Adaptive Interference Suppression of Phase-Only Thinned Arrays via Convex Optimization

For antenna arrays, the conventional beamformer (CB) is perhaps the simplest kind of beamforming techniques with only phase shifters connected to the antenna elements. However, the output signal to interference and noise ratio (SINR) performance is usually quite away from the optimum. Besides, as the number of antenna elements increases, the hardware cost and power consumption are still too high for many applications. In this article, based on the phase-only CB, we propose different ways for thinning this kind of antenna arrays, which can be efficiently solved via common convex-optimization solvers. Specifically, the first-derived phase-only thinned array can largely increase the output SINR with fewer active antenna elements than the CB. The second one can keep the output SINR performance as that of the CB with a significantly smaller number of active antenna elements than that required by the CB. The third criterion is to add an additional constraint on the number of active antenna elements while maximizing the output SINR, which may benefit the hardware planning. Simulation results confirm the feasibilities of these methods, and the above-stated tasks are all fulfilled. Therefore, the proposed phase-only thinned arrays have the flexibility to provide various levels of array performance with reduced hardware resources.