Markov models and cross-layer design for IEEE 802.11 DCF networks

• Main Author: You, Zihao
• Format: Dissertation
• Language: English
• Published: ProQuest Dissertations & Theses 01-01-2016
• Subjects: Electrical engineering
• ISBN: 9781339562957; 1339562952

Distributed coordination function (DCF) is a fundamental media access control (MAC) technique. It is defined in the IEEE 802.11 standard, which is the base of the widely deployed wireless local area network (WLAN) communication. The thesis works on two parts regarding DCF. The first is the Markov chain models that analyze the channel access mechanism of DCF. The second is the cross-layer design, which is based on the model in the first part. Two models are proposed in the paper to analyze the packet collision probability and throughput performances of the stations (STAs). The first model is a two-dimensional (2-D) model that is applied to the STAs working under saturated condition, which means that the STA always has a packet to transmit. The two dimensions are the retransmission stage and the backoff stage. The model extends the previous 2-D models by letting the packet error rates (PERs) be different for different STAs and different retransmission stages. The second model is also a 2-D model. It can be applied to the STAs working under either saturated or unsaturated condition. Compared with the previous 2-D models, it takes the buffer size of the STA into consideration. Compared with the 3-D models (the increased dimension is the number of packets in the buffer), it has the advantage of low complexity by avoiding matrix operations. The cross-layer design works on the physical (PHY) and MAC layers. Based on the saturated model, the STAs choose their modulation and coding schemes (MCSs) and packet lengths. The design aims to maximize the total throughput of all STAs, while some other important issues are also addressed, including the fairness, delay, etc.