Sub 100 nW Volatile Nano-Metal-Oxide Memristor as Synaptic-Like Encoder of Neuronal Spikes

Abstract

Advanced neural interfaces mediate a bioelectronic link between the nervous system and microelectronic devices, bearing great potential as innovative therapy for various diseases. Spikes from a large number of neurons are recorded leading to creation of big data that require online processing under most stringent conditions, such as minimal power dissipation and on-chip space occupancy. Here, we present a new concept where the inherent volatile properties of a nano-scale memristive device are used to detect and compress information on neural spikes as recorded by a multielectrode array. Simultaneously, and similarly to a biological synapse, information on spike amplitude and frequency is transduced in metastable resistive state transitions of the device, which is inherently capable of self-resetting and of continuous encoding of spiking activity. Furthermore, operating the memristor in a very high resistive state range reduces its average in-operando power dissipation to less than 100 nW, demonstrating the potential to build highly scalable, yet energy-efficient on-node processors for advanced neural interfaces.

Mitigating noise effects in volatile nano-metal oxide neural detector

“…In this work devices were accessed directly on-wafer via a probe card [13]. The electrical characterisation of devices was carried out in two stages: (a) the pristine devices were electroformed using pulses of positive polarity in the range of +4 V -+6 V [8] and (b) the devices are biased with volatility characterisation algorithm to determine the safe volatile region of operation of devices (i.e. up to approximately -4 V) [14].…”

“…Recently, we proposed a new spike-detection approach exploiting the volatile properties of emerging metal-oxide resistive switching memory devices [8], commonly known as 'memristors' [9]. When operated in volatile region, the devices tend to undergo metastable memory state transitions above an inherent threshold following which they inherently relax to their initial resistive state (RS) region [10].…”

“…When operated in volatile region, the devices tend to undergo metastable memory state transitions above an inherent threshold following which they inherently relax to their initial resistive state (RS) region [10]. Using nano devices, we demonstrated that the spiking events can be encoded into transient RS changes of device-under-test (DUT) reducing the power dissipated to less than 100 nW per channel [8]. However, the study did not examine the effect of subthreshold events and operational parameters such as signal amplification and offset on the response of memristive devices in volatile region.…”

Compact memristor‐based ultra‐wide band chirp pulse generator

“…Since that time, memristor device has gained an intensive attention due to its unique characteristics. Memristors have shown revolutionary potential in various fields, including neural networks, reconfigurable logic, nonvolatile memory, chaotic circuits, cryptography, circuit modeling, and programmable analog circuits . The memristor's resistance (memristance) can be easily programmed by applying either DC voltage or voltage pulses to achieve the required resistance.…”

“…Memristors have shown revolutionary potential in various fields, including neural networks, reconfigurable logic, nonvolatile memory, chaotic circuits, cryptography, circuit modeling, and programmable analog circuits. [6][7][8][9][10] The memristor's resistance (memristance) can be easily programmed by applying either DC voltage or voltage pulses to achieve the required resistance. Only two transistors alongside the memristor are needed to switch between programming and operating modes.…”