A new skill called monotonous reference convergence method used to accelerate the iteration convergence in a multiple interferences

In order to accelerate the convergence of the genetic algorithm iteration for the system of more than one interfering signal, a new method for controlling the iteration is proposed. As there is a reference convergence line being formed for determining which chromosome should be kept for the next iteration, it is named the reference convergence method. For one desired signal and two interfering signals, the regular criteria for determining which chromosome should be kept for the next iteration is that it should make the output power of the desired signal increase and the output power of two interfering signals decrease monotonically. Sometimes, this method results in one chromosome being kept for many iteration steps because no better one can be found due to the strict criteria. Especially, it is very difficult to get convergence when the number of interfering signals is large in the communication system. The reference convergence method makes it adjustable. The characteristic of the reference convergence method is its criteria for determining which chromosome should be kept for the next iteration. It should make the output power of desired signal increase monotonically to the greatest extent and the output powers of all the biggest interfering signals decrease monotonically to the greatest extent in the current iteration. The paper gives detailed explanations.