Abstract
Inter-Cell Interference (ICI) from neighboring cells is a major challenge that severely degrade the performance of Orthogonal Frequency Division Multiple Access based cellular mobile systems, particularly for cell-edge users. An efficient technique to mitigate ICI is interference coordination. The most commonly Inter-Cell Interfere Coordination technique is Fractional Frequency Reuse (FFR). Furthermore in order to effectively improve cell-edge performance in terms of coverage extension and throughput, the 3rd Generation Partnership Project introduced the use of relays in Long Term Evolution-Advanced (LTE-A) networks to achieve self-backhauling of radio signals between Evolved NodeBs (eNBs) and UEs. This paper introduces a Self-Organized Dynamic FFR Resource Allocation scheme (SODRA-FFR) which dynamically allocates frequency resources to cell inner and outer regions in relay based LTE-A networks to improve cell edge performance and maximize fairness among UEs. In this scheme, the downlink frequency resources are dynamically allocated to cell inner and outer regions and the outer region frequency resources are dynamically distributed between eNB and relay stations in each cell based on coordination between neighboring eNBs and relay stations through a message passing approach over LTE-X2 interfaces. The performance of the proposed SODRA-FFR scheme without and with relays is evaluated using MATLAB simulations and compared with different combinations of frequency resources allocation to cell inner and outer regions as well as with other frequency reuse scheme (i.e., frequency reuse-1 and frequency reuse-3). The results show that the proposed SODRA-FFR scheme improves cell-edge performance and achieves high degree of fairness among UEs compared to reference resource allocation schemes. The results also show that the proposed SODRA-FFR scheme with RSs improves the fairness performance by 30 and 13 % compared to that of frequency reuse-1 and frequency reuse-3 schemes, respectively. In addition, the proposed SODRA-FFR scheme with RSs achieves 55 and 26 % increase in cell-edge throughput compared to that of frequency reuse-1 and frequency reuse-3 schemes, respectively.