Pre-trained CNNs on ImageNet have been widely used in object tracking for feature extraction. However, due to the domain mismatch between image classification and object tracking, the submergence of the target-specific features by noise largely decreases the expression ability of the convolutional features, resulting in an inefficient tracking. In this paper, we propose a robust tracking algorithm with low-dimensional target-specific feature extraction. First, a novel cascaded PCA module is proposed to have an explicit extraction of the low-dimensional target-specific features, which makes the new appearance model more effective and efficient. Next, a fast particle filter process is raised to further accelerate the whole tracking pipeline by sharing convolutional computation with a ROI-Align layer. Moreover, a classification-score guided scheme is used to update the appearance model for adapting to target variations while at the same time avoiding the model drift that caused by the object occlusion. Experimental results on OTB100 and Temple Color128 show that, the proposed algorithm has achieved a superior performance among real-time trackers. Besides, our algorithm is competitive with the state-of-the-art trackers in precision while runs at a real-time speed.

To reconstruct low-resolution facial photographs which are in focus and without motion blur, a novel algorithm based on local similarity preserving is proposed. It is based on the theories of local manifold learning. The innovations of the new method include mixing point-based entropy and Euclidian distance to search for the nearest points, adding point-to-patch degradation model to restrict the linear weights and compensating the fusing patch to keep energy coherence. The compensation reduces the algorithm dependence on training sets and keeps the luminance of reconstruction constant. Experiments show that our method can effectively reconstruct 1612 images with the magnification of 88 and the 3224 facial photographs in focus and without motion blur.

This investigation proposed two array beamformers SPFDBB (Soft Penalty Frequency Domain Block Beamformer) and FDABB (Frequency Domain Adjustable Block Beamformer). Compared with the conventional beamformers, these frequency-domain methods can significantly reduce the computation power requirement in ASR (Automatic Speech Recognition) based applications. Like other reference signal based techniques, SPFDBB and FDABB minimize microphone's mismatch, desired signal cancellation caused by reflection effects and resolution due to the array's position. Additionally, these proposed methods are suitable for both near-field and far-field environments. Generally, the convolution relation between channel and speech source in time domain cannot be modeled accurately as a multiplication in the frequency domain with a finite window size, especially in ASR applications. SPFDBB and FDABB can approximate this multiplication by treating several frames as a block to achieve a better beamforming result. Moreover, FDABB adjusts the number of frames on-line to cope with the variation of characteristics in both speech and interference signals. A better performance was found to be achievable by combining these methods with an ASR mechanism.

A novel divide-by-3 injection-locked frequency divider (ILFD) is proposed. The ILFD circuit is realized with a cross-coupled n-core MOS LC-tank oscillator embedded with a push-push signal generator and two injection MOSFETs for coupling the injection signal into the resonator. The ILFD uses the linear mixer to extend the locking range and has been implemented in a standard 0.18 µm CMOS process. The core power consumption of the ILFD core is 3.12 mW. The divider's free-running frequency is tunable from 4.26 GHz to 4.9 GHz by tuning the varactor's control bias, and at the incident power of 0 dBm the locking range of the ILFD used as a divide-by-3 divider is 1.5 GHz, from 12.5 GHz to 14.0 GHz.

Selective attention mechanism, plays an important role in human visual perception, can be investigated by developing an approach to perceiving the multi-meaningful-dotted-pattern in a color blindness plate (CBP). In this Letter, a perception model driven by a simple active vision mechanism is presented for the image segmentation and understanding of a CBP. Experiments show that to understand one meaningful pattern in an image containing multi-meaningful patterns, the active visual search (i.e., pattern attention) is a very useful function.

In this paper we present hybrid positioning schemes that combine time of arrival (TOA) and angle of arrival (AOA) measurements from only two base stations (BSs) to locate the mobile station (MS) in non-line-of-sight (NLOS) environments. The proposed methods utilize two TOA circles and two AOA lines to find all the possible intersections to locate the MS without requiring a priori information about the NLOS error. The commonly known Taylor series algorithm (TSA) and the hybrid lines of position algorithm (HLOP) have convergence problems, and the relative positioning between the MS and the BSs greatly affects the location accuracy. The resulting geometry creates a situation where small measurement errors can lead to significant errors in the estimated MS location. Simulation results show that the proposed methods always perform better than TSA and HLOP for different levels of NLOS errors, particularly when the MS/BSs have an undesirable geometric layout.

