Researches On Random Medium Access Control Schemes In Wireless Local Area Networks

As the Internet continues to be extended and expanded ubiquitously, the application of wireless local area networks (WLANs) has got its rapid growth in recent years. In a WLAN, a medium access control (MAC) protocol is employed by stations to perform access to the limited and shared transmission medium. A given MAC defines its specific mode of medium access operation and it fundamentally affects the performance of the whole network. Among various MAC protocols, the ones that employ random medium access schemes with no need of central controller are most widely used in WLANs. For these protocols, there are many notable advantages, including convenient networking, simple operation, easy implementation, flexible medium utilization, and so on. However, there are also many challenges and open problems, and consequently the related research has always been one of the hottest focuses for MAC in WLANs. In this dissertation, we conduct some innovative studies on the random medium access MAC schems.IEEE 802.11 standard is the widely accepted specification of technologies for WLANs, where the proposed multiple random medium access protocol is developed based on the well known CSMA/CA (Carrier Sense Multiple Access/Collision Avoidance) mechanism. Nowadays, IEEE 802.11 MAC has occupied the dominant position in WLANs. Moreover, large numbers of studies for wireless networks are conducted in the context of this MAC. In this dissertation, we also choose the IEEE 802.11-based WLAN as the main object of our study, focusing on how to effectively carry out the model-based performance analysis, increase the channel utilization, optimize the whole system performance, improve the support of QoS differentiation guarantee, enhance the performance of rate adaptation, and so on. The contribution of our works in this dissertation consists of two aspects, including the proposal of analytical model and the design of optimization scheme. Specifically, our works are summaried as follows:Chapter 1 firstly gives an introduction to the development of WLANs and specifies the significance and major objectives of conducting studies on MAC for WLANs. Then, several typical MAC schemes and the related standards are introduced. Last, the existing research works on random medium access MAC are reviewed and summaried, where we primarily focus on several study-related aspects including model-based analysis for IEEE 802.11 MAC, contention resolution for random medium access, rate adaptation for IEEE 802.11 multirate transmission and the emerging dynamic spectrum access technology.In Chapter 2, an adaptive optimization scheme for IEEE 802.11 DCF is proposed to enhance the network performance. The scheme is designed based on the channel sensing result for network state information and thus it is named CSB (Channel Sensing Backoff). The key idea to approach optimal performance dynamically without remodeling the DCF is that the transmission attempt from the DCF is filtered by an adjustable proba- bility P_T, which is dynamically adapted to reflect the current channel competing level among the network stations. Unlike other proposals for the DCF optimization, CSB does not need to perform complex on-line estimation of the number of active stations in the network, and can make adaptive tuning always toward a certain optimization object under various network states. Detailed performance evaluation results show that the scheme can effectively adapt to both network size and packet length changes in the network, and simultaneously achieve performance improvements on several aspects including system throughput, collision probability, transmission delay, delay jitter, fairness and so on.In Chapter 3, we propose a novel Markov chain based model analysis for the IEEE 802.11 EDCA. To properly characterize the differentiated backoff contention channel access for different priorities in EDCA, a new 3-dimensional Markov chain model is introduced. By employing the model, we present the performance evaluations of the EDCA analytically, including each priority's transmission throughput, channel access delay, packet loss rate, etc. We not only analyze the saturated EDCA operation case, but also analyze the non-saturated EDCA operation case. Moreover, the proposed model analysis incorporates all the three major QoS-supporting features of the EDCA, i.e., W min W max, AIFS, and TXOP. Simulation results confirm the accuracy of the model analysis. Based on the model analysis, we further propose an admission control scheme named D-TXOP (Dynamic TXOP). While performing admission decision, the scheme adjusts different priorities'TXOP parameter settings adaptively according to their QoS requirements, which greatly enhances the network capacity as a result.In Chapter 4, we aim to provide QoS differentiation among different traffic classes while maximizing the system throughput performance. An adaptive optimization scheme, named QATC (QoS-supporting Adaptive Transmission Control), is proposed for IEEE 802.11 to achieve both design goals. The new scheme gets around the difficulty in the station number estimation. As an extension, we also exploit the scheme to optimize the IEEE 802.11 multirate networks with airtime fairness constraint. Based on the proposed QATC, we further consider the optimization for WLANs where AIFS is applied. A novel mechanism that takes differentiated channel access control in different contention zones is introduced into the system, and we propose an adaptive scheme EA-QATC (Enhanced AIFS-based QATC) to perform the optimizing control. EA-QATC inherits the advantages of QATC and gives an effective solution to the optimization problem in AIFS application.In Chapter 5, to analyze the performance of IEEE 802.11 multirate networks with ARF (Automatic Rate Fallback) rate adaptation algorithm, a discrete-time Markov chain model for the ARF is proposed. For the analytical model, the ability to differentiate the transmission failures caused by link errors from those caused by collisions according to the choice of transmission modes between the Basic and the RTS/CTS is considered. By exploiting the model, we study two enhanced application cases of ARF, where a node chooses the transmission modes with a random probability and according to the current status of ARF rate adaptation, respectively. Subsequently, combining with the study of IEEE 802.11 DCF backoff mechanism, we obtaine the system throughput analysis.In Chapter 6, we study the MAC schemes based on dynamic spectrum access technology, which aims at improving spectrum utilization by allowing so-called secondary users to transmit in already assigned primary users'bands provided that no significant amount of interference is generated. Two spectrum sharing schemes, preemptive and nonpreemptive, are presented for the dynamic spectrum access system. Under the two schemes, both the impact on the primary user's link-layer transmission and the transmi- ssion performance that could be acquired by the secondary user are investigated using a variety of queueing system models. By optimizing the transmission of the secondary user, it can improve the secondary user's goodput and enhance the channel utilization. The analysis is also further extended to the multi-channel system and the performance improvement from additional available channels for the secondary user is investigated.