Distributed and Self-organizing Heterogeneous Wireless Access Networks: Design, Analysis, and Optimization

下一代蜂窝无线通信网络将需要支持不断增加的用户数量以及各种富多媒体应用的数据速率的显著增加。这将需要在此类网络的设计和实施方式上进行一场革命。特别是，应对不断增长的用户数量和流量需求的两个关键解决方案如下：i)大规模部署流量热点，例如通过不同类型的小型蜂窝、中继和无线电头，以及（ii）使位于同一附近的设备之间实现直接连接，例如通过设备对设备通信。网络单元将使用不同的无线电接入技术，并在许可和/或未经许可的频段上运行。通过这种方式，蜂窝网络将演变为连接大量智能无线电设备的大规模和高度复杂的异构系统。随着新的网络元素和功能的大规模部署，分布式和自组织的网络操作对于实现可扩展性、稳定性、健壮性和敏捷性至关重要。有了自组织，外部实体的控制和人工干预将会减少，因此，网络运营商的运营支出将会降低。由于分布式和自组织蜂窝网络架构包含不同类型的网络元素、多个网络层和无线接入技术，因此需要重新审视用于遗留/传统蜂窝网络的所有网络功能和优化技术（例如，用于干扰管理、功率控制、准入控制、信道分配），并使其适应新的网络拓扑。特别是，需要开发可扩展和健壮的分布式算法，以实现频谱效率、功率效率和吞吐量方面的最佳网络运行。本研究的短期目标是设计、评估和优化分布式和鲁棒资源分配、流量卸载和负载平衡、自组织的认知和合作方法、节能技术和经济激励机制等先进技术，使下一代分布式和自组织无线网络具有实用性和商业可行性。本研究的长期目标是为下一代及以后的分布式和自组织无线网络技术的进步做出贡献。除了在蜂窝无线网络的建模、分析和优化方面做出新的理论贡献外，拟议研究所产生的技术还具有对第五代（5G）蜂窝网络新兴技术标准的演变产生重大影响的潜力。拟议研究的结果还将直接向加拿大工业提供潜在的商业和技术转让。科学的方法将基于健全的研究原则，包括文献调查和分析建模，结合广泛的数值分析和计算机模拟。这项研究将培养一批高素质的人才，以满足加拿大日益增长的信息和通信技术行业的需要。拟议研究的重要性来自：(i)对低成本无处不在的无线宽带服务的需求不断增长，（ii）需要培养专业知识和熟练的毕业生，将相关技术转移到加拿大的电信/IT行业，以获得我们的经济优势，以及（iii）目标是提高加拿大在无线/移动网络研究方面的世界领先地位。

The forthcoming generations of cellular wireless communications networks will be required to support the ever-increasing number of user populations along with significant increases in data rates for various multimedia-rich applications. This will require a revolution in the way such networks are designed and implemented. In particular, two key solutions to cope with the ever-increasing user population and traffic demand are as follows: i) massive deployment of traffic hotspots, for example, through different types of small cells, relays, and radio heads, and (ii) enabling direct connectivity between devices located in the same vicinity, for example, through device-to-device communications. The network elements will use different radio access technologies and operate in licensed and/or unlicensed bands. In this way, the cellular networks will evolve to large-scale and highly complex heterogeneous systems interconnecting a massive number of intelligent radio devices. With a large-scale deployment of new network elements and functionalities, a distributed and self-organizing network operation will be essential to achieve scalability, stability, robustness, and agility. With self-organization, minimum control from an external entity and manual intervention will be required, and hence, the network operators will incur low operation expenditure. With a distributed and self-organizing cellular network architecture incorporating different types of network elements, multiple network tiers and radio access technologies, all the network functionalities and optimization techniques (e.g., for interference management, power control, admission control, channel allocation) used for the legacy/traditional cellular networks need to be revisited and adapted to the new network topology. In particular, scalable and robust distributed algorithms will need to be developed to achieve optimal network operation in terms of the spectral efficiency, power efficiency, and throughput. The short-term objective of this research is to design, evaluate, and optimize advanced techniques for distributed and robust resource allocation, traffic offloading and load balancing, cognitive and cooperative methods for self-organization, energy-efficient techniques, and economic incentive mechanisms that make future generation distributed and self-organizing wireless networks practical and commercially viable. The long-term objective of this research is to contribute to the advancement of distributed and self-organizing wireless networks technologies through the next generation and beyond. In addition to novel theoretical contributions on modeling, analysis and optimization of cellular wireless networks, the technologies resulting from the proposed research have the potential of making significant impact on the evolution of emerging technical standards for fifth generation (5G) cellular networks. The outcome of the proposed research will also provide potential commercial and technology transfer directly to the Canadian industry. The scientific approach will be based on sound research principles involving literature survey and analytical modeling combined with extensive numerical analysis and computer simulations. The research will prepare a number of highly qualified personnel to serve the needs of the growing information and communications technologies industry in Canada. The importance of the proposed research stems from: (i) the growing demand for low-cost ubiquitous wireless broadband services, (ii) the need to produce the expertise and the skilled graduates to transfer the relevant technologies to the Canadian telecom/IT industry for our economic advantage, and (iii) the goal to enhance Canada's position as a world leader in wireless/mobile networking research.