Search

Privacy-preserving traffic flow classification has attracted a significant amount of research interest because of its increasing importance to both network management and privacy protection. In this paper, we propose novel methods for effectively protecting network flow identifiers and attributes against privacy inference attacks in port-based and payload-based classifications respectively, and analyze their performance guarantee on data utility for flow classification and privacy (security). For protection of flow identifiers, we propose a partial identifier protection approach applying randomization and anonymization respectively on desired bit positions to conceal sensitive information, and show their expected-case performance guarantee. For protection of flow attributes, we propose a perturbation-based scheme that first selects the representative attributes by deploying an entropy-based attribute selection method to filter out redundant and insignificant attributes and reduce the problem space, then partitions the attribute domains into either equal-depth or equal-width intervals and perturbs attribute values in these intervals by swapping them with those in adjacent intervals and intervals with same value distribution respectively. We analyze the performance guarantee of the proposed methods and show the experiment results of classification accuracy obtained by implementing popular machine-learning based benchmark classifiers on our selected attributes against that on raw attributes, and on our perturbed attribute values against that on raw values, respectively. The experiment results show that our proposed methods for attribute selection and perturbation retain a high degree of data utility under the desired privacy guarantee for network traffic classification. [ABSTRACT FROM AUTHOR]

Highlights • A full scheme aiming to achieve all necessary demands for shared-data public auditing. • A novel lazy-revocation strategy for efficient and secure user revocation. • Enable identity traceability of malicious users using a modification record table. • Enable dynamic operations of shared data using an extended dynamic hash table. • The scheme outperforms the state-of-the-art ones in computational costs. Abstract With increasing popularity of collaboration in clouds, shared data auditing has become an important issue in cloud auditing field, and attracted extensive attention from the research community. However, none of the state of the arts can fully achieve all indispensable functional and security requirements. Thus, in this paper, we present a comprehensive public auditing scheme for shared data. Specifically, to preserve users' identity privacy, signatures on the challenged blocks are converted to the ones signed by the group manager during proof generation; to protect data privacy, a random masking is adopted to blind data proof; a modification record table is designed to record operation information to support identity traceability; we further extend the dynamic hash table to support shared-data dynamics, and present a batch auditing strategy. Moreover, we design a lazy-revocation based group management mechanism to achieve efficient group dynamics, which can resist collusion attacks while significantly reducing computational costs. We formally prove the security of our scheme, and evaluate its performance by comprehensive experiments and comparisons with the state-of-the-art ones. The results demonstrate that our scheme can effectively achieve secure auditing and outperforms the previous ones in computational overhead while maintaining relatively low communication costs. [ABSTRACT FROM AUTHOR]

We address the problem of restoring an original image from its blurry and noisy observation together with inaccurate information of the blurring process. For this purpose, we propose an enhanced regularized structured total least squares (RSTLS) model that can estimate the latent image and blur kernel simultaneously. In the proposed model, both the image and the blur kernel are characterized by a group-based low-rank prior, which assumes that a group of vectorized similar data patches can be well approximated by a low-rank matrix. We develop an alternating minimization algorithm to solve the proposed model efficiently. Numerical experiments demonstrate the effectiveness of our method in terms of both quantitative measures and visual quality. [ABSTRACT FROM AUTHOR]

We propose a novel image decomposition method to decompose an image into its cartoon and texture components. To exploit the nonlocal self-similarity of cartoon-plus-texture images, we construct groups by stacking together similar image patches into 3-D arrays and consider group as the basic unit of decomposition. We decompose each group via a convex optimization model consisting of 3-D cartoon and texture priors. These priors characterize the local properties of the cartoon and texture components and the nonlocal similarity within each component in a unified and natural manner. We develop the alternating direction method of multipliers (ADMM) to efficiently solve the proposed model. For further improvement, we investigate an adaptive rule for the estimation of the regularization parameter. The proposed method is also extended to tackle noisy images. Numerical experiments confirm that the performance of the proposed method is competitive with some of the state-of-the-art schemes. [ABSTRACT FROM AUTHOR]

• Define a new tensor unfolding to unfold an N -way tensor into a three-way tensor. • Propose a novel tensor rank for N -way tensors based on the new tensor unfolding. • Establish a convex relaxation for efficiently minimizing the proposed tensor rank. • Apply the proposed relaxation to tensor recovery problems with ADMM-based solver. The recent popular tensor tubal rank, defined based on tensor singular value decomposition (t-SVD), yields promising results. However, its framework is applicable only to three-way tensors and lacks the flexibility necessary tohandle different correlations along different modes. To tackle these two issues, we define a new tensor unfolding operator, named mode- k 1 k 2 tensor unfolding, as the process of lexicographically stacking all mode- k 1 k 2 slices of an N -way tensor into a three-way tensor, which is a three-way extension of the well-known mode- k tensor matricization. On this basis, we define a novel tensor rank, named the tensor N -tubal rank, as a vector consisting of the tubal ranks of all mode- k 1 k 2 unfolding tensors, to depict the correlations along different modes. To efficiently minimize the proposed N -tubal rank, we establish its convex relaxation: the weighted sum of the tensor nuclear norm (WSTNN). Then, we apply the WSTNN to low-rank tensor completion (LRTC) and tensor robust principal component analysis (TRPCA). The corresponding WSTNN-based LRTC and TRPCA models are proposed, and two efficient alternating direction method of multipliers (ADMM)-based algorithms are developed to solve the proposed models. Numerical experiments demonstrate that the proposed models significantly outperform the compared ones. [ABSTRACT FROM AUTHOR]

