Transmission strategy for interference alignment in a system-level perspective

Abstract

The paper builds up on the idea of separating, in the time dimension, between conventional uncoordinated MIMO and IA transmission. The former usually utilizes all available dimensions (antennas) to send useful information, while the latter mitigates inter-cell interference. Due to the multitude of users in cellular networks, the transmission schemes are suitable for different groups of users. Therefore, in order to take advantage of both schemes, the users are classified into two groups: one with those users that supposedly benefit from cooperation and the other with those that should not cooperate. The proposed method for grouping the users has the advantage of being based on a simple decision metric, which avoids having to calculate precoders for all different possible combinations. Since each group must apply the proper transmission scheme, a fraction of the temporal resources is dedicated to each type of transmission. An automatic reservation of the temporal resources and an adaptive adjustment of the grouping threshold for each user, which avoids having to empirically determine a single threshold, are also proposed. These proposed mechanisms are shown to provide performance gains of up to 30% over the conventional transmissions.