Entropy-guided micro-genetic algorithm for multiuser detection in CDMA communications

Hypotheses Control-based Strategies for the Development of Near-optimum Bayesian Detectors for CDMA Systems

“…Figure 5 shows the average BER of users 1 and 2 for the Bayesian MUDs v.3 and v.5, and compares it with that obtained with both a NN-based detector [28], which is implemented with 12 hidden layers and uses the backpropagation algorithm for training, and a microGenetic Algorithm (" -GA) for multiuser detection which is described in [25]. This experiment is run considering K ¼ 8 users, all of them with equal energy, E (E i ¼ E j , i 6 ¼ j , 1 i; j K ).…”

“…Figure 8 shows the BER of the UOI versus the SIR=E 1 =E k (2 k K ) for four different multiuser detectors. The error probabilities of traditional detectors based on matched filters and decorrelator [2] as well as on a Genetic Algorithm [25] are shown for comparison.…”

“…Finally, another branch of MUDs is constituted by those schemes which make use of Natural Computation techniques, such as Genetic Algorithms [23][24][25][26][27] and Neural Networks [28][29][30][31]. Most of these methods make use of heuristic search strategies and have been extensively proven to be effective nonlinear MUDs.…”

A Particle Swarm Optimization Based Multiuser Detector for DS-CDMA Communication Systems

“…Vj,k (1) is generated for the kth dimension of thejth particle by randomly selecting a value with uniform probability over [v"" --10, vm2ax = 10] . Each particle in the initial population is evaluated using the objective function F by F(X) = 2y XT XTRXJ (10) For each particle, set XP(1) = Xj(1) and FJP = FJ, j 1, 2,..., n . Search for the best value of the objective function F .…”

“…Many heuristic search algorithms such as genetic algorithm [8][9][10], tabu search algorithm [11][12], and evolutionary programming (EP) [13] have recently been utilized to solve multiuser detection problems. Previous works [9] have revealed that the GA, which manipulates a group of possible solutions with crossover, mutation and selection operators to search for the optimal result, is a valuable method for solving the multiuser detection (MUD) problem.…”