Differential evolution (DE) is an efficient and robust evolutionary algorithm (EA), that has been widely and successfully applied to solve global optimization problems in diverse real-world applications. As the population structure has a major influence on the behavior of an EA, effectively incorporating population topology into DE has recently attracted increasing attention. Previous works have shown the effectiveness of different topologies in improving the performance of DE and revealed that different topologies can have different effects on the population’s ability to solve optimization problems. However, the synergy of different topologies for the problems being solved has not been systematically investigated in most DE variants. Moreover, individuals with different fitness values play different roles in guiding the search during the evolutionary process. Nevertheless, the individual-dependent roles are not considered in most DE variants that consider the population topology. To overcome these drawbacks and utilize the information that is derived from the differences between the fitness values of individuals for topology adaption, we propose a multi-topology-based DE (MTDE) algorithm that includes an ensemble of multiple population topologies (MPT), an individual-dependent adaptive topology selection (ITS) scheme, and a topology-dependent mutation (TDM) strategy. In the ensemble of MPT, multiple population topologies with different degrees of connectivity are employed. In the ITS scheme, each individual adaptively selects the topology that is most compatible its role in guiding the search based on its fitness value. In the TDM strategy, the parents for mutation are chosen from the neighborhood of the current individual based on the corresponding topology to generate offspring. The effectiveness of the proposed algorithm is extensively evaluated on a suite of benchmark functions. Experimental results demonstrate the competitive performance of MTDE when compared with other state-of-the-art DE variants and EAs. [ABSTRACT FROM AUTHOR]

Abstract Differential evolution (DE) has attracted much attention in the field of evolutionary computation and has proved to be one of the most successful evolutionary algorithms (EAs) for global optimization. Mutation, as the core operator of DE, is essential for guiding the search of DE. In this study, inspired by the phenomenon of social learning in animal societies, we propose an adaptive social learning (ASL) strategy for DE to extract the neighborhood relationship information of individuals in the current population. The new DE framework is named social learning DE (SL-DE). Unlike the classical DE algorithms where the parents in mutation are randomly selected from the current population, SL-DE uses the ASL strategy to intelligently guide the selection of parents. With ASL, each individual is only allowed to interact with its neighbors and the parents in mutation will be selected from its neighboring solutions. To evaluate the effectiveness of the proposed framework, SL-DE is applied to several classical and advanced DE algorithms. The simulation results on forty-three real-parameter functions and seventeen real-world application problems have demonstrated the advantages of SL-DE over several representative DE variants and the state-of-the-art EAs. [ABSTRACT FROM AUTHOR]

Currently, 3G and 4G networks provide customers with high-speed wireless services almost everywhere. However, the wireless connection is often unstable and unreliable, especially for fast-moving end users (e.g., those on trains and buses). To investigate the severity of this problem, we conducted real experiments on fast-moving trains to investigate the quality of 3G connections. From the results, we found that 1) from the temporal perspective, the 3G connections were not stable and suffered from frequent disruptions of connectivity, and 2) from the spatial perspective, the connections that were established in different train compartments were largely independent. These two findings motivate us to propose a brand-new fog computing structure, which acts as an intermediate layer between the end users and the 3G infrastructure. This new fog structure introduces a series of mutually chained network gateways that are located in different compartments. This structure addresses the aforementioned problem of unstable connectivity and thus ensures reliable wireless service for fast-moving users, such as passengers on trains. We performed a series of theoretical and empirical analyses to evaluate the performance of the newly proposed structure. All of the experimental results suggest that our proposed fog structure greatly improves the reliability of wireless connections on fast-moving trains. [ABSTRACT FROM AUTHOR]

In this paper, we study the problem of recovering a tensor with missing data. We propose a new model combining the total variation regularization and low-rank matrix factorization. A block coordinate decent (BCD) algorithm is developed to efficiently solve the proposed optimization model. We theoretically show that under some mild conditions, the algorithm converges to the coordinate wise minimizers. Experimental results are reported to demonstrate the effectiveness of the proposed model and the efficiency of the numerical scheme. [ABSTRACT FROM AUTHOR]