Towards an Appropriate Receiver Beamforming Scheme for Millimeter Wave Communication: A Power Consumption Based Comparison

TL;DR

This paper compares the power consumption of analog, digital, and hybrid beamforming schemes in millimeter wave receivers, focusing on ADC power, and identifies conditions where digital beamforming can be more power-efficient.

Contribution

It provides a detailed power consumption analysis of different beamforming schemes considering ADC power, antenna count, bandwidth, and ADC resolution, guiding optimal scheme selection.

Findings

Digital beamforming can be more power-efficient than analog or hybrid under certain conditions.

Power consumption varies significantly with the number of antennas, ADC bits, and bandwidth.

Choice of beamforming scheme depends on the signal-to-noise ratio regime.

Abstract

At millimeter wave (mmW) frequencies, beamforming and large antenna arrays are an essential requirement to combat the high path loss for mmW communication. Moreover, at these frequencies, very large bandwidths are available to fulfill the data rate requirements of future wireless networks. However, utilization of these large bandwidths and of large antenna arrays can result in a high power consumption which is an even bigger concern for mmW receiver design. In a mmW receiver, the analog-to-digital converter (ADC) is generally considered as the most power consuming block. In this paper, primarily focusing on the ADC power, we analyze and compare the total power consumption of the complete analog chain for Analog, Digital and Hybrid beamforming (ABF, DBF and HBF) based receiver design. We show how power consumption of these beamforming schemes varies with a change in the number of antennas, the number of ADC bits (b) and the bandwidth (B). Moreover, we compare low power (as in [1]) and high power (as in [2]) ADC models, and show that for a certain range of number of antennas, b and B, DBF may actually have a comparable and lower power consumption than ABF and HBF, respectively. In addition, we also show how the choice of an appropriate beamforming scheme depends on the signal-to-noise ratio regime.