Research On Massive Machine Type Communication Access Mechanism Based On Services Classification

With the widespread coverage of cellular networks and the popularity of smart devices,mobile network provides comprehensive services for people's lives.Communication has evolved from simple human-to-human communication to human-to-machine communication and even machine-to-machine communication.The 3rd Generation Partnership Project(3GPP)defines massive Machine Type of Communication as one of the three major scenarios in 5th-Generation(5G).With the proliferation of a large number of machine type communications,the access network has presented enormous challenges.To cope with this kind of challenge,this thesis studies the characteristics of random access,finds its resource bottleneck,and proposes an access mechanism based on service classification,which has high theoretical significance and practical value.This thesis takes the smart grid as an example to study the characteristics of machine type communication.Firstly,the number of machine type communication nodes is huge,the data is periodically uploaded,and the number of access requests is larger than that of human type communication.Existing preamble resources cannot meet the access requirements of machine type communication.This thesis proposes to reduce the number of user access requests through clustering.Secondly,the machine node transmits general information and critical events in the process of information transmission.The way to classify users in the original access network cannot meet the access requests of the machine nodes.Based on this,this thesis innovatively proposes service classification and adopts different transmission methods for different services.This thesis first clusters machine users through the Affinity Propagation(AP)clustering algorithm to reduce the number of users requesting access at the same time.The classified user needs to transmit the information to the cluster head and then initiate an access request.In this thesis,different transmission modes are used for different priority services in the cluster,and the Priority Channel Access(PCA)algorithm is used for high priority services,and Carrier Sense Multiple Access with Conflict Avoidance(CSMA/CA)is adopted for low priority services.Ensure the transmission of high priority services.Then,this thesis introduces the random access process and analyzes the resource bottleneck in the random access process,that is,the fixed preamble resource.Based on the preamble resource strategy of resource partitioning,this thesis proposes a preamble resource partitioning strategy based on service classification.By comparing and simulating the classification of services,the access success rate of high-priority services in machine-type communication can be improved in a targeted mannerBased on the above research,this thesis proposes a resource partitioning scheme for clustered machine users and human users.First,the access success rate of human communication is ensured by resource division.Second,access to high-priority services and low-priority services for machine type communication in different ways.Access low-priority services through random access and access high-priority services through distributed queues.The demand for preambles of distributed queues is dynamically changed.If the high-priority service does not use the corresponding preamble resource,the resource will be released to the low-priority service for transmission to improve the utilization of the preamble resource.In this thesis,the priority of high-priority services in machine type communication is higher than that of human type communication,so the preamble resource with the maximum number of Collision Resolution Queues(CRQ)is reserved in the resource allocation process.Therefore,in order to balance the high-priority services and human communication of machine type communication,the number of CRQs needs to be dynamically changed with the number of human type communication users and the number of machine type communication in the transmission process.