Search

Shadow detection is an important branch of computer vision. Recently, convolutional neural network (CNN)-based methods for shadow detection have achieved better performance than methods based on manually designed features. However, CNNs are extremely hungry for data and the training of CNN-based shadow detector requires time-consuming and expensive pixel-level annotations. To alleviate this problem in shadow detection, a method of data augmentation based on generative adversarial network (GAN), named ShadowGAN, has been proposed. Given a shadow mask and a shadow-free image, our ShadowGAN can generate shadow images with labels. To guide the training of ShadowGAN and get more realistic shadow images, (Formula presented.) loss is further implemented to impose a restriction between real shadow images and generated shadow images. The effectiveness of ShadowGAN is demonstrated by training existing shadow detectors on enlarged dataset. In addition, to better make use of shadow-free images in shadow detection, shadow image classification task is added for the shadow detectors. Experiments show that this task can guide the feature extraction network to learn more robust shadow features. At last, these two methods are combined and a better performance of shadow detection is achieved. National Key Research and Development Program of China. Grant Number: 2019YFC1511300; National Natural Science Foundation of China. Grant Number: 61971079; Basic Research and Frontier Exploration Project of Chongqing. Grant Number: cstc2019jcyj-msxmX0666; Innovative Group Project of the National Natural Science Foundation of Chongqing. Grant Number: cstc2020jcyj-cxttX0002; Department of Science and Technology of Sichuan Province. Grant Number: 2020YFQ0025.

This paper proposes a new robust truncated L $$_2$$ -norm twin support vector machine (T $$^2$$ SVM), where the truncated L $$_2$$ -norm is used to measure the empirical risk to make the classifiers more robust when encountering lots of outliers. Meanwhile, chance constraints are also employed to specify false positive and false negative error rates. T $$^2$$ SVM considers a pair of chance constrained nonconvex nonsmooth problems. To solve these difficult problems, we propose an efficient iterative method for T $$^2$$ SVM based on difference of convex functions (DC) programs and DC Algorithms (DCA). Experiments on benchmark data sets and artificial data sets demonstrate the significant virtues of T $$^2$$ SVM in terms of robustness and generalization performance.

Near-infrared spectroscopy (NIRS) is an outstanding detection technology with fast, nondestructive, on-line characteristics, which has been used in many fields to achieve qualitative/quantitative detection of products. This study proposes an analysis method that combines a one-dimensional convolution neural network (1-CNN) with NIRS to select a characteristic wavelength for a subsequent qualitative/quantitative analysis, while considering fully dimensional NIRS information. Batch normalization (BN) and an exponential decay learning rate are embedded into the training stage of the 1-CNN which improves performance and reduces the risk of overfitting. The proposed method is used for NIRS of dried matsutake samples from four regions in China, and a geographical origin identification model is established for the matsutake. The developed 1-CNN model achieves better classification accuracy than traditional models. The experimental results demonstrate that the proposed 1-CNN achieves good generalization and can act as an alternative method for analyzing NIRS data.

Proximal support vector machine (PSVM), as a variant of support vector machine (SVM), is to generate a pair of non-parallel hyperplanes for classification. Although PSVM is one of the powerful classification tools, its ability on feature selection is still weak. To overcome this defect, we introduce l 0 -norm regularization in PSVM which enables PSVM to select important features and remove redundant features simultaneously for classification. This PSVM is called as a sparse proximal support vector machine (SPSVM). Due to the presence of l 0 -norm, the resulting optimization problem of SPSVM is neither convex nor smooth and thus, is difficult to solve. In this paper, we introduce a continuous nonconvex function to approximate l 0 -norm, and propose a novel difference of convex functions algorithms (DCA) to solve SPSVM. The main merit of the proposed method is that all subproblems are smooth and admit closed form solutions. The effectiveness of the proposed method is illustrated by theoretical analysis as well as some numerical experiments on both simulation datasets and real world datasets.

No doubt incorporating deep learning in security public transportation is an attractive and challenging research direction. In this article, we incorporate multiple attention mechanisms with high-resolution and high-semantic features into neural networks to further classify and localize threat items automatically. We propose our efficient model located in different network layers. To be specific, our model is involved in three modules called the spatial attention module (SAM), the channel attention module (CAM), and the high-resolution and high-semantic module (HRHSM). For shallow and deep layers, we adapt SAM and CAM to mine the long-range spatial and channel information of the network, respectively. With regard to the last few layers with different sizes, we leverage HRHSM to fuse their features so that the output contains high-resolution and high-semantic features. Due to the generalizability, our model can be broadly applied to many commonly used networks. Experimental results on the SIXray dataset that is consisted of six classes of threat item in X-ray images, CIFAR-10, and CIFAR-100 datasets demonstrate that our modules can improve the performance of networks in image classification and object localization without ground-truth bounding boxes.

