2nd International ICST Conference on Cognitive Radio Oriented Wireless Networks and Communications

Research Article

Disruptive CSMA with Credit Payback (CP) Protocols for Multi-Radio Network

  • @INPROCEEDINGS{10.1109/CROWNCOM.2007.4549798,
        author={Jing Zhu and Xue Yang and Xingang Guo},
        title={Disruptive CSMA with Credit Payback (CP) Protocols for Multi-Radio Network},
        proceedings={2nd International ICST Conference on Cognitive Radio Oriented Wireless Networks and Communications},
        publisher={IEEE},
        proceedings_a={CROWNCOM},
        year={2008},
        month={6},
        keywords={CSMA fairness multi-radio device performance analysis protocol design},
        doi={10.1109/CROWNCOM.2007.4549798}
    }
    
  • Jing Zhu
    Xue Yang
    Xingang Guo
    Year: 2008
    Disruptive CSMA with Credit Payback (CP) Protocols for Multi-Radio Network
    CROWNCOM
    IEEE
    DOI: 10.1109/CROWNCOM.2007.4549798
Jing Zhu1,*, Xue Yang1,*, Xingang Guo1,*
  • 1: Communication Technology Lab, Intel Corporation, Hillsboro, OR 97124, USA
*Contact email: jing.z.zhu@intel.com, xue.yang@intel.com, xingang.guo@intel.com

Abstract

In this paper, we study a multi-radio network, where at least one station has multi-standard transceivers operated simultaneously. A radio at the multi-radio device can suffer service disruptions due to interference from collocated radios or due to constraints from shared resources. In this paper, we study the impact of temporal disruption on fairness for CSMAbased network, such as WLAN. We propose a novel credit payback (CP) mechanism to mitigate the impact of such disruptions. Credits are saved during disruptions and used to compensate the lost transmission opportunities of disruptive stations. We study the key parameter— credit payback ratio via theoretical analysis and simulations, considering two specific cases: random and regular. We also propose a local adaptive credit payback algorithm to achieve desired fairness performance without global information. Numerical results from both analytical models and simulations are presented to illustrate several interesting observations and to demonstrate the benefits of our solution.