Particle swarm optimization (PSO) is a population based meta-heuristic search algorithm that has been widely applied to a variety of problems since its advent. In PSO, the inertial weight not only has a crucial effect on its convergence, but also plays an important role in balancing exploration and exploitation during the evolution. However, PSO is easily trapped into the local optima and premature convergence appears when applied to complex multimodal problems. To address these issues, we present a modified particle swarm optimization with chaos-based initialization and robust update mechanisms. On the one side, the Logistic map is utilized to generate uniformly distributed particles to improve the quality of the initial population. On the other side, the sigmoid-like inertia weight is formulated to make the PSO adaptively adopt the inertia weight between linearly decreasing and nonlinearly decreasing strategies in order to achieve better tradeoff between the exploration and exploitation. During this process, a maximal focus distance is formulated to measure the particle's aggregation degree. At the same time, the wavelet mutation is applied for the particles whose fitness value is less than that of the average so as to enhance the swarm diversity. In addition, an auxiliary velocity-position update mechanism is exclusively applied to the global best particle that can effectively guarantee the convergence of MPSO. Extensive experiments on CEC′13/15 test suites and in the task of standard image segmentation validate the effectiveness and efficiency of the MPSO algorithm proposed in this paper.

MPSO: Modified particle swarm optimization and its applications

In this paper a new meta-heuristic search method, called Cat Swarm Optimization (CSO) algorithm is applied to determine the best optimal impulse response coefficients of FIR low pass, high pass, band pass and band stop filters, trying to meet the respective ideal frequency response characteristics CSO is generated by observing the behaviour of cats and composed of two sub-models In CSO, one can decide how many cats are used in the iteration Every cat has its' own position composed of M dimensions, velocities for each dimension, a fitness value which represents the accommodation of the cat to the fitness function, and a flag to identify whether the cat is in seeking mode or tracing mode The final solution would be the best position of one of the cats CSO keeps the best solution until it reaches the end of the iteration The results of the proposed CSO based approach have been compared to those of other well-known optimization methods such as Real Coded Genetic Algorithm (RGA), standard Particle Swarm Optimization (PSO) and Differential Evolution (DE) The CSO based results confirm the superiority of the proposed CSO for solving FIR filter design problems The performances of the CSO based designed FIR filters have proven to be superior as compared to those obtained by RGA, conventional PSO and DE The simulation results also demonstrate that the CSO is the best optimizer among other relevant techniques, not only in the convergence speed but also in the optimal performances of the designed filters

Cat Swarm Optimization algorithm for optimal linear phase FIR filter design.

Using the generalized Gaussian probability density function as the kernel, a generalized correntropy has been proposed. A generalized maximum correntropy criterion (GMCC) algorithm is then developed by maximizing the generalized correntropy. However, the GMCC algorithm has a high steady-state misalignment and involves a high calculation cost of the exponential term (generalized Gaussian kernel). In this brief, we propose a maximum Versoria criterion (MVC) algorithm, which is derived by maximizing the generalized Versoria function, to reduce steady-state misalignment and computational effort as compared to the GMCC algorithm. The MVC algorithm is then tested in system identification and acoustic echo cancellation scenarios, which have demonstrated that the proposed algorithm is robust against non-Gaussian impulsive noises and performs much better than the LMP and GMCC algorithms.

Maximum Versoria Criterion-Based Robust Adaptive Filtering Algorithm

In this paper, an improved particle swarm optimization technique called Craziness based Particle Swarm Optimization (CRPSO) is proposed and employed for digital finite impulse response (FIR) band stop filter design The design of FIR filter is generally nonlinear and multimodal Hence gradient based classical optimization methods are not suitable for digital filter design due to sub-optimality problem So, global optimization techniques are required to avoid local minima problem Several heuristic approaches are available in the literatures The Particle Swarm Optimization (PSO) algorithm is a heuristic approach with two main advantages: it has fast convergence, and it uses only a few control parameters But the performance of PSO depends on its parameters and may be influenced by premature convergence and stagnation problem To overcome these problems the PSO algorithm has been modified in this paper and is used for FIR filter design In birds' flocking or fish schooling, a bird or a fish often changes directions suddenly This is described by using a “craziness” factor and is modelled in the technique by using a craziness variable A craziness operator is introduced in the proposed technique to ensure that the particle would have a predefined craziness probability to maintain the diversity of the particles The algorithm's performance is studied with the comparison of real coded genetic algorithm (RGA), conventional PSO, comprehensive learning particle swarm optimization (CLPSO) and Parks and McClellan (PM) Algorithm The simulation results show that the CRPSO is superior or comparable to the other algorithms for the employed examples and can be efficiently used for FIR filter design

