Multicell Optimization for Diversity and Interference Mitigation

We study diversity and interference mitigation with an emphasis on the interaction between the transmitter-receiver (Tx-Rx) pairs in a two-user, two-cell system in which the performance at each Rx depends on the actions of the both desired and the interfering Txs. We will propose two techniques. The first scheme is a greedy or competitive algorithm in which each Tx-Rx pair employs an orthogonal space-time (ST) block code combined with a multimode beamformer (a generalized ST code) to minimize the error probability to its own Rx till an equilibrium is reached. The second scheme is a cooperative algorithm, also based on a generalized ST code in which each Tx-Rx pair minimizes its error probability at its desired Rx but cooperates by limiting the amount of interference to the second pair´s Rx. We will model and characterize the equilibrium of the competitive scheme (called the Nash equlibrium). We will then show that by using the cooperative algorithm, it is possible to find Pareto dominant actions where both Tx-Rx pairs either equal or improve their performance when compared with the NE actions. We will analyze the conditions for Pareto dominance of the cooperative scheme and finally demonstrate the performance improvements of both schemes over the single Tx-Rx pair generalized ST code.