PolyU Shenzhen Research Institute

Shenzhen, China

PolyU Shenzhen Research Institute

Shenzhen, China

Time filter

Source Type

Meng W.,Institute for Infocomm Research | Meng W.,City University of Hong Kong | Luo X.,Hong Kong Polytechnic University | Luo X.,PolyU Shenzhen Research Institute | And 2 more authors.
Proceedings - 15th IEEE International Conference on Trust, Security and Privacy in Computing and Communications, 10th IEEE International Conference on Big Data Science and Engineering and 14th IEEE International Symposium on Parallel and Distributed Processing with Applications, IEEE TrustCom/BigDataSE/ISPA 2016 | Year: 2016

To encourage collaboration among single intrusion detection systems (IDSs), collaborative intrusion detection networks (CIDNs) have been developed that enable different IDS nodes to communicate information with each other. This distributed network infrastructure aims to improve the detection performance of a single IDS, but may suffer from various insider attacks like collusion attacks, where several malicious nodes can collaborate to perform adversary actions. To defend against insider threats, challenge-based trust mechanisms have been proposed in the literature and proven to be robust against collusion attacks. However, we identify that such mechanisms depend heavily on an assumption of malicious nodes, which is not likely to be realistic and may lead to a weak threat model in practical scenarios. In this paper, we analyze the robustness of challenge-based CIDNs in real-world applications and present an advanced collusion attack, called random poisoning attack, which derives from the existing attacks. In the evaluation, we investigate the attack performance in both simulated and real CIDN environments. Experimental results demonstrate that our attack can enables a malicious node to send untruthful information without decreasing its trust value at large. Our research attempts to stimulate more research in designing more robust CIDN framework in practice. © 2016 IEEE.

Feng W.-Q.,Hong Kong Polytechnic University | Yin J.-H.,Hong Kong Polytechnic University | Yin J.-H.,PolyU Shenzhen Research Institute | Tao X.-M.,Hong Kong Polytechnic University | And 2 more authors.
International Journal of Geomechanics | Year: 2017

A plasticine material exhibits the characterized viscous stress-strain behavior with some similarity to the behavior of clayey soils. This paper presents a series of experimental tests, which include oedometer tests, isotropic creep tests, and triaxial multistrain rate compression tests, on a plasticine material. The test study focuses on effects of time and strain rate on viscous stress-strain behavior of the plasticine material under one-dimensional (1D) straining, isotropic stressing, and triaxial compression conditions. Values of compression index (Ccε), rebounding index (Crε), and creep coefficient (Cαε) are obtained from the 1D straining and 1D stressing test data. The plasticine material has no primary consolidation period, and creep occurs from the beginning. Values of Ccε, Crε, and Cαε are smaller than those of the soft clays. The triaxial multistrain rate compression test data show that the stress-strain behavior of the plasticine depends on the strain rates and the confining pressures. A parameter of ρ0.01 is adopted to evaluate the strain-rate effects. The strain-rate effects on the stress-strain behavior of the plasticine material are obvious and significant. The values of ρ0.01 are larger than those of clays. Both friction angle and cohesion of the plasticine increase with strain rate. This is different from the friction angle and cohesion at the critical state for all soils. The friction angle of the plasticine is from 2.57° at a strain rate of 0.01%/min to 3.21° at a strain rate of 1%/min, which is much smaller than that of all clays. With the help of a scanning electron microscope, the microstructures of this plasticine material before and after oedometer and isotropic creep tests are visualized and compared. The compression of the plasticine material is mainly due to the irrecoverable porosity decrease of the material and the structural compression. © 2016 American Society of Civil Engineers.

Borana L.,Hong Kong Polytechnic University | Yin J.H.,Hong Kong Polytechnic University | Yin J.H.,PolyU Shenzhen Research Institute | Singh D.N.,Indian Institute of Technology Bombay | And 2 more authors.
International Journal of Geomechanics | Year: 2017

A series of pullout tests were conducted on a model pile in a soil with different initial water contents and different pile surface roughnesses to study their influences on pile skin friction. To measure axial strains and skin friction of the pile, fiber Bragg grating (FBG) strain sensors were used. Data were collected during the pullout tests and are analyzed in this paper. Test results indicate that the skin friction and axial strain of the model piles decrease with the initial water content of the soil but increase with the degree of surface roughness of the piles. The axial strain induced in the upper segment of the tested pile is considerably greater than that in the bottom segment of the pile. In addition, the skin friction is distributed nonlinearly along the length of the model piles. © 2016 American Society of Civil Engineers.

