Cross-layer adaptive techniques for throughput enhancement in wireless OFDM-based networks

  • Authors:
  • Iordanis Koutsopoulos;Leandros Tassiulas

  • Affiliations:
  • Department of Computer and Communications Engineering, University of Thessaly, Volos, GR, Greece;Department of Computer and Communications Engineering, University of Thessaly, Volos, GR, Greece

  • Venue:
  • IEEE/ACM Transactions on Networking (TON)
  • Year:
  • 2006

Quantified Score

Hi-index 0.00

Visualization

Abstract

Although independent consideration of layers simplifies wireless system design, it is inadequate since: 1) it does not consider the effect of co-channel user interference on higher layers; 2) it does not address the impact of local adaptation actions on overall performance; and 3) it attempts to optimize performance at one layer while keeping parameters of other layers fixed. Cross-layer adaptation techniques spanning several layers improve performance and provide better quality of service for users across layers. In this study, we consider a synergy between the physical and access layers and address the joint problem of channel allocation, modulation level, and power control in a multicell network. Since performance is determined by channel reuse, it is important to handle co-channel interference appropriately by constructing co-channel user sets and by assigning transmission parameters so that achievable system rate is maximized. The problem is considered for orthogonal frequency-division multiplexing, which introduces novel challenges to resource allocation due to different quality of subcarriers for users and existing transmit power constraints. We study the structure of the problem and present two classes of centralized heuristic algorithms. The first one considers each subcarrier separately and sequentially allocates users from different base stations in the subcarrier based on different criteria, while the second is based on water-filling across subcarriers in each cell. Our results show that the first class of heuristics performs better and quantify the impact of different parameters on system performance.