Kernel adaptive filters (KAFs) are a class of nonlinear adaptive filters developed in the reproducing kernel Hilbert space, and are particularly suitable for addressing signal processing issues involving data streams and unknown nonlinearities. However, KAFs endure the issue of network structure growth as the number of training samples increases. To this end, a novel structural sparsification method for KAFs, i.e., Gauss Hermite Fourier features (GHFF) method, is first proposed by combining Gauss Hermite quadrature integration rule and Fourier transform. Subsequently, the GHFF method is integrated with the maximum correntropy criterion in filter design, leading to the development of two new adaptive filtering algorithms, i.e., GHFF maximum correntropy (GHFFMC) algorithm and stochastic batch GHFFMC (SB-GHFFMC) algorithm. The proposed GHFFMC and SB-GHFFMC algorithms are expected to exhibit excellent capabilities in characterizing the unknown nonlinear relationships within the data, along with robustness to outliers. Meanwhile, SB-GHFFMC is anticipated to exhibit superior filtering performance in comparison with GHFFMC, as it leverages a general and flexible batch gradient descent method for model optimization. Simulations on nonlinear system identification and time-series prediction of Chua’s circuits confirm the performance superiorities of the proposed algorithms compared to other robust KAFs and RFF-based filters.

The conventional active noise control (ANC) system relies on the error-correction learning to design the adaptive controller. However, as the residual noise, the correction error is discarded after each iteration and is still not a perfect input to the human ear. Although many efforts have been devoted to improve the performance by reusing the reference signals, the reuse of the past errors of ANC has been rarely studied in the literature. In this paper, we first design a novel error reused ANC (ErANC) system by introducing the information of the past errors for compensating the gap between the primary noise and estimated output. Then, applying the filtered-x least mean square (FxLMS) algorithm to ErANC for designing the controller, a novel ErFxLMS algorithm is developed to improve the noise reduction performance of FxLMS. However, as the correction errors accumulate over time, the computational and storage burdens of ErFxLMS will grow linearly. To address this issue, a sliding window strategy is applied to the ErFxLMS to curb the size of the stored errors, generating the sliding window ErFxLMS (SW-ErFxLMS) algorithm accordingly. Finally, the performance analysis of mean square error is carried out to validate the universal approximation ability of ErFxLMS. Simulation and experimental results illustrate the effectiveness of the proposed algorithms.

The online conjugate gradient (CG) algorithm with error-correction learning discards the error after each iteration, which shows low filtering accuracy in prediction tasks. Moreover, the CG based on the mean-square error (MSE) criterion suffers from poor performance when dealing with impulsive noise. To address these issues, we first construct a novel batched augmented least lncosh model (ALLM) based on the least lncosh (Llncosh) criterion by fully using the discarded errors. Then, a new structure of adaptive filters is designed under the framework of ALLM for online application. Using the least lncosh CG (LLCG) algorithm for the design of controller, a novel robust augmented LLCG (ALLCG) adaptive filter is finally developed, which achieves better filtering performance than the traditional CG and LLCG. Simulations demonstrate the advantages of ALLCG on robustness and prediction accuracy.

The filtered-x least mean square (FxLMS) algorithm has been proposed for an active noise control (ANC) system. However, due to the used mean square error (MSE) criterion, FxLMS suffers from performance degeneration for non-Gaussian noises, dramatically. To address this issue, a novel robust generalized hyperbolic tangent (GHT) criterion is first constructed in this paper. Then, the random Fourier features (RFF) method and the conjugate gradient (CG) method are used to address the nonlinearity existing in ANC and solve the quadratic optimization problem induced by the GHT criterion, respectively. Finally, a novel robust random Fourier conjugate gradient filtered-x generalized hyperbolic tangent (RFCGFxGHT) algorithm is proposed for ANC. The theoretical analyses regarding the convergence and computational complexity of RFCGFxGHT are also derived. Simulation experiments on nonlinear ANC systems corrupted by the synthetic logistic chaotic and$\alpha$-stable noises, as well as real-world functional magnetic resonance imaging (fMRI) and server room noises, are conducted to confirm the effectiveness, robustness, and desirable nonlinear learning ability of the proposed algorithm.

Community detection is one of the most important contents of complex network research, and it faces the challenge of balancing accuracy and efficiency. In response to this challenge, the paper proposes a community detection algorithm based on feature propagation. The algorithm first randomly initializes a vector for each node of the graph to complete the feature initialization. Then, with the help of the node similarity matrix, the vector representation of each node in the graph is learned through feature propagation, and finally the K-means clustering algorithm is used to cluster to obtain the community detection result. We have conducted experiments on real datasets and LFR datasets with the metric of NMI, and the results show that our algorithm is accurate and efficient.

