Simulation and Optimization of Frequency Reuse in OFDMA Networks

Abstract: Efficient radio resource management is an important aspect when it comes to achieving high bit rates in technologies such as the 3GPP Long Term Evolution (LTE). This thesis aims at understanding existing frequency reuse schemes in OFDMA networks, and to develop an algorithm for frequency allocation for irregular cellular layouts. Previous work done in this field mostly covers schemes for regular cells, whereas in real life cellular layouts are mostly irregular. A comparison of the existing allocation schemes for the users near the boundary of the cells, also known as edge users, with our scheme is presented. Reuse-1, Fractional frequency reuse-3 with random frequency allocation (FFR3-RFA) (for edge users) and our algorithmic frequency allocation (FFR3-AFA) scheme are simulated and compared. FFR3-AFA algorithm assigns a frequency sub-band to a cell by considering the frequency allocations in the neighbor cell edges and the overlap area between those neighboring cells. Static simulations were performed with one user per base station and constant downlink traffic of 100 Kbits/sec, so that all the resources are utilized and there is maximum interference. This way, the difference between the frequency reuse schemes is much more evident. The throughput for our calculation is the ratio of the total successful packets sent in the network and the total packets sent in the network. Small scenarios are considered with different downlink data rates and the results show that FFR3-AFA has better performance than the other two techniques. There is also room for improvement in the algorithm by introducing other factors into the equation other than overlap area, such as user throughput.