An estimation based adaptive fairness algorithm for ad hoc networks

Due to the hidden terminal problem and non-fully connected topology in ad hoc networks, stations and streams in the network cannot equally contend with each other. This leads to the fact that some stations and streams may be starved, which is the so-called fairness problem in ad hoc networks. To tackle this problem, this paper exploits the issue of per-stream fairness and extends the work of fair share based media access (FSMA) to propose an estimation based adaptive fairness algorithm named adaptive fair share based media access (AFSMA), which exactly estimates each stream´s fair share and adjusts the contention window size dynamically based on ambient contending information. In addition, a power mapping scheme is introduced to improve the performance of networks when there are a great number of contending streams. It is shown by simulation results that the new algorithm achieves better per-stream fairness and higher efficiency than those of the original ones. Moreover, the proposed mapping scheme has improved per-stream fairness and performance of the network when the congestion is heavy.