Attention mechanism, which selectively focuses on source-side information to learn a context vector for generating target words, has been shown to be an effective method for neural machine translation (NMT). In fact, generating target words depends on not only the source-side information but also the target-side information. Although the vanilla NMT can acquire target-side information implicitly by recurrent neural networks (RNN), RNN cannot adequately capture the global relationship between target-side words. To solve this problem, this paper proposes a novel target-attention approach to capture this information, thus enhancing target word predictions in NMT. Specifically, we propose three variants of target-attention model to directly obtain the global relationship among target words: 1) a forward target-attention model that uses a target attention mechanism to incorporate previous historical target words into the prediction of the current target word; 2) a reverse target-attention model that adopts a reverse RNN model to obtain the entire reverse target words information, and then to combine with source context information to generate target sequence; 3) a bidirectional target-attention model that combines the forward target-attention model and reverse target-attention model together, which can make full use of target words to further improve the performance of NMT. Our methods can be integrated into both RNN based NMT and self-attention based NMT, and help NMT get global target-side information to improve translation performance. Experiments on the NIST Chinese-to-English and the WMT English-to-German translation tasks show that the proposed models achieve significant improvements over state-of-the-art baselines.

Compared with acoustic microphone (AM) speech, bone-conducted microphone (BCM) speech is much immune to background noise, but suffers from severe loss of information due to the characteristics of the human-body transmission channel. In this letter, a new method for the speaker-dependent BCM speech enhancement is proposed, in which we focus our attention on the spectra restoration of the distorted speech. In order to better infer the missing components, an attention-based bidirectional Long Short-Term Memory (AB-BLSTM) is designed to optimize the use of contextual information to model the relationship between the spectra of BCM speech and its corresponding clean AM speech. Meanwhile, a structural error metric, Structural SIMilarity (SSIM) metric, originated from image processing is proposed to be the loss function, which provides the constraint of the spectro-temporal structures in recovering of the spectra. Experiments demonstrate that compared with approaches based on conventional DNN and mean square error (MSE), the proposed method can better recover the missing phonemes and obtain spectra with spectro-temporal structure more similar to the target one, which leads to great improvement on objective metrics.

Stance detection in twitter aims at mining user stances expressed in a tweet towards a single or multiple target entities. Detecting and analyzing user stances from massive opinion-oriented twitter posts provide enormous opportunities to journalists, governments, companies, and other organizations. Most of the prior studies have explored the traditional deep learning models, e.g., long short-term memory (LSTM) and gated recurrent unit (GRU) for detecting stance in tweets. However, compared to these traditional approaches, recently proposed densely connected bidirectional LSTM and nested LSTMs architectures effectively address the vanishing-gradient and overfitting problems as well as dealing with long-term dependencies. In this paper, we propose a neural network model that adopts the strengths of these two LSTM variants to learn better long-term dependencies, where each module coupled with an attention mechanism that amplifies the contribution of important elements in the final representation. We also employ a multi-kernel convolution on top of them to extract the higher-level tweet representations. Results of extensive experiments on single and multi-target benchmark stance detection datasets show that our proposed method achieves substantial improvement over the current state-of-the-art deep learning based methods.

The-state-of-the-art neural quality estimation (QE) of machine translation model consists of two sub-networks that are tuned separately, a bidirectional recurrent neural network (RNN) encoder-decoder trained for neural machine translation, called the predictor, and an RNN trained for sentence-level QE tasks, called the estimator. We propose to combine the two sub-networks into a whole neural network, called the unified neural network. When training, the bidirectional RNN encoder-decoder are initialized and pre-trained with the bilingual parallel corpus, and then, the networks are trained jointly to minimize the mean absolute error over the QE training samples. Compared with the predictor and estimator approach, the use of a unified neural network helps to train the parameters of the neural networks that are more suitable for the QE task. Experimental results on the benchmark data set of the WMT17 sentence-level QE shared task show that the proposed unified neural network approach consistently outperforms the predictor and estimator approach and significantly outperforms the other baseline QE approaches.

