Lousy processing increases energy efficiency in massive MIMO systems

Abstract

Massive MIMO (MaMIMO) is a key technology for 5G wireless communication, enabling large increase in both spectral and energy efficiency at the same time. Before it can be deployed, it is important to find efficient implementation strategies. Because of the many antennas, an essential part of decreasing complexity, and further improving energy efficiency, is optimization of the digital signal processing (DSP) in the per-antenna functions. Assuming an orthogonal frequency-division multiplexing (OFDM) based MaMIMO system, this paper explores coarse quantization in the per-antenna digital transmit filters and inverse fast Fourier transforms (IFFTs) and evaluates it in terms of performance and complexity savings. Results show that DSP complexity can be greatly reduced per-antenna, and therefore significant power savings can be achieved, with limited performance degradation. More specifically, when going towards MaMIMO and therefore increasing the number of antennas from 8 to 64, it is possible to reduce the complexity in each transmit filter by 55%. Also, when using 6 bits to represent the input signal and 6 bits for the filter coefficients, this results in an SNR degradation of less than 0.5 dB compared to floating-point performance. Consequently, we conclude that the overall system energy greatly benefits from lousy per-antenna processing.