Cognitive Networking

Abstract

(U) The Cognitive Networking program will develop technologies that provide information systems and communication networks with the ability to maintain and self-optimize their own functionality, reliability and survivability. These technologies will allow the military to focus its critical manpower resources on the mission rather than on the maintenance of its information systems and network infrastructure. Cognitive information processing will be used to optimize networked communications based on current conditions, past experience and high-level user guidance. The Cognitive Networking program is also addressing the warfighter’s need for actionable situational awareness in complex radio frequency (RF) environments. This work leverages advances in software-defined radio technology to achieve specific military goals. The program has interest in machine learning techniques that can enhance the effectiveness of jamming and other RF countermeasures. So-called “cognitive jamming” has the potential to deny the enemy’s effective use of the RF spectrum. The Cognitive Networks effort funds three programs: SAPIENT, LANDroids, and BOSS. • The Situation-Aware Protocols in Edge Network Technologies (SAPIENT) effort will develop a new generation of cognitive protocol architectures to replace conventional protocols that fare poorly in extreme network conditions and do not provide adequate service for key applications. Technology developed in the SAPIENT effort will have military utility wherever tactical communications are deployed. SAPIENT architectures will represent awareness with a knowledge base that is updated based on specification and observation. SAPIENT technology enables the automatic adaptation of protocols to the operational environment to dramatically reduce the effect of network impairments on applications while demonstrating a positive trend in capability as new situations are encountered and learned. • The Local Area Network droids (LANdroids) effort will give warfighters reliable communications in urban settings. LANdroids will accomplish this by creating robotic radio relay nodes that move autonomously to configure and maintain a communications mesh by reasoning about their positions relative to one another and relative to the warfighters. LANdroids will move as the warfighters move with the goal of maintaining warfighter connectivity throughout their operations. LANdroids will be pocket-sized so warfighters can carry several and drop or deploy them as they move through an area. The effort is creating both the intelligent radio control software and the small radio platform on which it runs. The technologies will be tested in a physical setting and at an operationally relevant scale. • The Brood of Spectrum Supremacy (BOSS) effort will provide actionable situational awareness to the warfighter in complex radio frequency (RF) environments. BOSS adds collaborative processing capabilities to tactical software-defined radios to achieve specific military goals. BOSS exploits cooperative use of computational, communication and sensory capabilities in a software radio, in aggregate, to generate breakthrough capabilities in the warfighter knowledge of their surroundings, with a particular focus on RF-rich urban operations. Machine learning techniques will enable real-time characterization of an adversary’s radio dynamics and provide cognitive, networked responses to new enemy threats. Ultimately this effort will develop Software Communications Architecture (SCA)-compliant waveforms suitable for implementation on a tactical software radio system.

Document Details

Document Type
Accomplishment
Publication Date
Oct 01, 2011
Source ID
ae27f47a292b977feb8fccc85b45ad82

Tags

Fields of Study

  • Computer science

Readers

  • Military Science and Technology Research and Modernization.
  • Radio communications and signal processing.
  • Tactical Satellite Communications Systems Engineering.

Technology Areas

  • AI & ML
  • AI & ML - DoD AI Strategy
  • Autonomy
  • Autonomy - UAVs

Related Documents