Session: MNS-01 MEMS/NEMS Power Sources, Sensors and Actuators, and Computing

Paper Number: 71659

Start Time: August 17, 11:10 AM

71659 - Multi-Modeshape Reservoir Computing Using a Continuous MEMS Microbeam 

Neuro-inspired (neuromorphic) computing is a computing approach that draws inspiration from biological neuronal systems. This approach was founded due to the shortcomings of traditional digital computing schemes, namely their high power consumption and their latency problems. Among the neuromorphic computing approaches, delay-based Reservoir computing (RC) offers great potential in engineering time-series problems, especially when applied in hardware due to its low computational power and its compact nature. However, this approach suffers from a large computational delay because of the serial probing of virtual nodes. To address this disadvantage, this paper presents the use of a continuous MEMS arch for Delay-based RC. This novel approach reduces the computational delay by using fewer virtual nodes through maintaining sufficient virtual node coupling and nonlinear complexity by using a nonlinear geometry of a MEMS microbeam and exciting multiple modeshapes to facilitate inherent coupling within the MEMS device. As a demonstration, we show that a single MEMS arch is capable of performing a binary waveform classification task of a multifrequency square-and-triangle waveform problem with a success rate > 96% using only 10 virtual nodes compared to 40 virtual nodes in a typical implementation. The reduction in the number of virtual neurons is achieved by biasing the MEMS device using an AC source around its second modeshape.

Presenting Author: Mohammad H Hasan Columbus State University

Authors:

Mohammad H. Hasan Columbus State University
Fadi Alsaleem University of Nebraska - Lincoln
Mohammad H Hasan Columbus State University