in their AI projects.</p><p dir="ltr"><span contenteditable="false" draggable="true" class="fr-video fr-deletable fr-fvc fr-dvb fr-draggable"><iframe width="540" height="600" src="https://c8056756.sibforms.com/serve/MUIFAMyA6BGksOUPEokHuBnEmZRHhYgJ4RXb4BV0gvzN98ymNIvQdTcBFQnRVfviT6yZSFW_R2g1hENBu7p7Afnyv9Kt0unAAbfFU5mMQhBwI-8hQnKf2r1mmRKhUgDLmtk8bpVSZ7kZGpleCPqtYRv8_kkvODL7F4neGLNGgdpFDm9pPzoYpwbQkc0BU5GVOFoq8CSmD0DDvDjG" 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></span><br></p><h2 dir="ltr">The NVIDIA TAO Toolkit</h2><p dir="ltr">NVIDIA TAO Toolkit is a great tool for developers aiming to harness the power of AI without the complexity typically associated with such endeavors. As an open-source solution, it democratizes advanced technology, allowing a broad range of developers—from startups to large enterprises—to create customized, high-performance AI models that are both scalable and adaptable.</p><h3 dir="ltr">Overview of NVIDIA TAO</h3><p dir="ltr">At its core, the NVIDIA TAO Toolkit facilitates AI model development by leveraging NVIDIA's extensive expertise in AI and deep learning. Designed to streamline the process of model training and optimization, NVIDIA TAO empowers developers to build sophisticated AI models without deep specialization in AI technologies. Its open-source nature ensures that it is continually refined by contributions from the global development community, enhancing its capabilities and accessibility.</p><p dir="ltr"><span class="fr-img-caption fr-fic fr-dib" style="width: 790px;"><span class="fr-img-wrap"><img src="https://images.wevolver.com/eyJidWNrZXQiOiJ3ZXZvbHZlci1wcm9qZWN0LWltYWdlcyIsImtleSI6ImZyb2FsYS8xNzE1MjMyMTA3NTEyLWltYWdlMi5wbmciLCJlZGl0cyI6eyJyZXNpemUiOnsid2lkdGgiOjk1MCwiZml0IjoiY292ZXIifX19" style="width: 100%px;" class="fr-fic fr-dib" alt="nvidia-tao-workflow"><span class="fr-inner" spellcheck="false">Fig. 1: Working with Nvidia TAO toolkit. Source:&nbsp;<a href="https://developer.nvidia.com/tao-toolkit-usecases-whitepaper/1-introduction" target="_blank" rel="noopener noreferrer">Nvidia</a></span></span></span></p><h3 dir="ltr">The Power of Transfer Learning</h3><p dir="ltr">One of the most significant features of the NVIDIA TAO Toolkit is its utilization of transfer learning. This technique revolutionizes how models are built by allowing developers to start with an existing model—pre-trained by NVIDIA or third-party sources on vast and diverse data sets—and customize it to their specific needs with relatively little data. This method drastically reduces the time and resources required for model training and enhances the model’s accuracy and performance from the outset.</p><p dir="ltr">Transfer learning mimics human learning dynamics, where knowledge gained in one context is repurposed for another. For AI development, this means a developer can take a model that recognizes general objects and retrain it to recognize specialized equipment or scenarios specific to their industry. This flexibility opens up a plethora of applications, from enhancing manufacturing processes with visual inspection tools to improving safety protocols through real-time anomaly detection.</p><h2 dir="ltr"><span contenteditable="false" draggable="true" class="fr-video fr-deletable fr-fvc fr-dvb fr-draggable"><iframe width="640" height="360" src="https://www.youtube.com/embed/kErfKDZ50To?&amp;ab_channel=NVIDIADeveloper&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"><span class="fr-mk" style="display: none;">&nbsp;</span></iframe></span>Key Features and Benefits of NVIDIA TAO</h2><p dir="ltr">The NVIDIA TAO Toolkit offers a suite of capabilities that streamline the development process for AI models. Here, we explore how its features cater to efficiency, accuracy, customization, and deployment flexibility.</p><h3 dir="ltr">Efficient Model Training</h3><p dir="ltr">The NVIDIA TAO Toolkit simplifies the model training process significantly. Developers can leverage TAO’s AutoML capability, which automates much of the model tuning. This not only speeds up the model development cycle but also makes it more accessible to non-specialists. The AutoML feature efficiently navigates through the complexities of model tuning, allowing developers to focus more on application development rather than the nuances of model configuration.</p><h3 dir="ltr">High Accuracy and Customization</h3><p dir="ltr">NVIDIA is renowned for its high-performance computing and deep learning technologies, and TAO is no exception. It enables the creation of highly accurate AI models tailored to specific needs using state-of-the-art vision transformers and NVIDIA's pre-trained models. These capabilities allow for the development of bespoke AI solutions that are finely tuned to the unique requirements of different use cases, from healthcare diagnostics to automated surveillance systems.</p><h3 dir="ltr">Optimization for Inference and Deployment Flexibility</h3><p dir="ltr">Once a model is trained, ensuring it performs efficiently on the intended hardware is crucial. NVIDIA TAO facilitates this by enabling significant optimizations for inference, enhancing model accuracy and performance across GPUs, CPUs, and cloud servers. Additionally, Edge Impulse extends this adaptability to various deployment platforms, including microcontrollers and various other edge devices, ensuring that models integrate smoothly with existing infrastructures.</p><h3 dir="ltr">AI-Assisted Data Annotations</h3><p dir="ltr">Preparing data for training AI models is often labor-intensive. NVIDIA TAO eases this burden with AI-assisted data annotation capabilities. This feature accelerates the preparation of training data, making it faster and less costly.