An adaptive matched filter that compensates for I,Q mismatch errors

An approach to adaptively match filter the I and Q components of complex-valued inputs consisting of a desired signal embedded in correlated external noise is presented. This approach is tolerant of I,Q mismatch errors, i.e., the external noise is effectively rejected and the desired signal enhanced in the presence of significant receiver I,Q errors. I,Q adaptive weighting removes many of the deleterious effects of I,Q quadrature detection imbalance, which can severely limit the adaptive matched filter (AMF) performance. However, for the I,Q AMF, the unknown desired signal´s initial phase complicates the design procedure and even for a reasonable design criterion, the AMF performance can fluctuate significantly as a function of this phase. An I,Q AMF technique whose performance is almost phase invariant is developed, and example of its utility is shown