Comparison of the statistical properties of the LTE-A and IMT-A channel models

For the design and performance evaluation of advanced wireless communication systems employing multiple-input multiple-output (MIMO) technologies, realistic MIMO channel models with a good tradeoff between accuracy and complexity are indispensable. This paper compares the statistical properties of the two latest standardized MIMO channel models: Long Term Evolution-Advanced (LTE-A) and IMT-Advanced (IMT-A) channel models. Closed-from expressions are derived for the spatial cross-correlation function (CCF), temporal autocorrelation function (ACF), envelope level-crossing rate (LCR), average fading duration (AFD), power delay profile (PDP), and frequency correlation function (FCF) for both models. Simulation results are provided which can match the corresponding theoretical derivations very well, demonstrating the correctness of both theoretical and simulation results. The LTE-A channel model is simple but has significant flaws in terms of the accuracy. It can only support system bandwidths up to 50 MHz, not the claimed 100 MHz, and only describes the average spatial-temporal properties of MIMO channels. The IMT-A channel model is complex with more model parameters but has better accuracy. It allows us to simulate the variations of different MIMO channel realizations and can indeed support system bandwidths up to 100 MHz.