The MAX78000 is a new breed of AI microcontroller built to enable neural networks to execute at ultra-low power and live at the edge of the IoT. This product combines the most energy-efficient AI processing with Maxim’s proven ultra-low power microcontrollers. The hardware-based convolutional neural network (CNN) accelerator enables even battery-powered applications to execute AI inferences.
The advanced system-on-chip design features an Arm Cortex-M4 with FPU CPU for efficient system control with an ultra-low-power deep neural network accelerator. The CNN engine has a weight storage memory of 442 kB, and can support 1-, 2-, 4-, and 8-bit weights (supporting networks of up to 3,5 million weights). The CNN weight memory is SRAM-based, so AI network updates can be made on the fly. The CNN engine also has 512 kB of data memory and its architecture is highly flexible, allowing networks to be trained in conventional toolsets like PyTorch and TensorFlow, then converted for execution on the MAX78000 using tools provided by Maxim.
In addition to the memory in the CNN engine, the MAX78000 has large on-chip system memory for the microcontroller core, with 512 kB flash and up to 128 kB SRAM. Multiple high-speed and low-power communications interfaces are supported, including I2S and a parallel camera interface.
The device is available in an 81-pin CTBGA (8 x 8 x 0.8 mm pitch) package.
Typical applications for the MAX78000 include audio processing (multi-keyword recognition, sound classification, noise cancellation), facial recognition, object detection and classification, and time-series data processing such as heart rate signal analysis, multi-sensor analysis, and predictive maintenance.
Cost-effective microcontroller series Altron Arrow
DSP, Micros & Memory
The STM32C5 series from STMicroelectronics delivers an excellent balance of performance, efficiency, and affordability for embedded designs that require more capability without increasing bill of materials cost.
Read more...Understanding two key tools for cleaner serial data Altron Arrow
Editor's Choice Analogue, Mixed Signal, LSI
Understanding how pre-emphasis and equalisation works, and when to use one over the other, is critical when designing reliable high-speed systems.
Read more...Ultra-low jitter clocks Altron Arrow
Telecoms, Datacoms, Wireless, IoT
Skyworks has introduced a new family of ultra-low jitter programmable clocks designed to meet the increasing demands of next-gen connectivity.
Read more...Efficient Bluetooth SoC Altron Arrow
Telecoms, Datacoms, Wireless, IoT
The EFR32BG29 wireless SoC from Silicon Labs is a highly efficient, high memory, low-power, and ultra compact SoC designed for secure and high-performance wireless networking for IoT devices.
Read more...Compact edge platform for AI iCorp Technologies
AI & ML
Built around the Qualcomm QCS6490 octa-core SoC, Quectel’s QuecPi Alpha delivers up to 12 TOPS of on-device AI performance.
Read more...Axon NPU powers smarter edge RF Design
AI & ML
The nRF54LM20B from Nordic Semiconductor is an ultra-low-power wireless SoC that combines advanced edge AI capabilities with robust radio connectivity and rich peripheral support.
Read more...FPGA platform designed for the edge
AI & ML
Sundance Multiprocessor Technology has introduced the SMT135-C, an Efinix Titanium Ti135 evaluation platform designed for edge computing, real-time control, and vision-centric workloads.
Read more...AI-controlled swarms: Algorithmic warfare Technews Publishing
AI & ML
The rapid proliferation of Unmanned Aerial Systems (UAS), ranging from hobbyist quadcopters to sophisticated munitions-carrying military drones, has fundamentally altered the security landscape and come to the fore with the current war in the Middle East.
While every effort has been made to ensure the accuracy of the information contained herein, the publisher and its agents cannot be held responsible for any errors contained, or any loss incurred as a result. Articles published do not necessarily reflect the views of the publishers. The editor reserves the right to alter or cut copy. Articles submitted are deemed to have been cleared for publication. Advertisements and company contact details are published as provided by the advertiser. Technews Publishing (Pty) Ltd cannot be held responsible for the accuracy or veracity of supplied material.
printer friendly version