Model-Based and AI-Assisted Autonomous Interference-Avoiding Directional Networking
Abstract
In contested and degraded electromagnetic environments that extend to hundreds of miles, U.S. Air Force and other DoD personnel and assets deployed on land, air, and space must be capable of maintaining operational wireless connectivity at all times within the laws of physics and hardware limitations. Directional connectivity of tactical networks not only enables effective and efficient use of the available space-time-frequency continuum but also safeguards signals from would-be eavesdroppers. In this project, we develop, implement and evaluate model-based and AI-assisted joint space-time-frequency waveform design algorithms for directional connectivity of multipleinput multiple-output (MIMO) network nodes that enable resilient, self-sustained directional wireless networking of distributed ground/air assets that operate at maximum throughput. The effort is founded on principled algorithms for joint space-time-frequency waveform design for optimal interference avoidance and autonomous network operation in all disturbance conditions (local or global, persistent or intermittent, narrowband or wideband.) Naturally, joint treatment of the space-time-frequency continuum is achieved with the design of one or more basic pulses that occupy the entire continuum of the readable frequency spectrum (all-spectrum). Optimized coded repeats of the basic pulses over a finite alphabet and optimization of their phase at each antenna provide the space-time waveform fabric (signature) on which we can carry information symbols with any desirable level of spatial directivity and interference avoidance within the physics laws of the medium and our hardware limitations. The research workplan is organized in the form of three technical thrusts that support the development, implementation and testing of autonomous directional software-defined MIMO systems for multi-hop networking. Thrust 1 combines model-based principled algorithms with AI tools to maximize performance gains and minimize network reaction time. To introduce awareness of the neighbors and channel usage at each hop and overcome the deafness introduced by directional links, in Thrust 2 we control the interaction of key network-layer functionalities with all-spectrum waveform optimization at PHY and MAC layers. Thrust 3, informed by Thrusts 1-2, is dedicated to implementation and performance evaluation of the developed dynamic network algorithms/protocols, first over network-level simulators and then over software-defined radio hardware.
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- Aug 10, 2021
- Source ID
- FA87502110500
Entities
People
- Dimitris A. Pados
Organizations
- Florida Atlantic University
- Rome Laboratory
- United States Air Force