Distributed Online Learning for Coexistence in Cognitive Radar Networks

Abstract

This work addresses the coexistence problem for radar networks. Specifically, we model a network of cooperative, independent, and non-communicating radar nodes which must share resources within the network as well as with non-cooperative nearby emitters. We approach this problem using online Machine Learning (ML) techniques. Online learning approaches are specifically preferred due to the fact that each radar node has no prior knowledge of the environment nor of the positions of the other radar nodes, and due to the sequential nature of the problem. For this task we specifically select the multi-player multi-armed bandit (MMAB) model, which poses the problem as a sequential game, where each radar node in a network makes independent selections of center frequency and waveform with the same goal of improving tracking performance for the network as a whole. For accurate tracking, each radar node communicates observations to a fusion center on set intervals. The fusion center has knowledge of the radar node placement, but cannot communicate to the individual nodes fast enough for waveform control. Every radar node in the network must learn the behavior of the environment, which includes rewards, interferer behavior, and target behavior. Each independent and identical node must choose one of many waveforms to transmit in each Pulse Repetition Interval (PRI) while avoiding \emph{collisions} with other nodes and interference from the environment. The goal for the network as a whole is to minimize target tracking error, which relies on obtaining high SINR in each time step. Our contributions include a mathematical description of the MMAB framework adapted to the radar network scenario. We conclude with a simulation study of several different network configurations. Experimental results show that iterative, online learning using MMAB outperforms the more traditional sense-and-avoid (SAA) and fixed-allocation approaches.

Hybrid Cognition for Target Tracking in Cognitive Radar Networks

“…As covered above, the problem where SINR is constant in space and time and the CRN is completely distributed with no feedback has been studied in [8]. We instead consider the case where the spectrum is interference limited in every channel, and the SINR varies by node and over time due to target motion and range from each radar node location.…”

“…Fully distributed cognition [8] [9] is useful when there is a desire for the parts of a CRN to be entirely disjoint and independent. Fully distributed approaches rely on consensus techniques [10] [11] to exchange information between the parts of a network and to determine optimal actions.…”

“…Our previous work [12] [8] considered strictly decentralized techniques, and did not assume the presence of a CC. While our approach was effective under these circumstances, the technique resulted in a relatively slow convergence rate.…”

Adversarial Multi-Player Bandits for Cognitive Radar Networks

“…As mentioned above, the problem where SINR is constant over time and node has been studied [8]. However, we will consider the case where SINR varies over time and by node.…”

“…Cognitive radar networks have also been of interest in the literature. The authors of [7] and [8] provided the first description of multi-player bandit algorithms applied to the CRN problem. In their approach, the reward for a given action was assumed to be constant between nodes and over time.…”