Nowadays IEEE 802.11 Wireless Local Area Networks (WLANs) are becoming more and more popular in our daily lives. Since 802.11 standards provide multiple data rates for transmission but do not specify algorithms for automatic rate selection, rate adaptation is used extensively to maximize system throughput by adjusting transmission rates dynamically based on channel conditions. To the best of our knowledge, existing rate adaptation schemes mostly make use of packet statistics or the signal strengths of received packets to adapt transmission rates, regardless of the packet priorities....[ Read more ]

Nowadays IEEE 802.11 Wireless Local Area Networks (WLANs) are becoming more and more popular in our daily lives. Since 802.11 standards provide multiple data rates for transmission but do not specify algorithms for automatic rate selection, rate adaptation is used extensively to maximize system throughput by adjusting transmission rates dynamically based on channel conditions. To the best of our knowledge, existing rate adaptation schemes mostly make use of packet statistics or the signal strengths of received packets to adapt transmission rates, regardless of the packet priorities.

In this thesis, we first find out that it is beneficial for rate adaptation schemes to consider packet priorities when adapting transmission rates. We then propose a novel priority-based rate adaptation scheme named SRAG (Stackelberg Rate Adaptation Game) which jointly considers channel conditions and packet priorities using game theory. SRAG focuses on vehicular networks where we can easily find a variety of applications with different priorities, such as file downloading, VoIP (Voice over IP), traffic querying, collision warning, etc. In SRAG, rate adaptation is modeled as a game for different priorities of users and each user tries to select an optimal transmission rate that maximizes its utility regardless of others’ utilities. SRAG adopts a Stackelberg game model to regulate behaviors of self-interested users and gives priority to high priority users in order to achieve better overall performance. Extensive ns-2 simulations demonstrate that SRAG can provide much better performance for high priority users than existing rate adaptation schemes, while maintaining good performance for low priority users.