Decentralized Bandits with Feedback for Cognitive Radar Networks

Howard, William W.; Buehrer, R. Michael

doi:10.1109/milcom55135.2022.10017685

Cited by 4 publications

(1 citation statement)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Further, Haykin provides the following dichotomy of cognitive networks: 1) Distributed Cognition, where observations from individual nodes are combined at a fusion center 1 but no feedback is provided to the nodes. 1 Fusion centers in this type of network are assumed to perform no decision functions; i.e., they simply combine measurements and provide data to operators.…”

Section: Introductionmentioning

confidence: 99%

Hybrid Cognition for Target Tracking in Cognitive Radar Networks

Howard¹,

Buehrer²

2023

Preprint

View full text Add to dashboard Cite

This work investigates online learning techniques for a cognitive radar network utilizing feedback from a central coordinator. The available spectrum is divided into channels, and each radar node must transmit in one channel per time step. The network attempts to optimize radar tracking accuracy by learning the optimal channel selection for spectrum sharing and radar performance. We define optimal selection for such a network in relation to the radar observation quality obtainable in a given channel. This is a difficult problem since the network must seek the optimal assignment from nodes to channels, rather than just seek the best overall channel. Since the presence of primary users appears as interference, the approach also improves spectrum sharing performance. In other words, maximizing radar performance also minimizes interference to primary users. Each node is able to learn the quality of several available channels through repeated sensing. We model the independent radar nodes as cognitive agents as well as the central coordinator and assume that information can be exchanged between the radars and the coordinator. Importantly, each part of the network acts as an online learner, observing the environment to inform future actions. We show that in interference-limited spectrum, where the signal-to-interference-plus-noise ratio can vary across channels and nodes, a cognitive radar network is able to use information from the central coordinator in order to reduce the amount of time necessary to learn the optimal channel selection. We also show that even limited use of a central coordinator can eliminate collisions, which occur when two nodes select the same channel. We provide several reward functions which capture different aspects of the dynamic radar scenario and describe the online machine learning algorithms which are applicable to this structure. In addition, we study varying levels of feedback, where central coordinator update rates vary. We compare our algorithms against baselines and demonstrate dramatic improvements in convergence time over the prior art. A network using hybrid

show abstract

Section: Introductionmentioning

confidence: 99%