</p><h2 dir="ltr">Integrating NVIDIA TAO with Edge Impulse</h2><p dir="ltr">The collaboration between NVIDIA TAO and Edge Impulse presents a synergistic approach to edge AI development, significantly reducing time-to-market while increasing the scalability and efficiency of AI solutions. This integration facilitates a streamlined workflow from model creation to deployment, addressing key challenges in edge AI development.</p><h3 dir="ltr">Enhancing Edge AI Development</h3><p dir="ltr">By incorporating NVIDIA TAO into the Edge Impulse development platform, developers gain access to a robust set of tools designed to expedite the creation of efficient models. This integration leverages the strengths of both platforms—TAO’s powerful model training and Edge Impulse’s sophisticated deployment capabilities. Together, they simplify the development cycle, enabling developers to rapidly prototype, iterate, and deploy AI models that are both robust and scalable.</p><h3 dir="ltr">Comprehensive Edge AI Solution</h3><p dir="ltr">The combined power of NVIDIA TAO and Edge Impulse provides a comprehensive solution for edge AI. From collecting data on edge devices to training and validating models and finally optimizing libraries to run efficiently on target hardware, this integration covers all bases. It supports a wide range of devices, from low-power microcontrollers to more capable GPUs, ensuring that developers can deploy their AI solutions across diverse modern IoT and edge computing environments.</p><p dir="ltr">Developers can tailor their solutions to specific hardware and performance needs, optimizing resource usage and operational efficiency. Additionally, the combined ecosystem supports a variety of silicon and sensors, making it an all-in-one solution for edge AI applications.</p><h2 dir="ltr">Practical Applications and Success Stories</h2><p dir="ltr">The integration of NVIDIA TAO with Edge Impulse offers a powerful platform for practical applications in various sectors. Here, we explore some specific use cases that demonstrate the effectiveness of this technology.</p><p dir="ltr">PCB Classification:&nbsp;In electronics manufacturing, ensuring that all components are correctly placed and functional on a PCB is crucial. A minor error can lead to a malfunctioning device, leading to increased costs and reduced consumer trust. The PCBClassification model, developed using NVIDIA TAO, addresses this challenge effectively.<a href="https://catalog.ngc.nvidia.com/orgs/nvidia/teams/tao/models/pcb_classification" target="_blank"><sup>[1]</sup></a></p><p dir="ltr">This model utilizes an advanced AI architecture known as global context vision transformer (GCViT), initially trained on a comprehensive image dataset (NVImageNet). The model's training involves transfer learning, where a model developed for one task is partially retrained to perform another. In this case, the base model, proficient in general image recognition, is fine-tuned to specifically identify missing or defective components in PCB images.</p><p dir="ltr">The process is straightforward yet powerful: images of PCBs are fed into the model, which has been trained to recognize various components like resistors, capacitors, and inductors. The model assesses each component's presence or absence and checks for defects. This automated inspection helps quickly spot errors during the manufacturing process, reducing the likelihood of defective products reaching the market.</p><p dir="ltr">The accuracy of this model is notable, achieving near-perfect precision in tests. This high level of performance is crucial for industries where the cost of errors can be exceptionally high, both financially and in terms of brand reputation.</p><p dir="ltr">Reducing Baggage Delays at Airports:&nbsp;The case of Trifork using NVIDIA TAO for baggage tracking at airports highlights the transformative potential of these technologies in enhancing operational efficiency.<a href="https://resources.nvidia.com/en-us-metropolis-software-success-stories/trifork" target="_blank"><sup>[2]</sup></a> By deploying vision AI solutions, Trifork has been able to track millions of pieces of luggage, significantly reducing baggage delays and lost baggage incidents. This application not only improves the passenger experience but also optimizes the workflow and resource allocation at airports.</p><p dir="ltr">Industrial Defect Detection:&nbsp;Another compelling application is in the field of industrial defect detection.<a href="https://developer.nvidia.com/blog/transforming-industrial-defect-detection-with-nvidia-tao-and-vision-ai-models/" target="_blank"><sup>[3]</sup></a> Using NVIDIA TAO, developers have created models that can identify defects in various industrial products with high precision. This application is particularly beneficial for quality control in manufacturing, where detecting and addressing defects early in the production process can lead to substantial cost savings and maintain high-quality standards.</p><p dir="ltr">Retail Recognition:&nbsp;In the retail sector, the application of NVIDIA TAO for retail recognition tasks illustrates how edge AI can transform inventory management. The models could help with identifying a vast array of retail Stock Keeping Units (SKUs), facilitating better stock management, customer service, and logistical efficiency. By automating the recognition process, retailers can ensure more accurate inventory tracking and faster response times, leading to enhanced operational efficiency and customer satisfaction.