We consider a physical-layer network coding (PNC) scheme based on M-ary continuous phase frequency shift keying (M-CPFSK) modulation for a bidirectional relay network. In this scheme, the maximum-likelihood sequence detection (MLSD) algorithm for the relay receiver over Rayleigh fading channels is discussed. Moreover, an upper bound on the minimum Euclidean distance for the superimposed signals is analyzed and the corresponding lower bound for the average symbol error rate (SER) at the relay is derived. Numerical results are also sustained by simulations which corroborate the exactness of the theoretical analysis.

Wireless technologies that offer high data rate are generally energy-consuming ones while low-energy technologies commonly provide low data rate. Both kinds of technologies have been integrated in a single mobile device for different services. Therefore, if the service does not always require high data rate, the low energy technology, i.e., Bluetooth, can be used instead of the energy-consuming one, i.e., Wi-Fi, for saving energy. It is obvious that energy savings are maximized by turning the unused technology off. However, when active sessions of ongoing services migrate between different technologies, the network-layer connectivity must be maintained, or a vertical handover (VHO) between different networks is required. Moreover, when the networks are not interconnected, the VHO must be fully controlled by the device itself. The device typically navigates traffic through the firmware of the wireless network interface cards (WNIC) using their drivers, which are dependent on the vendors. To control the traffic navigation between WNICs without any modification of the WNICs' drivers, Software-Defined Networking (SDN) can be applied locally on the mobile device, the so called local SDN. In the local SDN architecture, a local SDN controller (SDNC) is used to control a virtual OpenFlow switch, which turns WNICs into its switch ports. Although the SDNC can navigate the traffic, it lacks the global view of the network topology. Hence, to correctly navigate traffic in a VHO process, an extended SDN controller (extSDNC) was proposed in a previous work. With the extSDNC, the SDNC can perform VHO based on a link layer trigger but with a significant packet loss rate. Therefore, in this paper, a framework named esVHO is proposed that executes VHO at the network layer to reduce the packet loss rate and reduce energy consumption. Experiments on VHO performance prove that esVHO can reduce the packet loss rate considerably. Moreover, the results of an energy saving experiment show that esVHO performs high energy saving up to 4.89 times compared to the others.

This paper addresses power saving for STAs (Wireless Stations) in WLANs (Wireless LANs). Mobile devices are increasingly used in situations in which they access WLANs. However, mobile devices consume large amounts of power when they communicate through a WLAN, and this shortens their battery lifetime. IEEE 802.11 specifies PSM (Power-Saving Mode) as the power-saving method for standard WLANs. However, the sleep conditions specified by PSM for STAs are not optimal in terms of power saving, except when the number of STAs is small, and this increases packet transfer delay. In this paper, we propose a power-saving method in which STAs reduce power consumption by sleeping for a period specified by the NAV (Network Allocation Vector) duration, which is set by an RTS/CTS handshake, and the duration of the NAV is extended by bidirectional burst transmission. To suppress the transfer delay caused by the bidirectional burst transmission, an AP (Access Point) manages the transmission deadline of each downlink packet on the basis of its acceptable value of delay and adapts the number of packets transferred in the bidirectional burst transmission. Although another existing method also uses the NAV duration to manage STA sleeping, the bidirectional burst transmission can only be initiated by the STAs themselves and the NAV is of an extremely limited duration. On the other hand, the proposed method specifies generalized bidirectional burst transmission without the limitations of the transmission initiator and the burst length within acceptable packet transfer delay. Moreover, we investigate the combination of the proposed method with PSM in order to improve the performance in situations in which the number of STAs is small by taking advantage of the combined properties of PSM and the proposed method. The evaluation results demonstrate that these proposed methods can reduce the power consumption of wireless stations and suppress packet transfer delay.

Improvement of conventional networks with an incremental approach is an important design method for the development of the future internet. For this approach, we are developing a future aggregation network based on passive optical network (PON) technology to achieve both cost-effectiveness and high reliability. In this paper, we propose a timeslot (TS) synchronization method for sharing a TS from an optical burst mode transceiver between any route of arbitrary fiber length by changing both the route of the TS transmission and the TS control timing on the optical burst mode transceiver. We show the effectiveness of the proposed method for exchanging TSs in bidirectional bufferless wavelength division multiplexing (WDM) and time division multiplexing (TDM) multi-ring networks under the condition of the occurrence of a link failure through prototype systems. Also, we evaluate the reduction of the required number of optical interfaces in a multi-ring network by applying the proposed method.

