Full duplex (FD) communication can potentially double the throughput of a point-to-point link in wireless communication. Additionally, FD communication can mitigate the hidden node collision problem. The MAC protocols for FD communications are classified into two types; synchronous FD MAC and asynchronous one. Though the synchronous FD MAC mitigates hidden node collisions by using control frame, overhead duration for each data frame transmission may be a bottleneck for the networks. On the other hand, the asynchronous FD MAC mitigates the hidden node collisions by FD communication. However, it wastes more time due to transmission failure than synchronous FD MAC. Clarifying the effect of two major FD MAC types on networks requires a quantitative evaluation of the effectiveness of these protocols in networks with hidden node collisions. This paper proposes performance analysis of FD MAC protocols for wireless local area networks with hidden node collisions. Through the proposed analytical model, the saturated throughputs in FD WLANs with both asynchronous and synchronous FD MAC for any number of STAs and any payload size can be obtained.

Driver drowsiness estimation is one of the important tasks for preventing car accidents. Most of the approaches are binary classification that classify a driver is significantly drowsy or not. Multi-level drowsiness estimation, that detects not only significant drowsiness but also moderate drowsiness, is helpful to a safer and more comfortable car system. Existing approaches are mostly based on conventional temporal measures which extract temporal information related to eye states, and these measures mainly focus on detecting significant drowsiness for binary classification. For multi-level drowsiness estimation, we propose two temporal measures, average eye closed time (AECT) and soft percentage of eyelid closure (Soft PERCLOS). Existing approaches are also based on a time domain convolutional neural network (CNN) as deep neural network models, of which layers are linked sequentially. The network model extracts features mainly focusing on mono-temporal resolution. We found that features focusing on multi-temporal resolution are effective to multi-level drowsiness estimation, and we propose a parallel linked time-domain CNN to extract the multi-temporal features. We collected an own dataset in a real environment and evaluated the proposed methods with the dataset. Compared with existing temporal measures and network models, Our system outperforms the existing approaches on the dataset.

A pre-trained deep convolutional neural network (DCNN) is adopted as a feature extractor to extract the feature representation of vein images for hand-dorsa vein recognition. In specific, a novel selective deep convolutional feature is proposed to obtain more representative and discriminative feature representation. Extensive experiments on the lab-made database obtain the state-of-the-art recognition result, which demonstrates the effectiveness of the proposed model.

End-to-end (E2E) multilingual automatic speech recognition (ASR) systems aim to recognize multilingual speeches in a unified framework. In the current E2E multilingual ASR framework, the output prediction for a specific language lacks constraints on the output scope of modeling units. In this paper, a language supervision training strategy is proposed with language masks to constrain the neural network output distribution. To simulate the multilingual ASR scenario with unknown language identity information, a language identification (LID) classifier is applied to estimate the language masks. On four Babel corpora, the proposed E2E multilingual ASR system achieved an average absolute word error rate (WER) reduction of 2.6% compared with the multilingual baseline system.

High frame rate and ultra-low delay matching system plays an increasingly important role in human-machine interactions, because it guarantees high-quality experiences for users. Existing image matching algorithms always generate mismatches which heavily weaken the performance the human-machine-interactive systems. Although many mismatch removal algorithms have been proposed, few of them achieve real-time speed with high frame rate and low delay, because of complicated arithmetic operations and iterations. This paper proposes a temporal constraints and block weighting judgement based high frame rate and ultra-low delay mismatch removal system. The proposed method is based on two temporal constraints (proposal #1 and proposal #2) to firstly find some true matches, and uses these true matches to generate block weighting (proposal #3). Proposal #1 finds out some correct matches through checking a triangle route formed by three adjacent frames. Proposal #2 further reduces mismatch risk by adding one more time of matching with opposite matching direction. Finally, proposal #3 distinguishes the unverified matches to be correct or incorrect through weighting of each block. Software experiments show that the proposed mismatch removal system achieves state-of-the-art accuracy in mismatch removal. Hardware experiments indicate that the designed image processing core successfully achieves real-time processing of 784fps VGA (640×480 pixels/frame) video on field programmable gate array (FPGA), with a delay of 0.858 ms/frame.

An offline sensor gain-phase errors calibration method for a linear array using a source in unknown location is proposed. The proposed method is realized through three steps. First, based on the observed covariance matrix, we construct a function related to direction, and it is proved that when the function takes the minimum value, the corresponding value should be the direction of the calibration source. Second, the direction of calibration source is estimated by locating the valley from the constructed function. Third, the gain-phase errors are obtained based on the estimated direction. The proposed method offers a number of advantages. First, the accurate direction measurement of the calibration source is not required. Second, only one calibration source needs to be arranged. Third, it does not require an iterative procedure or a two-dimensional (2D) spectral search. Fourth, the method is applicable to linear arrays, not only to uniform linear arrays (ULAs). Numerical simulations are presented to verify the efficacy of the proposed method.

