In a One-Bit Rush: Low-Latency Wireless Spectrum Monitoring with Binary Sensor Arrays

TL;DR

This paper presents a low-latency wireless spectrum monitoring method using binary sensor arrays, employing a sequential hypothesis test with a reduced model to efficiently detect wireless sources with minimal delay.

Contribution

It introduces a likelihood-based sequential testing framework for binary sensors in spectrum monitoring, with a novel reduced model and performance analysis.

Findings

Achieves low-latency detection with minimal samples

Demonstrates effectiveness through simulations

Provides analytical performance insights

Abstract

Detecting the presence of a random wireless source with minimum latency utilizing an array of radio sensors is considered. The problem is studied under the constraint that the analog-to-digital conversion at each sensor is restricted to reading the sign of the analog received signal. We formulate the resulting digital signal processing task as a sequential hypothesis test in simple form. To circumvent the intractable probabilistic model of the multivariate binary array data, a reduced model representation within the exponential family in conjunction with a log-likelihood ratio approximation is employed. This approach allows us to design a likelihood-based sequential test and to analyze its analytic performance along Wald's classical arguments. In the context of wireless spectrum monitoring for satellite-based navigation and synchronization systems, we study the achievable processing latency, characterized by the average sample number, as a function of the binary sensors in use. The practical feasibility and potential of the discussed low-complexity sensing and decision-making technology is demonstrated via simulations.