Resource allocation (RA) is a key technique to guarantee the quality of service of users and maximize the system capacity in cognitive nonorthogonal multiple access (NOMA) networks. However, traditional RA approaches have been proposed under perfect channel state information (CSI) which is too ideal in practical systems due to the impact of link delays, quantization errors, etc. In this article, with imperfect CSI, a downlink robust RA algorithm is proposed for robust transmission to maximize the sum energy efficiency (EE) of secondary users (SUs) under channel uncertainties. Meanwhile, it protects the minimum data rates of SUs, satisfying the maximum transmission power constraints of base stations and the maximum interference temperature constraint of each primary user (PU). The formulated problem is nonconvex, thus challenging to solve. For delay-tolerant services, to deal with the intractability of the problem caused by outage probability constraints, the closed-form expressions of outage probabilities of SUs and PUs are derived under Gaussian CSI error models. For delay-sensitive services, the robust constraints with bounded uncertainty sets are transformed into convex ones. Based on successive convex approximation and the parameter transformation approach, the original problem is converted into a closed-form geometric programming problem solved by dual decomposition methods and subgradient methods. Additionally, users’ outage probabilities and the minimum required transmit power under the two modeling approaches are analyzed. Computational complexity and sensitivity analysis are provided. The simulation results show the proposed algorithm serves good robustness and EE.

Conventional non-orthogonal multiple access (NOMA)-based spatial modulation (SM) systems utilize only the real part of amplitude-phase modulation (APM) symbols for transmission, but quadrature spatial modulation (QSM) can extend this to two orthogonal components to improve the spectral efficiency. In this way, we propose a NOMA-based QSM system in multiple-input multiple-output scenarios, where APM symbols are divided into real and imaginary components for transmission and then these two components are sent by real and imaginary transmit antennas, respectively. To further improve the performance of our proposed system, we also propose a user grouping and a power allocation with relatively low complexity. Different from previous schemes, the user grouping scheme is designed for considering channel conditions, user locations, and channel correlations jointly. Moreover, the power allocation scheme is performed in each group to meet users' quality-of-service requirements, i.e., target-rate requirements. Simulation results demonstrate the effectiveness of proposed schemes by comparing with existing ones.

Heterogeneous network (HetNet) with energy harvesting is a promising technique to provide perpetual power supplies and ubiquitous coverage as well as high data rate for next-generation wireless communications. In this article, we consider a robust power allocation and power splitting (PS) problem for downlink simultaneous wireless information and power transfer (SWIPT)-enabled HetNets. The robust energy-efficiency (EE) maximization problem of femtocell users (FUs) is formulated under the outage-probability interference power constraint of macrocell user (MU), the maximum allowable transmission power of FU, and the EE-based outage constraint of each FU. The originally fractional optimization problem with the probabilistic constraint is NP-hard and difficult to solve. Without knowing the distribution of uncertain parameters, a min–max probability machine approach is first introduced to convert the semi-infinite optimization problem into a deterministic one which is transformed into a deterministic convex one by using the Dinkelbach method and the quadratic transformation approach. An iterative power allocation and PS scheme is obtained based on convex optimization methods. Finally, the effectiveness of the proposed algorithm is demonstrated by simulation results from the perspective of EE and robustness.

Body area network (BAN) is a key technology of solving telemedicine, where protecting security of vital signs information becomes a very important technique requirement. The traditional encryption methods are not suitable for BANs due to the complex algorithm and the large consumption. This paper proposes a new encryption method based on the QRS complex of the ECG signal, which adopts the vital signs from the BAN system to form the initial key, utilizes the LFSR (Linear Feedback Shift Register) circuit to generate the key stream, and then encrypts the data in the BANs. This new encryption method has the advantages of low energy consumption, simple hardware implementation, and dynamic key updating.

To improve energy efficiency and reduce the interference to macro users (MUs), this paper investigates the robust resource allocation (RA) problem for maximizing the interference efficiency of users (i.e., total rate/sum interference) in heterogeneous networks. Under perfect channel state information (CSI), the considered RA problem is formulated as a multivariate nonlinear programming problem with constraints on the maximum interference power of MUs, the minimum rate requirement of each femto user, and the maximum transmit power of the femto base station. The original fractional programming problem is converted into a convex optimization problem by using Dinkelbach's method, the logarithmic transformation method, and the successive convex approximation method, where the closed-form solution is obtained with the Lagrangian dual approach. Moreover, with imperfect CSI consideration, the original problem is reformulated as a robust RA problem that is transformed into a deterministic and convex optimization problem by using an inequality transformation approach. In addition, the computational complexity and the cost of robustness are also analyzed. The simulation results verify that the proposed algorithm has better interference efficiency and robustness by comparing with the existing algorithms.

In this paper, based on the characteristics of inertial data collected by smart terminals carried by pedestrians, a dynamic estimation algorithm of stride length using motion mode recognition is proposed. The motion mode includes pedestrian motion status and a smart terminal attitude. As pedestrians move forward, different motion modes will have a corresponding impact on the stride length estimation model, resulting in the stride length estimation model directly affecting the accuracy of the pedestrian dead reckoning system (PDR). Therefore, different motion modes need to make corresponding adjustments to the stride length estimation model to give a specific stride-length adjustment gain. In order to avoid errors caused by the change of the motion mode, a dynamic time warping (DTW) algorithm is proposed to identify the motion mode of the smart terminal, so as to select an appropriate gain to adjust the stride length estimation model, thereby improving the positioning accuracy of the pedestrian dead reckoning system. Experimental results show that the average positioning error of the stride-length estimation model based on DTW motion mode recognition is 1.78 meters in five motion modes. In performance, this algorithm is superior to other traditional stride length estimation models.

