基于BP神经网络的分布式传感器网络的可靠性分析

针对分布式传感器中的故障点多，导致估计系统可靠性参数困难的特点，提出了一种基本原则于BP三层神经网络的Markov可靠性模型。仿真结果表明，神经网络收敛时的可用度与Markov模型计算值的相对误差的数量级为10-4，从而验证了神经网络模型的正确性，另一方面与用两层神经网络的可靠性计算相比，其收敛速度加快。     

Nonlinear inverse modeling of sensor based on back-propagation fuzzy logical system

Objective To correct the nonlinear error of sensor output, a new approach to sensor inverse inodeling based on Back-Propagation Fuzzy Logical System （BP FS） is presented. Methods The BP FS is a computationaiiy efficient nonlinear universal approximator, which is capable of implementing complex nonlinear mapping from its input pattern space to the output with fast convergence speed. Results The neuro-fuzzy hybrid system, i.e. BP FS, is then applied to construct nonlinear inverse model of pressure sensor. The experimental results show that the proposed inverse modeling method automatically compensates the associated nonlinear error in pressure estimation, and thus the performance of pressure sensor is significantly improved. Conclusion The proposed method can be widely used in r nonlinearity correction of various kinds of sensors to compensate the effects of nonlinearity and temperature on sensor output.     

Recognizing frequency characteristics of gas sensor array

A novel method based on independent component analyzing （ICA） in frequency domain to distinguish the frequency characteristics of multi-sensor system is presented. The conditions of this type of ICA are considered and each step of resolving the problem is discussed. For a two gas sensor array, the frequency characteristics including amplitude-frequency and phase-frequency are recognized by this method, and cross-sensitivity between them is also eliminated. From the principle of similarity, the recognition mean square error is no more than 0.085.