The conventional particle swarm optimisation algorithm has proved very sucessful at finding a good optimum in problem spaces of low to medium complexity. However problem spaces with many optima can prove difficult, especially if the dimensionality of the problem space is high. The probability that the conventional particle swarm algorithm will converge to a sub-optimal position is unacceptably high. In this chapter an adaption of the conventional particle swarm algorithm is introduced that converts the behaviour from the conventional search and converge to an endless cycle of search, converge and then diverge to carry on searching. After introducing this new waves of swarm particles (WoSP) algorithm, its behaviour on a number of problem spaces is presented. The simpler of these problem spaces have been chosen to explore the parameters of the new algorithm, but the last problem spaces have been chosen to show the remarkable performance of the algorithm on highly deceptive multi dimensional problem spaces with extreme numbers of local optima.