Small group detection is still a challenging problem in crowds. Traditional methods use the trajectory information to measure pairwise similarity which is sensitive to the variations of group density and interactive behaviors. In this paper, we propose two types of information by simultaneously incorporating trajectory and interaction information, to detect small groups in crowds. The trajectory information is used to describe the spatial proximity and motion information between trajectories. The interaction information is designed to capture the interactive behaviors from video sequence. To achieve this goal, two classifiers are exploited to discover interpersonal relations. The assumption is that interactive behaviors often occur in group members while there are no interactions between individuals in different groups. The pairwise similarity is enhanced by combining the two types of information. Finally, an efficient clustering approach is used to achieve small group detection. Experiments show that the significant improvement is gained by exploiting the interaction information and the proposed method outperforms the state-of-the-art methods.

Function speculation design with error recovery mechanisms is quite promising due to its high performance and low area overhead. Previous work has focused on two-stage function speculation and thus lacks a systematic way to address the challenge of the multistage function speculation approach. This paper proposes a multistage function speculation with adaptive predictors and applies it in a novel adder. We deduced the analytical performance and area models for the design and validated them in our experiments. Based on those models, a general methodology is presented to guide design optimization. Both analytical proofs and experimental results on the fabricated chips show that the proposed adder's delay and area have a logarithmic and linear relationship with its bit number, respectively. Compared with the DesignWare IP, the proposed adder provides the same performance with 6-17% area reduction under different bit lengths.

In this paper, a novel design of planar dual-band multi-way lossless power dividers (PDs), namely Bagley Polygon PDs, is presented. The proposed PDs use Π-type dual-band transformers as basic elements, whose design formulas are analyzed and simplified to a concise form. The equivalent circuit of the dual-band Bagley Polygon PD is established, based on which design equations are derived mathematically. After that, the design procedure is demonstrated, and special cases are discussed. To verify the validity of the proposed design, 3-way and 5-way examples are simulated and fabricated at two IMT-Advanced bands of 1.8 GHz and 3.5 GHz, then simulation and measurement results are provided. The presented PDs have good performances on the bandwidths and phase shifts. Furthermore, the planar configuration leads to convenient design procedure and easy fabrication.

This paper presents a miniaturized transformer-based ultra-low-power (ULP) LC-VCO with embedded supply pushing reduction techniques for IoT applications in 65-nm CMOS process. To reduce the on-chip area, a compact transformer patterned ground shield (PGS) is implemented. The transistors with switchable capacitor banks and associated components are placed underneath the transformer, which further shrinking the on-chip area. To lower the power consumption of VCO, a gm-stacked LC-VCO using the transformer embedded with PGS is proposed. The transformer is designed to provide large inductance to obtain a robust start-up within limited power consumption. Avoiding implementing an off/on-chip Low-dropout regulator (LDO) which requires additional voltage headroom, a low-power supply pushing reduction feedback loop is integrated to mitigate the current variation and thus the oscillation amplitude and frequency can be stabilized. The proposed ULP TF-based LC-VCO achieves phase noise of -114.8dBc/Hz at 1MHz frequency offset and 16kHz flicker corner with a 103µW power consumption at 2.6GHz oscillation frequency, which corresponds to a -193dBc/Hz VCO figure-of-merit (FoM) and only occupies 0.12mm2 on-chip area. The supply pushing is reduced to 2MHz/V resulting in a -50dBc spur, while 5MHz sinusoidal ripples with 50mVPP are added on the DC supply.

This paper investigates data uploading in cellular networks with the consideration of device-to-device (D2D) communications. A generalized data uploading scheme is proposed by leveraging D2D cooperation among the devices to reduce the data uploading time. In this scheme, we extend the conventional schemes on cooperative D2D data uploading for cellular networks to a more general case, which considers D2D cooperation among both the devices with or without uploading data. To motivate D2D cooperation among all available devices, we organize the devices within communication range by offering them rewards to construct multi-hop D2D chains for data uploading. Specifically, we formulate the problem of chain formation among the devices for data uploading as a coalitional game. Based on merge-and-split rules, we develop a coalition formation algorithm to obtain the solution for the formulated coalitional game with convergence on a stable coalitional structure. Finally, extensive numerical results show the effectiveness of our proposed scheme in reducing the average data uploading time.

