Power Scalable Angle of Arrival Estimation Using Pilot Design With Orthogonal Subsequences

Abstract

Energy efficient array processing is critical to implement feasible solutions for directional communication in a mmWave channel. MmWave channels are highly susceptible to blockage and require frequent angle of arrival (AoA) estimation. An AoA estimation solutions with a fully digital architecture offers a low latency, high performance and flexible solution suitable for the stringent requirements of <b xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">5</b> G. However, the large number of high speed converters in a digital receiver are the dominant power consuming elements. Alternative analog or hybrid architectures use fewer high speed converters, but require sweeping measurements to estimate AoAs over the angular space, and thus adds latency to the estimation process. In this paper, we present a variable rate sub-Nyquist decoupling solution that leverages pilot design. The pilot's subsequence properties allow decoupling the source waveforms at fractions of the Nyquist rate. We leverage this concept to scale power consumption by the converters. We preprocess the received signals at the antenna array with the variable rate sub-Nyquist decoupling algorithm and use a few well known digital estimators for AoA estimation including DEML, KR-MUSIC based two level nested array and coprime filter bank. In addition to scalable power consumption, our research indicates some other benefits of the decoupling, including reduced complexity algorithm implementation and improved performance estimation for the non maximum likelihood estimators.