Search

when chaotic systems are implemented on the finite precision devices, some dynamical properties will disappear and lead to the problem of dynamical degradation. In this paper, based on the external anti-control theory, a novel compensatory method is proposed for improving the dynamical degradation. A continuous chaotic system is introduced to compensate the digital chaotic system with proper control function. Simulation results are given to show the validity of the proposed method. Furthermore, we propose a new pseudorandom number generator (PRNG) based on the controlled digital system. Statistical test results show that the proposed PRNG has good randomness, uniformity and complexity, thus it can be used for cryptography and other potential applications.

With the deregulation and development of power market, some new features and trends about electric power demand forecasting issues come into being, which mainly includes the following two aspects: Firstly, the forecasting has developed into a complicated problem associated with users' behavior and psychology. Secondly, short and medium-term forecasting used for bidding and clearing in power market are discussed more than long-term forecasting in order to guide power grid planning. Considering the new features of forecasting and the great progress in artificial intelligence (AI), this paper puts forward the idea and framework of forecasting problem under the power market environment by integrated model and Deep Learning technology.

Frontalization typically represents the process of generating frontal views of faces from single unconstrained images. Previous researches have suggested that face frontalization may considerably improve the performance of current face recognition systems in the wild. While many frontalization methods focuses on 3D reconstruction, approximating facial surface for each input image could be a hard problem and lead to facial misalignment. Therefore, more straightforward and simple approaches are proposed. A single unmodified 3D surface is used to approximate the shape of all input faces, and the frontal view is obtained through two specific steps, hard frontalization and soft symmetry. However, we find that large pose variations and landmark localization error may result in extreme artifacts in frontalized faces. To improve the performance of this strategy, we firstly proposed an adaptive method to take the posture of head into consideration, and use it to adjust the soft symmetry step. Then the quality of frontalization is assessed so we are able to adaptively compensate for the landmark localization error. Experiments on the Labeled Faces in the Wild (LFW) benchmark and CAS-PEAL-R1 dataset verify the effectivity of our methods to unconstrained conditions.

A safe real-time imitation framework for humanoid robots to human beings action is proposed. Postures of an actor are divided into three categories: robust postures, normal postures and dangerous postures. To deal with the robust postures, direct mapping from skeleton data gathered by 3D cameras to each angle of robot joint motors is taken because the postures are robust enough for imitation in high speed. When it comes to a normal posture, one specific safe robot posture is designed to correspond to each part of separated human postures. As the actor changes his posture from one to another, the robot will also switch from the corresponding posture to another in the designed way. Meanwhile, when there is no need to switch, the direct mapping is taken to fulfill the task. The imitation strategy for dangerous postures is the same as the normal postures except that the robot can not use direct mapping when there is no switch task because some changes of the designed posture may lead to tumble.

This research examines the influence of logistic service supply chain network (hereinafter referred to as the LSSC network) to the growth of logistic service integrator (LSI) from the perspective of social network. We divided the LSSC network into five sub-networks, they are social network, reputational network, marketing network, logistic collaboration network and logistic innovation network. We tested how the five sub-networks affecting the growth performance of LSI with the empirical methods. Eventually we concluded that the development of social network, logistic innovation network and marketing network has a significant impact on the growth of LSI; reputational network has an indirect impact on the firm growth by means of marketing network; and up to now, the development of logistic collaboration network is not very perfect, so it has little impact on the growth performance of LSI.

A birefringence compensation technique based on angled star couplers in arrayed waveguide grating (AWG) is experimentally investigated in silica-on-silicon waveguide with B-Ge codoped upper cladding. A 16-channel AWG with 100-GHz channel spacing is designed and fabricated. The star couplers are designed according to Rowland circle construction with an oblique incident/diffraction angle. Five different polarization compensation parameters corresponding to different incident/diffraction angles are employed to investigate the stress-induced birefringence. The experimental results confirm that the polarization-dependent wavelength shift (PDλ) can be tuned by varying the incident/diffraction angle at the star couplers, and a birefringence-free AWG can be achieved without any additional fabrication process.

A cyclic arrayed waveguide grating (AWG) router based on silicon nanowire waveguides with uniform channel insertion loss is designed and experimentally demonstrated. Different from conventional design, the directions of arrayed waveguides are no longer pointed to the central output waveguide, but are adjusted according to a distribution function. The experimental results show that the channel insertion loss non-uniformity is reduced from 2.7 dB to 1.1 dB. This uniform loss design does not change the overall construction of AWGs, and does not increase the device size.

A compact arrayed waveguide grating triplexer based on small SU-8 strip waveguides is designed by using cross-order method. The AWG triplexer has a very compact size of as small as 1.3 × 0.402 mm2 since the high index contrast between SU-8 core layer and SiO2 and air cladding, and can be easily fabricated by using direct ultraviolet photolithography process without additional etching process. The simulation results show that the 3-dB bandwidths of three channels (1310nm, 1490nm and 1550nm) for TE polarization are greater than 11 nm and each channel has a low polarization dependence wavelength shift less than 0.65nm.

Social Delay Tolerant Networks (Social DTNs) are promising by exploiting transfer opportunities between mobile devices to forward data for future network applications. Efficient data forwarding protocols are needed to adapt to the node mobility and intermittent connectivity of social DTNs. Previous forwarding algorithms present the node mobile patterns to promote opportunistic routing performance. In this work we introduce the node mobility period into the packet routing. A mobility prediction scheme based on node behavior period can effectively help routing decision. Realistic trace-based simulation shows that the proposed algorithm achieves performance improvement comparing with Epidemic with desirable delivery ratio and fewer delivery cost.

