The nature of statistical learning theory
The nature of statistical learning theory
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
Hi-index | 0.00 |
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.