In this paper, deep learning is combined to learn the bit posterior probability (soft information) of iterative decoding for a bit-interleaved coded modulation with iterative decoding (BICM-ID) system. The deep neural network (DNN) is adopted to learn and replace multiple modules of the receiver, which can jointly deal with multiple problems and improve the efficiency of the whole system. The output of direct learning iteration reduces the computational cost of iteration to a certain extent. Simulation is carried out under Rayleigh channel and multiple modulation modes and results show that the proposed scheme without iteration is better than traditional BICM systems, and very close to traditional BICM-ID systems which needs iteration between demodulation and decoding.

Non-dominated sorting is a critical component of all multi-objective evolutionary algorithms (MOEAs). A large percentage of computational cost of MOEAs is spent on non-dominated sorting. So the complexity of non-dominated sorting method in a large extent decides the efficiency of the MOEA. In this paper, we present a novel non-dominated sorting method called the dynamic non-dominated sorting (DNS). It is based on the sorting of each objective instead of dominance comparisons. The computational compelxity of DNS is $O(mN\log N)$ ($m$ is the number of objectives, $N$ is the population size), which equals to the best record so far. Based on DNS, we introduce a novel multi-objective genetic algorithm (MOGA) called the dynamic non-dominated sorting genetic algorithm (DSGA). Then, some numerical comparisons between different non-dominated sorting method are presented. The results shows that DNS is efficient and promising. Finally, numerical experiments on DSGA are also given. The results show that DSGA outperforms some other MOEAs both on general-scale and large-scale multi-objective problems.

The design and improvement of feature descriptor is always a prevalent research point. A strong design of feature descriptor can be applied to many industrial fields. However, in many successful designs, there are still have some shortcomings that need to be improved like the flipping operation in image. Although there are many techniques to deal with this problem recently, they still exist some uncertainty factors because they do not compare with different images which have the same but flipped object in their experiments. In this paper, we design an unsupervised learning structure algorithm based on SIFT to achieve flipping invariant and change object to get as many correct matches as possible. Then we will use this algorithm to match in UKBench dataset and indicate that this approach not only solves the problem of flipping operation but also ensures high matching accuracy. Compared with previous other algorithms, our algorithm is extremely simple to implement and does not destroy the structure of SIFT.

Electrocardiogram (ECG) is the physical interpretation of the electrical conductivity of the heart resulting from the muscular activities. It provides an effective way to observe heart diseases during medical treatments and studies. During the acquisition and transmission of ECG signal in the medical environment, the signal is often corrupted by various types of noises. The transmission of ECG regularly introduces noise due to terrible channel conditions. Moreover, there are different types of noises inherent in the data collection process. In order to reduce the impact of noise, in this paper, we proposed an algorithm based on the EMD and the piecewise polynomial fitting to remove the baseline wander for ECG signals. Here, the effect of this algorithm on wandered ECG signals is discussed elaborately with graphical interpretations. It is also analyzed that why this algorithm works better to remove baseline wander by using EMD and combine the piecewise polynomial fitting to denoise the signal processed by the EMD.

Template matching is widely used in computer vision and pattern recognition for its efficiency. However, conventional matching algorithms fail to match same templates of a same scene when illumination changes, especially when shadows exist. To achieve higher accuracy of template matching under different illuminations, a template matching algorithm is proposed based on gradient-chromatic illumination-invariant image. The matching results on Middlebury 2014 stereo dataset shows that our algorithm is illumination-invariant and achieve higher accuracy of rectangle template matching.

In this paper, in order to solve poor communication problems under the long-distance transmission and secure communication, an energy-efficient maximization based resource allocation (RA) algorithm is proposed to improve system energy efficiency (EE) and information security for secure intelligent reflecting surface-based cellular networks under an active eavesdropper. Firstly, a nonlinear RA model with coupled variables (i.e., beamforming vector and phase shift matrix) is formulated for a multiple-input single-output cellular system with the help of IRS, where the minimum secure rate constraint of user, the maximum transmit power constraint of base station (BS), and continuous phase shift constraint are considered simultaneously. Then, the fractional objective function is transformed into a parameter-subtractive form by using Dinkelbach's approach, meanwhile the coupled beamforming vector and the phase shift parameter are decoupled by using an alternating iteration method. Moreover, the non-convex problem is converted into a semi-definite programming problem solved by using convex optimization tools. Finally, simulation results demonstrate the proposed algorithm has good EE and security.

