This paper proposes a natural facial and head behavior recognition method using hybrid dynamical systems. Most existing facial and head behavior recognition methods focus on analyzing deliberately displayed prototypical emotion patterns rather than complex and spontaneous facial and head behaviors in natural conversation environments. We first capture spatio-temporal features on important facial parts via dense feature extraction. Next, we cluster the spatio-temporal features using hybrid dynamical systems, and construct a dictionary of motion primitives to cover all possible elemental motion dynamics accounting for facial and head behaviors. With this dictionary, the facial and head behavior can be interpreted into a distribution of motion primitives. This interpretation is robust against different rhythms of dynamic patterns in complex and spontaneous facial and head behaviors. We evaluate the proposed approach under natural tele-communication scenarios, and achieve promising results. Furthermore, the proposed method also performs favorably against the state-of-the-art methods on three benchmark databases.

Fault localization is essential for solving the issue of software faults. Aiming at improving fault localization, this paper proposes a deep learning-based fault localization with contextual information. Specifically, our approach uses deep neural network to construct a suspiciousness evaluation model to evaluate the suspiciousness of a statement being faulty, and then leverages dynamic backward slicing to extract contextual information. The empirical results show that our approach significantly outperforms the state-of-the-art technique Dstar.

The exchanged hypercube, denoted by EH(s,t), is a graph obtained by systematically removing edges from the corresponding hypercube, while preserving many of the hypercube's attractive properties. Moreover, ring-connected topology is one of the most promising topologies in Wavelength Division Multiplexing (WDM) optical networks. Let Rn denote a ring-connected topology. In this paper, we address the routing and wavelength assignment problem for implementing the EH(s,t) communication pattern on Rn, where n=s+t+1. We design an embedding scheme. Based on the embedding scheme, a near-optimal wavelength assignment algorithm using 2s+t-2+⌊2t/3⌋ wavelengths is proposed. We also show that the wavelength assignment algorithm uses no more than an additional 25 percent of (or ⌊2t-1/3⌋) wavelengths, compared to the optimal wavelength assignment algorithm.

In this paper, we propose a novel error recovery method for massive multiple-input multiple-output (MIMO) signal detection, which improves an estimate of transmitted signals by taking advantage of the sparsity and the discreteness of the error signal. We firstly formulate the error recovery problem as the maximum a posteriori (MAP) estimation and then relax the MAP estimation into a convex optimization problem, which reconstructs a discrete-valued sparse vector from its linear measurements. By using the restricted isometry property (RIP), we also provide a theoretical upper bound of the size of the reconstruction error with the optimization problem. Simulation results show that the proposed error recovery method has better bit error rate (BER) performance than that of the conventional error recovery method.

This paper presents an extremely low-profile front-end configuration for a base station at quasi-millimeter wave band. It consists of integrated modules of patch antennas and substrate integrated waveguide filters using two printed circuit boards, and transmitter modules using compact GaAs pHEMT three-dimensional monolithic millimeter-wave integrated circuits. The transmitter modules are located around the integrated modules. This is because the proposed front-end configuration can attain extremely low profile, and band-pass filtering performance at quasi-millimeter wave band. As a demonstration of the proposed configuration, 26-GHz-band 4-by-4 elements front-end module is fabricated and tested. The fabricated module has the thickness of about 1 cm, while that offers the attenuation of more than 30 dB with 2 GHz offset from 26 GHz. The proposed configuration can provide base station that can be effective in offering sub-millimeter wave and millimeter-wave bands broadband services for 5G mobile communications systems.

By using Password-based Authenticated Key Exchange (PAKE), a server can authenticate a user who has only the same password shared with the server in advance and establish a session key with the user simultaneously. However, in the real applications, we may have a situation where a user needs to share a session key with server A, but the authentication needs to be done by a different server B that shares the password with the user. Further, to achieve higher security on the server side, it may be required to make PAKE tolerant of a server breach by having multiple authentication servers. To deal with such a situation, Abdalla et al. proposed a variant of PAKE called Gateway Threshold PAKE (GTPAKE) where a gateway corresponds to the aforementioned server A being an on-line service provider and also a potential adversary that may try to guess the passwords. However, the schemes of Abdalla et al. turned out to be vulnerable to Undetectable On-line Dictionary Attack (UDonDA). In this paper, we propose the first GTPAKE provably secure against UDonDA, and in the security analysis, we prove that our GTPAKE is secure even if an adversary breaks into parts of multiple authentication servers.

