Soft constraint satisfaction (SCS) blind channel equalization algorithms

Abstract

SUMMARYThe constant modulus adaptive blind equalization algorithms presented in this paper are shown to correspond to an error performance surface which is much improved upon that of existing algorithms such as the wellknown constant modulus (or Godard) algorithm. Many undesirable local solutions (ULSs) are avoided by careful derivation. We use a deterministic optimization criterion with a soft constraint to obtain an update equation which contains a normalized gradient vector and a particular continuous non-linearity. This approach is extended to multiple constraints to yield faster converging algorithms. An autoregressive (AR) channel model is studied to demonstrate analytically the absence of a class of ULSs. Finally, the ÿndings are veriÿed experimentally for various AR and moving-average (MA) channels. ? 1998 John Wiley & Sons, Ltd.

A new family of blind adaptive equalization algorithms

“…A family of nonlinear functions is proposed as an extension of the SCS approach [1] which form a new set of blind adaptive algorithms. They are observed to maintain the advantages of the SCS approach.…”

A Blind Adaptive Technique for Cancelling ICI in OFDM-WPAN Systems

“…The CM functions proposed in [13,14] are only suitable for real number, and can't be directly used in our receiver. Thus, we propose the following modified functions to restore the CM of QPSK signals.…”

“…It is obvious from (14) that / has the unique stationary point: u=0. The Hessian matrix at the stationary point is given by:…”

“…The soft-constraint-satisfaction (SCS) based algorithms were proposed in [13,14], which have been proven to have better performance than the CMA. Hence, in this paper, we use the SCS algorithm to adjust the weight vector W. Due to the QPSK modulation used in each subcarrier, the original SCS algorithm cannot be directly used in our receiver.…”

“…The rationale of SCS is based on a deterministic optimization criterion where the CM property is imposed as a soft constraint, which can be expressed as [13,14] min W m…”