Intelligent reflecting surface (IRS)-enabled communication systems provide higher system capacity and spectral efficiency by reflecting the incident signals from transmitters in a low-cost passive reflecting way. However, it poses new challenges in resource allocation due to surrounding interference and phase shift, especially when IRS is employed in heterogeneous networks (HetNets). In this paper, a joint power allocation and phase shift optimization problem is studied for the downlink IRS-enabled HetNet, in which the IRS is deployed to enhance the communications between small cell users (SCUs) and associated base station (BS). The signal-to-interference-plus-noise ratio (SINR) received at the SCU is maximized by jointly optimizing the transmit power of the small-cell BS and the phase shift of the IRS, subject to the constraints on the minimum SINR requirement of the macro-cell user (MCU) and the phase shift. Although the formulated problem is non-convex, we develop an optimal power allocation and the IRS's passive array coefficient solution for the single-user scenario. For the multi-user scenario, we propose an iterative algorithm to maximize the total rates of SCUs for obtaining a suboptimal solution by an alternating iteration manner, where the sum of multiple-ratio fractional programming problem is converted into a convex semidefinite program (SDP) problem. Simulation results show that the proposed algorithm significantly improves the achieved transmission rates of SCUs compared to the case without the IRS.

In Vehicle-to-Vehicle (V2V) communication, the rapid movement of vehicles brings fast attenuation of signal propagation and serious shift effect of Doppler frequency, leading to serious transmission performance deterioration. Multiple-Input Multiple-Output (MIMO) technique can be well applied in V2V communication by exploiting diversity and multiplexing gain to improve the system reliability and increase the data rate. In an MIMO system, the achieved diversity and multiplexing gain is closely related to its detection method. Integer-Forcing (IF) detection is one of the linear detections proposed in recent years. Compared to the traditional linear and Maximum Likelihood (ML) detection, IF has a good tradeoff between the complexity and system performance. In this paper, considering its good detection performance, IF is applied to the detection of a MIMO V2V communication system and the performance is further analyzed theoretically by employing Alamouti code at the transmitters. The diversity gain proves to be full and is confirmed by the simulation results.

With the development of fifth generation wireless communication technology, how to improve system capacity and spectrum efficiency is a crucial problem in resource sharing of heterogeneous networks (HetNets). Most of existing resource allocation (RA) schemes in HetNets focus on perfect channel state information, however, exact channel information is difficult to obtain under link delay and stochastic channel condition. In order to resolve the RA issues under channel uncertainties, a robust RA algorithm is proposed to maximise the sum data rate of microcell users where the users are subjected to the individual transmission power constraint and the cross-tier interference constraint of macrocell users. The multiuser RA problem in HetNets is formulated under the consideration of bounded channel gain uncertainties where both the cross-tier channel and intra-tier channel are simultaneously considered. The non-linear optimisation problem is converted into a geometric programming problem that is solved by using Lagrange dual methods in a distributed way. Simulation results show that the proposed algorithm can well restrain the effect of channel uncertainty and achieve a good robustness.

Body area networks (BANs) are systems of wearable computing devices for long-term monitoring of personal health care. BAN is an emerging technology for the worldwide ageing population. In the BAN system, the transceiver is the most energy-consuming part of a sensor node and radio transmission in the vicinity of the human body is highly lossy and inefficient. Therefore, the energy of the sensor node constraints the life cycle and quality of service of the network; consequently, low-cost protocol shaves attracted wide interest. This study proposes a frame structure model of a self-adaptive guard band protocol, which introduces a GB in each time slot according to the allowed maximum time drift of the crystal, adaptively adjusts the value of the GB based on the actual time drift, and then ensures that the node simultaneously maintains the sleeping state and synchronisation with the coordinator during beacon transmission, thus reducing the energy consumption.

Body area networks (BAN) are at the forefront of technologies for long-term monitoring of personal healthcare, which is intended be an effective strategy to address the aging population worldwide. The transceiver is the most energy-consuming part of a sensor node, and radio transmission in the vicinity of the human body is highly lossy and inefficient. Therefore, the energy of the sensor node constrains the life cycle and quality of service (QoS) of the network; consequently, low-cost protocol shave attracted wide interest. This paper proposes a frame structure model of a self-adaptive guard band (SAGB) protocol, which introduces a guard band (GB) in each time slot according to the allowed maximum time drift of the crystal, adaptively adjusts the value of the GB based on the actual time drift, and then ensures that the node simultaneously maintains the sleeping state and synchronization with the coordinator during beacon transmission, thus reducing the energy consumption.

In this paper we consider the minimization of the sum of local convex component functions distributed over a multi-agent network. We first extend the Nesterov's random gradient-free method to the incremental setting. Then we propose the incremental gradient-free methods, including a cyclic order and a randomized order in the selection of component function. We provide the convergence and iteration complexity analysis of the proposed methods under some suitable stepsize rules. To illustrate our proposed methods, extensive numerical results on a distributed $l_1$-regression problem are presented. Compared with existing incremental subgradient-based methods, our methods only require the evaluation of the function values rather than subgradients, which may be preferred by practical engineers.

