As the mainstream approach in the field of machine translation, neural machine translation (NMT) has achieved great improvements on many rich-source languages, but performance of NMT for low-resource languages ae not very good yet. This paper uses data enhancement technology to construct Mongolian-Chinese pseudo parallel corpus, so as to improve the translation ability of Mongolian-Chinese translation model. Experiments show that the above methods can improve the translation ability of the translation model. Finally, a translation model trained with large-scale pseudo parallel corpus and integrated with soft context data enhancement technology is obtained, and its BLEU value is 39.3.

In this letter, we propose a deep neural network and semi-supervised learning based dehazing algorithm. The dehazing network uses a pyramidal architecture to recover the haze-free scene from a single hazy image in a coarse-to-fine order. To faithfully restore the objects with different scales, we incorporate cascaded multi-scale convolutional blocks into each level of the pyramid. Feature fusion and transfer in the network are achieved using the paths constructed by interleaved residual connections. For better generalization to the complicated haze in real-world environments, we also devise a discriminator that enables semi-supervised adversarial training. Experimental results demonstrate that the proposed work outperforms comparative ones with higher quantitative metrics and more visually pleasant outputs. It can also enhance the robustness of object detection under haze.

In this letter, the almost binary sequence (sequence with a single zero element) is considered as a special class of binary sequence. Four new bounds on the cross-correlation of balanced (almost) binary sequences with period Q ≡ 1(mod 4) under the precondition of out-of-phase autocorrelation values {-1} or {1, -3} are firstly presented. Then, seven new pairs of balanced (almost) binary sequences of period Q with ideal or optimal autocorrelation values and meeting the lower cross-correlation bounds are proposed by using cyclotomic classes of order 4. These new bounds of (almost) binary sequences with period Q achieve smaller maximum out-of-phase autocorrelation values and cross-correlation values.

The non-orthogonal multiple access (NOMA) approach has been developed in the fifth-generation mobile communication systems (5G) and beyond, to improve the spectrum efficiency and accommodate a large number of IoT devices. Although power domain NOMA is a promising candidate, it is vulnerable to the nonlinearity of RF circuits and cannot achieve high-throughput transmission using high-level modulations in nonlinear environments. This study proposes a novel post-reception nonlinear compensation scheme consisting of two blind nonlinear compensators (BNLCs) and a frequency-domain equalizer (FDE) to reduce the effect of nonlinear distortion. The improvement possible with the proposed scheme is evaluated by using the error vector magnitude (EVM) of the received signal, which is obtained through computer simulations. The simulation results confirm that the proposed scheme can effectively improve the quality of the received downlink power-domain NOMA signal and enable high-throughput transmission under the transmitter (Tx) and receiver (Rx) nonlinearities via a frequency-selective fading channel.

In the fifth-generation mobile communications system (5G), it is critical to ensure wireless security as well as large-capacity and high-speed communication. To achieve this, a chaos modulation method as an encrypted and channel-coded modulation method in the physical layer is proposed. However, in the conventional chaos modulation method, the decoding complexity increases exponentially with respect to the modulation order. To solve this problem, in this study, a hybrid modulation method that applies quadrature amplitude modulation (QAM) and chaos to reduce the amount of decoding complexity, in which some transmission bits are allocated to QAM while maintaining the encryption for all bits is proposed. In the proposed method, a low-complexity decoding method is constructed by ordering chaos and QAM symbols based on the theory of index modulation. Numerical results show that the proposed method maintains good error-rate performance with reduced decoding complexity and ensures wireless security.

Convolutional Neural Network (CNN) has made extraordinary progress in image classification tasks. However, it is less effective to use CNN directly to detect image manipulation. To address this problem, we propose an image filtering layer and a multi-scale feature fusion module which can guide the model more accurately and effectively to perform image manipulation detection. Through a series of experiments, it is shown that our model achieves improvements on image manipulation detection compared with the previous researches.

