Paper Title
Reinforcement Learning Model for Electromagnetic Signature Reduction of Naval Vessel

Abstract
This paper proposes a very efficient reinforcement learning based approach for signature reduction of naval vessels commonly known as degaussing (DG). As most of the naval vessels are made up of ferromagnetic materials, electromagnetic signatures are generated due to induced magnetization. As a result, local disturbances in the form of electromagnetic signatures are generated in the earth’s magnetic field due to an underwater naval vessel. If these disturbances are detectable then they may trigger weapon systems and will pose a security threat to the naval vessel. Degaussing helps to protect the naval vessels made from ferromagnetic material and hence ensures stealth mode of operation of the naval vessel. The main contribution is to propose a reinforcement learning based method to solve the DG problem of naval vessels by using Q learning algorithm. The advantages of the proposed method are high accuracy of signature reduction, ability to self-learn and obtaining the optimal values of currents to be applied at each degaussing coil. Keywords - Degaussing, Magnetic Signatures, Reinforcement Learning, Q Learning