Research On Capacity Estimation And Optimization Technology Of NB-IoT Network

NarrowBand Internet of Things(NB-IoT)is a new-generation cellular IoT access technology defined by 3 GPP for low-power wide-area coverage services with wide coverage,large connections,and delay sensitivity.Low profile and low power consumption.With the introduction of NB-IoT technology in areas such as smart homes,smart hotels,and shared bicycles,NB-IoT networks can improve the requirements for QoS such as network throughput and system average delay in pursuit of a better user experience.Still maintaining a massive connection has become an important research topic.The 3 GPP analysis shows that the NB-IoT network has the characteristics of massive connection,but the test data of Nokia Networks shows that the performance of network throughput and system delay is not ideal under the current access mechanism.Due to its narrowband characteristics,the NB-IoT network has very limited preamble resources.In a complex business environment,it will inevitably lead to a fierce leading collision,which directly affects QoS indicators such as network throughput and delay.The network capacity studied in this paper refers to the QoS standard based on a certain network throughput and system average delay,and the maximum number of terminals that the NB-IoT network can accommodate.In this paper,based on a single-service,the working state of a terminal in NB-IoT network is divided into transmission state,active state,idle state,access state and backoff state,and a semi-Markov model is established.Calculate the steady-state probability and jump probability of a terminal,and then derive the NB-IoT network capacity estimation method for a single-service scenario,and analyze the impact factors such as the impact probability,network throughput and system time delay and other QoS characteristics and a capacity estimation model of NB-IoT network for the service.Considering the QoS standard,how to properly allocate limited preamble resources to optimize network capacity under complex service scenarios is another key point of this paper.In this paper,the NB-IoT network capacity estimation of complex business model is studied,and a cluster-based enhanced random access reusable preamble allocation algorithm is proposed.The clustering algorithm is designed to match the terminal in the network according to the packet report rate and the base station.The distance is clustered to allocate the same preamble set for the father clusters,and the total number of preambles is determined by the random access intensity of the clusters.Then the network capacity optimization algorithm proposed in this paper is simulated and analyzed.The results show that the proposed algorithm can effectively reduce the random access preamble collision probability,thus obtaining greater network throughput and lower system average delay,achieving NB.-IoT network capacity optimization.