Planar Near-Field Array Antenna Diagnostics for Production Testing Using Compressive Sensing and a Fixed Sparce Sensing Array

In massive multiple input multiple output (MIMO) arrays, Compressive Sensing (CS) has been proposed to rapidly verify the array’s excitation in a production environment. All follow the general approach of creating the sparsity needed for CS by subtracting the measured far-field (FF) or near-field (NF) of the test array from that of a “gold standard” array measured under identical conditions. In several previous papers the authors have applied CS to planar near-field (PNF) measurements with a view to building a compact test facility well suited to the production environment for these antennas. In this article, we identify a fundamental flaw in the creation of the difference pattern between a gold standard antenna and the test antenna, in that the electrical path length difference between probe and gold standard antenna and that of probe to test antenna must be no more than <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$0.002\lambda $ </tex-math></inline-formula> or 0.7°. Even if mechanical tolerance can achieve this, thermal drift in the radio frequency (RF) subsystem will easily reach the 0.7° figure. This article then describes an analysis process that overcomes this problem, restoring the CS process to very nearly the levels of performance described in our previous publications and those of other workers. To achieve this, we have used a fixed set of NF probes located randomly over a region confined to the central part of the NF measurement plane, along with a PIN switch matrix to offer a very rapid and very consistent RF measurement. We have extensively simulated a massive MIMO array of <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$8\times 24$ </tex-math></inline-formula> elements and found that just 25 NF probes can detect up to 2% failure rate. In addition, we present a technique to improve the accuracy of the array excitation reconstruction by using several partial excitations of the array. In this work we consider both array element amplitude and phase reconstruction performance and demonstrates that it is possible to detect a single element, phase only, fault of just 22.5°.