Clustering Based Fractional Frequency Reuse and Fair Resource Allocation in Multi-cell Networks

Fractional frequency reuse (FFR), using different frequency reuse factors for cell center and edge regions, is able to effectively improve spectrum efficiency in multi-cell OFDMA networks. However, optimal performance is hard to achieve in practice as the efficiency of resource allocation could drop drastically due to the constraint from the frequency partitions formed by FFR. Since the radio resource is pre-partitioned for cell edge and center, fair resource allocation in a cell is also difficult to implement. Conventional frequency partition adjustment either has high complexity due to global optimization or suffers from heavy performance degradation due to absence of effective control on inter-cell interference (ICI). To solve this issue, we create models for analyzing geographical distribution of interference in multi-cell networks. Based on the observed non-uniform distributed ICI, we redefine the zones for fractional reuse and propose clustering based FFR, which offers resource allocation higher flexibility and better fairness with additional spatial dimension. Extensive simulation has been performed to validate practicality and effectiveness of our proposed scheme.