On the Theoretical Analysis of Optimal Cellular Systems design with Multi-user Detection in slow flat fading channel - Uplink Analysis

In this paper, we investigate the optimal multi-cell systems design involving multi-user detection. Specifically, we analyze the optimal intra-cell and inter-cell multi-access strategies in slow fading channels and the role of multi-user detection (MUD) in the multi-cell system design based on an information theoretical framework. We adapt Wyner's model and consider a multi-cell system of M cells with K single antenna homogeneous users. At each of the base stations, both zero forcing and the optimal MUD are considered as the base station receiver. We consider slow fading channels where the instantaneous channel capacity is unknown to the transmitters. Hence, packet outage will occur whenever the data rate of packet transmissions is less than the instantaneous channel capacity. Conventional performance measure by ergodic capacity may not be useful in this situation and we consider the overall network goodput, which measures the average number of bits/sec/Hz successfully delivered to the base station by the K users, as the capacity measure. Based on the theoretical framework, we derive closed-form expressions on the total network goodput with respect to different multi-cell system design schemes. The optimal intra-cell and inter-cell channel partitioning schemes with respect to the network goodput in various scenarios is found. We conclude that the multi-user detection plays a important role in multi-cell systems by improving the efficiency resource reuse within a cell