Fine-grained image classification is one of the key basic tasks of computer vision. The appearance of traditional deep convolutional neural network (DCNN) combined with attention mechanism can focus on partial and local features of fine-grained images, but it still lacks the consideration of the embedding mode of different attention modules in the network, leading to the unsatisfactory result of classification model. To solve the above problems, three different attention mechanisms are introduced into the DCNN network (like ResNet, VGGNet, etc.), including SE, CBAM and ECA modules, so that DCNN could better focus on the key local features of salient regions in the image. At the same time, we adopt three different embedding modes of attention modules, including serial, residual and parallel modes, to further improve the performance of the classification model. The experimental results show that the three attention modules combined with three different embedding modes can improve the performance of DCNN network effectively. Moreover, compared with SE and ECA, CBAM has stronger feature extraction capability. Among them, the parallelly embedded CBAM can make the local information paid attention to by DCNN richer and more accurate, and bring the optimal effect for DCNN, which is 1.98% and 1.57% higher than that of original VGG16 and Resnet34 in CUB-200-2011 dataset, respectively. The visualization analysis also indicates that the attention modules can be easily embedded into DCNN networks, especially in the parallel mode, with stronger generality and universality.

Facial Beauty Prediction (FBP) is a significant pattern recognition task that aims to achieve consistent facial attractiveness assessment with human perception. Currently, Convolutional Neural Networks (CNNs) have become the mainstream method for FBP. The training objective of most conventional CNNs is usually to learn static convolution kernels, which, however, makes the network quite difficult to capture global attentive information, and thus usually ignores the key facial regions, e.g., eyes, and nose. To tackle this problem, we devise a new convolution manner, Dynamic Attentive Convolution (DyAttenConv), which integrates the dynamic and attention mechanism into convolution in kernel-level, with the aim of enforcing the convolution kernels adapted to each face dynamically. DyAttenConv is a plug-and-play module that can be flexibly combined with existing CNN architectures, making the acquisition of the beauty-related features more globally and attentively. Extensive ablation studies show that our method is superior to other fusion and attention mechanisms, and the comparison with other state-of-the-arts also demonstrates the effectiveness of DyAttenConv on facial beauty prediction task.

This letter proposes a gradient-enhanced softmax supervisor for face recognition (FR) based on a deep convolutional neural network (DCNN). The proposed supervisor conducts the constant-normalized cosine to obtain the score for each class using a combination of the intra-class score and the soft maximum of the inter-class scores as the objective function. This mitigates the vanishing gradient problem in the conventional softmax classifier. The experiments on the public Labeled Faces in the Wild (LFW) database denote that the proposed supervisor achieves better results when compared with those achieved using the current state-of-the-art softmax-based approaches for FR.

In this paper, we propose a decentralized strategy to find out the linear precoding matrices for a two-hop multiuser relay communication system. From a game-theoretic perspective, we model the source allocation process as a strategic noncooperative game for fixing relay precoding matrix and the multiuser interference treating as additive colored noise. Alternately, from the global optimization perspective, we prove that the optimum relay precoding matrix follows the transceiver Winner filter structure for giving a set of source transmit matrices. Closed-form solutions are finally obtained by using our proposed joint iterative SMSE algorithm and numerical results are provided to give insights on the proposed algorithms.

The shift-and-add property (SAP) of a p-ary m-sequence {ak} with period N=pn-1 means that this sequence satisfies the equation {ak+η}+{ak+τ}={ak+λ} for some integers η, τ and λ. For an arbitrarily-given p-ary m-sequence {ak}, we develop an algebraic approach to determine the integer λ for the arbitrarily-given integers η and τ. And all trinomials can be given. Our calculation only depends on the reciprocal polynomial of the primitive polynomial which produces the given m-sequence {ak}, and the cyclotomic cosets mod pn-1.

