Embedded Computing Design

Imagine a police body cam. We’ve all seen, or, at least heard of those, right? Now, for first responders, what if those on-body systems were artificial intelligence-enabled and cloud- connected, capable of detecting and tracking faces, live streaming to support services, and saving footage into secure cloud storage. Things would certainly be more … transparent.

I recently came across an 8-bit PIC MCU data sheet that predates 1982. That means it also predates Microchip as a company (incorporated in 1989), making it a remnant of bygone General Instruments days.

Everyone has had a conversation with that experienced engineer or tenured member of the technical staff who’s said, “IoT is nothing new. I’ve been doing that for [insert number of decades], we just called it [insert connected embedded/M2M/other term implying networked equipment].” It’s eye roll-inducing. And it’s wrong.

Protocol analysis isn’t going out of style anytime soon, and neither are the oscilloscopes you need to perform it. In fact, they’re modernizing.

Let’s get back in the swing of things by starting with something tangible. This Nucleo-L4A6ZG development board from STMicroelectronics is based on a 32-bit Arm Cortex-M4 CPU core that runs at 80 MHz and, importantly, contains 320 KB of SRAM and 1 MB flash. More on that in a minute.

The new Jetson AGX Orin Developer Kit delivers 275 TOPS of performance at 15-60 W, and is pairing with advances in the NVIDIA Isaac and JetPack SDKs to accelerate AI development, deployment, and bring the output of Omniverse tools to autonomous robots at the edge.

Data is the foundation of smart factory and Industry 4.0 value, and that data is captured by sensors at the edge. Transforming that data into intelligence requires the convergence of the IT and OT domains, which begins with smart sensors.

PICMG IoT.1-compliant firmware must be user-friendly but also machine readable. In Part 4 we discuss the next link in the firmware development chain: the Builder.

PICMG IoT.1-compliant smart sensors run on a configured data model. In Part 3, we take a look at the process of configuring the smart sensor firmware using a sample configuration tool provided by PICMG.

To accelerate the development and deployment of smart sensors in Industry 4.0 applications, the PICMG IoT.1 specification outlines a standard data model for sensor manufacturers and systems integrators. In Part 1, we look at the requirements of a smart sensor and outline the tools required to make your own.

Percepio has improved Tracealyzer support for Microsoft Azure and the Azure ThreadX RTOS by integrating its trace recorder in ThreadX so that hours or days of trace data can be streamed to and stored on the host system for profiling or burn-in testing. The Percepio DevAlert IoT device monitoring service also now runs on Azure.

RISC-V is known as an open-standard instruction set architecture (ISA) whose base instructions have been frozen to minimize complexity. But more recently it has added a wide range of custom extensions and enhancements that are making it increasingly popular amongst SoC designers building application-specific systems.

SiFive announced the Essential 6-Series of real-time and mid-range application processors ahead of this week’s RISC-V Summit. The company also made its 21G3 release available, which includes a series of updates to their processor IP portfolio.

ROHM Semiconductor has added a new thermal function to its Solution Simulator, a free web-based simulation circuit for coupled electrical and thermal analysis in near-actual application conditions. The new function allows automotive and industrial engineers to collectively verify thermal issues on driver ICs and power devices on different circuits, and can be implemented for designs like PTC heaters for electric vehicles, DC/DC converter ICs, and LED drivers where heat dissipation issues are likely.

Support for RV64I, RV64E, and RV64GC with floating-point unit RISC-V CPUs has been added to the latest version of SEGGER Embedded Studio. The 64-bit CPU IP cores will benefit from SEGGER emRun C/C++ runtime, emFloat floating-point libraries, SEGGER Linker, and SEGGER Compiler, which combine in the environment to help developers generate compact, efficient code.

SiFive, Inc. has announced the Performance P650 processor, a coherent multicore processor that scales to 16 cores and is what the company calls the “fastest licensable RISC-V processor IP core.” The new modular 64-bit CPU core uses a 13-stage out-of-order core pipeline and advanced branch prediction to deliver a 40% performance improvement per clock cycle over its predecessor, the P550.

Both Xilinx and Intel are prioritizing the data center, which is driving programmable logic R&D towards high-performance sockets. In some cases, these trends are influencing the evolution of applications in the embedded electronics sector as well. But there are countless applications for which power consumption and heat dissipation are just as, if not more, important than performance. These use cases are often found at the edge and encompass everything from sensor hubs to display controllers to image processing and AI inferencing platforms.

The RDK-852 is a reference design for a 200 W three-phase inverter that delivers 200 W of continuous output power and a continuous current of .62 Amps per phase to brushless DC motors. Therefore, the kit comes with field-oriented control capabilities while also being incredibly efficient.

True or false? Edge devices, a web-based development environment, and a network connection are all you need to update embedded software as frequently as you want.