This letter treats the problem of lossless fixed-to-variable length source coding in moderate deviation regime. We investigate the behavior of the overflow probability of the Bayes code. Our result clarifies that the behavior of the overflow probability of the Bayes code is similar to that of the optimal non-universal code for i.i.d. sources.

In this letter, three constructions of perfect Gaussian integer sequences are constructed based on cyclic difference sets. Sufficient conditions for constructing perfect Gaussian integer sequences are given. Compared with the constructions given by Chen et al. [12], the proposed constructions relax the restrictions on the parameters of the cyclic difference sets, and new perfect Gaussian integer sequences will be obtained.

As device sizes are downscaled to nanometer, Random Telegraph Noise (RTN) becomes dominant. It is indispensable to accurately estimate the effect of RTN. We propose an RTN simulation method for analog circuits. It is based on the charge trapping model. The RTN-induced threshold voltage fluctuation are replicated to attach a variable DC voltage source to the gate of a MOSFET by using Verilog-AMS. In recent deca-nanometer processes, high-k (HK) materials are used in gate dielectrics to decrease the leakage current. We must consider the defect distribution characteristics both in HK and interface layer (IL). This RTN model can be applied to the bimodal model which includes characteristics of the HK and IL dielectrics. We confirm that the drain current of MOSFETs temporally fluctuates in circuit-level simulations. The fluctuations of RTN are different in MOSFETs. RTN affects the frequency characteristics of ring oscillators (ROs). The distribution of RTN-induced frequency fluctuations has a long-tail in a HK process. The RTN model applied to the bimodal can replicate a long-tail distribution. Our proposed method can estimate the temporal impact of RTN including multiple transistors.

In this work, two new proposed NMF models are developed for facial expression recognition. They are called maximum volume constrained nonnegative matrix factorization (MV_NMF) and maximum volume constrained graph nonnegative matrix factorization (MV_GNMF). They achieve sparseness from a larger simplicial cone constraint and the extracted features preserve the topological structure of the original images.

Multivariate Public Key Cryptography (MPKC) is one of the main candidates for secure communication in a post-quantum era. Recently, Yasuda and Sakurai proposed at ICICS 2015 a new multivariate encryption scheme called SRP, which offers efficient decryption, a small blow up factor between plaintext and ciphertext and resists all known attacks against multivariate schemes. However, similar to other MPKC schemes, the key sizes of SRP are quite large. In this paper we propose a technique to reduce the key size of the SRP scheme, which enables us to reduce the size of the public key by up to 54%. Furthermore, we can use the additional structure in the public key polynomials to speed up the encryption process of the scheme by up to 50%. We show by experiments that our modifications do not weaken the security of the scheme.

Secret sharing is a method to protect information for security. The information is divided into n shares, and the information is reconstructed from any k shares but no knowledge of it is revealed from k-1 shares. Physical layer security is a method to yield a favorable receive condition to an authorized destination terminal in wireless communications based on multi-antenna transmission. In this study, we propose wireless packet communications protected by the secret sharing based on Reed Solomon coding and the physical layer security based on vector coding, which implements a single-antenna system and a multi-antenna system. Evaluation results show the validity of the proposed scheme.

Radio signals show characteristics of minute differences, which result from various idiosyncratic hardware properties between different radio emitters. A robust detector based on exponentially weighted distances is proposed to detect the exact reference instants of the burst communication signals. Based on the exact detection of the reference instant, in which the radio emitter finishes the power-up ramp and enters the first symbol of its preamble, the features of the radio fingerprint can be extracted from the transient signal section and the steady-state signal section for radiometric identification. Experiments on real data sets demonstrate that the proposed method not only has a higher accuracy that outperforms correlation-based detection, but also a better robustness against noise. The comparison results of different detectors for radiometric identification indicate that the proposed detector can improve the classification accuracy of radiometric identification.

In this paper, we propose filter-and-forward beamforming (FF-BF) for cognitive two-way relay networks in which secondary users employ an orthogonal frequency-division multiplexing (OFDM) system. Secondary transceivers communicate with each other through multiple relays to obtain BF gain as well as to suppress the interference between the primary and secondary users who share the same spectrum. We consider two FF-BF design methods to optimize the relay filter. The first method enhances the quality of service of the secondary network by maximizing the worst subcarrier signal-to-interference-plus-noise ratio (SINR) subject to transmit power constraints. The second method suppresses the interference from the secondary network to the primary network through the minimization of the relay transmission power subject to subcarrier SINR constraints. Simulation results show that the proposed FF-BF improves system performance in comparison to amplify-and-forward relay BF.

