Fig. 2: Overview of in-memory nonlinear ADC. | Nature Communications

Fig. 2: Overview of in-memory nonlinear ADC.

From: Efficient nonlinear function approximation in analog resistive crossbars for recurrent neural networks

Fig. 2: Overview of in-memory nonlinear ADC.The alternative text for this image may have been generated using AI.

a The concept of traditional ramp-based ADC. b The schematic and timing of in-memory computing circuits with embedded nonlinear activation function generation. c The Inverse of the sigmoid function illustrates the shape of the required ramp voltage. d The value of each step of the ramp voltage Vramp denoted by ΔVk is proportional to memristor conductances Gadc,k used to program the nonlinear ramp voltage. The desired conductances for a 5-bit implementation of a sigmoid nonlinear activation is shown. e Comparison of used cell numbers between 5-bit and 4-bit in-SRAM with 5-bit in-RRAM nonlinear function. The RRAM-based nonlinear function has an approximation error between the two SRAM-based ones due to write noise while using a smaller area due to its compact size.

Back to article page