Rate Adaptation And Channel Allocation In Wave And Implementation Of VANET Communication Experiment Platform

Under the background of fast development of Internet of Things(IOT),the development of Internet of Vehicles(IOV)under the framework of IOT has aroused the attention of academia,industry and standardization organizations at home and abroad.In the future,the vehicle as a mobile terminal can also realize voice communication,information broadcasting,road monitoring and so on.These functions all rely on large amounts of dynamic traffic data,images,audio,video and other information transmission.Therefore,how to transmit data in real time and effectively become an urgent problem in the wireless network environment.Thus,from the point of improving the throughput of vehicle network data transmission,this paper studies rate adaptation and channel allocation.In this paper,we first study rate adaptation problem in Wireless Access in the Vehicular Environment Protocol(WAVE).In different vehicle environment,different transmission bitrates result in different throughputs.Thus,selecting an appropriate transmission bitrate in the specific scene can greatly improve vehicle data transmission throughput.In this paper,we first introduce the current rate adaptation algorithms,including ARF,AARF,ONOE,SampleRate,Minstrel and etc.Current rate adaptation algorithms mainly focus on indoor WiFi protocol(IEEE 802.11a/b/g/n),they can not adapt to channel estimation delay caused by high dynamic environment and channel test blank caused by high density of vehicles in IEEE 802.11 p.In the outdoor experiment,we found that the throughput of one bitrate and location are strongly correlated.Thus,we establish the location-based database,which is used to record data throughput in different locations by different sending bitrates.Next,we propose a location-based rate adaptation algorithm LRRA.Based on the historical data in the database and the current channel quality given by Minstrel,LRRA predicts the throughputs of different data bitrates,and gives the recommended sending bitrate.Simulation in NS3 shows that LRRA can improve the whole system’s throughput because it can adapt to the channel changes quickly.In this paper,multi-channel allocation algorithm in WAVE protocol is studied.WAVE protocol uses a multi-channel mechanism,and sets 6 Service Channels(SCH).Sending devices first need to determine which SCH to access before sending messages.However,WAVE protocol does not specify how to choose the SCH.Due to the unbalance of each SCH,random access will inevitably lead to unbalance.Some SCHs will be too crowded while some are too vacant.Thus,by appropriate channel access,each SCH can be fully used and the overall system throughput can be improved.In this paper we propose a collision probability based algorithm CPCA.We first model the Enhanced Distributed Channel Access Transmission Mechanism(EDCA)in vehicle network.The model considers four kinds of Access Categories(AC).When the sending queues are saturated or not saturated,the coexistence of internal and external conflicts are all considered.In this model,according to the number of stations in a SCH and the amount of data in each AC,the data collision probability of the four ACs can be calculated.Therefore,when a new station wants to join the SCHs.Respectively,we calculate collision probability of each AC.Combining the data coming rules,we calculate the throughput in one AC,and further calculate the predicted throughput in one SCH.Finally,the algorithm selects the SCH that can send the maximum amount of data.This mechanism can distribute the SCH reasonably and increase the overall throughput of the SCH greatly.The simulation in NS3 illustrates the effectiveness of the proposed algorithm.At last,we construct a VANET communication experiment platform.This system is composed of two On Board Units(OBU)and one Rode Side Unit(RSU).OBU uses NI Universal Software Radio Peripheral(USRP)and RSU uses PCI Extensions for Instrumentation(PXI).Based on Orthogonal Frequency Division Multiplexing(OFDM)technology,channel estimation and frequency domain equalization,we successfully realize Vehicle to Vehicle(V2V)and Vehicle to Infrastructure(V2I)communication.By the experiment on campus,we laid the foundation of a VANET prototype.