2-Layer Interference Coordination Framework Based on Graph Coloring Algorithm for a Cellular System With Distributed MU-MIMO

Abstract

In this study, a cellular system with a large-scale distributed multi-user multi-input multi-output (MU-MIMO) is considered, in which a large number of distributed antennas are deployed spatially over each base station coverage area (cell) and user clusters are formed in each cell to perform cluster-wise distributed MU-MIMO in parallel. In such a cellular system, the intercell and intracell interferences coexist and limit the link capacity. In this study, a 2-layer interference coordination (IC) framework that can effectively mitigate the two types of interferences simultaneously is proposed. In the 1 <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">st</sup> layer, the intercell IC is performed in a centralized manner by the non-real-time (non-RT) radio access network intelligent controller (RIC), and then in the 2 <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">nd</sup> layer, under the condition of the results in the 1 <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">st</sup> layer, intracell IC is done by each near-RT RICs in a decentralized manner. Furthermore, a restricted conditional graph coloring algorithm (RCGCA) suitable for this 2-layer IC framework is proposed. The proposed RCGCA is designed to be applied on a partial pre-colored graph, such that when it is applied in the 2-layer IC framework, it satisfies the requirement that the 2 <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">nd</sup> layer coloring must be applied under the condition of the pre-coloring results of the 1 <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">st</sup> layer. In addition, by restricting the total number of colors, the RCGCA can tradeoff between improving the capacity due to interference mitigation and degrading the capacity due to bandwidth partition, thereby maximizing the link capacity. We compare the link capacity achievable by the proposed 2-layer IC framework based on RCGCA with that achievable by the well-known fractional frequency reuse (FFR) scheme, no interference coordination case, fully centralized framework, and fully decentralized framework. Computer simulations confirm that our proposed 2-layer IC framework based on RCGCA can significantly improve the link capacity.