Reducing the complexity of Resource Allocation for underlaying Device-to-Device communications

Sharing resources between Long Term Evolution (LTE) cellular users and Device-to-Device (D2D) pairs comes at the cost of increased interference. In this paper, we propose a Minimum Knapsack-based Interference-aware Resource Allocation (MIKIRA) approach to show that a channel sharing approach can be used to increase system sum rate, minimize the interference both at the evolved Node B (eNB) and the receiver of the D2D pairs, and most importantly achieve a good signal quality at the eNB in lower time complexity in an LTE cellular network. We compare the system sum rate, interference and Signal-to-Interference and Noise Ratio (SINR) obtained by MIKIRA with a bipartite graph-based approach (GRA) for resource allocation and a random allocation approach. ¿From the results we find that MIKIRA shows improved performance when compared to other approaches in all of the metrics we used in the simulation. MIKIRA is more efficient in terms of time complexity (O(n² log(n))) when compared with the GRA (O(n³)); and also results in higher quality signals as the SINR achieved at the eNB is much better than the other two approaches.