Deepwave: Automated Radio Signal and Protocol Classification Through Deep Learning for Waveform Vulnerability Discovery
Abstract
The research objective of this project is to enable a fundamental leap forward in the design, development, evaluation and experimentation of high-throughput wireless networking with guaranteed security in the presence of adversarial attacks. To this end, University at Buffalo (UB) and General Electric Aviation Systems (GEAS) propose to i) design new radio signal sensing and protocol classification techniques for automated discovery of the vulnerabilities of wireless systems; ii) simulate Unmanned Aerial System (UAS) networks in a contested, degraded, and operationally limited (CDO) environment in the UBs Airborne Networking and Communications (UB-ANC) Emulator using Swarm Control for dynamic network management and control of UAS swarms; and iii) integrate in-phase and quadrature (IQ) sample-level fidelity radio frequency (RF) simulation into the Advanced Framework for Simulation, Integration, and Modeling (AFSIM) to provide a complete common operating picture for swarm operations.
Document Details
- Document Type
- Technical Report
- Publication Date
- May 01, 2022
- Accession Number
- AD1168538
Entities
People
- Nicholas Mastronarde
- Zhangyu Guan
Organizations
- University at Buffalo