</p><p dir="ltr">These practical applications and success stories demonstrate the broad utility and effectiveness of NVIDIA TAO when integrated with tools like Edge Impulse. The technology not only drives innovation and efficiency in existing processes but also opens up new possibilities for smart manufacturing, retail management, and beyond.</p><p dir="ltr"><em><strong>Further reading:&nbsp;</strong></em><a href="https://www.wevolver.com/article/a-practical-guide-to-understanding-sensor-fusion-with-edge-impulse-building-a-smart-running-jacket" target="_blank" id="isPasted"><strong><em>A Practical Guide to Understanding Sensor Fusion with Edge Impulse: Building a Smart Running Jacket</em></strong></a></p><h2 dir="ltr">Getting Started and Resources</h2><p dir="ltr">To begin utilizing NVIDIA TAO and Edge Impulse to develop and deploy AI models on edge devices, it's essential to engage with a variety of resources and steps designed to streamline this process.</p><p dir="ltr">Firstly, a thorough exploration of the available documentation is crucial. Both <a href="https://docs.edgeimpulse.com/docs/edge-impulse-studio/learning-blocks/nvidia-tao" target="_blank">Edge Impulse</a> and <a href="https://developer.nvidia.com/tao-toolkit" target="_blank">NVIDIA</a> provide comprehensive guides and tutorials on their respective websites. These documents are invaluable for understanding how to effectively integrate NVIDIA TAO models into Edge Impulse projects.</p><p dir="ltr">Additionally, Edge Impulse and Nvidia provide YouTube videos that detail the application and integration of NVIDIA TAO. These resources complement the written documentation and provide a medium through which users can understand how to work with these technologies.<span contenteditable="false" draggable="true" class="fr-video fr-deletable fr-fvc fr-dvb fr-draggable"><iframe width="640" height="360" src="https://www.youtube.com/embed/bwquofcF94g?&amp;ab_channel=EdgeImpulse&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"></iframe></span></p><p dir="ltr">Support doesn’t end with documentation and tutorials. NVIDIA and Edge Impulse have active forums like - <a href="https://forum.edgeimpulse.com/" target="_blank">Edge Impulse Forums</a> and <a href="https://forums.developer.nvidia.com/c/accelerated-computing/intelligent-video-analytics/tao-toolkit/17" target="_blank">NVIDIA Developer Forums</a>, where users can engage with a community of like-minded individuals. These forums are a great place to seek advice, ask questions, and share experiences related to edge AI development.</p><p dir="ltr">By leveraging these resources, developers can effectively navigate the initial learning curve and start applying NVIDIA TAO and Edge Impulse to real-world projects, gradually building up to more complex applications as familiarity with the tools increases.</p><p dir="ltr"><em><strong>Further reading:&nbsp;</strong></em><a href="https://www.wevolver.com/article/building-high-performance-edge-ai-solutions-with-nvidia-jetson-and-edge-impulse" target="_blank" rel="noopener noreferrer"><strong><em>Building High-Performance Edge AI Solutions with Nvidia Jetson and Edge Impulse</em></strong></a></p><h2 dir="ltr">Conclusion</h2><p dir="ltr">The integration of NVIDIA TAO and Edge Impulse brings in a new era of edge AI development, offering developers powerful tools to enhance the efficiency and effectiveness of their AI models. This combination brings together NVIDIA's advanced AI modeling capabilities with Edge Impulse's robust customization and deployment framework, enabling rapid development, optimization, and deployment of AI applications across a variety of platforms and devices.</p><p dir="ltr">For developers and companies venturing into edge AI, the synergy between NVIDIA TAO and Edge Impulse provides a competitive edge. By simplifying complex processes and reducing the time-to-market, this integration democratizes advanced AI technologies and empowers a broader range of users to innovate and solve real-world problems.</p><p dir="ltr"><span contenteditable="false" draggable="true" class="fr-video fr-deletable fr-fvc fr-dvb fr-draggable"><iframe width="540" height="600" src="https://c8056756.sibforms.com/serve/MUIFAMyA6BGksOUPEokHuBnEmZRHhYgJ4RXb4BV0gvzN98ymNIvQdTcBFQnRVfviT6yZSFW_R2g1hENBu7p7Afnyv9Kt0unAAbfFU5mMQhBwI-8hQnKf2r1mmRKhUgDLmtk8bpVSZ7kZGpleCPqtYRv8_kkvODL7F4neGLNGgdpFDm9pPzoYpwbQkc0BU5GVOFoq8CSmD0DDvDjG" 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></span><br></p><h2 dir="ltr">References</h2><p dir="ltr">[1] PCBClassification Model Card, Nvidia, 17 Oct 2023, [Online], Available from: <a href="https://catalog.ngc.nvidia.com/orgs/nvidia/teams/tao/models/pcb_classification" target="_blank">https://catalog.ngc.nvidia.com/orgs/nvidia/teams/tao/models/pcb_classification</a></p><p dir="ltr">[2] Trifork - Reducing baggage delay at airports with vision AI, Nvidia, [Online], Available from: <a href="https://resources.nvidia.com/en-us-metropolis-software-success-stories/trifork" target="_blank">https://resources.nvidia.com/en-us-metropolis-software-success-stories/trifork</a></p><p>[3] Nirmal Kumar Juluru, Zaid Pervaiz Bhat and Samuel Ochoa, ‘Transforming Industrial Defect Detection with NVIDIA TAO and Vision AI Models’, Nvidia, [Online], Available from: <a href="https://developer.nvidia.com/blog/transforming-industrial-defect-detection-with-nvidia-tao-and-vision-ai-models/" target="_blank">https://developer.nvidia.com/blog/transforming-industrial-defect-detection-with-nvidia-tao-and-vision-ai-models/</a></p>