In this paper, we investigate a downlink outage- constrained robust resource allocation algorithm for achieving the robust transmission in power-domain non-orthogonal multiple access based underlay cognitive radio networks. Our goal is to maximize the sum energy efficiency of secondary users (SUs) under the consideration of imperfect channel state information, while protecting the data rate requirement of each SU and satisfying the maximum transmission power constraint of the secondary base station as well as the interference temperature constraint of each primary user (PU). The proposed optimization problem is non-convex and difficult to solve. To deal with the intractability of the problem coming from the outage probability constraints, in particular, the closed-form expressions of the outage probabilities of SUs and PUs are derived. The original problem is converted into an equivalent convex problem by using successive convex approximation and parameter transformation. A robust subchannel and power allocation algorithm is developed to solve the problem. Simulation results show the effectiveness of the proposed algorithm by comparing with the existing algorithms.

Conventional downlink non-orthogonal multiple access (NOMA) with spatial modulation (SM) limits the number of chosen transmitting antennas. Therefore, a multiuser NOMA-based generalized spatial modulation (GSM) system is proposed for the downlink scenario, where more than one transmitting antenna are selected. Different channel gains are considered in the NOMA-GSM system to reduce the impact of intra-group interference. Moreover, to make good use of power domain and spatial domain resources, a dynamic user grouping is proposed, where different users are selected from different sets according to their effective channel gains. Moreover, a dynamic power allocation scheme is proposed to maximize the achievable sum rate of users, which is the suboptimal solution of the product of two-user signal-to-interference-plus-noise ratio. To extend to three-user systems, another power allocation scheme is proposed. The achievable sum rate maximization problem is transformed into a convex optimization problem by introducing the auxiliary variable. Simulation results show that proposed user grouping and power allocation schemes have advantages compared with existing ones.

To improve transmission efficiency and robustness, we investigate robust energy efficiency (EE)-based maximization power allocation and time optimization problem in orthogonal frequency division multiple access (OFDMA) based heterogeneous networks under channel uncertainties. The non-convex optimization problem is decomposed into three subproblems: subcarrier allocation, transmission time optimization and power allocation. The subcarrier allocation is obtained based on maximizing user's EE under the effective transmission time. The Taylor expansion method is used to obtain the optimal transmission time. Additionally, Bernstein approach and Dinkelbach's method are used to solve the power allocation problem. Simulation results indicate that the proposed algorithm has good robustness in terms of outage probability.

In this paper, the interference-efficient based resource allocation for uplink transmission in an OFDMA-based cognitive radio network is studied. The interference efficiency (IE) is defined as the total data rate of secondary users (SUs) over sum interference power imposed on primary users (PUs). The objective is to maximize the total IE of SUs subject to transmit power constraints of SUs, subcarrier assignment constraint and the interference power threshold constraints of SUs. The original mixed integer fractional programming problem is converted into the convex one which is solved by using the convex optimization technique. Simulation results show the effectiveness of the proposed algorithm in terms of the interference to PUs and transmission efficiency.

Mobile AdHoc Networks (MANET) can be fast implemented, and it is very popular in many specific network requirements, such as UAV (Unmanned Aerial Unit), Disaster Recovery and IoT (Internet of Things) etc. However, MANET is also vulnerable. AODV (Ad hoc On-Demand Distance Vector Routing) protocol is one type of MANET routing protocol and many attacks can be implemented to break the connections on AODV based AdHoc networks. In this article, aim of protecting the MANET security, we modeled the AODV protocol with one type of Automata and analyzed the security vulnerabilities of it; then based on the analyzing results, we proposed an enhancement to AODV protocol to against the Black Hole Attacks. We also implemented the proposed enhancement in NS3 simulator and verified the correctness, usability and efficiency.

Security Metrics help network administrators master the security status and strengthen security management for many years. Recently, with the usages of many new techniques and network structures, the cyber attacks become complex and the security measurement has received more and more attentions. However, existing methods usually focus on one aspect of security and the indicators used are usually difficult to quantify, which makes it difficult to understand network security status in some real circumstance. In this paper, we consider the network system security from the perspective of attack and defense and the changes of external security environment to propose a comprehensive and quantifiable index system for network security measurement. We illustrate the corresponding theories and the usages of each selected indicators and we also complete the real-time security measurement in various attacks and defenses by using NS3 simulator. The simulation results verify the correctness and rationality of the proposed Security Measurement Index System.

To improve user’s data rate and network coverage, multi-homing heterogeneous network has been considered as a key technology in recent years. However, traditional power allocation schemes cannot provide good robustness and suppress the impact of parameter perturbation due to the random interference and channel delays. A robust power allocation algorithm is proposed for an orthogonal frequency division multiple access based multi-homing heterogeneous network under energy harvesting. Considering the channel uncertainties from base stations to users, the total data rate of all mobile terminals (MTs) is maximized under the rate outage constraint of MT. Based on the uniform distribution of uncertainty, the originally robust counterpart problem is transformed to a deterministic and convex one. The Lagrange dual decomposition theory is used to solve it. Simulation results demonstrate that the proposed scheme has good convergence and robustness.