We have demonstrated crosstalk mitigation in single-mode MCFs using optical space coding. Four types of single-mode multicore fiber (MCF) models were evaluated by our scheme with the modified prime code and differential detection. Typically, intercore crosstalk was improved by 7-20 dB in 9-core fibers with an original crosstalk of 10-20 dB.

Many real world optimization problems involving sets of agents can be modeled as Distributed Constraint Optimization Problems (DCOPs). A DCOP is defined as a set of variables taking values from finite domains, and a set of constraints that yield costs based on the variables' values. Agents are in charge of the variables and must communicate to find a solution minimizing the sum of costs over all constraints. Many applications of DCOPs include multiple criteria. For example, mobile sensor networks must optimize the quality of the measurements and the quality of communication between the agents. This introduces trade-offs between solutions that are compared using the concept of Pareto dominance. Multi-Objective Distributed Constraint Optimization Problems (MO-DCOPs) are used to model such problems where the goal is to find the set of Pareto optimal solutions. This set being exponential in the number of variables, it is important to consider fast approximation algorithms for MO-DCOPs. The bounded multi-objective max-sum (B-MOMS) algorithm is the first and only existing approximation algorithm for MO-DCOPs and is suited for solving a less-constrained problem. In this paper, we propose a novel approximation MO-DCOP algorithm called Distributed Pareto Local Search (DPLS) that uses a local search approach to find an approximation of the set of Pareto optimal solutions. DPLS provides a distributed version of an existing centralized algorithm by complying with the communication limitations and the privacy concerns of multi-agent systems. Experiments on a multi-objective extension of the graph-coloring problem show that DPLS finds significantly better solutions than B-MOMS for problems with medium to high constraint density while requiring a similar runtime.

Nonnegative Matrix Factorization (NMF) with sparseness and smoothness constraints has attracted increasing attention. When these properties are considered, NMF is usually formulated as an optimization problem in which a linear combination of an approximation error term and some regularization terms must be minimized under the constraint that the factor matrices are nonnegative. In this paper, we focus our attention on the error measure based on the Euclidean distance and propose a new iterative method for solving those optimization problems. The proposed method is based on the Hierarchical Alternating Least Squares (HALS) algorithm developed by Cichocki et al. We first present an example to show that the original HALS algorithm can increase the objective value. We then propose a new algorithm called the Gauss-Seidel HALS algorithm that decreases the objective value monotonically. We also prove that it has the global convergence property in the sense of Zangwill. We finally verify the effectiveness of the proposed algorithm through numerical experiments using synthetic and real data.

The main contribution of this paper is to present an efficient GPU implementation of bulk computation of the CKY parsing for a context-free grammar, which determines if a context-free grammar derives each of a lot of input strings. The bulk computation is to execute the same algorithm for a lot of inputs in turn or at the same time. The CKY parsing is to determine if a context-free grammar derives a given string. We show that the bulk computation of the CKY parsing can be implemented in the GPU efficiently using Bitwise Parallel Bulk Computation (BPBC) technique. We also show the rule minimization technique and the dynamic scheduling method for further acceleration of the CKY parsing on the GPU. The experimental results using NVIDIA TITAN X GPU show that our implementation of the bitwise-parallel CKY parsing for strings of length 32 takes 395µs per string with 131072 production rules for 512 non-terminal symbols.

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.

We consider the intrinsic randomness problem for correlated sources. Specifically, there are three correlated sources, and we want to extract two mutually independent random numbers by using two separate mappings, where each mapping converts one of the output sequences from two correlated sources into a random number. In addition, we assume that the obtained pair of random numbers is also independent of the output sequence from the third source. We first show the δ-achievable rate region where a rate pair of two mappings must satisfy in order to obtain the approximation error within δ ∈ [0,1), and the second-order achievable rate region for correlated general sources. Then, we apply our results to non-mixed and mixed independently and identically distributed (i.i.d.) correlated sources, and reveal that the second-order achievable rate region for these sources can be represented in terms of the sum of normal distributions.