Probabilistic Adaptive Anonymous Authentication in Vehicular Networks

Vehicular networks have attracted extensive attention in recent years for their promises inimproving safety and enabling other value-added services. Mostprevious work focuses on designing the media access and physicallayer protocols. Privacy issues in vehicular systems have not beenwell addressed. We argue that privacy is a \it user-specificconcept, and a good privacy protection mechanism should allow usersto select the levels of privacy they wish to have. To address thisrequirement, we propose an adaptive anonymous authenticationmechanism that can trade off the anonymity level with computationaland communication overheads (resource usage). This mechanism, to ourknowledge, is the first effort on adaptive anonymous authentication.The resources used by our protocol are few. A high traffic volume of2000 vehicles per hour consumes about 60kbps bandwidth, whichis less than one percent of the bandwidth of DSRC (Dedicated ShortRange Communications). By using adaptive anonymity, the protocolresponse time can further be improved 2\sim4 times with less than20\% bandwidth overheads.