Cornell University Home Page

Allerton 1998 - Monticello, IL, 23-25 Sep 1998

Blind Adaptive Cross-Pole Interference Cancellation Using Fractionally-Spaced CMA

Wonzoo Chung, J. Treichler and C. R. Johnson, Jr.

Abstract

Cross-pole interference (CPI) cancellation is a specific type of source separation problem in a high-rate digital communication system which utilizes dual polarized carriers in order to achieve frequency reuse. In this paper, we analyze a fractionally-spaced (FS) blind adaptive cross-pole interference (CPI) cancellation scheme using the constant modulus algorithm (CMA). First, we study zero forcing solutions and minimum mean-squared error solutions of FS CPI equalizer on a cross coupled channel with a certain diversity. Second, we show close analogies between the FS-CMA CPI equalizer and the conventional FS-CMA blind adaptive equalizer. Thus existing results on FS-CMA can be applied to describe the important behavior of the FS-CMA CPI equalizer, for example, the agreement between CM-solutions and zero-forcing (ZF) solutions in the absence of channel noise and perfect modelling and the robustness of the FS-CMA CPI equalizer in more realistic situations (wchich include channel noise and undermodelling). Finally, we address the initialization of this blind algorithm by using a novel expression of the FS-CMA CPI cost function.