Modeling of Physical Carrier Sense in Multi-hop Wireless Networks and Its Use in Joint Power Control and Carrier Sense Adjustment

Abstract— In this paper, we extend both Bianchi’s and Kumar’s models and characterize the channel activities governed by IEEE 802.11 DCF in multi-hop wireless networks from the perspective of an individual sender. In particular, we incorporate the effect of PHY/MAC attributes (such as transmit power and physical carrier sense) that need not be considered in WLANs but become extraordinarily important in multi-hop wireless networks, and derive the throughput attained by each sender. With the use of the analytical model derived, we investigate the impact of transmit power and carrier sense threshold on network capacity, and identify a simple operating condition under which the network may attain throughput that is close to its optimal value. Based on the insight shed from the analytical model, we then propose a distributed and localized algorithm, called Local Minimum Spanning Tree with Carrier Sense Adjustment (LMST-CSA) that determines both the transmit power and the carrier sense threshold of a node. We evaluate LMST-CSA via J-Sim simulation [1]. Simulation results show that LMST-CSA achieves higher throughput as compared to conventional IEEE 802.11 DCF, LMST with no carrier sense adjustment, and LMST with static carrier sense adjustment.