Seongah JEONG Jinkyu KANG Hoojin LEE
In this letter, we investigate tight analytical and asymptotic upper bounds for bit error rate (BER) of constitutional codes over exponentially correlated Nakagami-m fading channels. Specifically, we derive the BER expression depending on an exact closed-form formula for pairwise error event probabilities (PEEP). Moreover, the corresponding asymptotic analysis in high signal-to-noise ratio (SNR) regime is also explored, which is verified via numerical results. This allows us to have explicit insights on the achievable coding gain and diversity order.
Ying JI Yu WANG Jien KATO Kensaku MORI
With the rapid development of multimedia, violent video can be easily accessed in games, movies, websites, and so on. Identifying violent videos and rating violence extent is of great importance to media filtering and children protection. Many previous studies only address the problems of violence scene detection and violent action recognition, yet violence rating problem is still not solved. In this paper, we present a novel video-level rating prediction method to estimate violence extent automatically. It has two main characteristics: (1) a two-stream network is fine-tuned to construct effective representations of violent videos; (2) a violence rating prediction machine is designed to learn the strength relationship among different videos. Furthermore, we present a novel violent video dataset with a total of 1,930 human-involved violent videos designed for violence rating analysis. Each video is annotated with 6 fine-grained objective attributes, which are considered to be closely related to violence extent. The ground-truth of violence rating is given by pairwise comparison method. The dataset is evaluated in both stability and convergence. Experiment results on this dataset demonstrate the effectiveness of our method compared with the state-of-art classification methods.
Tashpolat NIZAMIDIN Li ZHAO Ruiyu LIANG Yue XIE Askar HAMDULLA
As one of the popular topics in the field of human-computer interaction, the Speech Emotion Recognition (SER) aims to classify the emotional tendency from the speakers' utterances. Using the existing deep learning methods, and with a large amount of training data, we can achieve a highly accurate performance result. Unfortunately, it's time consuming and difficult job to build such a huge emotional speech database that can be applicable universally. However, the Siamese Neural Network (SNN), which we discuss in this paper, can yield extremely precise results with just a limited amount of training data through pairwise training which mitigates the impacts of sample deficiency and provides enough iterations. To obtain enough SER training, this study proposes a novel method which uses Siamese Attention-based Long Short-Term Memory Networks. In this framework, we designed two Attention-based Long Short-Term Memory Networks which shares the same weights, and we input frame level acoustic emotional features to the Siamese network rather than utterance level emotional features. The proposed solution has been evaluated on EMODB, ABC and UYGSEDB corpora, and showed significant improvement on SER results, compared to conventional deep learning methods.
Minhwan CHOI Hoojin LEE Haewoon NAM
This letter presents a comprehensive analytical framework for average pairwise error probability (PEP) of decode-and-forward cooperative network based on various distributed space-time block codes (DSTBCs) with antenna switching (DDF-AS) technique over quasi-static Rayleigh fading channels. Utilizing the analytical framework, exact and asymptotic PEP expressions can be effectively formulated, which are based on the Lauricella multiplicative hypergeometric function, when various DSTBCs are adopted for the DDF-AS system. The derived asymptotic PEP formulas and some numerical results enable us to verify that the DDF-AS scheme outperforms the conventional cooperative schemes in terms of error rate performance. Furthermore, the asymptotic PEP formulas can also provide explicit and useful insights into the full diversity transmission achieved by the DDF-AS system.
Donggu KIM Hoojin LEE Joonhyuk KANG
This paper derives highly accurate and effective closed-form formulas for the average upper bound on the pairwise error probability (PEP) of the multi-carrier index keying orthogonal frequency division multiplexing (MCIK-OFDM) system with low-complexity detection (i.e., greedy detection) in two-wave with diffuse power (TWDP) fading channels. To be specific, we utilize an exact moment generating function (MGF) of the signal-to-noise ratio (SNR) under TWDP fading to guarantee highly precise investigations of error probability performance; existing formulas for average PEP employ the approximate probability density function (PDF) of the SNR for TWDP fading, thereby inducing inherent approximation error. Moreover, some special cases of TWDP fading are also considered. To quantitatively reveal the achievable modulation gain and diversity order, we further derive asymptotic formulas for the upper bound on the average PEP. The obtained asymptotic expressions can be used to rapidly estimate the achievable error performance of MCIK-OFDM with the greedy detection over TWDP fading in high SNR regimes.
Sailan WANG Zhenzhi YANG Jin YANG Hongjun WANG
In general, semi-supervised clustering can outperform unsupervised clustering. Since 2001, pairwise constraints for semi-supervised clustering have been an important paradigm in this field. In this paper, we show that pairwise constraints (ECs) can affect the performance of clustering in certain situations and analyze the reasons for this in detail. To overcome these disadvantages, we first outline some exemplars constraints. Based on these constraints, we then describe a semi-supervised clustering framework, and design an exemplars constraints expectation-maximization algorithm. Finally, standard datasets are selected for experiments, and experimental results are presented, which show that the exemplars constraints outperform the corresponding unsupervised clustering and semi-supervised algorithms based on pairwise constraints.
