Spectrum Sensing In Cognitive Radio Based On Quantum Neural Network

As wireless communication emerging technologies continue to emerge, the demand for radio spectrum resources is also rising. The existing static spectrum usage policy makes allocable spectrum resources very rare and supply-and-demand issue of spectrum resources increasingly tense. As a new dynamic spectrum sharing technology, cognitive radio realize the "secondary use" of licensed spectrum. Spectrum sensing is the first step in cognitive radio, so accurate and rapid detection of weak signals realization will effectively improve the spectrum utilization.To overcome the shortcomings of traditional spectrum sensing and enhance performance at low SNR, this thesis proposes a sensing algorithm based on quantum neural network (QNN).The basic idea is to extract characteristic parameters by an authorized user signals and train quantum neural network, then to access authorization data signals uncertainty and store, to achieve ambient "spectrum opportunity" test. In addition, the thesis further analyzes the selection and design of the network model input key parameters such as characteristic values, hidden layer and so. Finally, establish the experimental simulation and results show that at low SNR, its detection performance is superior to cyclostationary feature detection and BP neural network detection. Convergence performance also has been greatly improved than the other. But clearly, we can see training error curve of QNN sensing algorithm in the training process jitter significantly and the stability is not good.In order to solve the above problems, this thesis improves QNN perception algorithm in the last, the new algorithm chooses the three Josephson function as transfer function to shorten excitation of the saturation zone and reduce the "false saturation" phenomenon occurred during training; With constraints added in the original learning objectives functions, the network weights adjustment and updating quantum interaction in the learning process decreases to a minimum. Through simulation experiment results, improved QNN algorithm has a more excellent convergence performance and has been significantly improved compared with that before improvements in detection performance.