Polarization-based zero forcing with channel estimation

A robust, low-complexity polarization-based zero forcing (ZF) filtering technique is proposed for the suppression of asynchronous, co-channel interference (ACCI) that cannot be resolved from the desired signal in angle. Suppression is achieved in the polarization-frequency dimension and requires either channel estimates or relative channel estimates for the orthogonally-polarized components of the interference signal. We consider the impact of background co-channel interference (BCCI) during channel estimation of the ACCI signal and show that the suppression performance of the ZF filter degrades with the quality of the ACCI channel estimates, i.e., the performance is approximately limited to the power ratio between the ACCI and the BCCI signals. Methods to counteract the impact of the BCCI are considered that exploit the correlation inherent in the polarization-frequency domain. One approach involves smoothing in the frequency domain, which is shown to yield improved channel estimates for ZF filtering and cancellation performance gains. Another strategy is also considered that monitors ACCI signal depolarization induced by the BCCI to identify the ACCI signal channel estimates with the least disturbance from the BCCI. The methods are demonstrated in a testbed using baseband signal generators, coherent receiver channels, and channel emulators with propagation impairment models derived from experimental testing.