Overlap-Save Minimum Mean Square Error Block Linear Equalization and Block Noise-Prediction Algorithms

Ang FENG,Qinye YIN,Wei HAN
2010-01-01
Abstract:Focusing on the problem that the overlap-save method can not be directly applied in the conventional frequency-domain equalizations, two schemes of block linear equalization and block noise-prediction to overlap-save minimum mean square error (MMSE) are proposed. In the design of the linear equalizer, the effective taps are restricted within a predefined range to avoid inducing colored noises when overlap-save method is used. As a result, the distortion of bit error rate curve will not appear in the high signal-to-noise ratio (SNR) area. The block noise-prediction is applied in the layered detection to combine the overlap-save method. Therefore, the feedforward filter and the feedback filter can be divided into two independent blocks, and the feedforward filter is exactly the same as the linear equalizer. Then, the overlap-save is employed by the feedforward filter, achieving both low complexity and high performance. Simulation results show that bit error rate curves of the proposed schemes do not rebound in the high SNR area, the overlap-save MMSE block linear equalization has only around 1dB SNR loss compared with the fast MMSE block linear equalization, and that the overlap-save MMSE block noise-prediction has only around 0.3dB SNR loss compared with the fast MMSE block decision-feedback equalization.
What problem does this paper attempt to address?