NVIDIA TAO and Edge Impulse integrate to streamline the development and deployment of edge AI models, empowering developers to quickly build and optimize AI applications for diverse hardware environments.

Accelerating Edge AI Model Development Using Nvidia TAO and Edge Impulse

NVIDIA Jetson AGX Orin Module and Developer Kit

Nvidia Jetson and Edge Impulse facilitate Edge AI development, offering computational power and model training platforms for real-world applications.

Building High-Performance Edge AI Solutions with Nvidia Jetson and Edge Impulse

<p id="isPasted">While embedded systems tend to lack the processing horsepower of servers or even modern personal computers, the sheer number of devices is making them an increasingly valuable target for bad actors looking to run illegal botnets and cryptocurrency mining operations. One of the first major security-related wake-up calls for embedded system designers was the 2016 Nest thermostat botnet attacks. Given the consumer-facing nature of the particular Internet of Things (IoT) coupled with an increased sensitivity regarding privacy and security; the Nest botnet caused a huge amount of discussion. Those discussions tended to center on how companies should build security into their low-cost IoT products and how consumers can safely operate the devices in their homes and businesses.</p><p>With the growing threat of cyberattacks, it is essential that developers keep security considerations in mind throughout the design process. By following some practical tips and recommendations, developers can guard against a wide range of attack scenarios. Read on for an outline of security measures developers can use in their embedded designs.</p><h2>Build Secure</h2><p>While there are numerous chip architectures, operating systems, and communications protocols; many IoT devices tend to be built around Arm<sup>®</sup>-based architectures, and if they run an OS it tends to be a Linux distribution. This commonality is good in many ways; lower costs and faster development times; it also comes with quite a few negatives. Attack vectors tend to become “one-size-fit-all”, especially for devices running a Linux-based OS. To mitigate the threats associated with widespread devices that share a common architecture, developers should implement the following “quick win” security design principles:</p><ul><li>Do NOT hardcode passwords into the firmware. Also, do not use a common default password for all devices. Require the user to create a custom username and password during device initialization.</li><li>Do NOT enable insecure protocols such as HTTP, FTP, or Telnet by default. Data going off the device via wired or wireless protocols must be strongly encrypted. Avoid “homebrewed” encryption solutions.</li><li>Ship the device with the most restrictive configuration possible and let the end-user make a proactive decision to reduce security-related settings.</li><li>All mechanisms used to access the devices should require authentication and authorization controls. Two-factor authentication (2FA) should be used if practical.</li><li>All user-facing inputs should be filtered to avoid injection-type attacks.</li><li>Implement a secure device management interface for the end-user that will allow them to manage their assets, update devices, monitor devices, and securely decommission the devices that have reached their end-of-life (EOL).</li><li>Over-The-Air (OTA) update mechanisms must be validated on-device. The update files must be sent encrypted en route to the device. Lastly, ensure there are anti-rollback features to prevent a device from being reverted to a previous, insecure firmware.</li><li>If third-party software libraries are used in the design of your device, they must be continually monitored to ensure that third-party updates are integrated and that they do not become deprecated. Abandoned software projects can become nasty vulnerabilities for your device. Change the third-party software default passwords before committing them to your project.</li><li>Limit what sensitive data should be stored on the device. Store such information in a secure enclave only.</li><li>Remember that with the IoT that embedded systems are only one part of a larger ecosystem. Ensure that security is built into the cloud, desktop, and mobile applications as well. Ecosystem security is only as strong as the weakest link.</li><li>Consider establishing a bug bounty program to encourage end-users and security researchers to submit flaws in a secure, responsible manner.</li></ul><p>Physical access to a device tends to be the game-over situation for devices. That doesn’t mean that there aren’t things that can be done to make it harder physically exploit these types of devices. Entire books have been written on making circuit boards and associated enclosures tamper-resistant but for a few “quick wins” consider the following physical design rules-of-thumb to harden your device:</p><ul><li>Pins for debugging ports such as JTAG and UARTs are very useful when developing and testing a device. They are also tempting targets for those seeking to reverse engineer a device with ill-intent. Obfuscating these pins and/or removing header pins for production units is recommended. Note that this will come at the expense of making it more difficult to troubleshoot once fielded. Designers must consider a balance between security and maintainability.</li><li>The use of adhesives, ultrasonic welding, and/or specialty security screws can make it more difficult to open a device.</li><li>Applying a non-conductive epoxy over sensitive components can obfuscate their identity and purpose.</li><li>You might be tempted to use older components that might have vulnerabilities. Be aware of the counterfeit components as well. If a deal seems too good to be true, it probably is. Balancing security and time-to-market considerations is not a decision to be taken lightly.</li><li>Multilayer boards can be used to route traces in a way that makes it harder to discern how the board functions.</li></ul><p><img src="https://s3-us-west-1.amazonaws.com/wevolver-project-images/froala%2F1723102912865-shutterstock_2343572457.jpg" style="width: 100%px;" class="fr-fic fr-dib"></p><p>Some of the following recommendations might be a bit much for consumer-grade IoT devices. However, as we will elaborate on later, industrial control systems and defense systems would likely benefit from these more robust physical security measures:</p><ul><li>Build security features into the board itself, such as microswitches, mercury switches, or magnetic switches that can detect if a board is being unintentionally handled or opened. Nichrome wire or fiber optics can also be used. If the wire or fiber is negatively impacted by someone trying to tamper with a device, then there will be a detectable change in the current flow of the wire or the behavior of photons through the fiber.</li><li>Side Channel or glitching attacks, while perhaps not common, provide a unique advantage to adversaries in that they use the laws of physics against a device and thus are very difficult to prevent though can be detected. By attacking the timing or restricting the flow of electrons to a CPU, it’s possible to get a device to act in unintended ways that negate security features. Voltage and current sensors can be implemented onboard circuit boards to detect if glitching may be occurring, though there is potential for false positives.</li><li>Significant adversaries can employ x-ray machinery to peer into microchips and discover the topography of the transistors to determine function and more. Sensors that detect x-rays can be added to detect tampering, though can do nothing to prevent an adversary from extracting useful information.</li></ul><p>It should be noted that there is quite a dichotomy between the paradigms of security and openness. Security values obfuscation. Open hardware values understanding. Regardless, keep in mind the old adage that locks only keep honest people honest, and so too with security-at-large. For more information on how to build secure IoT devices, visit the Open Web Application Security Project (OWASP) IoT Project.</p><h2>Operate Secure</h2><p>Even if a manufacturer could implement all the best secure design principles in their product, they would be mostly for naught if the end-user does not operate their device in a secure manner.</p><ul><li>Change the default router name, router password, network name (SSID), and network encryption key. Be sure to use strong password principles and do not use the same password for both networks.</li><li>Segment your home network into two “virtual” networks so that IoT devices cannot be “seen” by the desktop computers, Network Attached Storage (NAS) devices, etc. To do this quickly and easily, use the guest network feature for your IoT network.</li><li>Most IoT devices rely on a smartphone app to control the device. Keep the app up to date and use 2FA for login if available.</li><li>Disable any features of your IoT devices that you do not intend to use.</li><li>Update the firmware of both your router and IoT devices on a regular basis.</li><li>When an IoT device reaches the end of life and no longer receives updates consider replacing it with a newer model.</li></ul><h2>Industrial Strength Security</h2><p>Consumer-facing IoT products may be plentiful, but their industrial counterparts, collectively&nbsp;<iframe id="65f2ad2682d39cfbc40c772a" width="250" class="specs-iframe" height="440" src="https://wevolver.com/iframe/specs/microchip-technology-evb-pci11400-pcie-switch-evaluation-board?iframe=true" frameborder="0" scrolling="no" data-lf-form-tracking-inspected-kn9eq4roawkarlvp="true" data-lf-yt-playback-inspected-kn9eq4roawkarlvp="true" data-lf-vimeo-playback-inspected-kn9eq4roawkarlvp="true"></iframe> referred to as Industrial Control Systems (ICS) manage numerous extremely important and potentially dangerous processes. Everything from energy production to factories uses embedded digital technology (referred to as Operational Technology or OT; in contrast to office-centric Information Technology or IT) to control the facilities and associated machinery responsible for performing the various processes. The ICS environment is sufficiently different from a strict IT environment that special considerations for hardening OT devices and ICS networks are warranted. The most fundamental principle is that ICS should not have a connection to the internet. While this seems like a no-brainer, it is surprising how often this fundamental rule is violated. For more information on how to secure ICS networks and devices, there are two security frameworks you should review: MITRE ATT&amp;CK for ICS and the MITRE ATT&amp;CK for Enterprise.</p><p>The nature of where ICS systems are typically found (e.g., areas that are environmentally, chemically, or otherwise hazardous) means ICS has been designed to prioritize the availability of the systems over confidentiality. From a positive perspective, this typically means there are redundant systems, and those systems are designed to fail safely. However, ICS systems can be left in operations for several decades and may not always be kept up-to-date. In addition, many of the protocols are older and were built with efficiency, not security in mind. Bottom line, security in the ICS or IIoT space is uniquely challenging and best practices may be difficult to implement. However, embedded developers for such machinery should be cognizant of the need to modernize their design practices and incorporate security into future designs and not treat it as a bolt-on afterthought.</p>

