Adaptive IdentificationAlgorithms of Nonlinear Systems with their Applications

在无线通信的非线性控制系统设计和非线性信号处理中，非线性系统的辨识在基于模型的方法中占有重要地位。该项目的主要目的是为一类特定的系统构建和评估有效和系统的非线性辨识算法，以解决复杂的建模问题，将其描述在一个非线性系统辨识框架中。研究结果如下：(1)提出了一种新的全自适应隔振控制方案，在地面和底层之间安装一个由两个自适应控制器组成的半主动磁流变阻尼器。一种是自适应逆控制器，它可以向磁流变阻尼器提供必要的输入电压，以产生作用在受控结构上的指定参考阻尼力。另一种是自适应参考控制器，它可以使第一层结构的动力与期望的参考动力相匹配。提议中的…更多的完全自适应方法可以同时处理磁流变阻尼器模型和结构模型中的不确定性。讨论了包括两种自适应机制在内的整个系统的稳定性，并阐明了稳定性条件。实验结果也验证了所提出的自适应方案的有效性。(2)由于OFDM传输中的信息是复幅度的，因此功率放大器的非线性补偿是一个重要的问题。该项目提出了一种新的自适应预失真器，既能补偿非线性失真，又能最大限度地提高功率效率。提出了一种新的自适应耦合对消算法，该算法识别了所有传输路径的动态特性，并阐明了收敛条件。(3)将城市多径环境中移动台阵列天线定位问题描述为一个新的非线性参数辨识问题，通过阵列信号处理估计基站产生的散射源的方向和角度扩展，并估计源位置，从而使射线跟踪法计算的方向和角度扩展与阵列天线估计的方向和角度扩展之间的误差。通过室外实验验证了该方法的有效性。阵列天线的自校准也被视为一个非线性辨识问题，并将这一思想应用到Capon方法中。在对QR分解进行改进的基础上，提出了一种简单有效的阵列天线相关信号个数检测方法。较少

Identification of a nonlinear system plays an important role in model-based approaches in nonlinear control system design and nonlinear signal processing for wireless communication. The main purpose of the project is to construct and evaluate effective and systematic nonlinear identification algorithms for a specific class of systems to solve complicated modeling issues by formulating them in a nonlinear system identification framework. The results are summarized as follows :(1)A new fully adaptive control scheme is proposed for vibration isolation using a semi-active MR damper installed between ground and first floor, which is composed of two adaptive controllers. One is an adaptive inverse controller which can give necessary input voltage to MR damper so as to generate specified reference damping force acted on a controlled structure. The other is an adaptive reference controller which can match the dynamics of the first floor of structure to a desired reference dynamics. The propose … More d fully adaptive approach can deal with both uncertainties in models of MR damper and structure. The stability of the total system including the two adaptation mechanisms is discussed and the stability conditions are clarified. Experimental results are also shown to validate the effectiveness of the proposed adaptive scheme.(2)Since the information is taken on the complex amplitude in the OFDM transmission, compensation of the nonlinearity of power amplifier is a significant issue. The project proposed the new adaptive predistorter which can compensate for nonlinear distortion as well as can maximize the power efficiency. The new adaptive coupling cancellation algorithm was also proposed in which all of the transmission path dynamics are identified and convergence conditions are clarified.(3)Localization of mobile station in urban multipath environments by using an array antenna is formulated as a new nonlinear parameter identification issue, where the direction and angle spread of scattered sources produced by the station are estimated by the array signal processing, and the source location can also be estimated so that the error between the direction and angle spread calculated by the ray tracing method and those estimated by the array antenna. Its effectiveness was validated by outdoor experiments. Autocalibration of array antenna is also regarded as a nonlinear identification problem, and the idea was applied to Capon method. Detection of the number of correlated signals by array antenna is also investigated and a simple but effective method was proposed based on improvements of QR decomposition. Less