10.1 Communication and Human Networks: Towards Provably Timely and Reliable Battlefield Networks
Abstract
Battlefield networks enable soldiers at tactical edges to communicate with command centers and nearby soldiers in real time. An important requirement of battlefield networks is that most applications, such as tactical commands, voice and video calls, etc., require stringent per-packet delay and reliability guarantees. Similar real-time requirements are also critical for intra-tank wireless networks that connect sensors and actuators inside a tank. This project aims to establish a novel framework for developing solutions with provable timeliness and reliability guarantees for battlefield and intra-tank networks. The proposed approach differs significantly from most existing studies that aim to only develop heuristics to improve delay and throughput performance opportunistically without offering stringent performance guarantees. The proposed framework will explicitly address various practical behaviors of battlefield networks, which, in addition to stringent per-packet delay and reliability requirements, include stochastic and unreliable wireless transmissions, mobile soldiers, dynamic flows, and multihop transport. Specifically, the following research topics will be investigated: 1. Scheduling heterogeneous unicast flows: Multiple flows coexist in battlefield networks. Different flows generate packets with different traffic patterns, have different per-packet delay bounds, and require different reliability guarantees. Novel scheduling policies that fulfill the different requirements of heterogeneous flows will be proposed and evaluated. 2. Multicasting heterogeneous flows: Many flows in battlefield networks are multicast flows, such as commands for soldiers in a certain area. Multicasting differs from unicasting in that it incurs a high overhead to gather per-transmission feedback information from all receivers. Scheduling policies with and without feedback information will be evaluated. New policies that intelligently schedule packets to meet delay and reliability constraints, as well as obtain limited feedback information will be proposed. 3. End-to-end delay bounds for multi-hop networks: Multi-hop transmissions are usually needed to communicate with soldiers that are faraway from the command centers. Joint routing and scheduling policies that aim to provide end-to-end delay and reliability guarantees will be developed. 4. Direct soldier-to-soldier communications: Soldiers at the tactical edge can communicate with each other directly for fast information exchange. An important challenge for direct soldier-to-soldier communication is the lack of a centralized scheduler. Soldiers therefore need to communicate through random access. Distributed policies that achieve the optimal timeliness and reliability will be established. 5. Real-time self-configuration: Distributed rate control policies that quickly react to soldier mobility and flow dynamics will be introduced. In addition to timeliness and reliability, their convergence rate will be studied. 6. Intra-tank wireless communications: This project proposes a new architecture that connects sensors and actuators in a tank by wireless networks. To enable this architecture, new polices with provably safety properties will be developed. This architecture can potentially greatly reduce the cost of tanks and allow faster innovations. 7. Testbed implementations: All the policies developed in this project will be evaluated by testbeds. The PIs have built a wireless sensor network testbed as well as a software-defined radio testbed.
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- Jan 12, 2017
- Source ID
- W911NF1510279
Entities
People
- I-Hong Hou
Organizations
- Army Contracting Command
- Texas Engineering Experiment Station
- United States Army