Adaptive blind equalization for MIMO hybrid filter bank A/D converters

The hybrid filter bank (HFB) A/D converters permits implementing both high speed and high resolution A/D converters. However, its design requires an accurate knowledge of the analog filter bank coefficients, which is difficult to have due to analog imperfections. This paper proposes an adaptive blind equalization method for MIMO HFB A/D converters, which is able to adaptively estimate nonstationary input signals and analog filter bank coefficients. The estimated parameters are used to reconstruct the samples in a least mean squares (LMS) sense. Simulations are presented showing the good performance of the proposed method with 88 dB Distortion-to-Aliasing Ratio (DAR) and 14.3 bit resolution, which makes the MIMO HFB realistic for cognitive radio applications.