This paper is concerned with synthesizing VLSI array processors from iterative algorithms. Our primary objective is to obtain the highest processor efficiency but not the shortest completion time. Unlike most of the previous work that assumes the index space of the given iterative algorithm to be boundless, the proposed method takes into account the effects of the boundaries of the index space. Due to this consideration, the pseudo-dependence relations are excluded, and most of the independent computations can therefore be uniformly grouped. With the method described in this paper, the index space is partitioned into equal-size blocks and the corresponding computations are systematically and uniformly mapped into processing elements. The synthesized VLSI array processors possess the attractive feature of very high processor efficiency, which, in general, is superior to what is derived from the conventional linear transformation methods.

Reducing communication complexity is a viable approach to multilevel logic synthesis. A communication complexity based approach was proposed previously. In the previous works, only disjoint input decomposition was considered. However, for certain types of circuits, the circuit size can be reduced by using overlapped decomposition. In this paper, we consider overlapped decompositions. Some design issues for overlapped decompositions such as detecting globals" and deriving subfunctions are addressed. Moreover, the Decomposition Don't Cares (DDC) is considered for improving the decomposed results. By using these techniques together, the area and delay of circuits can be further minimized.

The objective of wireless location is to determine the mobile station (MS) location in a wireless cellular communications system. When signals are propagated through non-line-of-sight (NLOS) paths, the measurements at the base stations (BSs) contain large errors which result in poor detectability of an MS by the surrounding BSs. In those situations, it is necessary to integrate all available heterogeneous measurements to improve location accuracy. This paper presents hybrid methods that combine time of arrival (TOA) at three BSs and angle of arrival (AOA) information at the serving BS to obtain a location estimate for the MS. The proposed methods mitigate the NLOS effect by using the weighted sum of the intersections between three TOA circles and the AOA line without requiring the a priori knowledge of NLOS error statistics. Numerical results show that all positioning methods offer improved estimation accuracy over those which rely on the two circles and two lines. The proposed methods always achieve better location accuracy than the Taylor series algorithm (TSA) and the hybrid lines of position algorithm (HLOP) do, regardless of the NLOS error statistics.

Recent top-performing object detectors usually depend on a two-stage approach, which benefits from its region proposal and refining practice but suffers low detection speed. By contrast, one-stage approaches have the advantage of high efficiency while sacrifice their accuracies to some extent. In this paper, we propose a novel single-shot object detection network which inherits the merits of both. Motivated by the idea of semantic enrichment to the convolutional features within a typical deep detector, we propose two novel modules: 1) by modeling the semantic interactions between channels and the long-range dependencies between spatial positions, the self-attending module generates both channel and position attention, and enhance the original convolutional features in a self-guided manner; 2) leveraging the class-discriminative localization ability of classification-trained CNN, the semantic activating module learns a semantic meaningful convolutional response which augments low-level convolutional features with strong class-specific semantic information. The so called self-attending and semantic activating network (ASAN) achieves better accuracy than two-stage methods and is able to fulfil real-time processing. Comprehensive experiments on PASCAL VOC indicates that ASAN achieves state-of-the-art detection performance with high efficiency.

In this paper, we design a new color space YUskin Vskin from YUV color space, based on the principle of skin color with respect to the change of color temperature. Compared with previous work, this color space proved to be the optimal color space for hand segmentation with linear thresholds. We also propose a novel fingertip detection method based on the concomitance between finger and fingernail. The two techniques together improve the performance of hand contour and fingertip extraction in hand gesture recognition.

