Space division multiple access-orthogonal frequency division multiplexing system has become a potential wireless communication system by offering high spectral efficiency, performance and capacity. This article deals with minimum symbol error rate (MSER)-based multiuser detection (MUD) technique for the space division multiple access-orthogonal frequency division multiplexing system using an efficient invasive weed optimization (IWO) algorithm. The IWO algorithm is used for finding optimal weights such that the probability of error is directly minimized rather than minimizing the mean square error. Because of this, the MSER MUD is able to detect users even in overload scenario, where the number of users is more than the number of receiving antennas, unlike several classical detection techniques. The IWO is inspired from the nature of invasive colonization of weeds and relatively simple compared with other optimization techniques. The bit error rate performance of the proposed IWO-aided MSER MUD is found to be better than minimum means square error and differential evolution algorithm-aided MSER MUDs. Simulation results show that the proposed IWO MSER achieves faster convergence and lower complexity compared with the differential evolution MSER MUD. Copyright © 2013 John Wiley & Sons, Ltd.