Cloud manufacturing is a novel service-oriented networked manufacturing paradigm that aims to provide on-demand manufacturing cloud services to consumers. Scheduling is a critical means for achieving that aim. Currently, research on scheduling in cloud manufacturing is still in its infancy, and current frequently adopted meta-heuristic algorithm-based approaches have some shortcomings, e.g. they require complex design processes and lack adaptability to dynamic environments. Deep reinforcement learning (DRL) that combines advantages of reinforcement learning and deep learning provides an efficient, adaptive and intelligent approach for solving scheduling problems in cloud manufacturing. However, to the best of our knowledge, there has been no application of DRL to scheduling in cloud manufacturing. This work conducts a preliminary exploration over this issue. First, a DRL-based framework for scheduling in cloud manufacturing is proposed. Then a DRL model for online single-task scheduling in cloud manufacturing is presented to demonstrate the effectiveness of the framework. DRL as a promising technique will find wide applications in cloud manufacturing, and this work can provide some reference for future research on this.

Multi-disciplinary virtual prototype (MDVP) based on modeling and computer simulation technology has been applied in a wide range of engineering applications, especially the design, testing and evaluation of complex products. Generally, a large set of parameters need to be optimized to improve the performance of the prototype, which comprises a number of heterogeneous models from multiple domains. This calls for numerous collaborative simulation experiments in heterogeneous computing environments, which are difficult to build and configure. Cloud simulation is a promising solution with the extremely large resource pool and dynamic construction of virtual computing environments. Thus this paper firstly proposed an optimization framework of MDVP which formulated the domain models, optimizer and the distributed interactive environment on cloud for multi-disciplinary engineers. Then we designed the procedure for conducting parallel optimization of MDVP in the cloud simulation system by con-current execution of MDVP simulation system. The method proposed in the paper has been applied to a MDVP in the aerospace industry, which indicates that the proposed method can support an efficient engineering methodology that can transform the traditional centralized, serial simulation optimization to the distributed collaborative and parallel simulation optimization.

This paper proposes a coastline detection method for polarimetric synthetic aperture radar (SAR) images based on region-based and edge-based active contour models. It can be used to detect coastline accurately and fast. In this method, the region-based and edge-based active contour models are effectively combined by an important property of the likelihood ratio edge detector in polarimetric SAR images, which is proved by theory. Using low-resolution image obtained by multilook processing, we detect accurate and continued coarse coastlines by a region-based level set method. The property of the likelihood ratio edge detector of polarimetric SAR images along the coastline region is then analyzed. The coarse detection result is finally refined using a fast snake active contour model based on the edge property. Polarimetric SAR data acquired by RADARSAT-2 over a Singapore region and TerraSAR-X over a Berkeley region are both used to test the proposed algorithm. The experimental results show that the coastline is fast and accurately detected in different initial scales. The proposed method greatly reduces the data processing time compared with the coastline detection method based on a single scale.

Cloud Manufacturing can provide mass manufacturing resources and capabilities as services via the Internet. Undoubtedly, multiple manufacturing clouds (MCs) will have extremely abundant services in terms of function, price, etc. The ability to leverage ample services hosted in MCs has direct relation to the success or failure of a manufacturer. Meanwhile, various uncertainties in today's highly-dynamic business environment can easily disrupt manufacturing activities, rendering original schedules ineffective or even obsolete. IoT's real-time sensing ability can be used to detect those uncertainties. However, little work has been done to take advantage of abundant services from MCs and to effectively deal with uncertainties. In order to address this issue, we propose a mathematical formulation for IoT-enabled dynamic Service Selection (SS) across multiple MCs. We consider three kinds of uncertainties (fluctuation of completion time, choices of manufacturing services, and runtime changes made by users) that come from both the user and market sides. The formulation can guide the dynamic SS and enable users to continuously adjust SS to be more effective and efficient.

This letter proposes a harbor detection method by extracting specific paralleled contours of harbors in polarimetric synthetic aperture radar (SAR) images. First, land and sea are partitioned by the region-based level set segmentation algorithm which is applied to the extraction of the volume scattering component. Then, parallel curves are modeled by a series of parallel line segments. They are detected by finding out and connecting paralleled line segments given certain criteria. The detected parallel curves are merged according to a distance measurement so that the harbor contour is finally detected. RADARSAT-2 polarimetric SAR data covering an area in Singapore are used to test the harbor detection method. Experimental results show that the method improves sea-land segmentation accuracy and extracts all parallel curves along the coast correctly. The method enables an accurate detection of harbors.