The upcoming Third-Generation Partnership Project—Long-Term Evolution (3GPP-LTE) cellular standard will employ spatial multiplexing to significantly increase the data rates. Detection of the spatially multiplexed signals is an essential issue in the design of an LTE receiver. In this paper, we evaluate the performance–complexity tradeoffs for a set of low-complexity multiple input multiple output (MIMO) detection algorithms in a realistic LTE downlink system. Specifically, antenna correlation and channel estimation errors have been considered for a practical MIMO-LTE receiver. An LTE downlink model has been implemented in order to evaluate three types of detectors: linear, unsorted successive interference cancellation (SIC), and ordered SIC. Our simulation results show that the unsorted SIC detectors present a very poor performance–complexity tradeoff. Besides, linear detectors are shown to be the best candidates as the performance improvement for the ordered SIC detectors is not significant in a realistic scenario. Copyright q 2009 John Wiley & Sons, Ltd.