| The street lighting system,which is a component of the city infrastructure,exhibits a tendency toward digitalization,connectivity,and intelligence.This development is driven by the present advanced network communication and artificial intelligence technologies.A smart lamp post can detect environmental information,calculate and process data,send data steadily,safely,and consistently,as well as satisfy the demands of peripheral expansion with multi-bus interfaces,energy conservation,and intelligent monitoring.It is a key component of the intelligent city.Smart lamp posts in particular are network edge devices,and they have significant research and application value for urban intelligence.Their own data computing and processing skills are the fundamental assurances for urban security and intelligent regulation of street lights.This thesis explores the core processing algorithm theory and technology,studies the 5G smart gateway oriented to the application scenario of smart lamp posts as the carrier,conducts system requirements analysis and architecture design,and completes the core of data collection,calculation processing,and network transmission based on the edge gateway.This work builds an experimental framework for a multipurpose smart lamp post system and completes test verification.The following are the primary elements of this work:1.The research examined the intelligent lighting post peripherals’ interface bus characteristics.An MCU-based multi-bus data acquisition and lamp post control solution was created in response to the requirements of lamp post control and the development of the number of peripherals.Along with its embedded firmware,the hardware block diagram for MCU-based environmental sensor data acquisition and lamp post peripherals control was also developed.Additionally,the edge gateway’s acquisition of highresolution video images and environmental sensor data was implemented employing the epoll mechanism.2.It built and implemented the edge processing core module application program and software system,and conducted research on the adaptive street lamp control algorithm based on the edge gateway.The adaptive controller for street lights is created using the fuzzy control theory.According to the simulation and test results,there is a 14%energy savings compared to the street light control.The extraction of the license plate based on the open source framework and the identification of the density of people and cars in the area are realized in real-time at a frame rate of 30 frames per second using the deep neural network intelligent approach.processing done at the edge during target detection.The processing of environmental target detection and fuzzy control tasks are carried out concurrently using multi-process/multi-threading,and the output result of environmental target detection is used as the decision input of the fuzzy controller to complete the collaborative task processing.3.It establishes a data transmission system based on message queues and streaming media servers,designs a smart lamp post message data and high-definition video data transmission module based on 5G communication,and implements data collection,calculation processing,and 5G communication transmission based on edge gateways.Additionally,the design and system testing of the human-computer interaction interface for gateway state monitoring are realized using embedded Web technology.The 5G multifunctional smart lamp post system under research in this thesis employs the gateway as the carrier and executes cutting-edge technologies like 5G communication,deep learning,and the Internet of Things in smart lamp posts.It also features various capabilities such as data collection,calculation,caching,and transmission at the edge of the network.For further studies on the engineering utilization of edge computing processing in the fields of intelligent lighting strategy and real-time environment sensing,the system’s fusion application serves as a reference. |