Extremely-Fast, Energy-Efficient Massive MIMO Precoding with Analog RRAM Matrix Computing

TL;DR

This paper introduces an analog matrix computing approach for massive MIMO precoding, achieving ultra-fast, energy-efficient solutions that outperform traditional digital methods by two orders of magnitude.

Contribution

It pioneers the use of analog matrix computing for complex-valued massive MIMO precoding, including new circuit designs and processing techniques.

Findings

Zero-forcing precoding solved within 20 ns for 16x128 MIMO

Achieves 50x energy efficiency improvement

Outperforms conventional digital approaches by 100x in speed

Abstract

Signal processing in wireless communications, such as precoding, detection, and channel estimation, are basically about solving inverse matrix problems, which, however, are slow and inefficient in conventional digital computers, thus requiring a radical paradigm shift to achieve fast, real-time solutions. Here, for the first time, we apply the emerging analog matrix computing (AMC) to the linear precoding of massive MIMO. The real-valued AMC concept is extended to process complex-valued signals. In order to adapt the MIMO channel models to RRAM conductance mapping, a new matrix inversion circuit is developed. In addition, fully analog dataflow and optimized operational amplifiers are designed to support AMC precoding implementation. Simulation results show that the zero-forcing precoding is solved within 20 ns for a 16x128 MIMO system, which is two orders of magnitude faster than the conventional digital approach. Meanwhile, the energy efficiency is improved by 50x.