Sorting is one of the most fundamental problems in mathematics and computer science. Because high-throughput and flexible sorting is a key requirement in modern databases, this paper presents efficient techniques for designing a high-throughput sorting matrix that supports continuous data sequences. There have been numerous studies on the optimization of sorting circuits on FPGA (field-programmable gate array) platforms. These studies focused on attaining high throughput for a single command with fixed data width. However, the architectures proposed do not meet the requirement of diversity for database data types. A sorting matrix architecture is thus proposed to overcome this problem. Our design consists of a matrix of identical basic sorting cells. The sorting cells work in a pipeline and in parallel, and the matrix can simultaneously process multiple data streams, which can be combined into a high-width single-channel data stream or low-width multiple-channel data streams. It can handle continuous sequences and allows for sorting variable-length data sequences. Its maximum throughput is approximately 1.4 GB/s for 32-bit sequences and approximately 2.5 GB/s for 64-bit sequences on our platform.

This letter considers a dual-hop multiuser MIMO amplify-and-forward relay broadcast system with multi-antenna nodes. A unified scheme is addressed to jointly optimize the linear transceiver based on the sum mean-square error (MSE) and the sum rate criterion. The solutions are iteratively obtained by deriving the gradients of the objective functions for a gradient descent algorithm. Simulation results demonstrate the performance improvements in terms of the BER and the sum rate.

A method for constructing low-density convolutional (LDC) codes with the degree distribution optimized for block low-density parity-check (LDPC) codes is presented. If the degree distribution is irregular, the constructed LDC codes are also irregular. In this letter we give the encoding and decoding method for LDC codes, and study how to avoid the short cycles of LDC codes. Some simulation results are also presented.

Air Traffic Control (ATC) communication suffers from issues such as high electromagnetic interference, fast speech rate, and low intelligibility, which pose challenges for downstream tasks like Automatic Speech Recognition (ASR). This article aims to research how to enhance the audio quality and intelligibility of civil aviation speech through speech enhancement methods, thereby improving the accuracy of speech recognition and providing support for the digitalization of civil aviation. We propose a speech enhancement model called DIUnet_V (DenseNet & Inception & U-Net & Volume) that combines both time-frequency and time-domain methods to effectively handle the specific characteristics of civil aviation speech, such as predominant electromagnetic interference and fast speech rate. For model evaluation, we assess the denoising and enhancement effects using three metrics: Signal-to-Noise Ratio (SNR), Mean Opinion Score (MOS), and speech recognition error rate. On a simulated ATC training recording dataset, DIUnet_Volume10 achieved an SNR value of 7.3861, showing a 4.5663 improvement compared to the original U-net model. To address the challenge of the absence of clean speech in the ATC working environment, which makes it difficult to accurately calculate SNR, we propose evaluating the denoising effects indirectly based on the recognition performance of an ATC speech recognition system. On a real ATC speech dataset, the average word error rate decreased by 1.79% absolute and the average sentence error rate decreased by 3% absolute for DIUnet_V processed speech compared to the unprocessed speech in the built speech recognition system.

The purpose of Facial Beauty Prediction (FBP) is to automatically assess facial attractiveness based on human aesthetics. Most neural network-based prediction methods do not consider the ranking information in the task. For scoring tasks like facial beauty prediction, there is abundant ranking information both between images and within images. Reasonable utilization of these information during training can greatly improve the performance of the model. In this paper, we propose a novel end-to-end Convolutional Neural Network (CNN) model based on ranking information of images, incorporating a Rank Module and an Adaptive Weight Module. We also design pairwise ranking loss functions to fully leverage the ranking information of images. Considering training efficiency and model inference capability, we choose ResNet-50 as the backbone network. We conduct experiments on the SCUT-FBP5500 dataset and the results show that our model achieves a new state-of-the-art performance. Furthermore, ablation experiments show that our approach greatly contributes to improving the model performance. Finally, the Rank Module with the corresponding ranking loss is plug-and-play and can be extended to any CNN model and any task with ranking information. Code is available at https://github.com/nehcoah/Rank-Info-Net.

