Scheduling and link adaptation in the downlink of OFDM based Cellular networks
Independent thesis Advanced level (degree of Master (Two Years)), 20 credits / 30 HE creditsStudent thesis
Resource allocation and link adaptation are of major importance in radio communication systems. In this paper, we study them for the downlink of OFDM based cellular networks, using a transmission bandwidth of 20MHz.Concerning the resource allocation problem, we consider the performance of the Round Robin algorithm, the Maximum C/I algorithm and the Proportional Fair algorithm. Furthermore, we propose an iterative algorithm that can be viewed as an enhancement to the Maximum C/I algorithm. Upon that, we try to reallocate resources in order to provide fairness among the users. We develop a cost function that shows us the cost of an allocation scheme for all users, while we try within the iterative algorithm to minimize the total cost. All algorithms, but the Round Robin algorithm, adapt to both channel’s time and frequency variations. We propose a simple link adaptation algorithm, which adapts to channel conditions -both in time and frequency-and assigns modulation and coding scheme to a user, according to his average channel quality. It serves in maximization of the throughput, while maintaining the Packet Error Rate lower than 0.01.We study the above algorithms in a multi-cell radio network simulator. We apply a traffic model with Poisson arrival pattern. The performance measure studied is the user’s average packet delay and upon this we define capacity as the maximum load where 95% of users have an average packet delay lower than50msec. Results show that channel depended allocation can outperform the Round Robin algorithm by more than 80% in terms of capacity. The Maximum C/I algorithm performs better than the Proportional Fair algorithm by 7%. Moreover, implementing the proposed enhancement, we observe a further increase incapacity of 8%.
Place, publisher, year, edition, pages
Computer and Information Science
IdentifiersURN: urn:nbn:se:kth:diva-28022OAI: oai:DiVA.org:kth-28022DiVA: diva2:383812
Slimane, S. Ben
Slimane, S. Ben