With the development of mobile edge computing (MEC), some late-model application technologies, such as self-driving, augmented reality (AR) and traffic perception, emerge as the times require. Nevertheless, the high-latency and low-reliability of the traditional cloud computing solutions are difficult to meet the requirement of growing smart cars (SCs) with computing-intensive applications. Hence, this paper studies an efficient offloading decision and resource allocation scheme in collaborative vehicular edge computing networks with multiple SCs and multiple MEC servers to reduce latency. To solve this problem with effect, we propose a context-aware offloading strategy based on differential evolution algorithm (DE) by considering vehicle mobility, roadside units (RSUs) coverage, vehicle priority. On this basis, an autoregressive integrated moving average (ARIMA) model is employed to predict idle computing resources according to the base station traffic in different periods. Simulation results demonstrate that the practical performance of the context-aware vehicular task offloading (CAVTO) optimization scheme could reduce the system delay significantly.

In this paper, an automatic system of locating the indoor area with weak or no mobile signal was proposed and demonstrated experimentally by using the Internet of Things (IoT) technology. Nowadays, the technicians of mobile services providers need to go along with numerous heavy equipment to measure and record the mobile signal strength at outside environment. Recently, there are systems proposed to do such measurement at outdoor area by using the IoT technology automatically. However, these works could not be applied in the indoor area since there are difficulties to do the indoor mapping and positioning. In this work, the Bluetooth Low Energy (BLE) was used to tackle these two difficulties. After a proper placement of BLE in the testing site, while the technician walk around with a handheld analyzer, the data can be obtained accordingly for further analysis in the proposed system which includes the construction of floor plan, detection of mobile signal strength and suggestion of indoor base stations. The gift wrapping and centroid algorithms were used during the analysis. The experimental results showed that the proposed system successfully demonstrated the indoor mapping, positioning of weak mobile signal area and suggestion of indoor base stations for the normal rectangular rooms with an area of 100 m² on single floor.

Fruit detection in orchards is one of the most crucial tasks for designing the visual system of an automated harvesting robot. It is the first and foremost tool employed for tasks such as sorting, grading, harvesting, disease control, and yield estimation, etc. Efficient visual systems are crucial for designing an automated robot. However, conventional fruit detection methods always a trade-off with accuracy, real-time response, and extensibility. Therefore, an improved method is proposed based on coarse-to-fine multitask cascaded convolutional networks (MTCNN) with three aspects to enable the practical application. First, the architecture of Fruit-MTCNN was improved to increase its power to discriminate between objects and their backgrounds. Then, with a few manual labels and operations, synthetic images and labels were generated to increase the diversity and the number of image samples. Further, through the online hard example mining (OHEM) strategy during training, the detector retrained hard examples. Finally, the improved detector was tested for its performance that proved superior in predicted accuracy and retaining good performances on portability with the low time cost. Based on performance, it was concluded that the detector could be applied practically in the actual orchard environment.

Nowadays, recommendation systems (RSs) are applied to all aspects of online life. In order to overcome the problem that individuals who do not meet the constraints need to be regenerated when the many-objective evolutionary algorithm (MaOEA) solves the hybrid recommendation model, this paper proposes a many-objective particle swarm optimization algorithm based on multiple criteria (MaPSO-MC). A generation-based fitness evaluation strategy with diversity enhancement (GBFE-DE) and ISDE⁺ are coupled to comprehensively evaluate individual performance. At the same time, according to the characteristics of the model, the regional optimization has an impact on the individual update, and a many-objective evolutionary strategy based on bacterial foraging (MaBF) is used to improve the algorithm search speed. Experimental results prove that this algorithm has excellent convergence and diversity, and can produce accurate, diverse, novel and high coverage recommendations when solving recommendation models.

Compressed sensing in image reconstruction has attracted attention and many studies are proposed. As we know, adding prior knowledge about the distribution of the support on the original signal to CS can improve the quality of reconstruction. However, it is still difficult for a recovery framework adjusts its strategy for exploiting the prior knowledge efficiently according to the current estimated signals in serial iterations. With the theory of information geometry, we propose an adaptive strategy based on the current estimated signal in each iteration of the recovery. We also improve the performance of existing algorithms through the adaptive strategy for exploiting the prior knowledge according to the current estimated signal. Simulations are presented to validate the results. In the end, we also show the application of the model in the image.

