Low-Complexity Reconfigurable and Intelligent Ultrawideband Angular Sensing

Abstract

Paradigms such as multiantenna transceivers, spectrum sharing, and beamforming are critical to meet the desired throughput, reliability, latency, spectrum, and energy efficiency constraints in next-generation wireless networks. To enable these paradigms, base stations are undergoing major overhaul with complex digital signal processing, reconfigurable multistage architectures, as well as wideband sensing capability. In this article, we focus on the low-complexity multiantenna BS receiver to accomplish ultrawideband angular sensing. The proposed receiver is based on novel reconfigurable and intelligent sub-Nyquist sampling architecture. Here, reconfigurability allows the receiver to select and digitize noncontiguous frequency bands, while online learning-based intelligence allows the receiver to learn spectrum statistics and choose the frequency bands for sensing in order to maximize the throughput. Unlike the existing approaches, we show that the proposed approach does not need any prior knowledge of spectrum statistics. The efficacy of the proposed approach over state-of-the-art approaches is validated via hardware complexity as well as simulation results. We show that the proposed approach offers lower hardware complexity and achieves up to 23% lower estimation error than existing state-of-the-art methods.