CAREER: Theory and Algorithms for Efficient Control of Wireless Networks with Jointly Optimized Performance: High Throughput, Low Delay, and Low Complexity

随着智能设备和物联网的出现，无线技术催生了大量服务，涵盖了商业、科学和工程、娱乐、安全和安保、健康监测，并覆盖了我们大部分的社交活动。由于这些新服务的普及，今天的无线网络不仅见证了流量的空前增长，而且还见证了流量类型的显著变化(例如，具有更严格的延迟要求的语音/视频流量的比例要高得多)。这些新趋势要求下一代无线网络不仅要提供高数据速率(每秒数百吉比特)，还要提供超低延迟(亚毫秒)。此外，随着无线网络的发展和支持越来越多的用户，网络控制算法也必须产生较低的复杂性才能在实践中实现。然而，如何同时实现高吞吐量、低延迟和低复杂度的问题在很大程度上仍然悬而未决。解决这一重大研究挑战是该项目的主要目标。这项研究不仅有望大大提高我们对设计具有联合优化性能的无线网络的高效控制算法的理解，而且还将扩展/创建急需的理论基础，用于开发简单实用的协议，以优化下一代无线网络设计所需的关键性能指标。这项研究还将与一项全面的教育计划密切结合，该计划的重点是向本科生和K-12学生提供研究经验，招收和培训代表性不足的学生，并参与课程开发活动。本项目的目标是为设计在吞吐量、延迟和复杂性这三个方面都表现良好的高效网络控制算法创造新的理论基础。具体地说，本研究将围绕三个主要方面展开：(I)针对多信道蜂窝网络的小区内控制，旨在建立一个理论框架来设计具有可证明保证的最优吞吐量和最优(或接近最优)大偏差延迟率函数的低复杂度调度算法；(Ii)考虑对于较大系统(例如，密集多小区系统或自组织无线网络)更具挑战性的全网控制设置，并旨在开发一种新的基于节点的方法来设计具有可证明吞吐量和疏散时间性能的高效调度算法；它考虑分布式网络侧控制，旨在设计低复杂度的算法，以实现高吞吐量和低延迟。

With the advent of smart devices and the Internet of things, wireless technology has spawned a plethora of services that span business, science and engineering, entertainment, safety and security, health monitoring, and cover a large portion of our social interactions. Due to the prevalence of these new services, today's wireless networks are witnessing not only an unprecedented growth in the volume of traffic, but also a significant change in the types of traffic (e.g., a much higher percentage of voice/video traffic with more stringent delay requirements). These new trends require next-generation wireless networks to provide not only high data rates (tens of gigabits per second), but also ultra-low latencies (sub-millisecond). Moreover, as wireless networks grow and support an increasingly large number of users, network control algorithms must also incur low complexity in order to be implemented in practice. However, the question of how to simultaneously achieve high throughput, low delay and low complexity remains largely open. Addressing this major research challenge is a main goal of this project. Not only is this research expected to substantially advance our understanding of designing efficient control algorithms for wireless networks with jointly optimized performance, but it would also expand/create the much-needed theoretical foundations for developing simple and practical protocols to optimize the key performance metrics needed in the design of next-generation wireless networks. This research will also be closely integrated with a comprehensive educational plan, which is focused on providing research experiences to undergraduate and K-12 students, recruiting and training underrepresented students, and engaging in curriculum development activities. The goal of this project is to create new theoretical foundations for designing provably efficient network control algorithms that perform well in all three dimensions of throughput, delay, and complexity. Specifically, this research will be carried out around three main thrusts: (i) it focuses on intra-cell control for a multi-channel cellular network, and aims to build a theoretical framework for designing low-complexity scheduling algorithms with provably guaranteed optimal throughput and optimal (or near-optimal) large-deviations delay rate-function; (ii) it considers a more challenging setting of network-wide control for larger systems (e.g., a dense multi-cell system or an ad hoc wireless network), and aims to develop a new node-based approach for designing efficient scheduling algorithms with provable throughput and evacuation time performance; and (iii) it considers distributed network-side control and aims to design low-complexity algorithms that achieve high throughput and low delay.