Optimization And Research On Data Collection Strategies In Smart Cities Based On Vehicles And UAV

With the rapid development of Internet of Things and communication technology,the concept of smart city has gradually become a reality.The basis of building a smart city is huge numbers of smart devices with sensing and communication functions.These smart devices can adjust their functions according to the needs of different Io T applications to form an intelligent city sensing network,so as to improve the service level and operation efficiency of the city.However,there is a challenging problem,which is how to efficiently and reliably collect data from these widely deployed and huge numbers of smart devices.In response to this problem,based on the relevant urban data,this thesis studies the technologies such as wireless sensor network,intelligent sense,data mining,deep learning and machine learning,and designs the urban data collection scheme and related optimization strategies,so as to improve the efficiency and performance of data collection in the city.The main work and innovations of this thesis are as follows:(1)Aiming at the high cost of data collection in smart city and the difficulty of data collection of some smart devices,a scheme of data collection using the combination of vehicles and UAV as the data collection carrier is proposed.The scheme is mainly composed of vehiclesbased data collection task and UAV-based data collection task.For smart devices near urban roads,this scheme recruits some taxis to install data transceivers on them for opportunistic data collection at a low cost.When taxis pass by smart devices on the way to work,the data transceivers in them will establish a connection with smart devices for data collection.For smart devices in areas where vehicles cannot reach such as high altitudes,lakes,and parks,the scheme uses UAV to carry data transceivers to the locations of these smart devices for data collection.The experimental results show that the data collection rate of this scheme is 8.7% higher than that of other schemes.(2)Aiming at the problem of low data collection rate due to the limited storage of data transceivers in vehicles,a method of constructing interim data storage is proposed.When the vehicle passes near the interim data storage,all the collected data will be transmitted to the interim data storage,and delete the data after the data transmission to release the storage space.However,the construction of interim data storages requires a certain cost.Therefore,in order to maximize the benefits of data collection tasks at a relatively low cost,this thesis proposes an optimization strategy for interim data storages.Firstly,this strategy designs the performance evaluation indicators of data collection tasks.And then,based on the benefits of data collection rate brought by interim data storages and the impact of increased data collection costs,an optimization algorithm is constructed to calculate the optimal number of interim data storages.The experimental results show that the proposed method can improve the data collection rate by 39.3%while ensuring low cost.(3)In the above scheme,the smart device data transmitted from the vehicle can be received only when the interim data storage is within the communication range of the vehicle,while the communication range of the data transceiver in the vehicle is limited and the driving route of the vehicle is random.In view of the above problems,this thesis proposes an interim data storage location optimization strategy.Firstly,the strategy obtains the coordinate points with large traffic flow in the city by processing the historical trajectory data of vehicles.Next,the coordinates of these trajectory points are clustered based on the number of interim data storages and the communication range distance of the data transceiver in the vehicle.After that,an evaluation method for the trajectory point coordinate clustering problem is designed.Finally,multiple experiments are performed and the experimental results are comprehensively evaluated to determine the final location of the interim data storage.The experimental results based on the real vehicle trajectory data show that the proposed strategy can reduce the data collection time by 17.1%,and greatly improve the efficiency of data collection.(4)Aiming at the problems of high cost and limited energy of UAV,a UAV path optimization strategy based on CHNN model is proposed.Firstly,the UAV path optimization problem is mapped to a CHNN model.Next,the constraints and objective functions of the model are constructed according to the restriction rules for the UAV to perform data collection tasks.Afterwards,the energy function and equations of state are obtained by combining the constraints and the objective function to evaluate the convergence of the model.Finally,the state equation and energy function are optimized to solve the shortest path of UAV.The experimental results show that the proposed strategy is able to reduce data collection time by19.92% and data collection cost by 20%.Which verifies that the strategy can improve the efficiency of UAV data collection task.