With the growing threat of cyberattacks, it is essential that developers keep security considerations in mind throughout the design process. By following some practical tips and recommendations, developers can guard against a wide range of attack scenarios.

Why are security protocols crucial for your embedded systems?

Edge AI and sensor fusion revolutionize tech by enabling instant, informed decisions through multi-sensor data analysis. Explore their application by building a Smart Running Jacket with Sony Spresense and SensiEDGE CommonSense via Edge Impulse platform.

A Practical Guide to Understanding Sensor Fusion with Edge Impulse: Building a Smart Running Jacket

<h1 dir="ltr" id="isPasted">Introduction</h1><p dir="ltr">Artificial intelligence (AI) is experiencing a significant transition with the increasing adoption of edge computing, where data processing occurs closer to the data source. The aim is to enable real-time decision-making and reduce dependency on cloud services.&nbsp;</p><p dir="ltr">This transition is driven by advanced microprocessors, with the Texas Instruments AM68A being a notable example. It offers a combination of efficiency, power, and adaptability, making it well-suited for the requirements of Edge AI applications.&nbsp;</p><p dir="ltr">In this article, we uncover the capabilities and implications of the Texas Instruments AM68A within the Edge AI landscape. Our exploration will dive into the intricacies of its technological features and spotlight its integration with the&nbsp;<a href="https://edgeimpulse.com/" target="_blank">Edge Impulse</a> platform. We'll also feature some case studies that demonstrate the practical application of the AM68A in Edge AI projects focused on computer vision.<br><br><span contenteditable="false" draggable="true" class="fr-video fr-deletable fr-fvc fr-dvb fr-draggable"><iframe src="https://c8056756.sibforms.com/serve/MUIFAMyA6BGksOUPEokHuBnEmZRHhYgJ4RXb4BV0gvzN98ymNIvQdTcBFQnRVfviT6yZSFW_R2g1hENBu7p7Afnyv9Kt0unAAbfFU5mMQhBwI-8hQnKf2r1mmRKhUgDLmtk8bpVSZ7kZGpleCPqtYRv8_kkvODL7F4neGLNGgdpFDm9pPzoYpwbQkc0BU5GVOFoq8CSmD0DDvDjG" frameborder="0" scrolling="auto" allowfullscreen="" style="display: block; margin-left: auto; margin-right: auto; max-width: 100%; width: 766px; height: 603px;" 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></span></p><h1 dir="ltr"><br></h1><h3 dir="ltr">Overview of the Texas Instruments AM68A</h3><p dir="ltr">The Texas Instruments AM68A is a microprocessor that delivers great performance and features dedicated AI hardware acceleration. This enhancement is crucial for optimizing AI tasks directly on the device, enhancing efficiency and processing speed.&nbsp;</p><p dir="ltr">Before exploring its diverse uses and influence in Edge AI, it's essential to understand its core specifications and features. This groundwork will provide a clear perspective on the microprocessor's capabilities and how it can be utilized in various Edge AI contexts.</p><p dir="ltr"><span class="fr-img-caption fr-fic fr-dib" style="width: 304px;"><span class="fr-img-wrap"><img src="https://images.wevolver.com/eyJidWNrZXQiOiJ3ZXZvbHZlci1wcm9qZWN0LWltYWdlcyIsImtleSI6ImZyb2FsYS8xNzA0ODg1MDY3ODI1LWltYWdlMi5wbmciLCJlZGl0cyI6eyJyZXNpemUiOnsid2lkdGgiOjk1MCwiZml0IjoiY292ZXIifX19" style="width: 302px;" class="fr-fic fr-dib" alt="Texas-Instruments-TI-AM68A"><span class="fr-inner" spellcheck="false">Fig. 1: Texas Instruments AM68A development platform</span></span></span></p><p dir="ltr" id="isPasted">Here are some core specifications and features of the AM68A:</p><div dir="ltr" align="left"><table><tbody><tr><td><p dir="ltr"><strong>Feature</strong></p></td><td><p dir="ltr"><strong>Description</strong></p></td></tr><tr><td><p dir="ltr"><strong>Processor Core</strong></p></td><td><p dir="ltr">Dual 64-bit ARM Cortex-A72 microprocessor subsystems, up to 2 GHz, with a shared 1MB L2 cache.</p></td></tr><tr><td><p dir="ltr"><strong>Deep Learning Accelerators</strong></p></td><td><p dir="ltr">Deep Learning Accelerator capable of up to 8 TOPS, for efficient AI computations and real-time data processing.</p></td></tr><tr><td><p dir="ltr"><strong>Vision Processing</strong></p></td><td><p dir="ltr">Vision Processing Accelerators, including an Image Signal Processor and multiple vision assist accelerators for handling image and video data.</p></td></tr><tr><td><p dir="ltr"><strong>Additional Processing Units</strong></p></td><td><p dir="ltr">Dual-core ARM Cortex-R5F MCUs for general compute operations and device management.</p></td></tr><tr><td><p dir="ltr"><strong>Graphics Processing</strong></p></td><td><p dir="ltr">IMG BXS-4-64, a high-performance 3D GPU supporting OpenGL ES 3.1 and Vulkan 1.2 for graphic-intensive applications.</p></td></tr><tr><td><p dir="ltr"><strong>Connectivity Options</strong></p></td><td><p dir="ltr">Gigabit ethernet, USB 3.0, CSI 2.0, PCIe gen 3, and various memory interfaces.