Image stitching is one of the branches of computer vision. It combines two or more images for a scene to acquire a high-resolution panoramic image. An invariant local function often uses to stitch two images together. Since the flat plate of a digital radiography (DR) system does not cover all parts of the body, the whole bone structure image cannot seize in a single scan. To solved this problem, image stitching is broadly utilized by medical systems to stitch DR images, which can be helpful for scoliosis or lower extremity deformities in the diagnosis, and pre-operative planning are of great importance. In this paper, the stitching and retrieval of medical images planned. To conquer the background noise in medical images, and improve the recovery of quality and stitching rapidity of medical images, a random sample consensus (RANSAC) algorithm is useful to stitching the images of Chest digital radiography by scale-invariant feature transform (SIFT) and speeded-up robust features (SURF) feature extraction. Down-sampling utilizes to lessen the size of the images and reduction the measure of calculation. In the interim, the phase correlation is engaged to discover the overlapping region. After feature matching and perspective transformation, the stitched image is gotten dependent on the homography. At last, experimentation has finished showing the presentation.

The protection of information security is a very important technique requirement in Body area networks (BANs). According to the limitation of energy, the traditional encryption methods which adopt the complex algorithm and the large consumption are not suitable for BANs. This paper proposes a new encryption method based on the channel model of BANs, which adopts the path loss from the BAN systems to form the initial key, utilizes the LFSR (Linear Feedback Shift Register) circuit to generate the key stream, and then encrypts the data in the coordinator of the BAN system. This new encryption method has the advantages of low energy consumption, simple hardware implementation, and dynamic key updating.

For heterogeneous networks (HetNets), power allocation strategy with energy harvesting (EH) will be a promising technique in the green communication system. In this paper, considering the trade-off between nonrenewable energy consumption and robustness, a robust power allocation (RPA) problem is formulated to optimize energy-efficiency for EH-enabled HetNets under imperfect channel state information. Meanwhile, minimum required rate constraints and maximum interference constraints to macrocell users are guaranteed. Based on fractional programming and worst-case principle, the original non-convex optimization problem can be converted into a deterministic convex one which can be solved by using Lagrange dual decomposition theory. Numerical results show the efficiency and robustness of our proposed RPA algorithm comparing with non-robust algorithm.

In this work, a wearable, intelligent testing device based on nRF52832 microcontroller is designed to monitor the changes of life signs parameters of patients with depression and cardiovascular disease and determine disease seriousness and mood stability. The proposed device uses the ADS1292R chip and conductive silica gel for collection and preconditioning the electrocardiogram (ECG) and respiratory signals, which and then filtered by nRF52832. Since nRF52832 is a Bluetooth chip, it transmits the life signs signals to the mobile phone APP through Bluetooth to be displayed on a device screen. The wearable, intelligent detection devices have the advantages of small size, low power consumption, low cost, and convenience to be wearable.

In this paper, the trade-off among system sum energy consumption and robustness is studied. In this regard, a robust power allocation problem is formulated for a two-tier heterogeneous network with uplink transmission mode and consideration of imperfect channel state information. The objective is to minimize the total transmit power of femtocell users (FUs), while the interference to macrocell user receiver is limited to a predefined interference level, the transmit power of each FU transmitter is kept within their power budgets, and the actual signal-to-interference-plus-noise ratio of each femtocell receiver is above a minimum threshold. Considering the uncertainties of the interference links from FUs to macrocell base stations and forward transmission links of each FU, the robust power allocation problem is formulated as a semi-infinite programming problem (SIPP). By the worst-case approach, the SIPP is transformed into a convex optimization problem solved by the Lagrange dual decomposition method. Moreover, the feasible regions of constraints, computational complexity, and sensitivity degree of the proposed robust algorithm are also analyzed. Simulation results investigate the impact of channel uncertainties and the superiority of the proposed algorithm by comparing with non-robust algorithm.

Global Navigation Satellite System (GNSS) cannot achieve accurate positioning and navigation in the indoor environment. Bluetooth Beacon positioning is one of the most widely used technology. Because of the time-varying characteristics of the Bluetooth received signal strength indication (RSSI), traditional positioning algorithms have large ranging errors by using fixed path loss models. In this paper, we proposes a RSSI real-time correction method based on Bluetooth gateway, which is used to detect the RSSI fluctuations of surrounding Bluetooth nodes and upload them to the cloud server. The terminal to be located collects the RSSIs of surrounding Bluetooth nodes, and then adjusted by the RSSI fluctuation information stored on the server in real time. The adjusted RSSIs can be used for calculation and obtain a more accurate distance, i.e., smaller positioning error. Finally, the position of blind nodes is estimated by least squares algorithm. The experiment shows that the proposed method improves the positioning accuracy.

The Multiple-Input Multiple-Out (MIMO) technology can improve system capacity by using multiple antenna transceivers with high decode complexity at the receiver. The Integer-Forcing (IF) linear detection algorithm is used to reduce the complexity of implementation in MIMO system, and its performance is close to the ML detection algorithm. However, the decode complexity of the IF detection will increase with the increase of antenna number. In order to further reduce the decode complexity, we propose a new detection algorithm called Group Integer-Forcing detection algorithm in this paper. In this algorithm, the received symbols are divided into groups, then each group will be decoded by the IF detection respectively to recover transmitted codewords in their own group, and those recovered codewords will be gathered to obtain original transmitted information before grouping. The Group Integer-Forcing detection algorithm reduce the decode complexity via grouping, so it has great advantage on complexity compare with the IF detection. It also has better error performance than the ZF detection algorithm according to simulation results, and the error performance gap is more and more obvious with the increased antenna numbers.

