Fast Adaptive Microwave Beamforming Using Array Signal Estimation

Abstract

A new perturbation technique is proposed which enables adaptive beamforming (ABF) in microwave domain using a single-port beamformer. In this technique, for a system with antennas, the weight vector is independently perturbed times to obtain correlated outputs. These outputs are then used to find a noisy approximation of the antenna array signal for the gradient vector estimation. The process of weight perturbations is performed faster than the Nyquist rate mainly to increase the temporal correlation of the consecutive antenna array signal samples and to lower the perturbation error. Performance of the proposed perturbation technique with the adaptive unconstrained least mean square (ULMS) algorithm is investigated for different channel scenarios. The ULMS algorithm with single-port beamformer converges in very high noise and interference levels and with a convergence speed close to that of the multi-port receiver. After convergence, both the single-port and the multiport beamformer algorithms achieve the same steady state signal to interference plus noise ratio (SINR) gain. Effects of weight quantization are also investigated for the single-port beamformer with the proposed perturbation technique.Index Terms-Adaptive beamforming (ABF), digital beamforming (DBF), microwave beamforming (MBF), perturbation algorithms.

Single-port direction of arrival estimation using adaptive null-forming

“…In this technique, the array output power is minimized adaptively by nulling the incoming directional sources. The array signal vector is estimated using the perturbation technique proposed in [5] by making the consequent antenna array signals temporally correlated which is used in the adaptive algorithm. Temporal correlation is generated through fast weighting of the array signals in the RF domain or by lowering the receiver bandwidth.…”

Modified Microwave Sampling Beamformer for Fast Weighting Control and Image Rejection

“…However, the MBF structure requires multiple phase shifters and amplitude control elements. In addition, because only the beamformer output is transferred to the processor, perturbation techniques [4]- [9] should be used to estimate the gradient vector or array covariance matrix. In [10] the microwave sampling beamformer (MSBF) has been proposed to simplify the implementation of the phase shifters and amplitude control elements in the MBF structure.…”

“…Adaptive calculation of the weight vectors is another issue which is not addressed in this paper. In fact, there are different algorithms in [4]- [9] which may be used to calculate the weight vectors. Therefore, in all simulations it is assumed that the weight vector is known in the receiver.…”