Craziness based Particle Swarm Optimization algorithm for FIR band stop filter design

An exhaustive review on the use of structured stochastic search approaches towards system identification and digital filter design is presented in this paper. In particular, the paper focuses on the identification of various systems using infinite impulse response adaptive filters and Hammerstein models as well as on the estimation of chaotic systems. In addition to presenting a comprehensive review on the various swarm and evolutionary computing schemes employed for system identification as well as digital filter design, the paper is also envisioned to act as a quick reference for a few popular evolutionary computing algorithms.

Swarm and evolutionary computing algorithms for system identification and filter design: A comprehensive review

An adaptive inertia weight particle swarm optimization (AIW-PSO) algorithm was presented for designing IIR digital filter. In this algorithm, modified Versoria function was employed in the new relation of adaptive inertia weight factor function instead of Sigmoid function for avoiding the exponential computation and ensuring the small final misadjustment. Furthermore, the novel algorithm was well used in designing adaptive IIR digital filter about unknown system identification, and simulation results shown that the filter had more enhanced performance characteristics using the AIW-PSO algorithm and the complexity in calculation were improved greatly.

An Adaptive Inertia Weight Particle Swarm Optimization Algorithm for IIR Digital Filter

A novel Domain Adaptive Deep Recurrent Network for multivariate time series prediction

An independent central point OPTICS clustering algorithm for semi-supervised outlier detection of continuous glucose measurements

As continuous glucose monitoring (CGM) systems provide critical feedback information of blood glucose concentration to the artificial pancreas for patients with type 1 diabetes (T1D), faults in CGM may seriously affect the computation of insulin infusion rates which can lead to fatal consequences accompany with hypoglycemia or hyperglycemia. In the present work, the k-medoids clustering algorithm is modified by calculating cluster number with a Bayesian Information Criterion (BIC)-based cost function and the SAC (SSE-ASW Criterion) evaluation coefficient which considers both SSE (Sum of Square due to Error) and ASW (Average Silhouette Width) criteria. Then, the modified k-medoids clustering algorithm is proposed to detect sensor failures online with CGM measurements. Different from the qualitative model-based methods and quantitative model-based methods, sufficient clean data are the only requirement of the proposed method. During online monitoring, the new glycemic variability is then tracked against predefined confidence limits during training period to indicate abnormality. The feasibility of the proposed method is successfully assessed using CGM data collected from the UVa/Padova metabolic simulator.

Fault detection of continuous glucose measurements based on modified k-medoids clustering algorithm

Establishing relations between variables and real-time prediction of quality variables or other key indicators are critical for dynamic processes including industrial and biological processes. In this study, a novel multivariate statistical modeling method named “kernel-regularized latent variable regression (KrLVR) approach” is proposed for capturing the dynamics of a process by building KrLVR models with process and quality data. First, a regularization term based on a kernel matrix is incorporated into the objective of the latent variable regression model. Consequently, the proposed KrLVR method has the ability to overcome potential ill-conditioning resulting from collinearity in process data and has a stronger prediction power. Besides, the inner model is consistent with the outer model, which enables the proposed method to predict quality data with fewer latent variables. Second, the prior knowledge of dynamic processes such as exponential stability and smoothness can be integrated into the modeling process by using an appropriate kernel matrix. In addition, to meet the requirement of exponential stability of the model, the weights of the model should decay exponentially, that is, coincident with determining the number of historical observations for data augmentation (identification of the model structure). Therefore, the problem of tuning model complexity is eluded, and it becomes finding appropriate hyper-parameters of the kernel matrix. Moreover, the empirical Bayesian method is utilized for estimating hyper-parameters of the kernel matrix from augmented process and quality data. Three case studies illustrate the performance of the proposed KrLVR method by comparing with several other relevant methods. 

Xia Yu

Papers

An Adaptive Inertia Weight Particle Swarm Optimization Algorithm for IIR Digital Filter

A novel Domain Adaptive Deep Recurrent Network for multivariate time series prediction

An independent central point OPTICS clustering algorithm for semi-supervised outlier detection of continuous glucose measurements

Fault detection of continuous glucose measurements based on modified k-medoids clustering algorithm

Kernel-Regularized Latent-Variable Regression Models for Dynamic Processes