Performance Evaluation And Resource Allocation For DSRC Based Vehicular Safety Communication

As the economy develops rapidly in the whole world,global car ownership continues to increase.Though severity amelioration technologies such as airbags,seat belts,and automatic braking system have been used for many years to offer passive protection to vehicle occupants,the fatalities and injuries rate of traffic accidents have increased.Traffic safety has become a serious problem restricting the rapid development of modern society.Intelligent transportation system(ITS)is one of the most effective technical methods to settle the traffic safety problem.Dedicated short range communication(DSRC)enables vehicles to communicate with each other leading to vehicular ad hoc networks(VANETs),which is the most crucial component of ITS.Based on VANET,vehicular safety communication(VSC)exchanges safety messages among vehicles in real-time to remind drivers of potential hazards and improve traffic safety.Usually,safety applications have strict performance requirements on VSC while many factors existing in the vehicular communication environment affect the performance of VSC.To verify whether the existing protocols can meet the timeliness and reliability requirements of safety applications,this work analyzes the performance of DSRC-based vehicular safety communication mathematically.Moreover,DSRC is allocated a fixed 10 MHz control channel specified for safety-related communication and the growing number of vehicles will eventually result in channel congestion,thereby degrading the network performance.To alleviate the contradiction between the excessive demand for spectrum and the shortage of radio resources,this work has researched the resource allocation in DSRC-based vehicular safety communication.On the one hand,by designing more efficient congestion control algorithms,congestion control can be avoided;on the other hand,by utilizing cognitive radio(CR)technology,vehicles can opportunistically access the white spectrum.The main works are as follows:(1)Based on the previous discrete-time Markov chain(DTMC)models,a more comprehensive analytical model is proposed to evaluate the performance of basic safety message broadcasting adopting single-class IEEE 802.11 p Enhanced distributed channel access(EDCA)protocol.The model takes the impacts of various incoming traffic loads,packet length distribution,hidden terminal effects,node mobility,the MAC layer queuing system,and the faulty radio channels into account.Three average metrics among all receivers within the sender's transmission range are defined including packet delay,packet delivery rate,and packet reception rate are defined,and their closed-form expressions are derived.(2)To overcome the approximation existing in DTMC based models,a more general yet accurate analytic model is constructed to investigate the performance and reliability of multiple safety messages broadcasting in DSRC based VSC with IEEE 802.11 p EDCA.The newly developed model takes all the factors such as the IEEE 802.11 p EDCA backoff counter process,unsaturated packet arrivals,limited MAC queue length,hidden terminals,Nakagami-m fading channel into consideration.Three distance-related metrics including packet delay,packet delivery rate,and packet reception rate are defined to provides a deeper insight into the reliability as the range changes,and their closed-form expressions are derived.(3)For a platoon adopting a synchronized P-persistent repetition(SPR)medium access control(MAC)protocol,an optimization problem aiming at searching for an optimal beacon transmission rate allocation for platoon management is developed based on a network utility maximization framework and then solved to obtain optimal beacon transmission rate allocations of the platoon at any certain cruise velocities.(4)In the study of spectrum sensing for cognitive vehicular networks,a maximum eigenvalue based primary user(PU)signal detection algorithm with multiple antennas is proposed considering the existing maximum eigenvalue-based spectrum sensing with a single antenna is vulnerable to noise level,and the closed-form expressions of threshold and false alarm probability are derived.Furthermore,a relay-based individual primary receiver(PR)detection for cognitive radio(CR)networks is designed to find the opportunistic spectrum access that is identified as invalid in conventional spectrum sensing algorithms and its miss detection probability and false alarm probability are derived.Moreover,the proposed individual spectrum sensing scheme is further developed to a new collaborative one to combat the hidden terminals problem and channel fading effect in VANET,and the corresponding miss detection probabilities and false alarm probabilities are derived under fusion rules “AND”,“OR” and “K-out-of-n”.