A low-complexity signal detection approach based on the Kaczmarz algorithm (KA) is proposed to iteratively realize minimum mean square error (MMSE) detection for uplink massive multiple-input multiple-output (MIMO) systems. While KA is used for straightforward matrix inversion, the MMSE detection requires the computation of the Gram matrix with high complexity. In order to avoid the Gram matrix computation, an equivalent augmented matrix is applied to KA-based MMSE detection. Moreover, promising initial estimation and an approximate method to compute soft-output information are utilized to further accelerate the convergence rate and reduce the complexity. Simulation results demonstrate that the proposed approach outperforms the recently proposed Neumann series, conjugate gradient, and Gauss-Seidel methods in complexity and error-rate performance. Meanwhile, the FPGA implementation results confirm that our proposed method can efficiently compute the approximate inverse with low complexity.

Bistatic multi-input multi-output (MIMO) radar has the capability of measuring the transmit angle from the receiving array, which means the existence of information redundancy and benefits data association. In this paper, a data association decision for bistatic MIMO radar is proposed and the performance advantages of bistatic MIMO radar in data association is analyzed and evaluated. First, the parameters obtained by receiving array are sent to the association center via coordinate conversion. Second, referencing the nearest neighbor association (NN) algorithm, an improved association decision is proposed with the transmit angle and target range as association statistics. This method can evade the adverse effects of the angle system errors to data association. Finally, data association probability in the presence of array directional error is derived and the correctness of derivation result is testified via Monte Carlo simulation experiments. Besides that performance comparison with the conventional phased array radar verifies the excellent performance of bistatic MIMO Radar in data association.

An earthquake is a destructive natural disaster, which cannot be predicted accurately and causes devastating damage and losses. In fact, many of the damages can be prevented if people know what to do during and after earthquakes. Earthquake education is the most important method to raise public awareness and mitigate the damage caused by earthquakes. Generally, earthquake education consists of conducting traditional earthquake drills in schools or communities and experiencing an earthquake through the use of an earthquake simulator. However, these approaches are unrealistic or expensive to apply, especially in underdeveloped areas where earthquakes occur frequently. In this paper, an earthquake drill simulation system based on virtual reality (VR) technology is proposed. A User is immersed in a 3D virtual earthquake environment through a head mounted display and is able to control the avatar in a virtual scene via Kinect to respond to the simulated earthquake environment generated by SIGVerse, a simulation platform. It is a cost effective solution and is easy to deploy. The design and implementation of this VR system is proposed and a dormitory earthquake simulation is conducted. Results show that powerful earthquakes can be simulated successfully and the VR technology can be applied in the earthquake drills.

The major weakness of global navigation satellite system receivers is their vulnerability to intentional and unintentional interference. Frequency domain interference suppression (FDIS) technology is one of the most useful countermeasures. The pseudo-range measurement is unbiased after FDIS filtering given an ideal analog channel. However, with the influence of the analog modules used in RF front-end, the amplitude response and phase response of the channel equivalent filter are non-ideal, which bias the pseudo-range measurement after FDIS filtering and the bias varies along with the frequency of the interference. This paper proposes an unbiased interference suppression method based on signal estimation and spectrum compensation. The core idea is to use the parameters calculated from the tracking loop to estimate and reconstruct the desired signal. The estimated signal is filtered by the equivalent filter of actual channel, then it is used for compensating the spectrum loss caused by the FDIS method in the frequency domain. Simulations show that the proposed algorithm can reduce the pseudo-range measurement bias significantly, even for channels with asymmetrical group delay and multiple interference sources at any location.