Android operating system occupies a high share in the mobile terminal market. It promotes the rapid development of Android applications (apps). However, the emergence of Android malware greatly endangers the security of Android smartphone users. Existing research works have proposed a lot of methods for Android malware detection, but they did not make the utilization of apps' functional category information so that the strong similarity between benign apps in the same functional category is ignored. In this paper, we propose an Android malware detection scheme based on the functional classification. The benign apps in the same functional category are more similar to each other, so we can use less features to detect malware and improve the detection accuracy in the same functional category. The aim of our scheme is to provide an automatic application functional classification method with high accuracy. We design an Android application functional classification method inspired by the hyperlink induced topic search (HITS) algorithm. Using the results of automatic classification, we further design a malware detection method based on app similarity in the same functional category. We use benign apps from the Google Play Store and use malware apps from the Drebin malware set to evaluate our scheme. The experimental results show that our method can effectively improve the accuracy of malware detection.

Searchable symmetric encryption (SSE) enables clients to search encrypted data. Curtmola et al. (ACM CCS 2006) formalized a model and security notions of SSE and proposed two concrete constructions called SSE-1 and SSE-2. After the seminal work by Curtmola et al., SSE becomes an active area of encrypted search. In this paper, we focus on two unnoticed problems in the seminal paper by Curtmola et al. First, we show that SSE-2 does not appropriately implement Curtmola et al.'s construction idea for dummy addition. We refine SSE-2's (and its variants') dummy-adding procedure to keep the number of dummies sufficiently many but as small as possible. We then show how to extend it to the dynamic setting while keeping the dummy-adding procedure work well and implement our scheme to show its practical efficiency. Second, we point out that the SSE-1 can cause a search error when a searched keyword is not contained in any document file stored at a server and show how to fix it.

In this letter, we propose a novel technique to increase the visibility of the hazy image. Benefiting from the atmospheric scattering model and the invariance principle for scene structure, we formulate structure constraint equations that derive from two simulated inputs by performing gamma correction on the input image. Relying on the inherent boundary constraint of the scattering function, the expected scene albedo can be well restored via these constraint equations. Extensive experimental results verify the power of the proposed dehazing technique.

Aiming at the problem of low accuracy of target selection in indirect touch input, an indirect multi-touch input device was designed and built. We explored here four indirect touch input techniques which were TarConstant, TarEnlarge, TarAttract, TarEnlargeAttract, and investigated their performance when subjects completing the target selection tasks through comparative experiments. Results showed that TarEnlargeAttract enabled the shortest movement time along with the lowest error rate, 2349.9ms and 10.9% respectively. In terms of learning effect, both TarAttract and TarEnlargeAttract had learning effect on movement time, which indicated that the speed of these two techniques can be improved with training. Finally, the strategy of improving the accuracy of indirect touch input was given, which has reference significance for the interface design of indirect touch input.

In this letter, a statistical uncorrelated near class discriminant (SUNCD) approach is proposed for face recognition. The optimal discriminant vector obtained by this approach can differentiate one class and its near classes, i.e., its nearest neighbor classes, by constructing the specific between-class and within-class scatter matrices and using the Fisher criterion. In this manner, SUNCD acquires all discriminant vectors class by class. Furthermore, SUNCD makes every discriminant vector satisfy locally statistical uncorrelated constraints by using the corresponding class and part of its most neighboring classes. Experiments on the public AR face database demonstrate that the proposed approach outperforms several representative discriminant methods.

Recognizing the different segments of speech belonging to the same speaker is an important speech analysis task in various applications. Recent works have shown that there was an underlying manifold on which speaker utterances live in the model-parameter space. However, most speaker clustering methods work on the Euclidean space, and hence often fail to discover the intrinsic geometrical structure of the data space and fail to use such kind of features. For this problem, we consider to convert the speaker i-vector representation of utterances in the Euclidean space into a network structure constructed based on the local (k) nearest neighbor relationship of these signals. We then propose an efficient community detection model on the speaker content network for clustering signals. The new model is based on the probabilistic community memberships, and is further refined with the idea that: if two connected nodes have a high similarity, their community membership distributions in the model should be made close. This refinement enhances the local invariance assumption, and thus better respects the structure of the underlying manifold than the existing community detection methods. Some experiments are conducted on graphs built from two Chinese speech databases and a NIST 2008 Speaker Recognition Evaluations (SREs). The results provided the insight into the structure of the speakers present in the data and also confirmed the effectiveness of the proposed new method. Our new method yields better performance compared to with the other state-of-the-art clustering algorithms. Metrics for constructing speaker content graph is also discussed.