When wind speed is below the rated value, the efficiency of captured wind energy must be maximized and the mechanical oscillation be guaranteed to be small. To deal with these problems, This essay describes the advantages of the direct-driven permanent magnet synchronous wind power system, and then introduces the two-frequency-loop model based on frequency separation principle, The LPV model and LPV control method are suggested for the high-frequency part of the system. The output of the high-frequency part is used to compensate the mechanical torque. The mathematical model is built with MATLAB, and the online test is carried out by dSpace. The simulation results show that the controller reduces mechanical oscillation effectively, and enhances the system reliability.

Based on analyzing the features of transverse modes of LD beam and experiment with quartz cylindrical lenses of different diameters, using certain cylindrical lenses to compress the ellipse shaped LD beam in the direction of the fast axis can result in the best pumping efficiency. The research findings can be applied to the practical production to manufacture highly efficient Nd:YVO 4 /KTP optical-contact crystal lasers at a low cost.

Targeted for poor security, low efficiency of the existing equipment in a certain factory, a new automatic bolts typewriter based on hydromechatronics technology, which is suiting for 1-3 inChes or even greater specifications, 200-800mm or longer bolts typewriting. The overall sCheme, feeding and discharging mechanism of this typewriter is described in this paper. The typewriter has simple structure, easily adjusting properties and can working automatically or semi-automatically, which has high efficiency, low manufacturing costs, good performance, and high cost-effective.

In this paper, from detailed experiment-based research on how ambient temperature influences the output power of optical-contact crystal green laser, it is showed that the output power of optical-contact crystal laser is directly affected by ambient temperature, and individual difference exists in the optical-contact crystals. The negative effect caused by low ambient temperature can be eliminated by temperature control sleeves. Most optical-contact crystal lasers work well at room temperature.

Frequency domain equalization (FDE) has shown its ability to improve the performance of single carrier system. For DS-CDMA system, MMSE based FDE can obtain frequency diversity gain through undergoing more fading paths. Meanwhile, delay time-domain windowing technique has been introduced to suppress the noise power which perturbs the channel estimation. In this paper, we analysis the impact of delay time-domain windowing technique, the fading rate of the channel and the number of Rayleigh fading paths on the BER performance of DS-CDMA system with MMSE-FDE under MMSE channel estimation. Simulations show that delay time windowing technique can improve the BER performance significantly and such improvement grows with the increase of Eb/N0.

To implement the interoperation of different ontologies in medical informatics domain and achieve the goal of the sharing and multiplexing of domain knowledge, the heterogeneous problem between ontologies must be solved. Now the best method of resolving ontology heterogeneous is establishing ontology mapping, that is founding the semantic relations between ontologies and making corresponding mapping rules firstly, and then integrating all independent ontologies into a whole to be operated. In order to produce mapping-pairs and implement ontology mapping and realizing knowledge sharing, a concept similarity algorithm combining semantic similarity with semantic relativity is adopted to estimate the similarity between concepts. It has been proved that the evaluation of concept similarity between ontologies is more accurate by considering both semantic similarity and semantic relativity, and it has laid a good foundation for establishing mapping-pairs and knowledge sharing model based on ontology mapping.

Spin-forming of cup-shaped thin-walled trapezoidal inner-gear is a new technology of the near-net forming in gear manufacturing field. Experiments and FEM simulations shows that the quality of workpiece is greatly influenced by the forming process method. The defects, such as the un-uniform distribution of tooth height along the axial and tangential directions and the wave-shaped opening-end of workpiece etc., occur easily. The mechanism for defects during spinning is analyzed, which is mainly caused by the partial thickening (in the area of gear-tooth) and thinning (in the area of tooth-groove). Improved forming methods, such as stagger spin-forming, positive-reverse rotation of main spindle and adopting a restraint ring at the opening-end of blank, are put forward. The forming methods are simulated by MSC. Marc and also investigated by experiments. The results show that the improved process methods overcome the forming defects effectively.

In this paper, we develop a distributed cooperative spectrum sensing algorithm based on network code to decrease the need for bandwidth in cognitive radios. The model of the distributed cooperative spectrum sensing is analyzed, and the algorithm of distributed scheme based on network code is described. The simulation results show that compared with the distributed scheme without network code and the non-cooperative one, the throughput of secondary users is increased apparently, and also, the probability of detection of the distributed scheme with network code is improved about 10% and 20%, respectively, further, the optimal sensing time is shorter about 82% and 88%, respectively.

Sensing coverage is defined as the ratio of the sensible area to the entire desired area. It is one of the critical measures of performance or service quality offered by a sensor network. In previous studies, the sensing range of a node is generally assumed to be a deterministic value in all directions. In realistic environments, the sensing range of a node is not a constant because of shadowing effects. Recent studies are focused on the sensing coverage for WSNs in shadowed environments. But they ignore the edge effect which will induce an overestimation on the sensing coverage. The objective of this paper is to examine the impact of the shadowing effects and edge effect on the sensing coverage. The theoretical formulations for the sensing coverage are provided. Furthermore, the multiple-sensing coverage is studied as well. Our simulation results indicate the correctness of our analytical model.

Proposed a method of detecting intrusion using incremental SVM based on key feature selection. A center SVM summarizes the distributed samples and incorporates them to build the incremental SVM for locals. By eliminating the redulldant features of sample dataset the space dimension of the sample data is reduced?Using this method it can overcome the shortages of SVM-time-consuming of training and massive dataset storage. The simulation experiments with KDD Cup 1999 data demonstrate that our proposed method achieves the increasing performance for intrusion detection.