Abstract:Aimed at compression characteristic of one-dimension vibration signals, a signal compression method based on improved quantum BP neural network is proposed. Based on variance of vibration signal, the signal had been divided into four areas: smooth area, half-smooth area, half-boundary area and boundary area. For different areas, different compression ratio was chosen to keep many details and compression quality. Meantime, using concurrent calculation and algorithm acceleration of quantum BP neural network, the convergence speed is improved and the compression time is shortened. The experimental results showed that, compared with current methods in the same compression ratio, the proposed method can improve SNR and shorten execution time.
王怀光;张培林;李 胜;吴定海;周云川. 量子BP神经网络的自适应振动信号压缩及应用[J]. , 2014, 33(19): 35-39.
Wang Huaiguang;Zhang Peilin;Li Sheng;Wu Dinghai;Zhou Yunchuan. Self Adaptive for Vibration Signal Compression Based on Quantum BP Neural Network and Its Application . , 2014, 33(19): 35-39.