Project: Optimum Subband Coder Design

PEOPLE

Rajeev Gandhi, S.K.Mitra

OBJECTIVE

Two specific problems were investigated under this project. It is well known that mean square error (MSE) is not preserved under non-orthogonal transformations. This poses a significant challenge to quantize the subband signals in coders that involve non-orthogonal filter banks, since nearest neighbor (NN) encoding rule can no longer be applied to quantize the coefficients. In this part of the project, techniques for the optimal quantization of the subband coefficients in non-orthogonal subband coders have been developed. An exhaustive-search based quantization approach is proposed. The complexity of this approach is shown to increase exponentially with the length of the signal. Next, a reduced complexity solution in the form of a trellis-based search is utilized. Simulation results indicate appreciable SNR gains over standard coding techniques for non-orthogonal filter banks. A paper based on the results of this project was presented at a conference [1].

Wavelet and tree-structured filter bank based coding schemes find applications in a number of image-coding algorithms. The efficiency of such coding schemes can be enhanced further by adapting the parameters of the coder to the statistics of the signal being compressed. In this part of the project, an algorithm has been developed to determine the coefficients of the filter bank matched to the signal statistics, at every node of a tree structured filter bank. In the proposed algorithm, each node is visited in a top-down fashion and determines the bit-allocation and the coefficients of the filters used to carry out the decomposition. An estimate of the coding gain provided by the later stages is used to account for the interaction between the filter bank at the current node and the filter banks at the later nodes. Papers based on the results of this project were presented at conferences [2],[3].

PUBLICATIONS

These materials are presented to ensure timely dissemination of scholarly and technical work. Copyright and all rights therein are retained by authors or by other copyright holders. All persons copying this information are expected to adhere to the terms and constraints invoked by each authors copyright. In most cases, these works may not be reposted without the explicit permission of the copyright holder.

1. R. Gandhi and S.K. Mitra, "Optimal quantization in non-orthogonal subband coders," Proc. Asilomar Conference on Circuits, Systems, and Computers , Pacific Grove, CA, October 1999, pp. 1188-1192 .

2. R. Gandhi, S.K. Mitra, "An efficient top-down approach for the design of tree-structured orthonormal filter banks," Proceedings of 1999 Asilomar Conference , Pacific Grove, CA, USA, October 1999, p.p.1627-1631.

3. R. Gandhi and S.K. Mitra, "Design of optimal orthogonal tree-structured  filter banks," Proc. Midwest Symposium on Circuits & Systems , Las Cruces, NM, August 1999, pp. 1057-1060. 

4. R. Gandhi and S. K. Mitra, "Aliasing cancelation in block filters and periodically time varying systems: a time-domain approach," Proceedings of the 1997 IEEE International Symposium on Circuits and Systems, ISCAS '97, Hong Kong, 9-12 June 1997, pp.2421-2424.