"Fading Channel Equalization and Video Traffic Classification Using Nonlinear Signal Processing Techniques"

by Qilian Liang

May 2000

We apply an unnormalized interval type-2 TSK FLS as a type-2 fuzzy adaptive filter (FAF) to equalization of nonlinear and fading channels, and demonstrate that it can implement the Bayesian equalizer for such a channel, has a simple structure, and provides fast inference. Two structures are used for the Bayesian equalizer respectively: transversal and decision feedback. We propose a decision tree structure to implement the decision feedback equalizer (DFE), and each leaf of the tree is a type-2 FAF.

This DFE vastly reduces the computational complexity compared to a transversal equalizer (TE). Simulation results show that Bayesian equalizers based on type-2 FAFs performs much better than nearest neighbor classifiers (NNC) and the Bayesian equalizers based on type-1 FAFs. We also present a method for overcoming time-varying co-channel interference (CCI) using type-2 FAF.

We apply unnormalized interval type-2 Mamdani FLSs as type-2 fuzzy classifiers (FC) to MPEG VBR video traffic modeling and classification. We demonstrate that a type-2 fuzzy membership function, i.e., a Gaussian MF with uncertain std, is most appropriate to model the log-value of I/P/B frame sizes in MPEG VBR video. The fuzzy c-means (FCM) method is used to obtain the mean and standard deviation (std) of I/P/B frame sizes when the frame category is unknown. Five fuzzy classifiers and a Bayesian classifier are designed for video traffic classification, and the fuzzy classifiers are compared against the Bayesian classifier. Simulation results show that a type-2 fuzzy classifier performs the best of the six classifiers.

Finally, we present our conclusions and some directions for future research.

Price: Domestic $17.00; International $25.50