CONJUGATE-GRADIENT METHOD IN ADAPTIVE BILINEAR FILTERING

The application of the Conjugate Gradient (CG) method for the identifi cation of bilinear systems is investigated. An algorithm based on the CG method is developed for adaptive bilinear digital filtering, In thi s algorithm, the optimization is done over blocks of input and output data rather than a single pair of data. However, only one iteration an d coefficient update is done for every sample of data. This, coupled w ith the fact that the CG method used-here does not require a line sear ch makes it very efficient in computation. Simulation results show tha t this algorithm outperforms the LMS and RLS algorithms in terms of sp eed of convergence. A preconditioning technique is also applied to fur ther accelerate the convergence in some cases.