</p></td></tr><tr><td><p dir="ltr"><strong>Security Features</strong></p></td><td><p dir="ltr">Secure boot and crypto hardware accelerators for robust security.</p></td></tr></tbody></table></div><p dir="ltr"><em>Table 1: Specifications and features of Texas Instruments AM68A</em></p><h1 dir="ltr">The Role of Deep Learning Accelerators in Edge AI</h1><p dir="ltr">The Texas Instruments AM68A has deep learning accelerators, which play a crucial role in Edge AI and computer vision applications. These are specialized hardware components designed to efficiently execute the computations required for deep learning algorithms. Unlike general-purpose processors, these accelerators are optimized for specific tasks like matrix multiplications and convolutions, which are fundamental in neural network operations.</p><p dir="ltr">The significance of deep learning accelerators in edge computing lies in their ability to bring advanced AI functionalities to devices operating at the edge of networks. By handling complex computations locally, these accelerators reduce the need to transmit large amounts of data to the cloud for processing. This cuts down on latency, making real-time processing feasible, and addresses concerns like bandwidth limitations and data privacy.</p><h3 dir="ltr">Real-World Applications and Case Studies</h3><p dir="ltr">The Texas Instruments AM68A has been central to a variety of computer vision projects, demonstrating its versatility and capability. These case studies showcase the practical use of the AM68A in diverse Edge AI applications.</p><h3 dir="ltr">1. American Sign Language (ASL) Translation</h3><p dir="ltr">One of the remarkable projects developed using a Texas Instruments board is an American Sign Language (ASL) Translation system. This application utilizes the AM62A, the entry-level alternative to the AM68A, and is designed to translate ASL into text in real-time. By leveraging machine learning algorithms and image processing capabilities, this system enhances communication accessibility for the deaf and hard-of-hearing community.&nbsp;</p><p dir="ltr"><span class="fr-img-caption fr-fic fr-dib" style="width: 302px;"><span class="fr-img-wrap"><img src="https://images.wevolver.com/eyJidWNrZXQiOiJ3ZXZvbHZlci1wcm9qZWN0LWltYWdlcyIsImtleSI6ImZyb2FsYS8xNzA0ODg1MDk3MjkxLWltYWdlNC5wbmciLCJlZGl0cyI6eyJyZXNpemUiOnsid2lkdGgiOjk1MCwiZml0IjoiY292ZXIifX19" style="width: 300px;" class="fr-fic fr-dib" alt="sign-language"><span class="fr-inner" spellcheck="false">Fig. 2: Training the model to recognize different signs</span></span></span></p><p dir="ltr" id="isPasted">The project's success on the AM62A highlights the potential advancements that could be achieved with the more powerful AM68A.&nbsp;</p><p dir="ltr"><em><strong>Read more about this project:&nbsp;</strong></em><a href="https://docs.edgeimpulse.com/experts/image-projects/asl-byom-ti-am62a" target="_blank"><strong><em>Import a Pretrained Model with "Bring Your Own Model" - Texas Instruments AM62A</em></strong></a></p><h3 dir="ltr">2. Motorcycle Helmet Detection</h3><p dir="ltr">Another innovative application is the Motorcycle Helmet Detection project. Also developed on the Texas Instruments AM62A, this system utilizes machine learning and computer vision to detect whether motorcyclists are wearing helmets. This application is crucial for enhancing road safety and can be deployed at traffic signals or highway checkpoints.&nbsp;</p><p dir="ltr"><span class="fr-img-caption fr-fic fr-dib" style="width: 302px;"><span class="fr-img-wrap"><img src="https://images.wevolver.com/eyJidWNrZXQiOiJ3ZXZvbHZlci1wcm9qZWN0LWltYWdlcyIsImtleSI6ImZyb2FsYS8xNzA0ODg1MTIwMzY2LWltYWdlNi5wbmciLCJlZGl0cyI6eyJyZXNpemUiOnsid2lkdGgiOjk1MCwiZml0IjoiY292ZXIifX19" style="width: 300px;" class="fr-fic fr-dib" alt="helmet-detection-edge-ai"><span class="fr-inner" spellcheck="false">Fig. 3: Testing the motorcycle helmet detection system prototype</span></span></span></p><p dir="ltr" id="isPasted">The implementation with the AM62A demonstrates the feasibility of such safety applications, which can be further optimized with the AM68A's enhanced capabilities.&nbsp;</p><p dir="ltr"><em><strong>Explore this project in detail:&nbsp;</strong></em><a href="https://docs.edgeimpulse.com/experts/image-projects/motorcycle-helmet-detection-smart-light-ti-am62a" target="_blank"><strong><em>Motorcycle Helmet Identification and Traffic Light Control - Texas Instruments AM62A</em></strong></a></p><h3 dir="ltr">3. Posture Detection</h3><p dir="ltr">The Texas Instruments TDA4VM board, another counterpart of the AM68A with fewer features, has been used in a posture detection project. This application focuses on identifying incorrect postures in real-time to prevent workplace injuries. Using advanced image processing and AI, the system provides immediate feedback for posture correction.