Blind multiuser detection based on kernel approximation

Authors:
Tao Yang;Bo Hu
Affiliations:
Department of Electronics Engineering, Fudan University, Shanghai, China;Department of Electronics Engineering, Fudan University, Shanghai, China
Venue:
ISNN'06 Proceedings of the Third international conference on Advances in Neural Networks - Volume Part III
Year:
2006

Citing 5
Cited 1

The nature of statistical learning theory

The nature of statistical learning theory
Fast Approximation of Support Vector Kernel Expansions, and an Interpretation of Clustering as Approximation in Feature Spaces

Mustererkennung 1998, 20. DAGM-Symposium
Input space versus feature space in kernel-based methods

IEEE Transactions on Neural Networks
Support vector machine multiuser receiver for DS-CDMA signals in multipath channels

IEEE Transactions on Neural Networks
Nonlinear blind source separation using kernels

IEEE Transactions on Neural Networks

Blind multiuser detector for chaos-based CDMA using support vector machine

IEEE Transactions on Neural Networks

Quantified Score

Hi-index	0.00

Visualization

Abstract

A kernel based multiuser detection (MUD) scheme in code-division multiple-access (CDMA) system is proposed. In this scheme, the support vector (SV) under support vector (SVM) framework is obtained through a kernel sparsity approximation, which regulates the kernel width parameter via a heuristic approach to obtain an approximate equivalent SV. The corresponding SV coefficient is attained through evaluation of generalized eigenvalue problem, which avoids the conventional costly quadratic programming (QP) computation procedure in SVM. Simulation results show that the proposed scheme has almost the same BER as standard SVM and is better than minimum mean square error (MMSE) scheme when sample set is relatively large, meanwhile the proposed scheme have a low computation complexity.