Li W.,City University of Hong Kong | Meng W.,City University of Hong Kong | Meng W.,Institute for Infocomm Research | Luo X.,Hong Kong Polytechnic University | And 2 more authors.
Computers and Security | Year: 2016

Network intrusion detection systems (NIDSs) have been developed for over twenty years and have been widely deployed in computer networks to detect a variety of network attacks. But one of the major limitations is that these systems would generate a large number of alarms, especially false alarms (positives) during the detection. To address this issue, many machine learning approaches have been applied to reduce NIDS false positives. However, we notice that multi-view based approach is often ignored by the literature, which uses one function to model a particular view and jointly optimizes all the functions to optimize and improve the learning performance. In addition, most existing studies have not implemented their algorithms into practical alam systems. In this paper, we thus develop MVPSys, a practical multi-view based false alarm reduction system to reduce false alarms more efficiently, where each view represents a set of features. More specifically, we implement a semi-supervised learning algorithm to construct two-view items and automatically exploit both labeled and unlabeled data. That is, this system can automatically extract and organize features from an incoming alarm into two feature sets: destination feature set and source feature set, where the former contains the features related to the target environment and the latter contains the features about the source environment. In the evaluation, we deploy our system into two real network environments besides using two datasets. Experimental results indicate that our system can achieve a stable filtration accuracy of over 95%, offering a significant improvement as compared with the state-of-the-art algorithms. © 2016 Elsevier Ltd.

Xu D.-S.,Huazhong University of Science and Technology | Yin J.-H.,PolyU Shenzhen Research Institute | Yin J.-H.,Hong Kong Polytechnic University
Engineering Geology | Year: 2016

In this work, a slope reinforcement system using glass fiber reinforced polymer (GFRP) anchors with pressure grouting was adopted in a field project in Hong Kong. The performance of the GFRP anchor during slope excavation was measured using a novel distributed strain sensing technology, known as Brillouin Optic Time Domain Analysis (BOTDA). The full strain profiles along the GFRP anchor under different excavation stages were obtained using specially protected fiber optic sensors. In addition to fiber optic sensors, traditional strain gauges were installed in the same GFRP anchor. Comparisons show that the BOTDA sensors have good accuracy. In addition, the measured results indicate that the maximum tensile strains and forces occurred at one-third of the GFRP anchor length from the slope surface. The tensile force distribution within the active zone is curvilinear which is confirmed by elastic theory analysis. Shear stress distributions along the GFRP anchors were obtained by differentiating the strain data numerically. The theoretical analysis results were consistent with the measured data at the initial excavation stage. However, the theoretical analysis underestimated the shear stress at the final excavation stage where the slope undergoes plastic deformation. Based on the field measurement results and theory analysis, we conclude that the BOTDA sensing technology provides an alternative and effective approach to identifying distributed strains along anchors and shear zones in reinforced slopes. © 2016 Elsevier B.V.

Yin J.-H.,PolyU Shenzhen Research Institute | Yin J.-H.,Hong Kong Polytechnic University
International Journal of Geomechanics | Year: 2015

In this paper, a number of fundamental concepts are presented and explained. These include (1) differences among an instant compression line, a normal consolidation line, and a true instant compression line; (2) the uniqueness of viscoplastic strain rates with a stress-strain state; (3) whether the creep compression is smaller than the instant compression; (4) the separation of the total strain rates; (5) the relation between elastic-plastic models and elastic viscoplastic (EVP) models, etc. The major conclusions are the following: (1) the elastic compression is the true instant compression; (2) the magnitude of a creep-strain rate at a stress-strain state point is unique, independent of the loading path to reach this point; (3) the true instant (elastic) compression is much smaller than the creep compression; (4) it is more appropriate that strain rates of geomaterials are composed of elastic strain rates and viscoplastic strain rates; (5) the one-dimensional (1D) EVP (1D EVP) is a genuine extension of Maxwell's linear rheological model for considering the nonlinear behavior of soils; (6) the EVP model is more general than an elastic-plastic model; (7) the nonlinear functions proposed by the author are good for fitting the creep compression and the compression under high stress of most soft soils in 1D straining; and (8) the three-dimensional EVP model is rigorously derived using the 1D EVP model approach and the modified Cam-Clay model, but further improvements of this model are still needed. At the end, a number of areas are presented for further study. © 2015 American Society of Civil Engineers.

Fu M.W.,Hong Kong Polytechnic University | Fu M.W.,PolyU Shenzhen Research Institute | Wang J.L.,Hong Kong Polytechnic University | Korsunsky A.M.,University of Oxford
International Journal of Machine Tools and Manufacture | Year: 2016

Plastic deformation at the macroscopic scale has been widely exploited in industrial practice in order to obtain desired shape and control the requested properties of metallic alloy parts and components. The knowledge of deformation mechanics involved in various forming processes has been systematically advanced over at least two centuries, and is now well established and widely used in manufacturing. However, the situation is different when the physical size of the workpiece is scaled down to the micro-scale (µ-scale). In such cases the data, information and insights from the macro-scale (m-scale) deformation mechanics are no longer entirely valid and fully relevant to µ-scale deformation behavior. One important reason for the observed deviation from m-scale rules is the ubiquitous phenomenon of Size Effect (SE). It has been found that the geometrical size of workpiece, the microstructural length scale of deforming materials and their interaction significantly affect the deformation response of µ-scale objects. This observation gives rise to a great deal of research interest in academia and industry, causing significant recent effort directed at exploring the range of related phenomena. The present paper summarizes the current state-of-the-art in understanding the geometrical and microstructural SEs and their interaction in deformation processing of µ-scale components. The geometrical and grain SEs in µ-scale deformation are identified and articulated, the manifestations of the SE are illustrated and the affected phenomena are enumerated, with particular attention devoted to pointing out the differences from those in the corresponding m-scale domain. We elaborate further the description of the physical mechanisms underlying the phenomena of interest, viz., SE-affected deformation behavior and phenomena, and the currently available explanations and modeling approaches are reviewed and discussed. Not only do the SEs and their interaction affect the deformation-related phenomena, but they also induce considerable scatter in properties and process performance measures, which in turn affects the repeatability and reliability of deformation processing. This important issue has become a bottleneck to the more widespread application of µ-scale deformation processing for mass production of µ-scale parts. What emerges is a panoramic view of the SE and related phenomena in µ-scale deformation processing. Furthermore, thereby the outstanding issues are identified to be addressed to benefit and promote practical applications. © 2016 Elsevier Ltd

