Second-Order Optimization for Multi-Hop Wireless Battlefield Networks
Abstract
This project aims to develop a new theoretical framework for the design, analysis, and optimization for future mobile multi-hop wireless battlefield networks. We envision that future battlefield networks will leverage networked vehicles, such as unmanned aerial vehicles, to form multi-hop wireless networks to provide connectivity to mobile entities outside the coverage area of infrastructures. This multi-hop wireless network will employ emerging millimeter wave (mmWave) communications to establish high-bandwidth wireless links, and it will serve a plethora of emerging real-time applications, including augmented reality (AR), real-time sensing and control, etc. Many challenges need to be addressed before this vision can become a reality, including the need to provide end-to-end application-specific performance guarantees, the lack of centralized controller in multi-hop networks, the dynamics of mobile networks, and the directionality of mmWave communications. This project will address all these challenges in a unified framework. The core of this theory is a second-order model for both application requirements and communication channel features. This second-order model characterizes every random process by its mean and temporal variance. Compared to traditional first-order performance metrics, this second-order framework is, quite literally, an order richer in characterizing battlefield networks in all timescales. With this second-order model, this project will sharply characterize the capacity of mobile multi-hop wireless networks in terms of the end-to-end mean and temporal variance of packet deliveries. It will also develop distributed and low-complexity network algorithms for optimizing second-order performance metrics. To demonstrate its utility, this model will be applied for the optimization of several emerging network performance metrics, such as age-of-information and timely-throughput. A new prototype of aerial mmWave multi-hop network will be built for demonstrating and evaluating this theoretical framework of second-order optimization. Building upon the PIsÕ existing testbeds of unmanned aerial vehicles (UAVs) and mmWave communications, this project will build a system where UAVs form a multi-hop mmWave wireless backhaul network that connects a gateway to mobile users outside its coverage. An important challenge for mmWave communications is that they are subject to high path loss and require beam alignment to maintain links. New MAC protocols for dynamic beam alignment, hand-offs, and multi-hop communications will be developed to address this challenge. The research efforts can be summarized in the following three thrusts: Thrust 1: Second-order optimization for real-time applications in wireless access networks: This thrust focuses on one-hop wireless access networks. It will establish a second-order framework that models the performance of real-time applications, determines the second-order capacity regions of wireless access networks, and develops the optimal scheduling policy. Thrust 2: End-to-end second-order optimization for complex multi-hop networks: This thrust extends the proposed second-order framework for multi-hop wireless networks. It will develop several hop-by-hop and flow-by-flow decomposition techniques, and use these techniques to design distributed and tractable policies with end-to-end performance guarantees. Thrust 3: Aerial mmWave multi-hop networks: This thrust will build a new prototype for evaluating and demonstrating theoretical results. It will develop MAC protocols to address challenges arisen with the usage of mmWave communications.
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- Aug 02, 2022
- Source ID
- W911NF2210151
Entities
People
- I-Hong Hou
Organizations
- Army Contracting Command
- Texas Engineering Experiment Station
- United States Army