Synchronization Algorithm And Power Management Research For Wireless Sensor Networks

Wireless sensor network (WSN) is a typical distributed wireless communication network system. The clocks of sensor nodes in WSN are not the same with each other. In order to ensure WSN performing normally, a clock synchronization algorithm is necessary to make sure the accuracy of the clock synchronization between sensor nodes in WSN.Low power consuming is one of the most critical characteristics for WSN. The synchronization processes consume a lot of energy because of frequent communications between sensor nodes. The goal of achieving both computing algorithm accuracy and energy efficiency in the same time is difficult to reach or is impractical in reality. Therefore, trading-off between computation quality and energy consumption must be make for synchronization algorithms running on wireless sensors.The researches of these two points show as follows:(1) Introduced the relevant knowledge of clock synchronization algorithm and some energy-saving strategies for WSN tobe foundation for following synchronization algorithm design and power management researches.(2) Proposed a novel probability of reference broadcast synchronization algorithm for WSN.Due to the limitations of algorithm and the uncertainty of delay measurement, the reference broadcast synchronization algorithm has to repeat the synchronization process to achieve higher precision of clock synchronization. Re-synchronization needs a large number of communications and computations, which will consume a large amount of energy. It is against the requirements of low power for wireless sensor networks.Probabilistic synchronization algorithm used in traditional distributed systems is mainly applied in the distribution network which can not know the exact transmission delay. This algorithm takes full account of communication failures, such as conflict of communication and packets loss to ensure clock synchronization accuracy with a certain probability. In this paper, we combine the reference broadcast synchronization algorithm with probabilistic synchronization algorithm to improve algorithm's performance, so that it can get higher synchronizing accuracy with lower the failure probability in the wireless sensor network, then verified the reliability and feasibility of the novel algorithm with theory. Finally, the improved MICA experimental platform is proposed, and the results of the experiment show that the probability of reference broadcast synchronization algorithm has higher accuracy than the reference broadcast synchronization.(3) Proposed a novel time series synchronization algorithm.Compared to traditional wireless synchronization methods, a novel time series synchronization algorithm is proposed to improve the synchronization precision for structure health monitor. This paper supplies a new synchronization algorithm which compares the energy difference between post-damage waveform and pre-damage waveform received by each sensor node in structural damage detection models to use time series auto-align method to make two waves synchronous. It is shown theoretically and experimentally that the time series synchronization algorithm introduced in this paper is able to acquire satisfactory accuracy for the application of wireless damage detection system.(4) A clock synchronization algorithm for optimization of energy consumption.This paper analyzes a number of factors for energy consumption in the clock synchronization process: the quality of wireless communications, synchronization objects, rate of clock skew, clock synchronization precision, and synchronization scope. Then several energy-saving strategies are proposed in these aspects.Usually, researchers reduce numbers of communications by switching different operation modes of these components to reduce the energy consumption. However, the precision of synchronization will fall during idle modes because of clock drift. The goal of achieving both synchronization algorithm accuracy and energy efficiency in the same time is difficult to reach or is impractical in reality. Therefore, trading-off between computation quality and energy consumption must be made for a real application synchronization algorithms running on wireless sensors. Then, a model of probabilistic clock synchronization for sensor nodes is built and a trading-off between synchronization precision and energy consumption is supplied to illustrate such trade-off policies. The results of the analysis show that the energy consumption can be reduced by sleep mode without drop the synchronization precision too much.