基于随机接入的大规模低功率广域网络建模与优化研究

With the rapid growth of Machine-to-Machine (M2M) communications, low power wide area (LPWA) technologies emerge as the times require. As one of the most popular medium access control (MAC) protocols, random access has been widely adopted in LPWA networks as the access mechanism due to its simplicity and low cost. However, the performance of traditional random access mechanism will sharply degrade in the presence of massive access requests. Therefore, it is of significant importance to investigate how to optimize the network to efficiently support massive access from huge number of end devices in LWPA networks. This project is devoted to the performance optimization problem of large-scale LPWA networks based on random access by studying four issues, including analytical modeling, performance optimization, proactive random access strategy with channel state information at transmitter (CSIT) and performance optimization of heterogeneous LPWA networks. First of all, based on the characteristics of LPWA networks i.e., huge number of end devices and dynamic changing of the traffic load, a new analytical framework will be established for the ALOHA-based wireless LPWA networks, which enables the studying of adaptive random access strategy for LPWA networks according to the dynamic changing of the traffic load. Based on the model, this project will investigate the effect of channel fading and the receiver structure on the network performance, characterize the limiting performance of ALOHA-based LPWA network and study how to properly adjust the system parameters to achieve it in wireless environment. Moreover, to further improve the network performance, this project will study the proactive random access strategy with CSIT by utilizing the channel-sensing ability of M2M end devices. Finally, the project will focus on the practical scenarios to investigate the performance optimization and QoS (quality-of-service) guarantee of heterogeneous LPWA networks with multi-service coexistence. This project will lay a solid theoretic basis for the development of future LPWA technologies.

随着M2M通信迅速发展，LPWA技术应运而生。随机接入以机制简单、低成本的特点被广泛应用于LPWA网络。面对传统随机接入机制在海量接入下性能显著下降的缺陷，如何优化网络性能以支持海量终端高效接入是亟待解决的问题。本项目以基于随机接入的大规模LPWA网络性能优化为核心问题，围绕建模分析，性能优化，基于信道感知的主动随机接入策略和异构环境下网络性能优化四个方面进行研究。首先，针对LPWA网络节点数目庞大、负载变化显著的特点建立基于ALOHA协议的分析框架，设计可根据网络负载变化自适应调节的随机接入策略；基于此框架，研究无线环境下信道衰落以及不同的接收机模型对网络性能的影响，揭示网络的极限性能；为进一步提升网络性能，利用终端可感知信道的能力，研究基于信道状态信息的主动随机接入策略；最后针对实际应用场景，研究多业务并存下的网络性能优化和服务质量保证。本项目可为LPWA技术的发展打下坚实的理论基础。

随着物联网技术的迅猛发展，越来越多的机器型设备会融入到现有网络架构中以满足各种应用需求。如何支持海量机器型设备高效接入成为了亟待解决的问题。本项目面向采用非授权频谱的基于随机接入的大规模低功率广域网络(LPWAN)，综合考虑传输环境设计适配的接入机制，并以网络吞吐率、接入时延和传输速率等网络性能的优化为核心问题。.针对M2M终端数目庞大、负载动态变化显著的特点，研究可根据节点负载动态变化自适应调节的随机接入机制；针对M2M终端面临的无线环境，建立包含无线信道特性和接收机特性等在内的分析模型研究衰落信道和不同的接收机结构对网络性能的影响，以网络吞吐率、接入时延以及和速率为研究对象，刻画网络的极限性能并提供最优的参数配置方案。为了进一步提升接入效率，本项目同时也研究了基于重传限制和自动组分页的随机接入机制，推到出了网络的极限性能。基于上述研究，本项目进一步考虑了在服务质量保证下的网络优化问题，刻画出了在速率约束下的最低接入时延，并将此优化方法应用于智能电网等实际应用场景中，为实际应用提供了指导方案。最后，本项目为解决单一基站覆盖范围有限的难题，探讨了中继辅助传输技术的可行性，设计了基于混合比特级编码的中继网络传输机制。