Recently, the Static Heterogeneous Particle Swarm Optimization (SHPSO) has been studied by more and more researchers. In SHPSO, the different search behaviours assigned to particles during initialization do not change during the search process. As a consequence of this, the inappropriate population size of exploratory particles could leave the SHPSO with great difficulties of escaping local optima. This motivated our attempt to improve the performance of SHPSO by introducing the dynamic heterogeneity. The self-adaptive heterogeneity is able to alter its heterogeneous structure according to some events caused by the behaviour of the swarm. The proposed triggering events are confirmed by keeping track of the frequency of the unchanged global best position (pg) for a number of iterations. This information is then used to select a new heterogeneous structure when pg is considered stagnant. According to the different types of heterogeneity, DHPSO-d and DHPSO-p are proposed in this paper. In, particles dynamically use different rules for updating their position when the triggering events are confirmed. In DHPSO-p, a global gbest model and a pairwise connection model are automatically selected by the triggering configuration. In order to investigate the scalability of and DHPSO-p, a series of experiments with four state-of-the-art algorithms are performed on ten well-known optimization problems. The scalability analysis of and DHPSO-p reveals that the dynamic self-adaptive heterogeneous structure is able to address the exploration-exploitation trade-off problem in PSO, and provide the excellent optimal solution of a problem simultaneously.
Chunxiao LIU Guijin WANG Xinggang LIN
Learning an appearance model for person re-identification from multiple images is challenging due to the corrupted images caused by occlusion or false detection. Furthermore, different persons may wear similar clothes, making appearance feature less discriminative. In this paper, we first introduce the concept of multiple instance to handle corrupted images. Then a novel pairwise comparison based multiple instance learning framework is proposed to deal with visual ambiguity, by selecting robust features through pairwise comparison. We demonstrate the effectiveness of our method on two public datasets.
Toru NANBA Tatsuhiro TSUCHIYA Tohru KIKUNO
This letter discusses the applicability of boolean satisfiability (SAT) solving to pairwise testing in practice. Due to its recent rapid advance, using SAT solving seems a promising approach for search-based testing and indeed has already been practiced in test generation for pairwise testing. The previous approaches use SAT solving either for finding a small test set in the absence of parameter constraints or handling constraints, but not for both. This letter proposes an approach that uses a SAT solver for constructing a test set for pairwise testing in the presence of parameter constraints. This allows us to make full use of SAT solving for pairwise testing in practice.
Bin TONG Weifeng JIA Yanli JI Einoshin SUZUKI
We propose a new method, called Subclass-oriented Dimensionality Reduction with Pairwise Constraints (SODRPaC), for dimensionality reduction. In a high dimensional space, it is common that a group of data points with one class may scatter in several different groups. Current linear semi-supervised dimensionality reduction methods would fail to achieve fair performances, as they assume two data points linked by a must-link constraint are close each other, while they are likely to be located in different groups. Inspired by the above observation, we classify the must-link constraint into two categories, which are the inter-subclass must-link constraint and the intra-subclass must-link constraint, respectively. We carefully generate cannot-link constraints by using must-link constraints, and then propose a new discriminant criterion by employing the cannot-link constraints and the compactness of shared nearest neighbors. The manifold regularization is also incorporated in our dimensionality reduction framework. Extensive experiments on both synthetic and practical data sets illustrate the effectiveness of our method.
Xianglan JIN Dong-Sup JIN Jong-Seon NO Dong-Joon SHIN
The probability of making mistakes on the decoded signals at the relay has been used for the maximum-likelihood (ML) decision at the receiver in the decode-and-forward (DF) relay network. It is well known that deriving the probability is relatively easy for the uncoded single-antenna transmission with M-pulse amplitude modulation (PAM). However, in the multiplexing multiple-input multiple-output (MIMO) transmission, the multi-dimensional decision region is getting too complicated to derive the probability. In this paper, a high-performance near-ML decoder is devised by applying a well-known pairwise error probability (PEP) of two paired-signals at the relay in the MIMO DF relay network. It also proves that the near-ML decoder can achieve the maximum diversity of MSMD+MR min (MS,MD), where MS, MR, and MD are the number of antennas at the source, relay, and destination, respectively. The simulation results show that 1) the near-ML decoder achieves the diversity we derived and 2) the bit error probability of the near-ML decoder is almost the same as that of the ML decoder.
Some statistical characteristics, including the means and the cross-correlations, of frequency-selective Rician fading channels seen by orthogonal frequency division multiplexing (OFDM) subcarriers are derived in this paper. Based on a pairwise error probability analysis, the mean vector and the cross-correlation matrix are used to obtain an upper bound of the overall bit-error rate (BER) in a closed-form for coded OFDM signals with and without inter-carrier interference. In this paper, the overall BER is defined as the average BER of OFDM signals of all subcarriers obtained by considering their cross-correlations. Numerical examples are presented to compare the proposed upper bound of the overall BERs and the overall BERs obtained by simulations.