A unified decision scheme for the classification and localization of cable faults is proposed based on a two-step procedure. Having basis in the time domain reflectometry (TDR), the proposed scheme is capable of determining not only the locations but also types of faults in a cable without an excessive additional computational burden compared to other TDR-based schemes. Results from simulation and experiments with measured real data demonstrate that the proposed scheme exhibits a higher rate of correct decision than the conventional schemes in localizing and classifying faults over a wide range of the location of faults.

This paper proposes a traveling maintenance method based on the resource pool concept, as a new network maintenance model. For failure recovery, the proposed method utilizes permissible time that is ensured by shared resource pools. In the proposed method, even if a failure occurs in a communication facility, maintenance staff wait for occurrence of successive failures in other communication facilities during the permissible time instead of immediately tackling the failure. Then, the maintenance staff successively visit the communication facilities that have faulty devices and collectively repair them. Therefore, the proposed method can reduce the amount of time that the maintenance staff take for fault recovery. Furthermore, this paper provides a system design that optimizes the proposed traveling maintenance according to system requirements determined by the design philosophy of telecommunication networks. Through simulation experiments, we show the effectiveness of the proposed method.

We propose an idea generation support system known as the “GUNGEN-Heartbeat” that uses heartbeat variations for creating high quality ideas during brainstorming. This system shows “An indication of a check list” or “An indication to promote deep breathing” at time beyond a value with variance of heart rates. We also carried out comparison experiments to evaluate the usefulness of the system.

The two issues of art image creation and data hiding are integrated into one and solved by a single approach in this study. An automatic method for generating a new type of computer art, called stained glass image, which imitates the stained-glass window picture, is proposed. The method is based on the use of a tree structure for region growing to construct the art image. Also proposed is a data hiding method which utilizes a general feature of the tree structure, namely, number of tree nodes, to encode the data to be embedded. The method can be modified for uses in three information protection applications, namely, covert communication, watermarking, and image authentication. Besides the artistic stego-image content which may distract the hacker's attention to the hidden data, data security is also considered by randomizing both the input data and the seed locations for region growing, yielding a stego-image which is robust against the hacker's attacks. Good experimental results proving the feasibility of the proposed methods are also included.

In resource-constrained network function virtualization (NFV) environments, it is expected that data throughput for service chains is maintained by using virtual network functions (VNFs) effectively. In this paper, we formulate an optimization problem for maximizing the total data throughput in resource-constrained NFV environments. Moreover, based on our formulated optimization problem, we propose a heuristic service chain construction algorithm for maximizing the total data throughput. This algorithm also determines the placement of VNFs, the amount of resources for each VNF, and the transmission route for each service chain. It is expected that the heuristic algorithm can construct service chains more quickly than the meta-heuristic algorithm. We evaluate the performance of the proposed methods with simulations, and we investigate the effectiveness of our proposed heuristic algorithm through a performance comparison. Numerical examples show that our proposed methods can construct service chains so as to maximize the total data throughput regardless of the number of service chains, the amount of traffic, and network topologies.

The increase in network traffic in recent years has led to increased power consumption. Accordingly, many studies have tried to reduce the energy consumption of network devices. Various types of data have become available in large quantities via large high-speed computer networks. Time-constrained file transfer is receiving much attention as an advanced service. In this model, a request must be completed within a user-specified deadline or rejected if the requested deadline cannot be met. Some bandwidth assignment and routing methods to accept more requests have been proposed. However, these existing methods do not consider energy consumption. Herein, we propose a joint bandwidth assignment and routing method that reduces energy consumption for time-constrained large file transfer. The bandwidth assignment method reduces the power consumption of mediate node, typically router, by waiting for requests and transferring several requests at the same time. The routing method reduces the power consumption by selecting the path with the least predicted energy consumption. Finally, we evaluate the proposed method through simulation experiments.

Communication networks are now an essential infrastructure of society. Many services are constructed across multiple network domains. Therefore, the reliability of multi-domain networks should be evaluated to assess the sustainability of our society, but there is no known method for evaluating it. One reason is the high computation complexity; i.e., network reliability evaluation is known to be #P-complete, which has prevented the reliability evaluation of multi-domain networks. The other reason is intra-domain privacy; i.e., network providers never disclose the internal data required for reliability evaluation. This paper proposes a novel method that computes the lower and upper bounds of reliability in a distributed manner without requiring privacy disclosure. Our method is solidly based on graph theory, and is supported by a simple protocol that secures intra-domain privacy. Experiments on real datasets show that our method can successfully compute the reliability for 14-domain networks in one second. The reliability is bounded with reasonable errors; e.g., bound gaps are less than 0.1% for reliable networks.

In this paper, we propose a salient region detection method with multi-feature fusion and edge constraint. First, an image feature extraction and fusion network based on dense connection structure and multi-channel convolution channel is designed. Then, a multi-scale atrous convolution block is applied to enlarge reception field. Finally, to increase accuracy, a combined loss function including classified loss and edge loss is built for multi-task training. Experimental results verify the effectiveness of the proposed method.

