Research On Real-time And Reliable Transmission Strategy Based On Load Balance Of Manufacturing Of Internet Of Things

As a large-scale integrated network for the whole process of manufacturing, the manufacturing of internet of Things includes data acquisition, transmission, processing, control and service, in order to realize optimal scheduling control and dynamic information service for complex industrial manufacturing process. In the real time transmission layer, there are the following characteristics of the manufacturing of internet of Things:the network structures are highly heterogeneous, wireless network is large scale multi hop, network topology changes dynamically and transmission data types are various. So the existing network transmission strategies are very difficult to meet the QoS requirements of different services in large scale heterogeneous hybrid networks.According to the existing problems that load imbalance, real-time and reliability could not be guaranteed in the process of data transmission for the manufacture of Internet of things and combining the current research status at home and abroad, taking into account the special environment characteristics and requirements of manufacturing industry, this paper carried on the thorough research for real time and reliable transmission strategy based on load balancing for manufacturing of internet of Things on the basis of analyzing the network architecture of it. The specific work is as follows:(1) Firstly, It constructed the network architecture of the manufacturing transmission network by analyzing the structure and characteristics of the transmission network, and introduced two common network data transmission strategies:cluster routing strategy and multipath routing strategy.(2) A dynamic adaptive clustering strategy based on load balancing and energy efficiency was designed. Through the cluster management of the access layer of the manufacturing of internet of Things, the heterogeneity of the terminal layer was shielded. It considered the two factors of load balancing and energy efficiency for clustering methods. The PPR discrete particle swarm optimization algorithm was used to search the optimal allocation scheme, which was reasonable to divide the cluster range of each wireless Mesh router. This algorithm had a very good inhibitory effect on the load imbalance and network congestion caused by the dynamic change of the data flow in the manufacturing industry.(3) It designed a new a real time and reliable multipath routing strategy based on genetic algorithm. According to the multi dimension QoS demand discrete manufacturing networking in dynamic business data stream transmission, it put forward a kind of real time reliable multi path routing strategy for manufacturing of internet of Things- MPCRA-GA. Firstly, the algorithm obtained initial population of genetic operation by improved DSR protocol, and then output the optimal path collection after the multiple genetic operations. According to the structure of the multi path, the source nodes allocated and transmitted date based on path cost of roulette wheel selection method. In this way, it ensured the delay, reliability and other requirements of the QoS and balanced the load. Genetic algorithm has good global search ability and non orientation, and it also has a good performance in the path optimization.Finally, the simulations were done to the proposed the two algorithms, and the simulation results were in accord with the expectation of the algorithm. The dynamic adaptive clustering strategy based on load balancing and energy effectively shielded the heterogeneity of the underlying terminals and achieved the balance of the cluster load at lower communication cost; The real time reliable multi path routing strategy based on genetic algorithm improved the real-time and reliability of data transmission, and balanced the communication load of multiple transmission paths.