This letter proposes a post-processing method to improve the smoothness and safety of the path for an autonomous vehicle navigating in an urban environment. The proposed method transforms the initial path given by local path planning algorithms using a stochastic approach to improve its smoothness and safety. Using the proposed method, the initial path is efficiently transformed by iteratively updating the position of each waypoint within it. The proposed method also guarantees the feasibility of the transformed path. Experimental results verify that the proposed method can improve the smoothness and safety of the initial path and ensure the feasibility of the transformed path.

The haze situation will seriously affect the quality of license plate recognition and reduce the performance of the visual processing algorithm. In order to improve the quality of haze pictures, a license plate recognition algorithm based on haze weather is proposed in this paper. The algorithm in this paper mainly consists of two parts: The first part is MPGAN image dehazing, which uses a generative adversarial network to dehaze the image, and combines multi-scale convolution and perceptual loss. Multi-scale convolution is conducive to better feature extraction. The perceptual loss makes up for the shortcoming that the mean square error (MSE) is greatly affected by outliers; the second part is to recognize the license plate, first we use YOLOv3 to locate the license plate, the STN network corrects the license plate, and finally enters the improved LPRNet network to get license plate information. Experimental results show that the dehazing model proposed in this paper achieves good results, and the evaluation indicators PSNR and SSIM are better than other representative algorithms. After comparing the license plate recognition algorithm with the LPRNet algorithm, the average accuracy rate can reach 93.9%.

Peer assessment in education has pedagogical benefits and is a promising method for grading a large number of submissions. At the same time, student reliability has been regarded as a problem; consequently, various methods of estimating highly reliable grades from scores given by multiple students have been proposed. Under most of the existing methods, a nonadaptive allocation pattern, which performs allocation in advance, is assumed. In this study, we analyze the effect of student-submission allocation on score estimation in peer assessment under a nonadaptive allocation setting. We examine three types of nonadaptive allocation methods, random allocation, circular allocation and group allocation, which are considered the commonly used approaches among the existing nonadaptive peer assessment methods. Through simulation experiments, we show that circular allocation and group allocation tend to yield lower accuracy than random allocation. Then, we utilize this result to improve the existing adaptive allocation method, which performs allocation and assessment in parallel and tends to make similar allocation result to circular allocation. We propose the method to replace part of the allocation with random allocation, and show that the method is effective through experiments.

Previous studies on anomaly detection in videos have trained detectors in which reconstruction and prediction tasks are performed on normal data so that frames on which their task performance is low will be detected as anomalies during testing. This paper proposes a new approach that involves sorting video clips, by using a generative network structure. Our approach learns spatial contexts from appearances and temporal contexts from the order relationship of the frames. Experiments were conducted on four datasets, and we categorized the anomalous sequences by appearance and motion. Evaluations were conducted not only on each total dataset but also on each of the categories. Our method improved detection performance on both anomalies with different appearance and different motion from normality. Moreover, combining our approach with a prediction method produced improvements in precision at a high recall.

This paper studies the performance of quantized massive multiple-input multiple-output (MIMO) systems with superimposed pilots (SP), using linear minimum mean-square-error (LMMSE) channel estimation and maximum ratio combining (MRC) detection. In contrast to previous works, arbitrary-bit analog-to-digital converters (ADCs) are considered. We derive an accurate approximation of the uplink achievable rate considering the removal of estimated pilots. Based on the analytical expression, the optimal pilot power factor that maximizes the achievable rate is deduced and an expression for energy efficiency (EE) is given. In addition, the achievable rate and the optimal power allocation policy under some asymptotic limits are analyzed. Analysis shows that the systems with higher-resolution ADCs or larger number of base station (BS) antennas need to allocate more power to pilots. In contrast, more power needs to be allocated to data when the channel is slowly varying. Numerical results show that in the low signal-to-noise ratio (SNR) region, for 1-bit quantizers, SP outperforms time-multiplexed pilots (TP) in most cases, while for systems with higher-resolution ADCs, the SP scheme is suitable for the scenarios with comparatively small number of BS antennas and relatively long channel coherence time.

