Feature: Medical


Figure 8: A KWS20 case study application showed that a higher clock speed resulted in lower energy consumption due to shorter loading times, particularly when only the Arm processor was used. (Image source: Analog Devices)


configure and control field instruments through the HART modem’s SPI connection to the Arm Cortex-M4. Along with documentation and


other development resources, Analog Devices offers the MAX32675EVKIT MAX32675C evaluation kit to help speed testing and prototype development.


Meeting emerging requirements for edge AI To build eff ective applications in a growing number of areas, developers must implement edge devices that effi ciently execute AI algorithms for intelligent time- series processing or recognition of objects, words or faces. Analog Devices’ MAX78000 is designed specifi cally to support these capabilities while maintaining the fundamental requirement for low power consumption. Like the ultra-low-power


microcontrollers described earlier, the 36 March 2025 www.electronicsworld.co.uk


MAX78000 (Figure 7) builds on an Arm Cortex-M4 with an FPU processor, 512 Kbytes of fl ash, 128 Kbytes of SRAM and 16 Kbytes of cache to meet the core application execution requirements. To support edge AI solutions, the MAX78000 augments its processing subsystem with a pair of additional resources, including: • A 32-bit RISC-V coprocessor that provides the system with ultra-low power consumption signal processing capabilities


• An integrated hardware-based convolutional neural network (CNN) accelerator to meet the emerging demand for edge AI devices T e MAX78000 supports the same low-


power operating modes and power-down mode described earlier for the MAX32655, with the CNN remaining available through sleep and low-power modes, state retention in micro power, standby and backup modes and a power-down mode for use during


end-product storage and distribution. As with the other microcontrollers


discussed here, the MAX78000’s high level of integration helps developers meet the requirements for a minimal bill of materials (BOM) and end-product size. With the device’s integrated ADC and signal processing capabilities, developers can use the MAX78000 with few additional components to quickly implement edge AI applications such as keyword spotting (KWS) or face identifi cation (FaceID). Besides simplifying edge AI


implementation, the MAX78000’s combination of multiple power modes, dual processors and hardware-based CNN allows developers to achieve fast inference speed with minimal power consumption. Analog Devices engineers closely examined performance in a study of power-optimised applications on the MAX78000. As part of this study, the engineering team measured energy


Page 1  |  Page 2  |  Page 3  |  Page 4  |  Page 5  |  Page 6  |  Page 7  |  Page 8  |  Page 9  |  Page 10  |  Page 11  |  Page 12  |  Page 13  |  Page 14  |  Page 15  |  Page 16  |  Page 17  |  Page 18  |  Page 19  |  Page 20  |  Page 21  |  Page 22  |  Page 23  |  Page 24  |  Page 25  |  Page 26  |  Page 27  |  Page 28  |  Page 29  |  Page 30  |  Page 31  |  Page 32  |  Page 33  |  Page 34  |  Page 35  |  Page 36  |  Page 37  |  Page 38  |  Page 39  |  Page 40  |  Page 41  |  Page 42  |  Page 43  |  Page 44  |  Page 45  |  Page 46  |  Page 47  |  Page 48  |  Page 49  |  Page 50