Feng W.-Q.,Hong Kong Polytechnic University | Feng W.-Q.,PolyU Shenzhen Research Institute | Liu Z.-Y.,Hong Kong Polytechnic University | Tam H.-Y.,Hong Kong Polytechnic University | And 2 more authors.
Measurement: Journal of the International Measurement Confederation | Year: 2016

The pore water pressure sensors with the six-hole suspended-core polarization-maintaining photonic crystal fiber (SC-PM-PCF) and commercial polarization-maintaining photonic crystal fiber (PM-PCF) are designed based Sagnac interferometer and calibrated in the laboratory. According to the theoretical analysis and calibration results, the transmission spectrum is very sensitive to the pore water pressure. It is found that the wavelength of the spectrum has a good linear relationship with variances of the surrounding pore water pressure, and the coefficient of wavelength-pressure of the commercial PM-PCF is 304.41 kPa/nm with the length of 35 cm as the sensing element while the coefficient of the SC-PM-PCF is 254.75 kPa/nm with the length of 100 cm. Finally, the two PM-PCF sensors are applied and compared with the conventional Pore water Pressure Transducers (PPTs) in a physical model test. It is found that measurements of the PM-PCF sensors are in good agreement with the results measured by the conventional PPTs. © 2016 Elsevier Ltd. All rights reserved.

Liu X.,Hong Kong Polytechnic University | Liu X.,PolyU Shenzhen Research Institute | Xiao B.,Hong Kong Polytechnic University | Xiao B.,PolyU Shenzhen Research Institute | And 2 more authors.
IEEE Transactions on Parallel and Distributed Systems | Year: 2015

Radio-Frequency Identification (RFID) technology brings revolutionary changes to many fields like retail industry. One important research issue in large RFID systems is the identification of unknown tags, i.e., tags that just entered the system but have not been interrogated by reader(s) covering them yet. Unknown tag identification plays a critical role in automatic inventory management and misplaced tag discovery, but it is far from thoroughly investigated. Existing solutions either trivially interrogate all the tags in the system and thus are highly time inefficient due to re-identification of already identified tags, or use probabilistic approaches that cannot guarantee complete identification of all the unknown tags. In this paper, we propose a series of protocols that can identify all of the unknown tags with high time efficiency. We develop several novel techniques to quickly deactivate already identified tags and prevent them from replying during the interrogation of unknown tags, which avoids re-identification of these tags and consequently improves time efficiency. To our knowledge, our protocols are the first non-trivial solutions that guarantee complete identification of all the unknown tags. We illustrate the effectiveness of our protocols through both rigorous theoretical analysis and extensive simulations. Simulation results show that our protocols can save up to 70 percent time when compared with the best existing solutions. © 2015 IEEE.

Meng B.,Hong Kong Polytechnic University | Fu M.W.,Hong Kong Polytechnic University | Fu M.W.,PolyU Shenzhen Research Institute | Shi S.Q.,Hong Kong Polytechnic University
Materials and Design | Year: 2016

With the increasing demand for meso/micro-scaledmedical products made of biocompatiblematerials, thermalaidedmesoforming is proposed to improvematerial formability and homogenize flow behavior of materials that are difficult to deform at room temperature. However, the unique material deformation behavior and the interactive effects of material microstructure and deformation temperature on forming quality of the fabricated micropart remain unknown. This study thus aims at addressing this issue in thermalmesoforming in terms of deformation load, material flow,microstructural evolution, dimensional accuracy, and defect formation. Accordingly, the fabrication of a titanium dental abutment by one-stroke mesoforming at elevated temperature is conducted and explored. The characteristic and quality of the mesoformed part are extensively examined. The surface grains on the square extrudate undergo severe deformation and generate an equiaxed structure, reflecting that mesoforming at elevated temperature facilitates the homogenization of material flow without coarsening grain size. In addition, the dimensional accuracy, surface quality and the sizes of burr and flash are associated with the initial grain size of pure titanium, and the surface finish is improved by using fine-grained titanium. The fine-grained material is thus desirable for achieving the optimal surface quality in the thermal-aided mesoformed parts. © 2015 Elsevier Ltd.

Loading PolyU Shenzhen Research Institute collaborators
Loading PolyU Shenzhen Research Institute collaborators