GreenWaves Technologies

Based in Grenoble, France, GreenWaves Technologies operates as a fabless semiconductor company focused on designing ultra-low-power processors for the edge of the Internet of Things (IoT). The company was established in late 2014 by Loic Lietar, who serves as CEO, and a team of industry veterans including Denis Mestdagh and Eric Flamand. The founding team aimed to address the growing need for significant computing power in battery-operated devices without compromising energy efficiency. Initially, the company explored high-data-rate wireless connections but pivoted to focus on co-founder Eric Flamand's concept for an ultra-low-power AI processor, which gained more market traction.

GreenWaves' core business revolves around the design and sale of application processors based on the open-source RISC-V instruction set architecture. As a fabless company, it designs the chips and outsources the manufacturing to foundries like GlobalFoundries. Its revenue is generated from the sale of these processors to original equipment manufacturers (OEMs) and product designers. The company targets clients in markets that require sophisticated AI and signal processing capabilities in power-constrained environments, such as hearables, wearables, medical devices, and industrial IoT sensors.

The company's product line includes the GAP processor family, notably the GAP8 and its successor, GAP9. GAP8 was introduced as one of the first RISC-V based processors to enable AI in battery-powered IoT applications. The subsequent GAP9 processor offers a substantial improvement, delivering up to 50 GOPS of computing power at a consumption of just 50mW, a fivefold reduction in power usage compared to its predecessor. This efficiency allows for the processing of rich data sources like images, sounds, and vibrations directly on the device. GAP9 incorporates a ten-core architecture, support for transprecision floating-point arithmetic, and enhanced security features like hardware cryptography. These processors are designed to run complex machine learning algorithms, enabling features like neural network-based noise filtering, voice control, and object detection on devices with limited battery life.

Keywords: fabless semiconductor, RISC-V, edge AI, IoT processors, ultra-low-power, signal processing, hearables, wearables, machine learning, embedded solutions