Adaptive MIMO Radar Architecture for Energy-Efficient Wireless Sensing in the D-Band

TL;DR

This paper introduces an adaptive MIMO radar architecture for energy-efficient wireless sensing in the D-band, utilizing a reconfigurable array, dynamic power scaling, and algorithm selection to optimize performance and power consumption.

Contribution

It proposes a novel adaptive MIMO radar system with a reconfigurable array, dynamic SNR and power scaling, and optimized DOA algorithm selection for energy-efficient D-band sensing.

Findings

Reconfigurable array reduces power consumption and computational complexity.

Power scaling laws improve resource allocation based on target distance.

Optimal DOA algorithm depends on the number of array elements.

Abstract

The D-band offering an untapped wide bandwidth is promising for high data rate communication and high-resolution wireless sensing. However, these potentials are hindered by the low performance and energy efficiency of the D-band circuits and systems. We present an adaptive multi-input multi-output (MIMO) radar architecture for energy-efficient wireless sensing in the D-band, leveraging a reconfigurable 2D array of radar transceiver front-ends, a scaling approach for the receiver (RX) signal-to-noise ratio (SNR) and the transmitter (TX) output power ($P_{\rm TX}$) with target distance, and dynamic selection of the direction-of-arrival (DOA) estimation algorithm. The reconfigurable radar array, providing an adaptive radar resolution, enhances the energy efficiency by reducing power consumption in the radar RF front-end and lowering the computational complexity in the radar back-end. The RX SNR and the TX output power are scaled with the distance as ${\rm SNR} \propto d^{-p}$ and $P_{\rm TX} \propto d^{4-p}$, where $0 < p < 4$, leading to more efficient resource allocation in varying target distance conditions. Additionally, DOA estimation results using MUSIC and MVDR algorithms indicate that the optimum algorithm, in terms of the accuracy and computational complexity, should be selected based on the number of radar array elements. Furthermore, we develop a hardware model for the MIMO radar RF front-end to evaluate the power consumption of the TX, RX, and local oscillator (LO) distribution network. It is shown that the power consumption of the LO distribution network, which can dominate the power consumption for a large MIMO radar, can be minimized through a distribution strategy for LO amplifiers employed for compensating passive losses. Performance of the adaptive MIMO radar is evaluated in the free-space and the through-wall indoor sensing scenarios in the D-band.