Research On Key Technologies In Cluster Oriented Resource Scheduling Optimization For IoT

With the development of computer and network technologies,Internet of Things(IoT)has been applied in all walks of life.Resource scheduling,as the premise and foundation for data transmission and management in IoT systems,plays a critical role in ensuring the performance of IoT and avoiding interference in data transmission.However,many resource scheduling solutions give little consideration to the problem of "energy hole" and ignore the timely recovery and reuse of the network resources allocated to nodes when they fail.In addition,many existing priority-based resource scheduling have to preset the priority of data,which is impractical in certain scenarios,and such approaches cannot detect multiple types of emergency data at the same time.In heterogeneous IoT,the time slot reuse rate is difficult to improve and the utilization rate of network resources is low.By dividing network into several clusters,the clustering technology manages multiple member nodes through the cluster head,thus reducing IoT energy consumption and prolonging the life of the network.To solve the above problems,this paper,combined with the clustering technology of IoT,proposes key technologies in cluster-oriented resource scheduling optimization for IoT.Its main research contents are as follows.(1)IoT resource scheduling based on unequal clustering:Through the research of energy-saving and efficient IoT unequal clustering scheme based on dynamic topology,the phenomenon of "energy hole" in multi-hop IoT is alleviated.Based on the unequal clustering scheme,a dynamic routing and cluster head layering mechanism is put forward to identify optimal relay nodes.On the basis of unequal clustering and hierarchical routing,an IoT resource scheduling based on unequal clustering(RSUC)is proposed;The intra-cluster communication scheduling of RSUC is purposed to adapt to the dynamic changes of topology caused by failed nodes,and recovery the resources from failed nodes.In intercluster communication scheduling of RSUC,different transmission time slots were used for the odd layer and even layer to make the data transmitted in parallel.Simulation results show that by reasonably allocating time slots and channels of intra-cluster and inter-cluster communication,the total number of time slots and energy consumption of the network are effectively reduced,and the network throughput is improved.(2)Resource scheduling of unequal clustering IoT based on priority of data change rate:On the basis of RSUC resource scheduling scheme,the emergency data scheduling scheme is introduced.By dynamically adjusting the time weight of each type of data change rate,the offset between the observed and normal values of the data change rate is minimized.Under the convex optimization theory,a normal data change rate model is acquired to demonstrate the normal operation state of IoT;The priority of data can be judged in real time through judging difference between the observed and normal data change rate,a resource scheduling of unequal clustering IoT based on priority of data change rates(UCPDR)is proposed.UCPDR can identify various types of emergency data,without making settings in advance,in which time slots and channel resources of intra cluster and inter cluster communication can be preferentially allocated to emergency data according to the real time priority of data change rate.Simulation results show that UCPDR has high accuracy in emergency data detection and can effectively reduce the network delay for emergency data.(3)Resource scheduling of unequal clustering IoT based on detection matrix:In heterogeneous IoT scenarios with base stations located far away from data acquisition area,different resource scheduling schemes are adopted by data acquisition nodes and routing nodes.In the routing and forwarding transmission phases,through resource scheduling based on routing tree(RSRT),the channel allocation and time slot allocation in a round can be obtained.by which,data in single round can be transmitted to base station without conflicts with the least time slot in RSRT.On basis of the RSRT,a resource scheduling of unequal clustering IoT based on detection matrixes(UCDM)is put forward.The minimum time slot interval of continuous periodic scheduling is determined to avoid wasting network resources in serial scheduling in UCDM.Simulation results show that time slot multiplexing technology can effectively improve the slot utilization of continuous transmission of the IoT and improve the network throughput.(4)System for unequal clustering IoT data monitoring:The resource scheduling proposed in this paper is deployed in the real-world IoT data monitoring project to verify its feasibility and effectiveness.This system provides a foundation for equipment-fault diagnoses.