This paper investigates the wireless information surveillance in a suspicious millimeter wave (mmWave) wireless communication system via the spoofing relay based proactive eavesdropping approach. Specifically, the legitimate monitor in the system acts as a relay to simultaneously eavesdrop and send spoofing signals to vary the source transmission rate. To maximize the effective eavesdropping rate, an optimization problem for both hybrid precoding design and power distribution is formulated. Since the problem is fractional and non-convex, we resort to the Dinkelbach method to equivalently reduce the original problem into a series of non-fractional problems, which is still coupling. Afterwards, based on the BCD-type method, the non-fractional problem is reduced to three subproblems with two introduced parameters. Then the GS-PDD-based algorithm is proposed to obtain the optimal solution by alternately optimizing the three subproblems and simultaneously updating the introduced parameters. Numerical results verify the effectiveness and superiority of our proposed scheme.

The electric properties of a field-effect transistor not only depend on gate surface sidewall but also on channel orientation when applying channel stain engineering. The change of the gate surface and channel orientation through the rotated FinFETs provides the capability to compare the orientation dependence of performance and reliability. This study characterized the <100> and <110> channels of FinFETs on the same wafer under tensile and compressive stresses by cutting the wafer into rectangular silicon pieces and evaluated their piezoresistance coefficients. The piezoresistance coefficients of the <100> and <110> silicon under tensile and compressive stresses were first evaluated based on the current setup. Tensile stresses enhance the mobilities of both <100> and <110> channels, whereas compressive stresses degrade them. Electrical characterization revealed that the threshold voltage variation and drive current degradation of the {100} surface were significantly higher than those of {110} for positive bias temperature instability and hot carrier injection with equal gate and drain voltage (VG=VD). By contrast, insignificant difference is noted for the subthreshold slope degradation. These findings imply that a higher ratio of bulk defect trapping is generated by gate voltage on the <100> surface than that on the <110> surface.

In this paper, a complete SSTA scheme is proposed to calculate the output waveform of a logic cell on any random selected point in the process variational space, or the mean value and variance of the output signal with very high accuracy and acceptable CPU cost. At first, Miller capacitances between the input nodes and internal nodes of a logic cell are introduced to construct the improved MCSM model so as to improve the modeling accuracy. Secondly, the stochastic collocation method jointed with the Modified Nested Sparse Grid technique is adopted for SSTA procedure to avoid the exponential increase of the collocation points number caused by tensor product. Thirdly, a Nominal waveform based Fast Simulation Method is developed to speedup the simulation on each collocation point. At last, Automatic Waveform Construction Technique is developed to construct the output waveform with the approximation points as little as possible to decrease the computational cost while guaranteeing high accuracy. Numerical results are also given to demonstrate the efficiency of the proposed algorithm.

In this letter, a simple dispersion matrix design method for generalized space-time shift keying is presented, in which the dispersion matrices are systematically constructed with cyclic identity matrix, without the need of computer search. The proposed scheme is suitable for any number of transmit antennas greater than two, and can achieve the transmit diversity order of two except two special cases. Simulation results are presented to verify our theoretical analysis and the performance of the proposed scheme.

To achieve more accurate measurements of the mobile station (MS) location, it is possible to integrate many kinds of measurements. In this paper we proposed several simpler methods that utilized time of arrival (TOA) at three base stations (BSs) and the angle of arrival (AOA) information at the serving BS to give location estimation of the MS in non-line-of-sight (NLOS) environments. From the viewpoint of geometric approach, for each a TOA value measured at any BS, one can generate a circle. Rather than applying the nonlinear circular lines of position (LOP), the proposed methods are much easier by using linear LOP to determine the MS. Numerical results demonstrate that the calculation time of using linear LOP is much less than employing circular LOP. Although the location precision of using linear LOP is only reduced slightly. However, the proposed efficient methods by using linear LOP can still provide precise solution of MS location and reduce the computational effort greatly. In addition, the proposed methods with less effort can mitigate the NLOS effect, simply by applying the weighted sum of the intersections between different linear LOP and the AOA line, without requiring priori knowledge of NLOS error statistics. Simulation results show that the proposed methods can always yield superior performance in comparison with Taylor series algorithm (TSA) and the hybrid lines of position algorithm (HLOP).

