Human somatosensory system based on sensor-memory integrated technology

Abstract

As a represent of the artificial neural network (ANN) for incorporating sensing-functions and memory functions into one system to realize the highly miniaturizated and highly integrated devices or systems, artificial sensory system (ASS) has a far-reaching influence on energy convention, electronic information industry, precise instrument, sensing, and automation engineering. The artificial sensory systems have made considerable progresses from low degree integrations to highly advanced sophisticated integrations, form single-modal perceptions to multimode-fused perceptions. Furthermore, the multimodal information is processed with a centralized and sequential computing architecture. Therefore, it is imperative to solve the issues of large hardware area, power consumption, and communication bandwidth, during the processes that multimodal sensing signals converted into digital mode before they can be processed by a digital processor. Therefore, deepening the research of sensory integration is of the great importance. In this review, we briefly introduce fundamental knowledge of the memristor mechanism, give a brief introduction of a myriad of sensors, enumerate representative human somatosensory system, and elucidate the relationship between the properties of the memeristor device and the structure, electronic character of the sensors, prospects and challenges of sensor-memory integrated technology.