The Global Navigation Satellite System (GNSS) cannot achieve accurate positioning and navigation in the indoor environment. Therefore, efficient indoor positioning technology has become a very active research topic. Bluetooth beacon positioning is one of the most widely used technologies. Because of the time-varying characteristics of the Bluetooth received signal strength indication (RSSI), traditional positioning algorithms have large ranging errors because they use fixed path loss models. In this paper, we propose an RSSI real-time correction method based on Bluetooth gateway which is used to detect the RSSI fluctuations of surrounding Bluetooth nodes and upload them to the cloud server. The terminal to be located collects the RSSIs of surrounding Bluetooth nodes, and then adjusts them by the RSSI fluctuation information stored on the server in real-time. The adjusted RSSIs can be used for calculation and achieve smaller positioning error. Moreover, it is difficult to accurately fit the RSSI distance model with the logarithmic distance loss model due to the complex electromagnetic environment in the room. Therefore, the back propagation neural network optimized by particle swarm optimization (PSO-BPNN) is used to train the RSSI distance model to reduce the positioning error. The experiment shows that the proposed method has better positioning accuracy than the traditional method.

Compared with the Multiple-Input Multiple-Output (MIMO) systems coded with Space-Time Block Code (STBC), Spatial Modulation (SM) has high spectral efficiency and low diversity gain. After combined with Space-Time Block Code (STBC), so called STBC-SM is able to obtain higher diversity gain at the same time. In this paper, we propose a new full-diversity STBC-SM scheme for spatial modulation systems with four transmit antennas. The transmitted information is partitioned into two parts for STBC-SM codeword selection and modulation symbol mapping, respectively. The proposed codeword matrices are constructed by simultaneously choosing the transmit antennas of one pair of orthogonal STBC instead of only one. Precoding and symbol-crossing of the modulated symbols are carried out to ensure higher diversity gain. By theoretical analysis, the designed scheme is proved to be full diversity. Finally, simulation results is presented and show that the proposed STBC-SM obviously improves the diversity gain and system performance.

Body area network (BAN) consists of intercommunicating sensors, which are either wearable, or can be implanted into the human body. Being a network that features critical information, BAN requires encryption, however common encryption methods found in computer networks are not ideal due to complex algorithms and high power consumption. The authors propose a new encryption method based on the human body channel. This new encryption method has the advantages of low-power, dynamic updating, rapid operation, and easy implementation, which is suitable for BAN systems.

In this paper, we investigate the robust power allocation problem in a two-tier orthogonal frequency division multiplexing (OFDM) access based heterogeneous networks, where the minimal rate requirement of each femtocell user and interference power constraint of macrocell user are properly controlled under channel uncertainties. We formulate an optimization problem to maximize the sum data rate of the femtocell users under the consideration of channel uncertainties of both cross-tier and intra-tier links while guaranteeing the rate requirement of femtocell users and controlling the interference to the macrocell users in terms of probability. According to the distribution models of estimation errors, we transform the original problem into a deterministic one and solve it by the subgradient method and Karush-Kuhn-Tuchker conditions. Simulation results show that our approach can fast converge to the optimal value and well protect the performance of macrocell users under channel uncertainties.

Heterogeneous wireless networks are a promising technology in next generation wireless communication networks, which has been shown to efficiently reduce the blind area of mobile communication and improve network coverage compared with the traditional wireless communication networks. In this paper, a robust power allocation problem for a two-tier heterogeneous wireless networks is formulated based on orthogonal frequency-division multiplexing technology. Under the consideration of imperfect channel state information (CSI), the robust sum-rate maximization problem is built while avoiding sever cross-tier interference to macrocell user and maintaining the minimum rate requirement of each femtocell user. To be practical, both of channel estimation errors from the femtocells to the macrocell and link uncertainties of each femtocell user are simultaneously considered in terms of outage probabilities of users. The optimization problem is analyzed under no CSI feedback with some cumulative distribution function and partial CSI with Gaussian distribution of channel estimation error. The robust optimization problem is converted into the convex optimization problem which is solved by using Lagrange dual theory and subgradient algorithm. Simulation results demonstrate the effectiveness of the proposed algorithm by the impact of channel uncertainties on the system performance.

Rapid progress of networking technologies leads to an exponential growth in the number of unauthorized or malicious network actions. As a component of defense-in-depth, Network Intrusion Detection System (NIDS) has been expected to detect malicious behaviors. Currently, NIDSs are implemented by various classification techniques, but these techniques are not advanced enough to accurately detect complex or synthetic attacks, especially in the situation of facing massive high-dimensional data. Besides, the inherent defects of NIDSs, namely, high false alarm rate and low detection rate, have not been effectively solved. In order to solve these problems, data fusion (DF) has been applied into network intrusion detection and has achieved good results. However, the literature still lacks thorough analysis and evaluation on data fusion techniques in the field of intrusion detection. Therefore, it is necessary to conduct a comprehensive review on them. In this article, we focus on DF techniques for network intrusion detection and propose a specific definition to describe it. We review the recent advances of DF techniques and propose a series of criteria to compare their performance. Finally, based on the results of the literature review, a number of open issues and future research directions are proposed at the end of this work.