</p><p dir="ltr"><span class="fr-img-caption fr-fic fr-dib" style="width: 302px;"><span class="fr-img-wrap"><img src="https://images.wevolver.com/eyJidWNrZXQiOiJ3ZXZvbHZlci1wcm9qZWN0LWltYWdlcyIsImtleSI6ImZyb2FsYS8xNzA0ODg1MTQxOTU2LWltYWdlNS5wbmciLCJlZGl0cyI6eyJyZXNpemUiOnsid2lkdGgiOjk1MCwiZml0IjoiY292ZXIifX19" style="width: 300px;" class="fr-fic fr-dib" alt="posture-detection-edge-ai"><span class="fr-inner" spellcheck="false">Fig. 4: Texas Instruments AM62A is used to efficiently distinguish between correct and incorrect postures, promoting ergonomic practices and preventing injuries.</span></span></span></p><p dir="ltr" id="isPasted">The AM68A, with its superior processing power, could elevate such health and safety applications to new levels.&nbsp;</p><p dir="ltr"><em><strong>Learn more about this project:&nbsp;</strong></em><a href="https://docs.edgeimpulse.com/experts/image-projects/ti-tda4vm-posture-detection" target="_blank"><strong><em>Correct Posture Detection and Enforcement - Texas Instruments TDA4VM</em></strong></a></p><h3 dir="ltr">4. Smart Factory Prototype</h3><p dir="ltr">Finally, the Smart Factory Prototype project, developed on the Texas Instruments TDA4VM, illustrates the potential of these boards in industrial applications. This system employs AI to monitor manufacturing processes, enhancing efficiency and predicting maintenance needs. The project demonstrates how edge computing can revolutionize industrial operations, and with the advanced features of the AM68A, these applications could achieve even greater efficiency and accuracy.&nbsp;</p><p dir="ltr"><em><strong>Read more about the Smart Factory Prototype:&nbsp;</strong></em><a href="https://docs.edgeimpulse.com/experts/image-projects/smart-factory-prototype-ti-tda4vm" target="_blank"><strong><em>Smart Factory Prototype - Texas Instruments TDA4VM</em></strong></a></p><h3 dir="ltr">Transitioning to the AM68A</h3><p dir="ltr">Each of these projects, developed on the AM62A and TDA4VM boards, lays the groundwork for what could be achieved with the AM68A. With its enhanced processing power and deep learning capabilities, the AM68A is poised to elevate these applications, offering more efficient processing, quicker response times, and the ability to handle more complex AI models. This transition represents a significant step forward in the capabilities of edge computing devices and their applications across various industries.</p><h2 dir="ltr">Conclusion</h2><p dir="ltr">The AM68A, with its advanced deep learning accelerators and enhanced processing capabilities, stands as a significant player in the rapidly evolving landscape of edge computing. Its integration with the Edge Impulse platform has opened new avenues for AI model development, making advanced AI functionalities more accessible and effective.</p><p dir="ltr">Throughout the article, we've seen how the AM68A is being utilized in various sectors, from enhancing road safety with motorcycle helmet detection to improving workplace ergonomics through posture detection. These applications, initially developed on the AM62A and TDA4VM, offer a glimpse into the possibilities with the more advanced AM68A. The chip's potential in driving innovations in smart factories, healthcare monitoring, environmental analysis, and beyond, underscores its versatility and efficiency in handling diverse Edge AI applications.</p><h1 dir="ltr"><em>About the sponsor: Edge Impulse</em></h1><p dir="ltr"><em>Edge Impulse is the leading development platform for embedded machine learning, used by over 1,000 enterprises across 200,000 ML projects worldwide. We are on a mission to enable the ultimate development experience for machine learning on embedded devices for sensors, audio, and computer vision, at scale.&nbsp;</em></p><p dir="ltr"><em>From getting started in under five minutes to MLOps in production, we enable highly optimized ML deployable to a wide range of hardware from MCUs to CPUs, to custom AI accelerators. With Edge Impulse, developers, engineers, and domain experts solve real problems using machine learning in embedded solutions, speeding up development time from years to weeks. We specialize in industrial and professional applications including predictive maintenance, anomaly detection, human health, wearables, and more.</em></p><p dir="ltr"><img src="https://images.wevolver.com/eyJidWNrZXQiOiJ3ZXZvbHZlci1wcm9qZWN0LWltYWdlcyIsImtleSI6ImZyb2FsYS8xNzA0ODg1MTczMTQ0LWltYWdlMS5wbmciLCJlZGl0cyI6eyJyZXNpemUiOnsid2lkdGgiOjk1MCwiZml0IjoiY292ZXIifX19" style="width: 300px;" class="fr-fic fr-dib" alt="edge-impulse"></p><p><span contenteditable="false" draggable="true" class="fr-video fr-deletable fr-fvc fr-dvb fr-draggable"><iframe width="540" height="600" src="https://c8056756.sibforms.com/serve/MUIFAMyA6BGksOUPEokHuBnEmZRHhYgJ4RXb4BV0gvzN98ymNIvQdTcBFQnRVfviT6yZSFW_R2g1hENBu7p7Afnyv9Kt0unAAbfFU5mMQhBwI-8hQnKf2r1mmRKhUgDLmtk8bpVSZ7kZGpleCPqtYRv8_kkvODL7F4neGLNGgdpFDm9pPzoYpwbQkc0BU5GVOFoq8CSmD0DDvDjG" 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></span><br></p>

