Research On Data Aggregation In Vanets

With the rapid development of wireless communication technology, information sensor technology and automatic control technology, Internet of Things(Io T) has been widely applied in many fields, such as agriculture, transportation, logistics, smart home and so on. As an important Io T application in the transportation field, Verhicular Ad-hoc NETworks(VANETs) provide one technical support platform for Intelligent Transportation Systems(ITS). Many applications of ITS need to collect and disseminate a lot of realtime data. However the mobility of the vehicles makes communication resources that are utilized during the processes of data collection and dissemination be changing frequently. At the same time these communication resources are extremely limited. Therefore,data collection and dissemination problems in VANETs are very challenging. In order to decrease the amount of transmission data in the processes of data collection and dissemination, data aggregation technology is introduced. This technology has been one of the hottest research points in the VANETs field by the academics. Data aggregation can save the communication resource consumption in the process of data transmission, but it also has an effect on the delay and accuracy of data, which directly affect the correctness and effectiveness of many applications in VANETs. Data aggregation problems in VANETs are researched in this thesis. The main research issues and results include the following.Firstly, this thesis researches the deadline constrained data aggregation problem in VANETs. Many applications in VANETs need low-delay requirement of sensing data.The delay of sensing data is greatly affected by the dynamic communication in VANETs.Currently, data aggregation algorithms in VANETs has not deeply analyzed and utilized the dynamic comunication information. They could not guarantee the delay of sensing data. Therefore, this thesis proposes the deadline constrained data aggregation scheduling problem in VANETs, which is to schedule the aggregation transmission time of nodes for optimizing the amount of aggregation information. The thesis defines this problem and proves that it is NP-hard. The thesis proposes an aggregation information optimal algorithm with delay guarantee. The key point of this problem is how to avoid the wireless conflicts in the process of data transmission. Wireless conflicts can be classified to two categories, one of which is the conflict caused by simultaneously sending data from multiple nodes to one node, called as direct conflict; the other of which is the conflict caused by the sending data from one node in the communication region of whose there is another node that is receiving data, called as indirect conflict. This thesis proposes the data aggregation transmission time scheduling algorithm under the candidate time set of nodes based on dynamic programming for direct conflict avoidance. The thesis proposes to filter the candidate time set based on match and independent set for indirect conflict avoidance.We evaluate the performance of the algorithms on real cab trajectory data set. The results show the data collection rate of our proposed algorithm is higher over 20 % than other algorithms, average data transmission rate lower to 1/3, average delay more stable.Secondly, this thesis researches the deadline constrained data aggregation problem in multi-channel VANETs. To improve the transmission capacity of the network, the standard protocol stack in VANETs introduce multi-channel mechanism. To compare with single channel, the different receiving nodes can utilize different channeles to decrease the wireless conflicts, but it need to solve the channel allocation problem. Therefore, this thesis proposes the aggregation information optimization algorithm with delay guarantee in multi-channel VANETs. The basic idea of this algorithm is to utilize multiple channels to avoid the indirect conflicts. The key point is to constraint the number of indirect conflicts, which is less than the number of channels k. Firstly, the thesis creates k-constrained optimal aggregation routing tree to schedule the route of aggregation data. Next, the thesis proposes aggregation transmission time scheduling algorithm. This algorithm utilizes dynamic programming to schedule the transmission times and channels of the nodes. The time scheduling can get optimal solution. At last, we conduct a lot of experiments on the real cab trajectory data set and simulation trajectory data set. The results indicate the data collection rate of our proposed algorithm is higher over 25 % than current algorithms, and surpasses 90 %. The efficiency of data aggregation has been improved above 3 times than other algorithms.Moreover, this thesis studies accuracy constrained data aggregation problem in VANETs. Different from deadline constraint, the key point of aggregation with accuracy guarantee is to utilize the correlation of data to decrease the amount of data transmission. Current researches introduce compressed sensing to compress the amount of data and guarantee the accuracy of recovered data. However, compressed sensing need to meet RIP condition,which means that data need to be aggregated by the specialized way. Because the data is taken by mobile vehicles, data aggregation is constrained by the mobility of the vehilces.Existing methods use artificial introduction of the vehicle to controll the trajectories of the vehicles, but they need to consume much resource and not to utilize wireless communication resource that it exists in VANETs. Therefore, this thesis proposes an aggregation transmission protocol based on compressed sensing in VANETs. The protocol utilizes wireless communication between vehicles to adjust data carried by mobile vehicles in order to realize data aggregation based on compressed sensing. Experimental results show that the proposed method can recover data that satisfies the requirement of accuracy by using only half of the origin data.Finally, this thesis studies multi-precision constrained data aggregation problem in VANETs. Because of the limited communication resource in VANETs, the realtime routing guidance system need to decrease the amount of transmission data in the process of real-time road information aggregation. The uneven distribution of guidance requirements makes that the accuracy of some roads affects the correctness of more guidance requirements. The accuracy of aggregation data decides the amount of transmission data. According to the accuracy requirements of each road, each road decreases the amount of aggregation data under satisfing many guidance requires. Different from single accuracy requirement, data aggregation with multi-precision guarantee need to get multiprecision requirements, distribute multi-precision aggregation queries and aggregate data with multi-precision guarantee. Therefore, this thesis proposes a data aggregation protocol with multi-precision guarantee. Simulation results show that data collection rate of our method reaches about 80 %, improved about 20 % than the naive method, and bandwidth consumption is as low as 2.5 times the amount of data gathered, 1/3 lower than the naive method.