Bench Talk

New Tech Tuesdays

Join Rudy Ramos for a weekly look at all things interesting, new, and noteworthy for design engineers.

In today’s digital transformation age, edge computing has emerged as a pivotal technology for processing data at its source. Central to this transformation are embedded system-on-chips (SoCs) and system-on-modules (SOMs), which are compact, efficient, and powerful computing solutions that bring speed and intelligence closer to the point of use.

This week, we investigate the benefits and applications of these embedded systems and why they're revolutionizing industries.

The Edge of Innovation

Edge computing is witnessing significant growth primarily due to the proliferation of Industrial Internet of Things (IIoT) and Internet of Things (IoT) devices and the increasing demand for low-latency processing.

Edge computing refers to processing data near the source rather than relying on a centralized data-processing center or cloud (Figure 1). This proximity helps create a synergy between embedded SoCs and edge that reduces latency, increases speed, and significantly enhances the performance of applications that require real-time processing, revolutionizing sectors from industrial automation and autonomous vehicles to healthcare and medical devices.

Figure 1: Edge computing has the potential to enhance data processing by reducing the proximity of the data to its source. (Source: VectorMine/stock.adobe.com)

Enhanced Performance, Efficiency, and Security

Embedded SoCs are designed to seamlessly integrate hardware and software functionalities, allowing for optimized performance and swift task execution. Software frameworks such as TensorFlow, PyTorch, and Edge Impulse are instrumental in deploying machine learning models onto these edge devices.

One of the standout benefits of using embedded SoCs in edge computing is their ability to deliver high performance per watt, ensuring that edge devices can operate longer without draining their power resources. Moreover, built-in security features such as encryption, secure boot, and hardware-based authentication provide robust protection against cyber threats and unauthorized access, which means safer, more reliable operations. In addition, modern high-performance SoCs with AI capabilities support advanced use cases like multi-class object detection and super-resolution video enhancement. These AI-enabled SoCs power applications that require compact components, which include smartphones, wearable devices, and sophisticated IIoT and IoT solutions.

A New Frontier for SoCs in Edge Computing

In IoT, edge computing transforms simple devices into smart ones, allowing them to perform complex tasks without constant cloud interaction. In the industrial sector, IIoT devices use embedded systems for predictive maintenance and automated quality control, even in unreliable connectivity environments. In healthcare, embedded SoCs enable real-time health monitoring and diagnostics and process patient data locally for immediate analysis and response.

In smart cities, embedded SoCs power various IoT devices, from smart meters to traffic management systems, enhancing urban living and resource management through real-time data processing. Additionally, autonomous vehicles rely on low-latency data processing for split-second decisions, especially as they advance from Level 2 to Level 3 automation. Embedded SoCs handle data from sensors and cameras, ensuring safe and efficient vehicle operation.

The Newest Products for Your Newest Designs^®

This week's New Tech Tuesday showcases pioneering products from AMD that are poised to shape the landscape of edge computing through seamless integration and optimal performance.

The AMD Kria^™ K24 SOM is a small, power-efficient computing device designed for cost-sensitive industrial and commercial edge applications. Part of the Kria portfolio of adaptive SOMs and developer kits, this SOM is based on the AMD Zynq™ UltraScale+™ MPSoC architecture. The K24 SOM is available in both commercial and industrial versions, with the industrial-grade SOM built for 10-year industrial lifecycles.

The K24 SOM delivers lower latency and optimized power efficiency to motor control and DSP edge applications. It provides deterministic performance to highly reliable end systems, helping to accelerate whole applications at the edge and a faster path to volume production deployment compared to chip-down.

As the development platform for Kria K24 SOMs, the AMD Kria KD240 Drives Starter Kit is built for motor control and DSP applications. It is complete with a variety of interfaces and features, like native Python support for ease of development. This kit enables embedded software and control systems developers without FPGA expertise to develop multiple target applications, such as robotics drives/actuators, industrial motors, industrial Ethernet gateways/sensors, EV charging stations, medical equipment, and aerial systems.

Tuesday’s Takeaway

The benefits of edge computing with embedded devices are immense and diverse. By embracing SoCs, leveraging powerful software, and optimizing for IoT and IIoT, these systems are transforming data processing and unlocking a new level of autonomy in technology.