Diagnosis of Antenna Array with Undefined Mutual Coupling via Optimization

Failure of element (s) in Antenna array degrades the performance of the radiation pattern, and consequently affect mobile communication systems. Hence, diagnosis of antenna array is an important operation whenever element (s) failure in the array occur. There are various diagnostic techniques found in literature that employ compressive sensing. Conversely, the techniques are based on easy formulation of array factor with no incorporation of mutual coupling existing between the radiators. This article shows how this deficiency lead to defective and bad diagnosis when there is presence of mutual coupling using port-level coupling matrix and average embedded antenna pattern. Furthermore, the element excitations are optimized to reduce the effect of mutual coherence of system measurement matrix, causing reduced measurements required for effective fault detection. Numerical simulation and experimental results demonstrate how the incorporation of mutual coupling generates an adequate and reliable array diagnosis. Finally, the implementation of the proposed framework via measurement probe in space shows motivating results for its practical applications.

I. Introduction

Present and upcoming technology use larger number of elements in the active arrays. For instance, large number of elements is employed in massive MIMO (multiple input multiple output), full MIMO systems, and telecommunication devices that employ sophisticated arrays. Hence, the demand for a reliable antenna array diagnosis is an inevitable task to rectify the distorted radiation characteristics because of element (s) failure [1]–[6]. In addition, fault diagnosis is important in 5G wireless systems, where a very large number of elements are needed to satisfy the required reconfigurability and high radiation behaviour [2]. However, the more the number of elements in beamforming configuration, the more the probability of failed element (s). Therefore, an effective and highly reliable fault diagnosis method remains important, because replacement operation and manual dismantling take a lot of time, costly, and not feasible in satellite communications [2], [4]. Compressive sensing (CS) technique has been adapted to fault diagnosis in antenna arrays, because the number of failed radiators is assumed and always smaller than the total number of radiators in the antenna array.