Research On M2M Radio Resource Management Based On Non-Orthogonal Multiple Access In Unlicensed Bands

With the continuous advancement of the technology research and industrialization process of the 5th Generation(5G)Mobile Communication System,the efficient and seamless communication scenarios for "human to things" and "things to things"gradually appear,which makes machine to machine(M2M)communication technology play an important role in the field of Internet of Things(IoT)for 5G because of its full automation and ready-to-use.However,the large-scale connections,periodic and frequent communication and requirements for low latency generated by the M2M services put higher demands on the communication system with the limited wireless resources.On the one hand,the limited licensed spectrum cannot meet the requirements of the communication for massive M2M equipment.One the other hand,the transmit power cannot exceed the maximum value because of low power consumption and inter-cell interference.At the same time,the development of 5G technology in the field of wireless access networks and multiple access has brought new opportunities to existing communication systems.Power domain-based non-orthogonal multiple access(NOMA)technology can achieve multi-user multiplexing of the same time-frequency resource through different power,so that it can significantly improve resource utilization.The Cloud Radio Access Network(C-RAN)improves the flexibility of the net by separating the baseband function and radio frequency function of the base station,and also brings favorable conditions for multi-point cooperative communication.Therefore,deploying the NOMA-based M2M communication system with the C-RAN architecture in the unlicensed spectrum and research on the radio resource management is a key method to solve the shortage of wireless resources and improve the performance of the M2M systems.Radio resource management for NOMA-based M2M communication systems deployed in unlicensed spectrums is studied in this thesis.The main contents and innovations are as follows:Firstly,a NOMA-based M2M communication system model with C-RAN architecture is proposed to improve system flexibility and resource utilization while providing a hardware foundation for efficient cooperative communication of multiple base stations.Secondly,based on the existing research on the coexistence of different systems,a coexistence method based on frequency hopping is proposed for the M2M system and Wi-Fi system in unlicensed bands.The synchronization of frequency hopping is realized by inputting the same time-varying parameters to the frequency hopping pattern generator of the sending device and the receiving device.The simulation proves that the method can reduce the impact of the existing method on the Wi-Fi system by more than 50%and has an ability to resist the inter-cell interferenceThirdly,a power allocation method based on fairness for multi-base stations is proposed.The multi-base stations' power allocation problem in the users' group is decomposed into multiple single base station's power allocation problems with the certain power allocation ratio between base stations,and the linear program is transformed.The linear problem is solved by the dichotomy method to determine the optimal solution.The sliding window method is used to iterate the above process to determine the power ratio between the base stations.The simulation result proves that the algorithm ensures the fairness and improves the users,rate and the system performance is better than both the orthogonal multiple access system and the system with a single base station.Fourthly,based on the above power allocation algorithm,a user grouping strategy based on matching algorithm is proposed.This paper transforms the user grouping problem into a bilateral matching problem based on bilateral preference,and explores the performance of different preference lists.The effectiveness of the algorithm and the overall performance of the system are verified by simulations.