Design And Simulation Of IEEE 802.15.4 Dynamic Adaptive And Energy-efficient MAC Protocol

IEEE 802.15.4 standard targets short-range wireless communication for low-rate, low-power and low-cost devices, and defines MAC and PHY specifications for low-rate WSN (Wireless Senor Network). After deeply studying on this standard, this thesis analyzed some shortcomings in adaptation to load-changing environment while keeping low-power consumption for existing WSN MAC protocol, proposed improved theme of MAC protocol based on dynamic adaptiveness for adjusting duty cycle of super-frame and back-off exponent, and designed the corresponding simulation program. Simulation tests have proved its feasibility. The comparative results indicate that the theme has the better performance in load-changing environment. The completed specific research and implementation of this thesis are listed as the following:Research on the core mechanism of mainstream WSN MAC protocols, analysis to their characteristics, advantages and disadvantages are conducted. The main influence factors to energy-consumption for WSN MAC protocols are discussed. The simulation tests and results analysis are made for IEEE 802.15.4 MAC protocol on NS network platform.To aim at improving non-adaptation between the fixed duty cycle and the changing traffic load, and herefrom the contradiction between packet delay and energy consumption in traditional IEEE 802.15.4 MAC protocol, by finding the best relationship between duty cycle and the network traffic, the paper proposed an adaptive mechanism that can dynamically adjust the duty cycle of super-frame based on the current network load condition. The node optimal duty cycle is obtained by repeated simulation tests in a variety of load conditions. In a relatively fixed topology, this mechanism can improve the performance of sensor networks by finding the best compromise between energy consumption and packet delay.For the shortcoming in adaptation to load changing in IEEE 802.15.4 CSMA/CA algorithms, the thesis analysed the relationship between traffic load and back-off exponent (herefrom the self-added value of retransmision). When the traffic load is high, back-off window shrinks to the minimum after a successful transmission, collision probability becomes high, which leads to increasing of the number of retransmissions, and the continuous retransmissions cause malignant cycles resulting in network congestion. Based on sensing the current network traffic, this paper proposed a new back-off algorithm called CSMA/CA-DA through the dynamic adjustment of the initial back-off exponent value of BE and the retransmission self-added value of I. The thesis has improved the protocol design, completed simulation tests and performance analysis. In relatively fixed topology conditions,by finding the optimal back-off exponent in the different traffic load cases, the improved algorithm has made the nodes keep in optimal state when competing channels.