Service composition optimization is a classic NP-hard problem. How to quickly select high-quality services that meet user needs from a large number of candidate services is a hot topic in cloud service composition research. An efficient second-order beetle swarm optimization is proposed with a global search ability to solve the problem of cloud service composition optimization in this study. First, the beetle antennae search algorithm is introduced into the modified particle swarm optimization algorithm, initialize the population bying using a chaotic sequence, and the modified nonlinear dynamic trigonometric learning factors are adopted to control the expanding capacity of particles and global convergence capability. Second, modified secondary oscillation factors are incorporated, increasing the search precision of the algorithm and global searching ability. An adaptive step adjustment is utilized to improve the stability of the algorithm. Experimental results founded on a real data set indicated that the proposed global optimization algorithm can solve web service composition optimization problems in a cloud environment. It exhibits excellent global searching ability, has comparatively fast convergence speed, favorable stability, and requires less time cost.

Relation extraction is one of the key basic tasks in natural language processing in which distant supervision is widely used for obtaining large-scale labeled data without expensive labor cost. However, the automatically generated data contains massive noise because of the wrong labeling problem in distant supervision. To address this problem, the existing research work mainly focuses on removing sentence-level noise with various sentence selection strategies, which however could be incompetent for disposing word-level noise. In this paper, we propose a novel neural framework considering both intra-sentence and inter-sentence relevance to deal with word-level and sentence-level noise from distant supervision, which is denoted as Sentence-Related Gated Piecewise Convolutional Neural Networks (SR-GPCNN). Specifically, 1) a gate mechanism with multi-head self-attention is adopted to reduce word-level noise inside sentences; 2) a soft-label strategy is utilized to alleviate wrong-labeling propagation problem; and 3) a sentence-related selection model is designed to filter sentence-level noise further. The extensive experimental results on NYT dataset demonstrate that our approach filters word-level and sentence-level noise effectively, thus significantly outperforms all the baseline models in terms of both AUC and top-n precision metrics.

Battery-powered wireless sensor networks are prone to premature failures because some nodes deplete their batteries more rapidly than others due to workload variations, the many-to-one traffic pattern, and heterogeneous hardware. Most previous sensor network lifetime enhancement techniques focused on balancing the power distribution, assuming the usage of the identical battery. This paper proposes a novel fine-grained cost-constrained lifetime-aware battery allocation solution for sensor networks with arbitrary topologies and heterogeneous power distributions. Based on an energy–cost battery pack model and optimal node partitioning algorithm, a rapid battery pack selection heuristic is developed and its deviation from optimality is quantified. Furthermore, we investigate the impacts of the power variations on the lifetime extension by battery allocation. We prove a theorem to show that power variations of nodes are more likely to reduce the lifetime than to increase it. Experimental results indicate that the proposed technique achieves network lifetime improvements ranging from 4–13 over the uniform battery allocation, with no more than 10 battery pack levels and 2-5 orders of magnitudes speedup compared with a standard integer nonlinear program solver (INLP).

The hybrid precoding (HP) technique has been widely considered as a promising approach for millimeter wave communication systems. In general, the existing HP structure with a complicated high-resolution phase shifter network can achieve near-optimal spectral efficiency, however, it involves high energy consumption. The HP architecture with an energy-efficient switch network can significantly reduce the energy consumption. To achieve maximum energy efficiency, this paper focuses on the HP architecture with switch network and considers a novel adaptive analog network HP structure for such mmWave MIMO systems, which can provide potential array gains. Moreover, a multiuser adaptive coordinate update algorithm is proposed for the HP design problem of this new structure. Simulation results verify that our proposed design can achieve better energy efficiency than other recently proposed HP schemes when the number of users is small.

