Research And Realization Of Wireless Sensor Network Algorithm For Low Power Based On Distributed Data Processing

Wireless sensor network is a kind of method that is cheap and convenient,with the characteristics of self-organization,flexible to build and low cost,traditional methods are so costly that can't match the modem one no matter in military or civilian areas.But because of the life-cycle constrained by the battery energy,an important objective of studying wireless sensor network is to meet the requirements of network applications under the premise as much as possible to prolong the network lifetime,thus extending the node life,which is the main content of the node-level low-power techniques.It's much more concerned than the traditional wireless sensor networks.In general,the nodes which rely on battery power often need the network with high rates and strong power of processing and will generate a huge amount of data,sensing energy is also of uncertainty and randomness,making the power supply of the nodes in dynamic changes.This thesis first introduces the wireless sensor network,describes the basic concepts and application environment.Then,the energy consumption of wireless sensor networks is described,and a comprehensive analysis of the energy consumption of wireless sensor network node sources and causes and characteristics is generated based on the features of the practical application of information sensing,information processing and transmission of information,it studies on a variety of methods to reduce system power consumption,focuses on low power requirements for wireless sensor networks,it analyzes and researches about energy consumption of each module of the wireless sensor network,then proposes a holistic strategy of low-power sensor networks.Based on the proposed low-power sensor networks program,focusing on energy-saving programs based on the communication module of the distributed data processing,a method of reducing the power consumption of the communication module is proposed based on independent processing data algorithm transplant.And on this basis,the power spectral density algorithm,for example,a number of classic power spectrum algorithms were analyzed and determined according to the actual needs of the low-power design software and the superiority of the algorithm was verified by experiment Build the hardware platform based on the overall strategy and build wireless sensor network power test system,use the policy applied to the Imote2 nodes to achieve low complexity and high efficiency transmission.Analysis the operating system and software language that applied to the nodes,according to the needs of low-power strategy of its structure and the actual node module,designed algorithm portable software.On this basis,it brought out a process based on power spectral density function the algorithm transplant module assembly design and a detailed analysis of the difference between centralized data processing model and distributed data processing model and the function of each module and mode of operation,focusing on the superiority of distribution-data processing method and the inter-relationship of the new modules.It achieved the transplantation of the algorithm(power spectral density function)on the wireless network node software platform.Finally,we tested the data acquisition of nodes in network,carried out comparative experiments and analysis of the two methods.Tested analysis strategies and methods,this paper makes the energy consumption of the wireless sensor network system effectively reduced,the system has a small size,low power consumption,stable performance,etc.The experimental results show the effectiveness of this strategy.