In this paper, a non-isolated bidirectional DC-DC converter with zero voltage switching and constant switching frequency is proposed. Unlike the active clamp bidirectional converters, to create soft switching condition in both direction, only one auxiliary switch is used, which reduces conduction losses and the complexity of the circuit. The proposed converter is controlled by pulse width modulation and the switches are gated complementary, thus the implementation of the control circuit is simple. Low switching losses, high efficiency, high power density, are the advantages of this converter. The simulation and experimental results of the converter verify theoretical analysis. Based on an implemented prototype of the proposed converter at 80 watts, the measured efficiency is 96.5%.

The impact of intersymbol interference (ISI) on single carrier frequency domain equalization with multiple input multiple output (MIMO-SCFDE) systems is severe. Most existing channel equalization methods fail to solve it completely. In this paper, given the disadvantages of the error propagation and the gap from matched filter bound (MFB), we creatively introduce a decision feedback equalizer with frequency-domain bidirectional noise prediction (DFE-FDBiNP) to tackle intersymbol interference (ISI) in MIMO-SCFDE systems. The equalizer has two-part equalizer, that is the normal mode and the time-reversal mode decision feedback equalization with noise prediction (DFE-NP). Equal-gain combining is used to realize a greatly simplified and low complexity diversity combining. Analysis and simulation results validate the improved performance of the proposed method in quasi-static frequency-selective fading MIMO channel for a typical urban environment.

The bidirectional DC-DC converters that are used in backup power supplies, energy storage systems, and electric vehicles, are described in this paper, because they have recently attracted a lot of attention. First, this paper shows the main use of the bidirectional DC-DC converter, the optimum circuit topology in accordance with its use, and the characteristic properties of the circuits. In addition, the expected characteristics for the next generations of power semiconductor devices for each bidirectional converter circuit are shown.

This paper introduces open-wire fault-resilient multiple-valued codes for reliable asynchronous point-to-point global communication links. In the proposed encoding, two communication modules assign complementary codewords that change between two valid states without an open-wire fault. Under an open-wire fault, at each module, the codewords don't reach to one of the two valid states and remains as “invalid” states. The detection of the invalid states makes it possible to stop sending wrong codewords caused by an open-wire fault. The detectability of the open-wire fault based on the proposed encoding is proven for m-of-n codes. The proposed code used in the multiple-valued asynchronous global communication link is capable of detecting a single open-wire fault with 3.08-times higher coding efficiency compared with a conventional multiple-valued code used in a triple-modular redundancy (TMR) link that detects an open-wire fault under the same dynamic range of logical values.

NAND multi-level cell (MLC) flash memories are widely used due to low cost and high capacity. However, the increased number of levels in MLC results in larger interference and errors. The errors in MLC flash memories tend to be directional and limited-magnitude. Many related works focus on asymmetric errors, but bidirectional errors also occur because of the bidirectional interference and the adjustment of the hard-decision reference voltages. To take advantage of the characteristics, we propose t bidirectional (lu,ld) limited-magnitude error correction codes, which can reduce errors more effectively. The proposed code is systematic, and can correct t bidirectional errors with upward and downward magnitude of lu and ld, respectively. The proposed method is advantageous in that the parity size is reduced, and it has lower bit error rate than conventional error correction codes with the same code rate.

In this paper, a novel feature named bidirectional local template patterns (B-LTP) is proposed for use in pedestrian detection in still images. B-LTP is a combination and modification of two features, histogram of templates (HOT) and center-symmetric local binary patterns (CS-LBP). For each pixel, B-LTP defines four templates, each of which contains the pixel itself and two neighboring center-symmetric pixels. For each template, it then calculates information from the relationships among these three pixels and from the two directional transitions across these pixels. Moreover, because the feature length of B-LTP is small, it consumes less memory and computational power. Experimental results on an INRIA dataset show that the speed and detection rate of our proposed B-LTP feature outperform those of other features such as histogram of orientated gradient (HOG), HOT, and covariance matrix (COV).

