Quantum-Inspired Evolutionary Membrane Computing For Emitter Signals

As a new branch of natural computing, membrane computing, employing various features specific to the structure and functionality of the living cells, is a parallel, distributed and nondeterministic computing model. Theoretically, most membrane computing systems are computationally universal. However, compared with the theoretical study, membrane computing is a rather new research direction with a well-defined practical interest, and therefore further studies are very necessary to extend the use of P systems for real-world applications. Consequently, membrane algorithms are researched to analyze radar emitter signals to expand the application of membrane computing. The main work and research fruits are as follows.1. The principle of time-frequency atom decomposition (TFAD) algorithm is introduced. Experiments are carried out on a linear-frequency modulated radar emitter signal to analyze the advantages and disadvantages of TFAD. To decrease the high computational complexity, the TFAD algorithm based on a quantum-inspired evolutionary algorithm (QIEA) is applied to analyze the radar emitter signals. Experimental results show that QIEA can release the computational burden to improve the efficiency of signal processing.2. The basic theory of membrane computing and its research status are discussed. Then, a quantum-inspired evolutionary algorithm based on P systems (QEPS) is used for radar emitter signals to promote the application of membrane computing. QEPS combines the framework and evolution rules of P systems with QIEA. In the elementary membrane, QIEA is employed to evolve the system; the communication rules are used to exchange the information among individuals. Experiments carried out on radar emitter signals with 10 dB signal-to-noise rate show that QEPS not only can extract specific information from radar emitter signals but also can reduce the computational complexity. It performs better than the greedy algorithm (GrA) and the counterpart QIEA.3. To overcome the shortcoming of QEPS's liability to fall into the local extreme value in searching the best atom, by introducing the tabu search algorithm into QEPS to search the best solution of the skin membrane, a modified variant of the quantum-inspired evolutionary algorithm based on P systems (MQEPS) is proposed. Extensive experiments conducted on 55 satisfiability problems and 16 radar emitter signals show that MQEPS can not only effectively extract the time-frequency atom features from radar emitter signals, but also can obtain better results compared with GrA, QIEA and QEPS. 4. To improve the convergence in searching atoms of quantum-inspired evolutionary membrane algorithms, by combining real-observation quantum-inspired evolutionary algorithm (RQEA) with active membranes, a real-observation QEPS (RQEPS) is proposed. RQEPS involves a dynamic structure including membrane merge and division to enhance the information exchange among individuals. In each elementary membrane, RQEA is employed to evolve the system to release the computational burden. Experiments conducted on radar emitter signals with the length of 256 and 15642 points show that RQEPS can effectively extract the best atom features from an over-complete time-frequency atom dictionary to decrease the computational complexity; it is superior to GrA, QIEA, QEPS and MQEPS in terms of search capability and computational complexity.This work was supported by the National Natural Science Foundation of China (60702026) and the Scientific and Technological Funds for Young Scientists of Sichuan (09ZQ026-040).