8054

7489

كارشناسي ارشد

معماري كامپيوتر

اصفهان: دانشگاه صنعتي اصفهان، دانشكده برق و كامپيوتر

1391

چهارده،94ص.: مصور،جدول،نمودار

ص.ع.به فارسي و انگليسي

مهدي مهدوي

30/7/92

رضا هاشمي، محمد حسين منشئي

مهندسي برق و كامپيوتر

ID7489

به فارسي و انگليسي: قابل رويت در نسخه ديجيتالي

A Prediction based Dynamic Tracking Algorithm For Wireless Sensor Networks Mehdi Heidari m heidary@ec iut ac ir 1391 12 19 Department of Electrical AND Computer Engineering Isfahan University Of technology Isfahan 84156 83111 Iran Degree M Sc Language farsi Dr Mehdi Mahdavi Assistant Prof m mahdavi@cc iut ac ir Abstract Recently advances in the fabrication and integration of sensing communication technologies and economical deployment of large scale sensor networks which are capable of large scale target tracking become possible therfore the wireless sensor networks are a fast growing and marvelous research area that has attracted considerable research attention in the recent past The creation of large scale sensor networks interconnencting several hundered to a few thousand sensor nodes opens up several wide range of application and technical challenges Sensor nodes have been deployed to play significant roles in battlefield disaster prone area traffic control habitat monitoring and intruder tracking One of the most important application of these networks is target tracking In this application wireless sensor networks consisted of many sensor nodes which are used to sense detect a target and track that target until it goes out of the monitored field Because of the nodes have limitation of energy consumption and Existance of multiple targets in real Enviroment Power consumption and multiple target tracking issues are of great important The target tracking algorithm can be mainly classify into four schems such as message based tracking tree based tracking prediction based tracking and cluster based tracking Among them the cluster based tracking protocols are more energy efficient and hence many protocols are reported to solve the energy consumption problem such as CRTA DPT and CDDTA The proposed algorithm uses a prediction based clustering approach for scalable and cluster based tracking mechanism to provide a distributed and energy efficient solution The algorithm is robust against both node and prediction failure which may result in temporary loss of the target Futher to that a novel Error Correction procedure is executed to recover lost target quickly The lost target occurs when the target changes it direction or speed so abruptly that it moves significant away from the predicted location and falls out of the detectable area The simulation results show that in comparison with the existing algorithm is able to track multiple target with random way point model more accurately over a wide of target speed Furthermore the proposed algorithm reduced communicatin that transmits between clusters and hence energy consumption in our algorithm is extermly efficient Therefore Key words Target tracking Network Scalability Error Correction Procedure Sensor Election Procedure Random Way Point model PDF created with pdfFactory trial version www pdffactory com

مهدي مهدوي

رضا هاشمي، محمد حسين منشئي