Study On Link Quality Estimation Parameters Optimization Mechanism For Wireless Sensor Networks Based On Uncertain Multi-Criteria

Wireless sensor networks (WSNs) are special ad-hoc networks which arecomposed of large numbers of low power consumption, energy restricted, cheapsensor nodes in wireless a self-organizing manner. All of sensor nodes are deployedin a way to spread the monitored areas, which mostly have poor environment orhumans beings could not reach, the sensor nodes communicate or exchangeinformation with others in the areas via radio frequency signals, and the informationcolleting by sensor nodes is gathered together and send to the sink node, which isresponsible for dealing with all the data, which produces a wide range ofapplications and prospects. As radio frequency signal during the transmissionprocess prone to reflection, scattering, refraction, and is susceptible to the influenceof the external environment, which brings a series of changes easily, leads to thedecrease of the link quality. Therefore, the studies of wireless link quality arebecoming a hot issue, which is not only the basis for system functions, but alsoaffecting the routing protocol selection in the upper link.Firstly, we analyze current status of the study of wireless sensor network linkquality, on the basis of this, the link estimation parameters are classified, that ishardware-based and software-based link estimation, and describing and comparingthe performance of two types in detail. The research shows that when the existinglink parameters separately evaluate the performance of link quality, which is notsatisfactory, so we introduce more link quality estimation parameters, and as there isa strong correlation between some of the parameters. At the same time, by takinginto account the not full links information and the unsure selected set of parametersin the real links, thus we bring into the uncertain multi-criteria decision makingtheory.Then, we carry out a detailed study about the uncertain multi-criteria decisiontheory in this paper, which is in conjunction with the multi-parameter and applied tothe estimation of link quality, we establish a parameter optimization model andselect the appropriate estimation parameters set. By making use of the Kalman filter model based on the wavelet transform denoising the selected parameters data, wepropose a link quality estimation mechanism based on the absolute weight error ratio,and achieve a good estimation effect. To some extent, which reduce the node energyconsumption and prolong the life of the wireless sensor networks, improving theutilization of self-organizing networks.In the case of the stable and unstable link quality, a large number of nodes arecarried out in this paper, by comparing the estimation effect between the link qualityestimation mechanism we proposed and the traditional link parameters, theexperiments show that the proposed mechanism has a smaller the estimation error,especially when the link quality is unstable, this mechanism has a better estimationeffect.