In this paper, we propose a method to reduce the labeling cost while acquiring training data for a malicious domain name detection system using supervised machine learning. In the conventional systems, to train a classifier with high classification accuracy, large quantities of benign and malicious domain names need to be prepared as training data. In general, malicious domain names are observed less frequently than benign domain names. Therefore, it is difficult to acquire a large number of malicious domain names without a dedicated labeling method. We propose a method based on active learning that labels data around the decision boundary of classification, i.e., in the gray area, and we show that the classification accuracy can be improved by using approximately 1% of the training data used by the conventional systems. Another disadvantage of the conventional system is that if the classifier is trained with a small amount of training data, its generalization ability cannot be guaranteed. We propose a method based on ensemble learning that integrates multiple classifiers, and we show that the classification accuracy can be stabilized and improved. The combination of the two methods proposed here allows us to develop a new system for malicious domain name detection with high classification accuracy and generalization ability by labeling a small amount of training data.

Feature learning based on deep network has been verified as beneficial for person re-identification (Re-ID) in recent years. However, most researches use a single network as the baseline, without considering the fusion of different deep features. By analyzing the attention maps of different networks, we find that the information learned by different networks can complement each other. Therefore, a novel Dual Network Fusion (DNF) framework is proposed. DNF is designed with a trunk branch and two auxiliary branches. In the trunk branch, deep features are cascaded directly along the channel direction. One of the auxiliary branch is channel attention branch, which is used to allocate weight for different deep features. Another one is multi-loss training branch. To verify the performance of DNF, we test it on three benchmark datasets, including CUHK03NP, Market-1501 and DukeMTMC-reID. The results show that the effect of using DNF is significantly better than a single network and is comparable to most state-of-the-art methods.

Heterogeneous networks (HetNets) technology is expected to be applied in next generation cellular networks to boost system capacity. However, applying HetNets introduces a significant amount of interference among different tiers within the same cell. In this paper, we propose a weighted rank constrained rank minimization (WRCRM) based interference alignment (IA) approach for three-tier HetNets. The concept of RCRM is applied in a different way to deal with the basic characteristic of different tiers: their different interference tolerance. In the proposed WRCRM approach, interference components at different tiers are weighted with different weighting factors (WFs) to reflect their vulnerability to interference. First, we derive an inner and a loose outer bound on the achievable degrees of freedom (DoF) for the three-tier system that is modeled as a three-user mutually interfering broadcast channel (MIBC). Then, the derived bounds along with the well-known IA feasibility conditions are used to show the effectiveness of the proposed WRCRM approach. Results show that there exist WF values that maximize the achievable interference-free dimensions. Moreover, adjusting the required number of DoF according to the derived bounds improves the performance of the WRCRM approach.

Regarding the enlarged array size for the 5G new radio (NR) millimeter-wave phased-array transceivers, an improved phase tuning resolution will be required to support the accurate beam control. This paper introduces a CMOS implementation of an active vector-summing phase shifter. The proposed phase shifter realizes a 6-bit phase shifting with an active area of 0.32mm2. To minimize the gain variation during the phase tuning, a gain error compensation technique is proposed. After the compensation, the measured gain variation within the 5G NR band n257 is less than 0.9dB. The corresponding RMS gain error is less than 0.2dB. The measured RMS phase error from 26.5GHz to 29.5GHz is less than 1.2°. Gain-invariant, high-resolution phase tuning is realized by this work. Considering the error vector magnitude (EVM) performance, the proposed phase shifter supports a maximum data rate of 11.2Gb/s in 256QAM with a power consumption of 25.2mW.

In this letter, we propose a novel deep domain-adaptive convolutional neural network (DDACNN) model to handle the challenging cross-corpus speech emotion recognition (SER) problem. The framework of the DDACNN model consists of two components: a feature extraction model based on a deep convolutional neural network (DCNN) and a domain-adaptive (DA) layer added in the DCNN utilizing the maximum mean discrepancy (MMD) criterion. We use labeled spectrograms from source speech corpus combined with unlabeled spectrograms from target speech corpus as the input of two classic DCNNs to extract the emotional features of speech, and train the model with a special mixed loss combined with a cross-entrophy loss and an MMD loss. Compared to other classic cross-corpus SER methods, the major advantage of the DDACNN model is that it can extract robust speech features which are time-frequency related by spectrograms and narrow the discrepancies between feature distribution of source corpus and target corpus to get better cross-corpus performance. Through several cross-corpus SER experiments, our DDACNN achieved the state-of-the-art performance on three public emotion speech corpora and is proved to handle the cross-corpus SER problem efficiently.

The surveillance problem is to find optimal trajectories of agents that patrol a given area as evenly as possible. In this paper, we consider multiple agents with fuel constraints. The surveillance area is given by a weighted directed graph, where the weight assigned to each arc corresponds to the fuel consumption/supply. For each node, the penalty to evaluate the unattended time is introduced. Penalties, agents, and fuels are modeled by a mixed logical dynamical system model. Then, the surveillance problem is reduced to a mixed integer linear programming (MILP) problem. Based on the policy of model predictive control, the MILP problem is solved at each discrete time. In this paper, the feasibility condition for the MILP problem is derived. Finally, the proposed method is demonstrated by a numerical example.