Performance of heterogeneous multiuser MIMO system with cross-layer design and multiple outdated estimations in Rayleigh fading channel

By combining adaptive modulation and automatic repeat request protocol as well as user scheduling, a cross-layer design (CLD) scheme for multiuser MIMO system with the normalized signal-to-noise ratio (SNR)-based scheduling under heterogeneous case is developed, where perfect and imperfect channel state information (CSI) are both considered. Based on this, by exploiting previous channel estimation information, the multiple outdated estimates method is presented to improve the system performance under imperfect CSI. According to the performance analysis and the normalized-SNR based scheduling scheme, the average spectral efficiency (SE) and packet error rate (PER) of the system are derived, respectively. As a result, closed-form SE and PER expressions are attained. These expressions include the ones with conventional single outdated estimate as special cases, and can provide good performance evaluation for the multiuser CLD system under heterogeneous case. Simulation results show that the derived theoretical SE and PER are in consistent with the corresponding simulations. Moreover, the presented CLD scheme with multiple estimates can provide higher SE and lower PER than that with conventional single estimate in the presence of imperfect CSI. Besides, multiuser MIMO-CLD has superior performance over the conventional single user counterpart due to the multiuser diversity.