Memristor-based in-situ convolutional strategy for accurate braille recognition

基于忆阻器的原位卷积策略用于精确盲文识别

Signal processing has entered the era of big data,and improving processing efficiency becomes crucial.Traditional computing architectures face computational efficiency limitations due to the separation of storage and computation.Array circuits based on multi-conductor devices enable full hardware convolutional neural networks(CNNs),which hold great potential to improve computational efficiency.However,when processing large-scale convolutional computations,there is still a significant amount of device redundancy,resulting in low computational power consumption and high computational costs.Here,we innovatively propose a memristor-based in-situ convolutional strategy,which uses the dynamic changes in the conductive wire,doping area,and polarization area of memristors as the process of convolutional operations,and uses the time required for conductance switching of a single device as the computation result,embodying convolutional computation through the unique spiked digital signal of the memristor.Our strategy reasonably encodes complex analog signals into simple digital signals through a memristor,completing the convolutional computation at the device level,which is essential for complex signal processing and computational efficiency improvement.Based on the implementation of device-level convolutional computing,we have achieved feature recognition and noise filtering for braille signals.We believe that our successful implementation of convolutional computing at the device level will promote the construction of complex CNNs with large-scale convolutional computing capabilities,bringing innovation and development to the field of neuromorphic computing.

大数据时代下,提高信号处理效率至关重要.由于传统计算架构的计算设备中存储单元和计算单元相互分离,未来将面临着计算效率的限制.基于多电导态器件的阵列电路可实现全硬件卷积神经网络(CNNs),具备提高计算效率的潜力.然而,在处理大规模卷积计算时,仍存在大量器件冗余,导致计算功耗低、计算成本高.本文创新性地提出了一种基于忆阻器的器件级原位卷积策略:以忆阻器的导电丝、掺杂面积和极化面积等的动态变化作为卷积运算过程,单个器件的电导切换所需的时间作为计算结果,通过忆阻器独特的尖峰数字信号体现卷积计算.通过忆阻器将复杂的模拟信号合理地编码为简单的数字信号,成功在单个器件上完成了卷积计算,这对于复杂信号处理和计算效率提高至关重要.在器件级原位卷积计算的基础上,本文进一步实现了盲文信号的特征识别和噪声过滤.本文所提的基于单个忆阻器的器件级原位卷积计算,将推动具有大规模卷积计算能力的复杂CNNs的构建,促进神经形态计算领域的创新和发展.