An Efficient Reconfigurable CAM and IMC Framework using Multistate RRAM Devices

The rapid growth of data-intensive applications has created a strong need for high-speed and energy-efficient memory architectures, as conventional CMOS-based CAM and von Neumann systems face limitations such as high power consumption, poor scalability, and data transfer latency. To overcome these issues, this work proposes a reconfigurable framework that integrates Content Addressable Memory (CAM) and In-Memory Computing (IMC) using multistate Resistive Random Access Memory (RRAM). By utilizing the multi-level resistance states of RRAM, the proposed design enables both data storage and computation within the same memory array, supporting parallel search and processing operations. This reduces data movement, improves computational speed, lowers energy consumption, enhances storage density, and provides adaptable functionality for different computing tasks.