Temporal Difference Learning Waveform Selection

Journal of Computers 09/2010; 5(9):1394-1401. DOI: 10.1109/CCCM.2009.5267516
Source: DBLP


The largest difference between cognitive radar and other adaptive radar is the adaptivity of transmitter in cognitive radar. How to optimally decide or select the radar waveform for next transmission based on the observation of past radar returns is one of the important issues. In this paper, with the stochastic dynamic programming model of waveform selection, we use the method of temporal difference learning to solve this problem and realize the adaptivity of waveform selection. The simulation results show that the uncertainty of state estimation using temporal difference learning is less than that using fixed waveform.

Full-text preview

Available from:
  • [Show abstract] [Hide abstract]
    ABSTRACT: In typical radar systems, the process of recognizing a target requires human involvement. This human element makes radar systems not fully reliable due to unstable performance that varies between operators. This paper describes an intelligent radar system which addresses this problem in a border surveillance environment. The proposed radar system is capable of automatically detecting and then classifying different targets using an artificial neural network trained with the Levenberg-Marquardt algorithm. The training and test sets presented to the neural network are composed by high-resolution Inverse Synthetic Aperture Radar pictures obtained by the radar's detection module. Simulation results show that the intelligent radar system can reliably detect and distinguish the different objectives. Moreover, the radar system can outperform human operators and another radar system that deals with similar objectives. These results indicate that future intelligent systems can potentially replace human radar operators in this critical security setting.
    No preview · Article · Nov 2012 · International Journal of Advanced Robotic Systems