| Along with the development of computer technology and communication network in recent years, the traditional communication and exchange between human to human (H2H) get more convenient. Moreover, the connections between machine to machine (M2M) are becoming more popular in the communication development. The goal of M2M technology is to make all the machines intelligent, have networking and communication capabilities. M2M applications have tremendous potential market, and its best bearing platform will be the wireless broadband data networks which are constructed with 3G as the main part and wireless local area network (WLAN) as an important complement by operators. WLAN has the significant advantages of small-range coverage, high-speed access and so on. So it has been widely used in people's work and life. Therefore, the research into quality of service of M2M services in WLAN has very important theoretical significance and application value.Through the in-depth analysis about M2M applications, this paper has summarized M2M services'characteristics, such as delay tolerance, low mobility, and small data frame. IPP distribution is used mathematically to describe the arrive process. Based on G. Bianchi's two-dimensional Markov model, we take the imperfect wireless channels into account, and analysis the unsaturated performance of IEEE 802.11 MAC protocol in detail. The buffer queue of each station (STA) is modeled as an IPP/G/1/K queuing system. To solve the model, we adopt embedded Makrov chain method, and obtain M2M services'QoS parameters including throughput and delay in WLAN.Considering that M2M services have many characteristics which are different from traditional H2H, We propose two improved strategies for IEEE 802.11 DCF protocol based on DCF mathematical model to improve the overall performance of the WLAN. One is the problem of a large number of stations. The number of M2M terminals is great, and the waste time of retransmission due to channel competition leading to conflict will be the main factor which affects network performance. In this paper, the binomial distribution replaces the uniform distribution to select the backoff time, effectively alleviating the impact of stations'competition collisions. The other is the problem of small data frame transmission. Many M2M applications generate very small data frames, such as control, alarm information etc. Sending lots of small data frames will reduce the channel utilization, especially in the wireless networks with limited bandwidth; thereby severely affect the performance of wireless networks. This scheme allows each STA to send a group of consecutive frames instead of just one after it has won the channel. We also set up the maximum transfer time to ensure fairness among the stations, and then radio channel resources can be made full use of to improve the M2M's quality of service.In summary, this paper mainly researches QoS analysis and improvement of M2M services in the IEEE 802.11 WLAN. Firstly, we not only analyze the M2M applications'categories, but also determine their services'characteristics and arrive distribution model; Secondly, DCF algorithm is modeled based on the two-dimensional Markov model so as to obtain MAC layer medium access delay, then to solve IPP/G/1/K queuing system; Thirdly, aiming at the M2M services'features, two kinds of QoS improvement schemes are proposed for MAC protocol; Finally, MATLAB software is used to simulate IEEE 802.11 WLAN system, and the corresponding system performance parameters are obtained. The simulation results are compared with the theoretical analysis in order to verify the correctness of mathematical model and simulation algorithm, also certain validity and practicability of improvement schemes. |
PDF Full Text Request