Cross-layer Lifetime Optimization In Wireless Sensor Network

Wireless sensor network is the combination of information collection, transmission and processing. Wireless sensor network combined the sensor technology, embedded technology, wireless communication, and distributed information processing and micro-electro-mechanism system. It can monitor environment and collect information in real time with all kinds of integrated micro-sensor. The micro-sensor would process the collected information by embedded unit and finally transport the information to terminal user by multi-hop wireless ad hoc network which constructed randomly. However, new incoming problems are followed with the birth of the wireless sensor network. As known that sensor network relies on the battery to supply energy, it is impossible to recharge and replace the battery in sensor. Therefore, the problem how to maximize lifetime of wireless sensor network attracts researchers' attention. They have proposed multiple optimization models and cross-layer strategies to deal with this problem.In this article, we analyze current layer-based optimal methods in wireless sensor network and summarize the physical, link and routing optimization techniques. Then we compare some strategies in cross-layer optimization algorithms.According to the analysis and summary of the current lifetime optimization algorithms in wireless sensor network, we firstly focus on synchronous small-scale networks with interference-free link scheduling and practical MQAM link transmission schemes. In this case, our model is TDMA model, and optimization problem can be closely approximated as a convex optimization problem that can be efficiently solved. We explore the modulation rate in physical layer how to affect the network lifetime. The result demonstrates that network lifetime will be prolonged with the adaptive modulation rate. There also exists a fixed uniform modulation rate in each link, the network lifetime under this rate can be close to the lifetime under adaptive modulation rate.Secondly, we extend our model to non-TDMA network. Depending on our prior analyses of modulation rate, we propose an iterative algorithm to fight against non-convex problem in non-TDMA optimal model. Through the analysis of performance about our algorithm, we get a conclusion that our algorithm can not only prolong the network lifetime comparing to TDMA network, but also decrease the complexity of computation. Then we quantify the best trade-off curve between delay and energy consumption, analyze how our algorithm affect the delay in wireless sensor network.