We propose a reduced complexity maximum likelihood sequence detection (MLSD) equalizer for wireless communications using bidirectional decision feedback equalizers (DFEs). We apply reduced-length two-level estimates produced by a bidirectional DFE. Therefore, the computationally expensive MLSD algorithm is applied sparingly for two-level signals with the effective channel length shorter than the original channel, regardless of the original constellation size of the symbols. Simulation results show that the proposed algorithm outperforms existing combination schemes based on bidirectional DFEs, especially for large constellations.

In this paper, a new skin detection method using pixel color and image regional information, intended for objectionable image filtering is proposed. The method consists of three stages: skin detection, feature extraction and image classification. Skin detection is implemented in two steps. First, a Sinc function, fitted to skin color distribution in the Cb-Cr chrominance plane is used for detecting pixels with skin color properties. Next, to benefit regional information, based on the theory of color image reproduction, it's shown that the scattering of skin pixels in the RGB color space can be approximated by an exponential function. This function is incorporated to extract the final accurate skin map of the image. As objectionable image features, new shape and direction features, along with area feature are extracted. Finally, a Multi-Layer Perceptron trained with the best set of input features is used for filtering images. Experimental results on a dataset of 1600 images illustrate that the regional method improves the pixel-based skin detection rate by 10%. The final classification result with 94.12% accuracy showed better results when compared to other methods.

The IEEE 802.15.4 protocol is considered a promising technology for low-cost low-power wireless personal area networks. Researchers have discussed the feasibility of voice communications over IEEE 802.15.4 networks. To this end, the personal area network (PAN) coordinator allocates guaranteed time slots (GTSs) for voice communications in the beacon-enabled mode of IEEE 802.15.4. Although IEEE 802.15.4 is capable of supporting voice communications by GTS allocation, it is impossible to accommodate voice transmission beyond two hops due to the excessive transmission delay. In this paper, we propose a GTS allocation scheme for bidirectional voice traffic in IEEE 802.15.4 multihop networks. The goal of our proposed scheme is to achieve low end-to-end delay and packet drop ratio without a complex allocation algorithm. Thus, the proposed scheme allocates GTSs to devices for successful completion of voice transmission in a superframe duration. The proposed scheme also considers transceiver switching delay. This is relatively large compared to a time slot due to the low-cost and low-gain antenna designs. We analyze and validate the proposed scheme in terms of average end-to-end delay and packet drop ratio. Our scheme has lower end-to-end delay and packet drop ratio than the basic IEEE 802.15.4 GTS allocation scheme.

A bidirectional amplification module has been proposed for use in IP-over-CWDM networks. The module is based on a bidirectional erbium-doped fiber amplifier. The loss compensation characteristics of the module obtained in a bidirectional IP transmission experiment confirmed that the losses of the optical node and the transmission fiber in the network can be compensated for effectively by the module making it possible to increase the number of nodes and the total fiber length of the network.

Provider authentication is necessary in bidirectional broadcasting services, and a digital signature scheme is often used to prevent an adversary from attempting impersonation. The cost of secure signing key management is extremely high. In addition, the key has to be updated very often, since it is frequently used. The result is that the verification key also has to be updated very often, and its redistribution cost is huge. These costs are real and substantive problems, especially when the number of users is large. In this paper, we propose a system that dramatically reduces these costs. In the system, the signing key is updated, but the corresponding verification key does not have to be updated. This means that the signing key can be updated without any cost for redistributing the verification key and that the system is secure against the threat of signing key leakage, since the key can be frequently updated. Moreover, we propose a new key management method that divides a conventional key management server's role into two. The use of a key-insulated signature (KIS) scheme enables low-cost and more secure key management with two servers. Finally, to make a bidirectional broadcasting service more secure even if the signing key is leaked, we developed a new strong KIS scheme. We performed an experiment that assessed the cost of our strong KIS scheme and found that it is sufficiently low. Accordingly, a provider authentication system employing this scheme would be more efficient and would have lower key redistribution and network costs in comparison with conventional authentication systems.

An improved bidirectional search algorithm for computing the weight spectrum of convolutional codes is presented. This algorithm does not employ the column distance function of a code which plays an important role in the original bidirectional search algorithm. We show the proposed algorithm can reduce computaion time for obtaining the weigth spectrum of convolutional codes significantly compared with that of the bidirectional search algorithm.