| With the vigorous development of the Internet of Things technology,all walks of life are also driven by the Internet of Things technology towards the direction of precision and convenience.In view of the existing parking system in the process of use exposed parking resources management is not fine enough,the function is not perfect,use is not convenient and so on.In order to solve the above problems,this design combined with the Internet of Things technology proposed a set of solutions.This design realizes a set of parking system based on ARM front-end hardware and mobile APP,so that the hardware and software cooperate with each other to improve the overall operating efficiency.The hardware part designs two kinds of ARM board.The NB-Io T communication module of the main ARM board communicates with the cloud server through the MQTT protocol,and the LORA communication module communicates with the sub-ARM board through the gateway,and completes the generation and parsing of QR codes.The camera of the sub-ARM board is responsible for collecting the QR code presented by the user,and the LORA communication module communicates with the LORA of the main ARM board for data.The two ARM boards work together to form the front-end hardware.The Android-based mobile APP with front-end hardware equipment and Baidu Map API realizes the functions of parking space reservation,code scanning and access car,billing by the hour,addressing outside the parking lot,addressing inside the parking lot and so on.Compared with the traditional parking billing system,this design realizes the field addressing through the improved Dijkstra shortest path algorithm,which to some extent makes up for the shortage of the commercial map resources without the field map resources,and improves the function of the parking system.Each parking space is controlled remotely by front-end hardware equipment,which improves the fine management of the parking system.QR code is used as a means to verify the identity of the car owner,which improves the convenience and security of the parking system.Considering the performance of this system,firstly combining with the existing system’s QR code bad code recognition rate,it puts forward the requirement that the bad code recognition rate can reach 100% in the case of QR code damage less than 30%.Secondly,Dijkstra shortest path algorithm is improved based on binary tree sorting algorithm,which reduces the time consumption of Dijkstra algorithm by about 8%.Finally,in view of the actual situation that there will be a large number of concurrent requests during the peak of parking lot usage,a design index that the main ARM can support 100 concurrent requests is proposed. |