High-performance microprocessors with AI hardware acceleration are revolutionizing Edge AI, enabling advanced computer vision capabilities directly on devices for real-time data processing and analysis.

Building Robust Edge AI Computer Vision Applications with High-Performance Microprocessors

Microcontrollers and microprocessors are fundamental components in electronics and computing. Both play crucial roles in operating various devices, from smartphones to embedded systems in automobiles and home appliances. This article studies their differences in architecture, performance, and more.

Microcontroller vs Microprocessor: A Comprehensive Guide to Their Differences and Applications

<p id="isPasted">In today's ever-evolving technological landscape, optimizing energy consumption in embedded devices is of paramount importance. This article sheds light on the pivotal role of debug UART logs in power profiling and explores how this synergy benefits developers and engineers alike.</p><p><strong>Debug UART Logs: A Power Profiling Asset</strong></p><p>By connecting your embedded device's debug UART to <a href="https://www.qoitech.com/products/" target="_blank" rel="noopener noreferrer">Qoitech's Otii box</a>, you unlock a potent capability – the ability to timestamp log data alongside power consumption measurements. This fusion of debugging and power profiling offers a formidable solution for scrutinizing power consumption during the development and testing phases.</p><p><strong>Why Debug UART Logs Matter</strong></p><ol><li><strong>Efficiency Optimization</strong>: Real-time insights into your embedded device's behavior, combined with power consumption data, enable you to identify power-hungry hardware components or inefficient firmware code. This knowledge is essential for optimizing your embedded device's energy efficiency, especially when refining hardware and firmware.</li><li><strong>Debugging Power-Related Issues</strong>: Synchronized debug UART logs and power measurements facilitate the pinpointing of energy-draining events or operations during development. This simplifies the task of identifying and addressing power-related issues, making debugging more efficient.</li><li><strong>Reduced Development Time</strong>: Merging debug and power profiling streamlines development. You can swiftly identify and rectify power issues, resulting in shorter development cycles and faster time-to-market.</li><li><strong>Cost Savings</strong>: Effective power management translates to extended battery life for portable embedded devices and reduced electricity costs for mains-powered devices. This can lead to substantial cost savings in the long run.</li></ol><p><strong>Benefits for IoT and Embedded Professionals</strong></p><ol><li><strong>Hardware Developers</strong>: This technique empowers you to create energy-efficient embedded devices, offering improved user experiences, longer battery life, and reduced environmental impact. It also expedites the identification and resolution of power-related bugs, saving valuable development time.</li><li><strong>Software Developers</strong>: Debug UART power profiling can enhance the performance of embedded and application software. Recognize that low-power consumption involves not only hardware but also the entire system and ecosystem, where applications or protocols could be energy drains.</li><li><strong>Product Managers</strong>: Ensure that embedded devices meet energy efficiency standards and deliver exceptional performance, resulting in higher customer satisfaction and potentially increased market share.</li><li><strong>Researchers</strong>: Utilize this technique for experiments, data collection, and contributing to the advancement of energy-efficient technologies.</li></ol><p><strong>Conclusion</strong></p><p>Understanding the factors that drain energy provides invaluable insights into your IoT device's operation, enabling informed decisions to develop low-power embedded devices. For a comprehensive, step-by-step guide on implementing this process using Otii Arc/Ace Pro, please consult the <a href="https://www.qoitech.com/docs/quick-start/setup-uart-recording" target="_blank" rel="noopener noreferrer">Otii documentation</a>.</p>

UART log sync with power measurements with Otii Ace Pro

Understanding what drains energy is a game-changer, whether you're a developer crafting the next IoT breakthrough, fine-tuning an embedded system, or simply interested in gauging the power efficiency of your devices.

Efficient IoT Design: Synchronized UART Logging and Power Measurement

IoT Stars

Bota Systems

embedded systems

embedded systems

Profiles