Hisashi KASHIMA Satoshi OYAMA Yoshihiro YAMANISHI Koji TSUDA
Pairwise classification has many applications including network prediction, entity resolution, and collaborative filtering. The pairwise kernel has been proposed for those purposes by several research groups independently, and has been used successfully in several fields. In this paper, we propose an efficient alternative which we call a Cartesian kernel. While the existing pairwise kernel (which we refer to as the Kronecker kernel) can be interpreted as the weighted adjacency matrix of the Kronecker product graph of two graphs, the Cartesian kernel can be interpreted as that of the Cartesian graph, which is more sparse than the Kronecker product graph. We discuss the generalization bounds of the two pairwise kernels by using eigenvalue analysis of the kernel matrices. Also, we consider the N-wise extensions of the two pairwise kernels. Experimental results show the Cartesian kernel is much faster than the Kronecker kernel, and at the same time, competitive with the Kronecker kernel in predictive performance.
Ha X. NGUYEN Ha H. NGUYEN Tho LE-NGOC
A stochastic quasi-gradient algorithm is applied to design linear dispersion (LD) codes for two-way wireless relay networks under Rayleigh fading channels. The codes are designed to minimize an upper bound of the average pairwise error probability. Simulation results show that the codes obtained by the optimization technique achieve a coding gain over codes that are randomly generated based on the isotropic distribution.
Waiyawut SANAYHA Yuttapong RANGSANSERI
In this paper, we propose a novel image projection technique for face recognition applications based on Fisher Linear Discriminant Analysis (LDA). The projection is performed through a couple subspace analysis for overcoming the "small sample size" problem. Also, weighted pairwise discriminant hyperplanes are used in order to provide a more accurate discriminant decision than that produced by the conventional LDA. The proposed technique has been successfully tested on three face databases. Experimental results indicate that the proposed algorithm outperforms the conventional algorithms.
Zhimeng ZHONG Shihua ZHU Gangming LV
In this letter, we analyze the pairwise error probability (PEP) behaviour of distributed space-time code (DSTC) with amplify-and-forward relaying over Nakagami-m multipath channels. An upper bound of PEP for DSTC is derived. From our analysis, it is seen that of the paths from the source to relays and from relays to the destination, those with smaller diversity order result in an overall system performance bottleneck. Numerical examples are provided to corroborate our theoretical analysis.
Boonsarn PITAKDUMRONGKIJA Kazuhiko FUKAWA Hiroshi SUZUKI Takashi HAGIWARA
This paper proposes a new MIMO-OFDM precoding technique that aims to minimize a bit error rate (BER) upper bound of the maximum likelihood detection (MLD) in mobile radio communications. Using a steepest descent algorithm, the proposed method estimates linear precoding matrices that can minimize the upper bound of BER under power constraints. Since the upper bound is derived from all the pairwise error probabilities, this method can effectively optimize overall Euclidean distances between signals received by multiple antennas and their replicas. Computer simulations evaluate the BER performance and channel capacity of the proposed scheme for 22 and 44 MIMO-OFDM systems with BPSK, QPSK, and 16 QAM. It is demonstrated that the proposed precoding technique is superior in terms of average BER to conventional precoding methods including a precoder which maximizes only the minimum Euclidean distance as the worst case.
Jangbok KIM Kyunghee CHOI Gihyun JUNG
This letter proposes a modified Pairwise test case generation algorithm. The proposed algorithm produces additional test cases that may not be covered by Pairwise algorithm due to the dependency between internal modules of software systems. The algorithm produces additional cases utilizing internal module dependencies. The algorithm effectively increases the coverage of testing without significantly increasing the number of test cases.
Weifeng LI Katsunobu ITOU Kazuya TAKEDA Fumitada ITAKURA
We address issues for improving hands-free speech enhancement and speech recognition performance in different car environments using a single distant microphone. This paper describes a new single-channel in-car speech enhancement method that estimates the log spectra of speech at a close-talking microphone based on the nonlinear regression of the log spectra of noisy signal captured by a distant microphone and the estimated noise. The proposed method provides significant overall quality improvements in our subjective evaluation on the regression-enhanced speech, and performed best in most objective measures. Based on our isolated word recognition experiments conducted under 15 real car environments, the proposed adaptive nonlinear regression approach shows an advantage in average relative word error rate (WER) reductions of 50.8% and 13.1%, respectively, compared to original noisy speech and ETSI advanced front-end (ETSI ES 202 050).
This paper investigates zero pronouns in Korean, especially focusing on the center transitions of adjacent utterances under the framework of Centering Theory. Four types of nominal entity (Epair, Einter, Eintra, and Enon) from Centering Theory are defined with the concept of inter-, intra-, and pairwise salience. For each entity type, a case study of zero phenomena is performed through analyzing corpus and building a pronominalization model. This study shows that the zero phenomena of entities which have been neglected in previous Centering works are explained via the center transition of the second previous utterance, and provides valuable results for pronominalization of such entities, such as p2-trans rule. We improve the accuracy of pronominalization model by optimal feature selection and show that our accuracy outperforms the accuracy of previous works.