تشخيص اطلاعات غلط در شبكه هاي اقتضايي خودرويي

چكيده انگليسي :

Bogus information detection in VANETs Abbas Mirheidari a mirheidari@ec iut ac ir May 2013 Department of Electrical and Computer Engineering Isfahan University of Thechnology Isfahan 84156 83111 Iran Degree M Sc Language Farsi Supervisor Dr Mohammad Ali Montazeri montazeri@cc iut ac ir Abstract The efficiency and success of wireless systems which make use of sensors to collect information is determined by the data quality of collected sensing data One of the most effective factors on the data quality is false information which may result from the malfunction of sensor boards unreliable wireless communication channel and compromised sensors Vehicular Ad hoc NETworks VANETs which are a particular application of highly dynamic sensing systems has emerged with the main purpose of improving the safety of passengers and roads In these networks equipping vehicles with computers global positioning system and wireless communication devices have enabled emergency information dissemination through vehicle to vehicle V2V and vehicle to RSU V2R communications Malicious vehicles in VANET may disseminate rumors and attempt to provide false information such as reporting fake events causing other vehicles to make incorrect decisions and wrong moves Due to safety concerns a crucial point is how to trust the information transmitted in other word how the false information could be detected To resolve this problem many researches have focused on trust management Some of proposed trust models have used reputation schemes applied to mobile ad hoc networks MANETs But due to the ephemeral nature of vehicular networks reputation schemes cannot be sufficiently suited to the VANETs Some others have used data centric trust models These models efficiency also clollapses when the network area is sparse or when collusion attack happens Based on the researches have been done so far in VANETs its almost unattainable to introduce a completely distributed trust management system which could be able to retain all its functionalities in highly dynamic parts of the network where the neighboring vehicles are rapidly changing and moving in and out of range or when collusion attack exists In this project we propose a bogus information detection system for VANETs in which vehicles honesty reputation is saved in the RSUs common database and every vehicle asks periodically for its reputation from the nearest RSU In this system when vehicles report a message they attach their honesty reputation and their neighboring vehicles ID to the message The message receiver evaluates the reliability of the message based on the reporter vehicle s class and reputation and also based on the percentage of its neighboring vehicles reporting the same message Since trust issue is inherently coupled to uncertainty fuzzy logic as an effective solution to uncertainty problems is utilized to evaluate the reliability of messages After evaluating messages the honesty reputation of the lieing vehicles will be degraded Simulation results demonstrate that the proposed bogus information detection system in this project is highly accurate even in the case of collusion attack or a highly dynamic topology Key words Vehicular ad hoc network malicious behaivior trust management fuzzy inference system

ميرحيدري، عباس

تشخيص اطلاعات غلط در شبكه هاي اقتضايي خودرويي

رفتارهاي بدخواهانه , مديريت اعتماد , سيستم استدلال فازي