Research On NB-IoT Uplink Resource Optimization And Energy Saving Strategy

Narrowband Internet of Things(NB-IoT)has received great attention since its inception,as one of the low power wide area network(LPWAN)technologies,it has the characteristics of low power consumption,narrow bandwidth,strong coverage,low cost,and massive connections.Compared with other wireless communication technologies,it has numerous advantages.With the beginning of 5G commercial,NB-IoT will have greater development space in the field of 5G Internet of Thing.While it has many advantages,and also faces the challenges of resource shortage and low energy consumption.In order to cope with these challenges,this thesis mainly studies the problems of resource and energy consumption in the NB-IoT uplink.First,based on the characteristic of narrow bandwidth in NB-IoT,this thesis proposes an access control algorithm that can reasonably allocate uplink resource to narrowband physical random access channel(NPRACH)and narrowband physical uplink shared channel(NPUSCH)to maximize throughput.Secondly,in order to fulfill the requirements of low energy consumption,prolong the life of the device and reduce the energy consumption of communication.This thesis proposes a random access energy saving algorithm based on Q-learning.The main research contents of this thesis are as follows:First,considering the existing allocation method lacks an uplink resource allocation algorithm that can cope with various situations,this thesis proposes an adaptive access control algorithm.In this algorithm,the base station first estimates the maximum number of accessible devices by performing exhaustive calculations on parameters such as the number of repetitions and preambles,and the size of data.Then,the base station adjusts the NPRACH and NPUSCH resource according to the current network load conditions,and controls the number of access devices to maximize throughput.The simulation results show that,compared with the traditional algorithm,the proposed algorithm can maximize the number of successful communication devices in various situations.Second,considering that NB-IoT does not have an energy saving solution for the random access process,this thesis proposes a random access energy saving algorithm based on Q learning.In this algorithm,the base station is acted as the agent,the access control factor is defined as the action,the energy consumption and throughput are defined as the state,The agent gets corresponding rewards by choosing appropriate actions.Based on the proposed Q-learning framework,the agent can continuously interact with the environment and learn the optimal access factor,which can save the energy consumption of the random access process under the premise of ensuring throughput.Simulation results show that the energy saving algorithm based on Q-learning can save about 20% of energy consumption compared with traditional algorithm.