Research On Applications And Location Algorithms For Wireless Sensor Networks

With the progress of wireless communication, embedded technology, sensor technology,micro-electromechanical technology and distributed information processing technology,wireless sensor networks (WSNs) developed a new access to information technology. In bothmilitary and civilian aspects which has a broad application prospect and in the specificapplication, location information is very important, only obtained location information, thedata which involved is valid. In addition, localization of all nodes of the wireless sensornetworks can effectively improve the efficiency of the whole wireless sensor networks rootingand provide the network's automatic implementation of system management, the balancedload and topology control with effective technology guarantee. So we can see that to design ahigh precision, low complexity of the wireless sensor node localization algorithm has alwaysbeen a technical hot topic to be resolved.In this paper, based on the specific performance of current mainstream WSNs nodelocalization algorithm, we adopt the technical research by means with the combination of theprinciple of wireless communication and mathematical model, starting from the essentialrequirement of scientific research with the low computational cost (low-power) and the actualproduct feasibility of high demand, to design a localization algorithm which uses a smallamount of anchor with wireless reference nodes, obtains better location performance andmeets the demand of a variety of scenarios for the final research goal, based on the currentmainstream signal arrival time TOA and received signal strength RSSI ranging technology,we put forward three new kinds of wireless sensor node localization algorithm. Under thebasis of the theory research, this paper will apply partly location technology to the actualwireless temperature measuring system, some meaningful results are obtained, and the mainresearch contents are as follow:1. For the current wireless sensor networks node localization algorithm has carried onthe comprehensive review. In the evaluation method of wireless location algorithmpositioning accuracy, the theoretical analysis was deduced the Cramer-rao lower bound(CRLB) under the condition of WSNs, and in the process of the positioning accuracylocalization algorithm assessment methods, we conclude that the distance effect, the angle ofarrival error, the error of anchor position are how to influence the CRLB.2. In general, the nonlinear relationship exist between the TOA measurements and theunknown parameter value in the observation equation corresponding to TOA-based sourcelocation, it normally results in the nonexistence of any efficient unbiased estimator that attainsthe (CRLB). This paper proposes a new method of translating a linear method i.e. by using thecoordinate system transformation corresponding to the formula Nonlinear Equations removedobserved variables. Then it proposes an appropriate method based on the use of lineartranslational coordinates of the source node TOA localization algorithm. From themeasurement error of time, computational complexity, the mean square error and CRLB fouraspects to evaluate the algorithm performance, the performance analysis and simulation study conducted show that our proposed algorithm can achieve CRLB when the zero-meanGaussian and independent measurement errors are sufficiently small.3. In general wireless sensor nodes transmit power source and the path loss coefficientsare unknown, we addresses the localization of an isotropically decaying source based on thereceived signal strength (RSS) measurements that are collected from nearby active sensorsthat are position-known and wirelessly connected. For such a source localization problem, wepropose an iterative algorithm, in which the unknown source position and two other unknownparameters (i.e. the source power and pathloss factor) are estimated in an alternating waybased on each other, with our proposed sub-optimum initial estimate on source positionobtained based on the RSS measurements. Analysis and simulation study show that ourproposed iterative algorithm guarantees globally convergence to the least-squares (LS)solution, where for our suitably assumed independent and identically distributed (i.i.d.)zero-mean Gaussian RSS measurement errors the converged localization performanceachieves the optimum that corresponds to CRLB.4. This paper deals with self-localization of wirelessly connected sensors, based onpartially available pairwise distance measurements among these randomly deployed sensorsand a few ad hoc deployed position-known anchors. We propose a distributed cooperativescheme that can be implemented at all individual sensors in such networks to self-identify andself-localize a dominant class of localizable sensors. The new scheme exploits successivesensor anchor distance estimations (i.e. prospective sensors successively, starting from theones neighboring to anchors, obtain their distances to these closest anchors). Analysis andsimulation study show that our proposed scheme can be used to identify the consideredlocalizable sensors, and also can perform accurate localization on these localizable sensors inthe absence of any measurement error or achieve optimum localization performance in thepresence of our suitably assumed independent zero-mean Gaussian measurement errors.5. Considering actual winery cellar pool temperature measurement needs and theadvantages of WSNs that in terms of environmental information monitoring, a low-powerindustrial monitoring system with location information is designed and implemented in thisarticle. Efficient low-power industrial monitoring systems and sensor network nodepositioning technology combined up and realized in the winery over1200ports pits real-timetemperature monitoring function, which has great practical value. In order to ensure thelong-term stability of system operation, designing a test experiment for the overall technologyperformance of wireless system and proceeding field test work accordingly. Then we analyzethe corresponding results of research work. The low-power monitoring system has more than1200industrial temperature sensing end nodes, and it has run stably more than two years at awinery, which is enough to prove the feasibility and rationality of the overall system solution.