Abstract
Many array signal processing algorithms have been developed to reject interfering signals and, therefore, increase the quality of the communication link. Some of them are reference-based and some others are blind, i.e., without any side information. A specific type of deterministic blind beamforming algorithms, named self-reference, is treated in the paper. Self-reference beamforming is suitable for signals presenting a diversity or a redundancy structure. This type of signals exhibit the property of exact self-prediction that allows the definition of an error signal to be minimized. In the present paper, the time diversity present in the well-known DS-CDMA communication system is analyzed. Blind approaches are derived from the property of exact self-prediction arising from the implicit time redundancy encountered in the spreading codes, and from the explicit time redundancy that can also be induced by the transmitter. The proposed methods do not require, unlike most blind subspace-based techniques, any kind of spectral matrix decomposition, with the consequent reduction of complexity and computational burden. The presented methods were tested via simulations and compared to the classical time reference beamformer (TRB) in terms of SINR and BER, showing a performance slightly better than a TRB scheme with 10% of the bits as a training sequence.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call