An artificial transmission line with variable capacitors as its shunt elements, also known as a nonlinear transmission line, can be used to generate pulsed waveforms with short durations. In this work, the variable capacitors are implemented using ferroelectric materials. Analysis and experimental results of such a ferroelectric-based artificial transmission line are presented. The differential equation that describes the nonlinear transmission line is derived and solved. The analytical expression for the solitary waves propagating along the line is found. An artificial transmission line is fabricated using thin-film barium--strontium--titanate capacitors and commercially available chip inductors. The fabrication process of the ferroelectric-based artificial transmission line is described. On-wafer characterization of the line is performed. Measurement results show that, with proper dc bias and substantial input power, a sinusoidal input waveform turns into a bell-shaped pulse train at the output, demonstrating the pulse-shaping capability of the ferroelectric-based artificial transmission line.

MapReduce is commonly used as a parallel massive data processing model. When deploying it as a service over the open systems, the computational integrity of the participants is becoming an important issue due to the untrustworthy workers. Current duplication-based solutions can effectively solve non-collusive attacks, yet most of them require a centralized worker to re-compute additional sampled tasks to defend collusive attacks, which makes the worker a bottleneck. In this paper, we try to explore a trusted worker scheduling framework, named VAWS, to detect collusive attackers and assure the integrity of data processing without extra re-computation. Based on the historical results of verification, we construct an Integrity Attestation Graph (IAG) in VAWS to identify malicious mappers and remove them from the framework. To further improve the efficiency of identification, a verification-couple selection method with the IAG guidance is introduced to detect the potential accomplices of the confirmed malicious worker. We have proven the effectiveness of our proposed method on the improvement of system performance in theoretical analysis. Intensive experiments show the accuracy of VAWS is over 97% and the overhead of computation is closed to the ideal value of 2 with the increasing of the number of map tasks in our scheme.

There are different types of social ties among people, and recognizing specialized types of relationship, such as family or friend, has important significance. It can be applied to personal credit, criminal investigation, anti-terrorism and many other business scenarios. So far, some machine learning algorithms have been used to establish social relationship inferencing models, such as Decision Tree, Support Vector Machine, Naive Bayesian and so on. Although these algorithms discover family members in some context, they still suffer from low accuracy, parameter sensitive, and weak robustness. In this work, we develop a Novel Family Relationship Recognition (NFRR) algorithm on telecom dataset for identifying one's family members from its contact list. In telecom dataset, all attributes are divided into three series, temporal, spatial and behavioral. First, we discover the most probable places of residence and workplace by statistical models, then we aggregate data and select the top-ranked contacts as the user's intimate contacts. Next, we establish Relational Spectrum Matrix (RSM) of each user and its intimate contacts to form communication feature. Then we search the user's nearest neighbors in labelled training set and generate its Specialized Family Spectrum (SFS). Finally, we decide family relationship by comparing the similarity between RSM of intimate contacts and the SFS. We conduct complete experiments to exhibit effectiveness of the proposed algorithm, and experimental results also show that it has a lower complexity.

An efficient algorithm is proposed to identify the active users and extracting their respective timing information in asynchronous direct sequence CDMA (DS-CDMA) communication system over Rayleigh fading channel. The joint identification and timing estimation algorithm is derived by performing discrete Fourier transform (DFT) on the observation vector and exploiting the uniqueness and nullity characteristics of the root-MUSIC test polynomial. The root-MUSIC based algorithm is shown to be asymptotically near-far resistant. Compared to the maximum a posteriori (MAP) or maximum likelihood (ML) based multiuser timing estimator, the complexity is greatly reduced by separating the multi-dimensional optimization problem into several polynomial rooting problems. Moreover, we characterize the dependence of system performance with respect to signature sequence length, number of active users, window size, desired user's signal-to-noise ratio (SNR) and crosscorrelation property of the code structure. The analytical results reveal that under the uncorrelated Rayleigh fading model, the root-MUSIC timing estimator tends to achieve the Cramer-Rao lower bound (CRLB) at interesting signature sequence length and desired user's SNR.