5th International ICST Conference on Broadband Communications, Networks, and Systems

Research Article

A Fast Channel Allocation Scheme Using Simulated Annealing in Scalable WLANs

  • @INPROCEEDINGS{10.1109/BROADNETS.2008.4769071,
        author={Jiayuan Chen and Sverrir Olafsson and Xuanye Gu and Yang Yang},
        title={A Fast Channel Allocation Scheme Using Simulated Annealing in Scalable WLANs},
        proceedings={5th International ICST Conference on Broadband Communications, Networks, and Systems},
        publisher={IEEE},
        proceedings_a={BROADNETS},
        year={2010},
        month={5},
        keywords={Distributed channel allocation Simulated annealing},
        doi={10.1109/BROADNETS.2008.4769071}
    }
    
  • Jiayuan Chen
    Sverrir Olafsson
    Xuanye Gu
    Yang Yang
    Year: 2010
    A Fast Channel Allocation Scheme Using Simulated Annealing in Scalable WLANs
    BROADNETS
    IEEE
    DOI: 10.1109/BROADNETS.2008.4769071
Jiayuan Chen1,*, Sverrir Olafsson2,3,*, Xuanye Gu2,*, Yang Yang4,*
  • 1: Department of Electronic and Electrical Engineering, UCL, London WC1E 7JE, UK BT, Adastral Park, Martlesham Heath, Ipswich, IP3 5RE, UK
  • 2: BT, Adastral Park, Martlesham Heath, Ipswich, IP3 5RE, UK
  • 3: University of Reykjavík, Ofanleiti 2, 103 Reykjavik, Iceland
  • 4: Department of Electronic and Electrical Engineering, UCL, London WC1E 7JE, UK
*Contact email: jiayuan.chen@bt.com, sverrir.olafsson@bt.com, xuanye.gu@bt.com, yyang@ee.ucl.ac.uk

Abstract

The primary difficulty with frequency reuse in scalable WLANs is interference mitigation. When deploying 802.11 devices in a close proximity, appropriate channel selection for each AP becomes one of the most challenging issues provided by limited usable frequencies. In this paper we introduce a distributed version of Simulated Annealing to solve the dynamic channel allocation problem in High-Density WLANs. The approach is generic and applicable to any access system where channels are not allocated in a prefixed or centralized manner. The simulation results show that the proposed algorithm scales well and approximates the optimal solutions under a wide range of different network topologies.