Radio resource management and evolutionary computation in CDMA cellular radio networks

The paper proposes distributed power control (PC) algorithms that employ two evolutionary computation (EC) or genetic algorithm (GA) techniques to solve systems of linear equations for power update in CDMA cellular radio systems. The proposed algorithms apply evolutionary computation algorithms from a phenotypic and a genotypic view to the CDMA power control problem. The major gain from the evolutionary computation algorithms is more rapid optimization on systems of linear equations compared with the simple genetic algorithm (SGA). Employing distributed constrained power control (DCPC) and bang-bang (BB) algorithms as the basic reference, we have designed and implemented computational experiments on the DS-CDMA system. The proposed EC-DCPC phenotypic algorithm is compared with the DCPC algorithm. The GA-DCPC genotypic algorithm is also compared with the BB algorithm used in IS-95 and W-CDMA systems. The simulation results indicate that the proposed EC-DCPC phenotypic and GA-DCPC genotypic algorithms significantly decrease mobile terminal power consumption compared with the DCPC and BB algorithms, respectively. The calculation results show that our proposed EC-DCPC phenotypic and GA-DCPC genotypic algorithms also have a high potential advantage for increasing CDMA cellular radio network capacity.