Generalized spatial modulation (GSM) activates more transmit antennas than spatial modulation to increase the spectral efficiency. Although the maximum likelihood (ML) detector is able to achieve the optimal performance, its exhaustive search leads to intractable computational complexity. In this paper, we design a detection scheme based on Integer-Forcing (IF) algorithm, which utilize IF receiver to decode the transmitted symbols for all possible antenna combinations, and then compare the Euclidean distances between the detected symbols and the received signals. Finally, the symbol vector with minimum Euclidean distance and the corresponding antenna combination are detected as the decided symbols and transmit antenna combination. Simulation results shows that the designed detection scheme achieves obviously better performance than the ZF and the MMSE detection with relatively lower complexity.

Mobile ad hoc network (MANET) is a kind of mobile multi-hop network which can transmit data through intermediate nodes, it has been widely used and become important since the growing of the market of Internet of Things (IoT). However, the transmissions on MANET are vulnerable, it usually suffered with many internal or external attacks, and the research on security topics of MANET are becoming more and more hot recently. Blackhole Attack is one of the most famous attacks to MANET. In this paper, we focus on the Blackhole Attack in AODV protocol, and use NS-3 network simulator to study the impact of Blackhole Attack on network performance parameters, such as the Throughput, End-to-End Delay and Packet Loss Rate. We further analyze the changes in network performance by adjusting the number of blackhole nodes and total nodes, and the movement speed of mobile nodes. The experimental results not only reflect the behaviors of the Blackhole Attack and its damage to the network, but also provide the characteristics of Blackhole Attacks clearly. This is helpful to the research of Blackhole Attack feature extraction and MANET security measurement.

In this paper, we present a new global optimization method to solve nonlinear systems of equations. We reformulate given system of nonlinear equations as a global optimization problem and then give a new auxiliary function method to solve the reformulated global optimization problem. The new auxiliary function proposed in this paper can be a filled function, a quasi-filled function or a strict filled function with appropriately chosen parameters. Several numerical examples are presented to illustrate the efficiency of the present approach.

Body area network (BAN) is a key technology of solving remote medical, where protecting security of vital signs information is a very important technique requirement. This paper presents a data security scheme based on ECG characteristics, investigating methods of the circuit-level data encryption and dynamic key refreshment due to the obtained ECG signal characteristic value. The proposed scheme is constructed from the inherent features of the BAN system, adopting vital information from security protection, which has advantages of high strength, low cost, and easy implementation. Therefore, the scheme helps to design low-power sensor nodes and provides theoretical support and engineering implementation.

Spatial modulation (SM) is a promising multiple-input multiple-output (MIMO) transmission technology, which exploits the indices of transmit antennas to encode information into the spatial dimension and needs less radio frequency chains. To detect the transmitted information, the maximum likelihood (ML) algorithm is often adopted which has high computational complexity. Some low-complexity detection algorithms only consider hard-decision cases with no channel coding. In this paper, a low-complexity soft output detection algorithm is proposed for coded spatial modulation systems. Its computational complexity doesn't increase with the modulation order and hence much lower than the soft output ML detection algorithm. Simulation results demonstrate that the bit error rate (BER) performance is near to the ML detection with obvious complexity reduction. By sorting the detection metrics and selecting less transmit antenna indices as search candidates to compute soft bit information, further reduction of computational complexity could be obtained.

Traditional power allocation algorithms in heterogeneous networks (HeNets) are given under the assumption of perfect channel state information (CSI). However, the perfect CSI is difficult to obtain due to the efficiency of channel fading and feedback delays. In this paper, a robust power allocation algorithm is proposed to minimize the total transmit power of femtocell users under the constraints of macrocell interference power, maximum transmit power and signal-to-interference-noise-ratio (SINR). Though the original optimization problem becomes a semi-infinite programming problem (SIP) when using ellipsoidal uncertainty model. The SIP can be converted into a deterministic convex optimization problem by the worst-case approach, which can be easily solved in sub-gradient method. Simulation results show our proposal has good robustness and energy efficiency.

With the development of the fifth generation communication technology, how to improve system capacity and spectral efficiency is a key issue, which has attracted more and more attention from industry and academia. Heterogeneous network has been considered as a new promising technique for enhancing the quality of service, energy and spectrum efficiency as well as coverage of network due to different radio access technology (RAT) and different network structures. In this paper, the rate maximization resource allocation problem for multiuser cognitive heterogeneous networks is formulated to flexibly use network resource and improve the overall capacity, which simultaneously considers cross-tier interference constraint and maximum transmit power of cognitive microcell base station. The non-convex optimization problem is converted into a geometric programming problem which can be solved by Lagrange dual method in a distributed way. Simulation results are given to show the performance of the proposed algorithm in terms of the achievable system capacity and the interference to the macrocell network.