Software-defined networking (SDN) decouples the control and forwarding of network devices, providing benefits such as simplified control. However, due to cost constraints and other factors, SDN is difficult to fully deploy. It has been proposed that SDN devices can be incrementally deployed in a traditional IP network, i.e., hybrid SDN, to provide partial SDN benefits. Studies have shown that better traffic engineering performance can be achieved by modifying the coverage and placement of SDN devices in hybrid SDN, because they can influence the behavior of legacy switches through certain strategies. However, it is difficult to develop and execute a traffic engineering strategy in hybrid SDN. This article proposes a routing algorithm to achieve approximate load balancing, which minimizes the maximum link utilization by using the optimal solution of linear programming and merging the minimum split traffic flows. A multipath forwarding mechanism under the same problem is designed to optimize transmission time. Experiments show that our algorithm has certain advantages in link utilization and transmission time compared to traditional distributed routing algorithms like OSPF and some hybrid SDN routing mechanisms. Furthermore, our algorithm can approximate the control effect of full SDN when the deployment rate of SDN devices is 40%.

We show that for any convex body Q in the plane, the average distance from the Fermat-Weber center of Q to the points in Q is at least Δ(Q)/6, where Δ(Q) denotes the diameter of Q. Our proof is simple and straightforward, since it needs only elementary calculations. This simplifies a previously known proof that is based on Steiner symmetrizations.

With the high development of computation requirements in Internet of Things, resource-limited edge servers usually require to cooperate to perform the tasks. Most related studies usually assume a static cooperation approach which might not suit the dynamic environment of edge computing. In this paper, we consider a dynamic cooperation approach by guiding edge servers to form coalitions dynamically. It raises two issues: 1) how to guide them to optimally form coalitions and 2) how to cope with the dynamic feature where server statuses dynamically change as the tasks are performed. The coalitional Markov decision process (CMDP) model proposed in our previous work can handle these issues well. However, its basic solution, coalitional Q-learning, cannot handle the large scale problem when the task number is large in edge computing. Our response is to propose a novel algorithm called deep coalitional Q-learning (DCQL) to solve it. To sum up, we first formulate the dynamic cooperation problem of edge servers as a CMDP: each edge server is regarded as an agent and the dynamic process is modeled as a MDP where the agents observe the current state to formulate several coalitions. Each coalition takes an action to impact the environment which correspondingly transfers to the next state to repeat the above process. Then, we propose DCQL which includes a deep neural network and so can well cope with large scale problem. DCQL can guide the edge servers to form coalitions dynamically with the target of optimizing some goal. Furthermore, we run experiments to verify our proposed algorithm's effectiveness in different settings.

Most existing object detection methods and text detection methods are mainly designed to detect either text or objects. In some scenarios where the task is to find the target word pointed-at by an object, results of existing methods are far from satisfying. However, such scenarios happen often in human-computer interaction, when the computer needs to figure out which word the user is pointing at. Comparing with object detection, pointed-at word localization (PAWL) requires higher accuracy, especially in dense text scenarios. Moreover, in printed document, characters are much smaller than those in scene text detection datasets such as ICDAR-2013, ICDAR-2015 and ICPR-2018 etc. To address these problems, the authors propose a novel target word localization network (TWLN) to detect the pointed-at word in printed documents. In this work, a single deep neural network is trained to extract the features of markers and text sequentially. For each image, the location of the marker is predicted firstly, according to the predicted location, a smaller image is cropped from the original image and put into the same network, then the location of pointed-at word is predicted. To train and test the networks, an efficient approach is proposed to generate the dataset from PDF format documents by inserting markers pointing at the words in the documents, which avoids laborious labeling work. Experiments on the proposed dataset demonstrate that TWLN outperforms the compared object detection method and optical character recognition method on every category of targets, especially when the target is a single character that only occupies several pixels in the image. TWLN is also tested with real photographs, and the accuracy shows no significant differences, which proves the validity of the generating method to construct the dataset.