Efficient extraction of interconnect parasitic parameters has become very important for present deep submicron designs. In this paper, the improved boundary element method (BEM) is presented for 3-D interconnect resistance extraction. The BEM is accelerated by the recently proposed quasi-multiple medium (QMM) technology, which quasi-cuts the calculated region to enlarge the sparsity of the overall coefficient matrix to solve. An un-average quasi-cutting scheme for QMM, advanced nonuniform element partition and technique of employing the linear element for some special surfaces are proposed. These improvements considerably condense the computational resource of the QMM-based BEM without loss of accuracy. Experiments on actual layout cases show that the presented method is several hundred to several thousand times faster than the well-known commercial software Raphael, while preserving the high accuracy.

We investigate an online problem of a robot exploring the outer boundary of an unknown simple polygon P. The robot starts from a specified vertex s and walks an exploration tour outside P. It has to see all points of the polygon's outer boundary and to return to the start. We provide lower and upper bounds on the ratio of the distance traveled by the robot in comparison to the length of the shortest path. We consider P in two scenarios: convex polygon and concave polygon. For the first scenario, we prove a lower bound of 5 and propose a 23.78-competitive strategy. For the second scenario, we prove a lower bound of 5.03 and propose a 26.5-competitive strategy.

This paper presents an efficient yet powerful codebook model, named classified codebook model, to categorize natural scene category. The current codebook model typically resorts to large codebook to obtain higher performance for scene categorization, which severely limits the practical applicability of the model. Our model formulates the codebook model with the theory of vector quantization, and thus uses the famous technique of classified vector quantization for scene-category modeling. The significant feature in our model is that it is beneficial for scene categorization, especially at small codebook size, while saving much computation complexity for quantization. We evaluate the proposed model on a well-known challenging scene dataset: 15 Natural Scenes. The experiments have demonstrated that our model can decrease the computation time for codebook generation. What is more, our model can get better performance for scene categorization, and the gain of performance becomes more pronounced at small codebook size.

A deeply-coupled system can feed the INS information into a GNSS receiver, and the signal tracking precision can be improved under dynamic conditions by reducing tracking loop bandwidth without losing tracking reliability. In contrast to the vector-based deep integration, the scalar-based GNSS/INS deep integration is a relatively simple and practical architecture, in which all individual DLL and PLL are still exist. Since the implementation of a deeply-couple system needs to modify the firmware of a commercial hardware GNSS receiver, very few studies are reported on deep integration based on hardware platform, especially from academic institutions. This implementation-complexity issue has impeded the development of the deeply-coupled GNSS receivers. This paper introduces a scalar-based MEMS IMU/GNSS deeply-coupled system based on an integrated embedded hardware platform for real-time implementation. The design of the deeply-coupled technologies is described including the system architecture, the model of the inertial-aided tracking loop, and the relevant tracking errors analysis. The implementation issues, which include platform structure, real-time optimization, and generation of aiding information, are discussed as well. The performance of the inertial aided tracking loop and the final navigation solution of the developed deeply-coupled system are tested through the dynamic road test scenarios created by a hardware GNSS/INS simulator with GPS L1 C/A signals and low-level MEMS IMU analog signals outputs. The dynamic tests show that the inertial-aided PLL enables a much narrow tracking loop bandwidth (e.g. 3Hz) under dynamic scenarios; while the non-aided loop would lose lock with such narrow loop bandwidth once maneuvering commences. The dynamic zero-baseline tests show that the Doppler observation errors can be reduced by more than 50% with inertial aided tracking loop. The corresponding navigation results also show that the deep integration improved the velocity precision significantly.

Multi-hypothesis prediction technique, which exploits inter-frame correlation efficiently, is widely used in block-based distributed compressive video sensing. To solve the problem of inaccurate prediction in multi-hypothesis prediction technique at a low sampling rate and enhance the reconstruction quality of non-key frames, we present a resample-based hybrid multi-hypothesis scheme for block-based distributed compressive video sensing. The innovations in this paper include: (1) multi-hypothesis reconstruction based on measurements reorganization (MR-MH) which integrates side information into the original measurements; (2) hybrid multi-hypothesis (H-MH) reconstruction which mixes multiple multi-hypothesis reconstructions adaptively by resampling each reconstruction. Experimental results show that the proposed scheme outperforms the state-of-the-art technique at the same low sampling rate.