A CMOS Distributed Sensor System for High-Density Wireless Neural Implants for Brain-Machine Interfaces

Abstract

Current state-of-the-art Brain-Machine Interfaces (BMIs) rely on invasive “passive” microelectrode technologies, which are prohibitively challenging to scale beyond several hundred channels due to physical constraints. Further performance enhancement in BMIs relies on the ability to develop implantable sensor technologies that would be scalable to thousands of channels without significant biological overhead. In this work., we describe prototype testing of a distributed sensor system of CMOS “Neurograins,” which provide a high density network of autonomous implantable neural sensors. These Neurograins are wirelessly powered using near-field RF at ~1 GHz and at densities up to 250 chips/cm2, A 3-coil system is used to enhance the wireless signal transfer function. Telemetry is accomplished using RF backscatter with TDMA networking at a data rate of 10 Mbps, and Bit Error Rates (BER) <0.1% (measurement limit). An asynchronous periodic, packetized multiple access network is demonstrated for initial Neurograin arrays. Benchtop measured results incorporate a physiologic model for RF attenuation - a “Brain phantom” - to accurately mimic the biomedical scenario.