In this paper, we present a scheme to compute either AB or AB2 multiplications over GF(2m) and propose a bit-parallel systolic architecture based on the proposed algorithm. The AB multiplication algorithm is derived in the same form as the formula of AB2 multiplication algorithm, and an architecture that can perform AB multiplication by adding very little extra hardware to AB2 multiplier is designed. Therefore, the proposed architecture can be effectively applied to hardware constrained applications that cannot deploy AB2 multiplier and AB multiplier separately.

In this paper, for improving the robustness of D2D-based SNS by avoiding the cascading failure, we propose an autonomous decentralized friendship management called virtual temporal friendship creation. In our proposed virtual temporal friendship creation, some virtual temporal friendships are created among users based on an optimization problem to improve the robustness although these friendships cannot be used to perform the message exchange in SNS. We investigate the impact of creating a new friendship on the node resilience for the optimization problem. Then we consider an autonomous decentralized algorithm based on the obtained results for the optimization problem of virtual temporal friendship creation. We evaluate the performance of the virtual temporal friendship creation with simulation and investigate the effectiveness of this method by comparing with the performance of a method with meta-heuristic algorithm. From numerical examples, we show that the virtual temporal friendship creation can improve the robustness quickly in an autonomous and decentralized way.

Through the concept of network slicing, a single physical network infrastructure can be split into multiple logically-independent Network Slices (NS), each of which is customized for the needs of its respective individual user or industrial vertical. In the beyond 5G (B5G) system, this customization can be done for many targeted services, including, but not limited to, 5G use cases and beyond 5G. The network slices should be optimized and customized to stitch a suitable environment for targeted industrial services and verticals. This paper proposes a novel Quality of Service (QoS) framework that optimizes and customizes the network slices to ensure the service level agreement (SLA) in terms of end-to-end reliability, delay, and bandwidth communication. The proposed framework makes use of network softwarization technologies, including software-defined networking (SDN) and network function virtualization (NFV), to preserve the SLA and ensure elasticity in managing the NS. This paper also mathematically models the end-to-end network by considering three parts: radio access network (RAN), transport network (TN), and core network (CN). The network is modeled in an abstract manner based on these three parts. Finally, we develop a prototype system to implement these algorithms using the open network operating system (ONOS) as a SDN controller. Simulations are conducted using the Mininet simulator. The results show that our QoS framework and the proposed resource allocation algorithms can effectively schedule network resources for various NS types and provide reliable E2E QoS services to end-users.

Cloud-based Global Navigation Satellite Systems (CB-GNSS) positioning architecture that offloads part of GNSS positioning computation to cloud/edge infrastructure has been studied as an architecture that adds valued functions via the network. The merits of CB-GNSS positioning are that it can take advantage of the abundant computing resources on the cloud/edge to add unique functions to the positioning calculation and reduce the cost of GNSS receiver terminals. An issue in GNSS positioning is the degradation in positioning accuracy in unideal reception environments where open space is limited and some satellite signals are blocked. To resolve this issue, we propose a satellite selection algorithm that effectively removes the multipath components of blocked satellite signals, which are the main cause of drop in positioning accuracy. We build a Proof of Concept (PoC) test environment of CB-GNSS positioning architecture implementing the proposed satellite selection algorithm and conduct experiments to verify its positioning performance in unideal static and dynamic conditions. For static long-term positioning in a multipath signal reception environment, we found that CB-GNSS positioning with the proposed algorithm enables a low-end GNSS receiver terminal to match the positioning performance comparable to high-end GNSS receiver terminals in terms of the FIX rate. In an autonomous tractor driving experiment on a farm road crossing a windbreak, we succeeded in controlling the tractor's autonomous movement by maintaining highly precise positioning even in the windbreak. These results indicates that the proposed satellite selection algorithm achieves high positioning performance even in poor satellite signal reception environments.

A triple loss based framework for generalized zero-shot learning is presented in this letter. The approach learns a shared latent space for image features and attributes by using aligned variational autoencoders and variants of triplet loss. Then we train a classifier in the latent space. The experimental results demonstrate that the proposed framework achieves great improvement.