Abstract:In-memory computing (IMC) paradigm has attracted extensive attention for future electronics to overcome the bottleneck and memory wall problem in the von Neumann systems. Nonvolatile logic based on resistive random-access memory (RRAM) is a promising route to implement such architecture. This paper presents the circuits and computing methodology to implement functionally complete Boolean logic and arithmetic block co-design using the CMOS-compatible 2-transistor-2-RRAM (2T2R) structure with reversely connected bipolar RRAM pairs. Arbitrary logic functions for two and multiple operands could be implemented in one-step operation, after which the computation results are in situ stored as the nonvolatile resistive states of the RRAM. In this design, the logic functionality is achieved by simply applying external operational voltages and/or corresponding RRAM initial states to the proposed 2T2R units. Compared with the traditional CMOS designs where the circuit topologies change with logic functions, the proposed 2T2R chain has a universal design with repeated unit blocks, independent of the logic functions. The highly regular and symmetric circuit structure makes it easy for design, integration, and fabrication. The proposed computing scheme/structure is intrinsic, clean, and efficient for IMC applications and presents superior performance in speed and area. The result of the study could build a technology cell library that can be potentially used as input to a technology-mapping algorithm for processing in-memory applications.

Time-domain computing for Boolean logic using STT-MRAM

Complementary Series-connected STT-MTJ for Time-based Computing-in-Memory

Efficient Time-Domain In-Memory Computing Based on TST-MRAM

Design of an Area-Efficient Computing in Memory Platform Based on STT-MRAM

Implementation of 16 Boolean Logic Operations Based on One Basic Cell of Spin-Transfer-torque Magnetic Random Access Memory

Computing-in-Memory Paradigm Based on STT-MRAM with Synergetic Read/Write-Like Modes

A Novel In-memory Computing Scheme Based on Toggle Spin Torque MRAM.

In-memory Processing Paradigm for Bitwise Logic Operations in STT-MRAM

A 9T-SRAM Based Computing-in-memory with Redundant Unit and Digital Operation for Boolean Logic and MAC

Boolean Logic Operations Based on Four-Terminal Magnetic Tunnel Junction for Computing in Memory

Low Power Computing Using STT-MRAM

Modified Peripheral MRAM Sensing for In-memory Boolean Logic

A 9T-SRAM In-Memory Computing Macro for Boolean Logic and Multiply-and-accumulate Operations

Functionally Complete Boolean Logic and Adder Design Based on 2T2R RRAMs for Post-CMOS In-Memory Computing

A High Throughput In-MRAM-Computing Scheme Using Hybrid P-Sot-mtj/gaa-cntfet

A Robust Dual Reference Computing-in-Memory Implementation and Design Space Exploration Within STT-MRAM

Reconfigurable Bit-Serial Operation Using Toggle SOT-MRAM for High-Performance Computing in Memory Architecture

Time-Domain Computing in Memory Using Spintronics for Energy-Efficient Convolutional Neural Network

A CFMB STT-MRAM-Based Computing-in-Memory Proposal with Cascade Computing Unit for Edge AI Devices

A novel dual-reference sensing scheme for computing in memory within STT-MRAM

Cross-coupled 4T2R Multi-Logic In-Memory Computing Circuit Design