A Completely Binary and Hardware-Efficient PSO for CDMA Multiuser Detection

The application of evolutionary algorithms to CDMA multiuser detection has been studied for more than a decade. But not all variants of these algorithms can be directly applied to binary sequence detection. In this paper, we proposed a novel Particle Swarm Optimization (PSO) method. PSO is transformed into a completely binary algorithm, which is ideal for hardware to run. The new approach is successfully tested in CDMA multiuser situation. Numerical experiments demonstrate that the proposed algorithm is effective and efficient, and outperforms existing comparable methods like the common discrete PSO, binary Ant Colony Optimization (ACO), and binary Genetic Algorithm (GA) to a large extent.