Root cause analysis of BGP updates is the key to debug and troubleshoot BGP routing problems. However, it is a challenge to precisely diagnose the cause and the origin of routing instability. In this paper, we are the first to distinguish link failure events from policy change events based on BGP updates from single vantage points by analyzing the relationship of the closed loops formed through intersecting all the transient paths during instability and the length variation of the stable paths after instability. Once link failure events are recognized, their origins are precisely inferred with 100% accuracy. Through simulation, our method is effective to distinguish link failure events from link restoration events and policy related events, and reduce the size of candidate set of origins.

With the development of wireless network technology and popularization of mobile devices, the Wireless Local Area Network (WLAN) has become an indispensable part of our daily life. Although the 802.11-based WLAN provides enormous convenience for users to access the Internet, it also gives rise to a number of security issues. One of the most severe threat encountered by Wi-Fi users is the evil twin attacks. The evil twin, a kind of rogue access points (RAPs), masquerades as a legitimate access point (AP) to lure users to connect it. Due to the characteristics of strong concealment, high confusion, great harmfulness and easy implementation, the evil twin has led to significant loss of sensitive information and become one of the most prominent security threats in recent years. In this paper, we propose a passive client-based detection solution that enables users to independently identify and locate evil twins without any assistance from a wireless network administrator. Because of the forwarding behavior of evil twins, proposed method compares 802.11 data frames sent by target APs to users to determine evil twin attacks. We implemented our detection technique in a Python tool named ET-spotter. Through implementation and evaluation in our study, our algorithm achieves 96% accuracy in distinguishing evil twins from legitimate APs.

In the design of Space Time Block Coding (STBC), for an arbitrary complex signal constellation with a size above 2 as well as a real signal matrix with a size above 8, it is difficult to acquire full code rate and full transmit diversity simultaneously. In this letter, an efficient selective receiver switching scheme is proposed for STBC with the full code rate and non-orthogonal design with the example of a 4-by-4 matrix. In the proposed scheme with the aid of beamforming, we divide the received signals into two groups according to the encoded matrix. By this way, we can eliminate the interference from the neighboring signals by more than half.

In an approximately synchronized (AS) code-division multiple-access (CDMA) communication system, zero correlation zone (ZCZ) sequences can be used as its spreading sequences so that the system suppresses multiple access interference (MAI) and multi-path interference (MPI) fully and synchronously. In this letter, the mutually orthogonal (MO) ZCZ polyphase sequence sets proposed by one of the authors are improved, and the resultant ZCZ sequences in each set arrive at the theoretical bound regarding ZCZ sequences under some conditions. Therefore, the improved MO ZCZ sequence sets are optimal.

In the classical computation theory, the language of a system features the computational behavior of the system but it does not distinguish the determinism and nondeterminism of actions. However, Milner found that the determinism and nondeterminism affect the interactional behavior of interactive systems and thus the notion of language does not features the interactional behavior. Therefore, Milner proposed the notion of (weak) bisimulation to solve this problem. With the development of internet, more and more interactive systems occur in the world, such as electronic trading system. Security is one of the most important topics for these systems. We find that different security policies can also affect the interactional behavior of a system, which exactly is the reason why a good policy can strengthen the security. In other words, two interactive systems with different security policies are not of an equivalent behavior although their functions (or business processes) are identical. However, the classic (weak) bisimulation theory draws an opposite conclusion that their behaviors are equivalent. The notion of (weak) bisimulation is not suitable for these security-oriented interactive systems since it does not consider a security policy. This paper proposes the concept of secure bisimulation in order to solve the above problem.

Based on the traffic predictability characteristic of Networks-on-Chip (NoC), we propose a pre-allocation based flow control scheme to improve the performance of NoC. In this scheme, routes are pre-allocated and the injection rates of all routes are regulated at the traffic sources according to the average available bandwidths in the links. Then, the number of packets in the network is decreased and thus, the congestion probability is reduced and the communication performance is improved. Simulation results show that this scheme greatly increases the throughput and cuts down the average latency with little area and energy overhead, compared with the switch-to-switch flow control scheme.