Although non-binary low-density parity-check (NB-LDPC) codes have better error-correction capability than that of binary LDPC codes, their decoding complexity is significantly higher. Therefore, it is crucial to reduce the decoding complexity of NB-LDPC while maintaining their error-correction capability to adopt them for various applications. The extended min-sum (EMS) algorithm is widely used for decoding NB-LDPC codes, and it reduces the complexity of check node (CN) operations via message truncation. Herein, we propose a low-cost CN processing method to reduce the complexity of CN operations, which take most of the decoding time. Unlike existing studies on low complexity CN operations, the proposed method employs quick selection algorithm, thereby reducing the hardware complexity and CN operation time. The experimental results show that the proposed selection-based CN operation is more than three times faster and achieves better error-correction performance than the conventional EMS algorithm.

Echocardiography, an ultrasound scan of the heart, is regarded as the primary physiological test for heart disease diagnoses. How an echocardiogram is interpreted also relies intensively on the determination of the view. Some of such views are identified as standard views because of the presentation and ease of the evaluations of the major cardiac structures of them. However, finding valid cardiac views has traditionally been time-consuming, and a laborious process because medical imaging is interpreted manually by the specialist. Therefore, this study aims to speed up the diagnosis process and reduce diagnostic error by providing an automated identification of standard cardiac views based on deep learning technology. More importantly, based on a brand-new echocardiogram dataset of the Asian race, our research considers and assesses some new neural network architectures driven by action recognition in video. Finally, the research concludes and verifies that these methods aggregating dynamic information will receive a stronger classification effect.

In order to solve the problems of the existing audio fingerprint method when extracting audio fingerprints from long speech segments, such as too large fingerprint dimension, poor robustness, and low retrieval accuracy and efficiency, a robust audio fingerprint retrieval method based on feature dimension reduction and feature combination is proposed. Firstly, the Mel-frequency cepstral coefficient (MFCC) and linear prediction cepstrum coefficient (LPCC) of the original speech are extracted respectively, and the MFCC feature matrix and LPCC feature matrix are combined. Secondly, the feature dimension reduction method based on information entropy is used for column dimension reduction, and the feature matrix after dimension reduction is used for row dimension reduction based on energy feature dimension reduction method. Finally, the audio fingerprint is constructed by using the feature combination matrix after dimension reduction. When speech's user retrieval, the normalized Hamming distance algorithm is used for matching retrieval. Experiment results show that the proposed method has smaller audio fingerprint dimension and better robustness for long speech segments, and has higher retrieval efficiency while maintaining a higher recall rate and precision rate.

Privacy vulnerability of social networks is one of the major concerns for social science research and business analysis. Most existing studies which mainly focus on un-weighted network graph, have designed various privacy models similar to k-anonymity to prevent data disclosure of vertex attributes or relationships, but they may be suffered from serious problems of huge information loss and significant modification of key properties of the network structure. Furthermore, there still lacks further considerations of privacy protection for important sensitive edges in weighted social networks. To address this problem, this paper proposes a privacy preserving method to protect sensitive weighted edges. Firstly, the sensitive edges are differentiated from weighted edges according to the edge betweenness centrality, which evaluates the importance of entities in social network. Then, the perturbation operations are used to preserve the privacy of weighted social network by adding some pseudo-edges or modifying specific edge weights, so that the bottleneck problem of information flow can be well resolved in key area of the social network. Experimental results show that the proposed method can not only effectively preserve the sensitive edges with lower computation cost, but also maintain the stability of the network structures. Further, the capability of defending against malicious attacks to important sensitive edges has been greatly improved.

The cloud storage provides flexible data storage services for data owners to remotely outsource their data, and reduces data storage operations and management costs for data owners. These outsourced data bring data security concerns to the data owner due to malicious deletion or corruption by the cloud service provider. Data integrity verification is an important way to check outsourced data integrity. However, the existing data verification schemes only consider the case that a verifier launches multiple data verification challenges, and neglect the verification overhead of multiple data verification challenges launched by multiple verifiers at a similar time. In this case, the duplicate data in multiple challenges are verified repeatedly so that verification resources are consumed in vain. We propose a duplicate data verification algorithm based on multiple verifiers and multiple challenges to reduce the verification overhead. The algorithm dynamically schedules the multiple verifiers' challenges based on verification time and the frequent itemsets of duplicate verification data in challenge sets by applying FP-Growth algorithm, and computes the batch proofs of frequent itemsets. Then the challenges are split into two parts, i.e., duplicate data and unique data according to the results of data extraction. Finally, the proofs of duplicate data and unique data are computed and combined to generate a complete proof of every original challenge. Theoretical analysis and experiment evaluation show that the algorithm reduces the verification cost and ensures the correctness of the data integrity verification by flexible batch data verification.

Software-Defined Networking (SDN) has three key features: separation of control and forwarding, centralized control, and network programmability. While improving network management flexibility, SDN has many security issues. This paper systemizes the security threats of SDN using spoofing, tampering, repudiation, information disclosure, denial of service, and elevation of privilege (STRIDE) model to understand the current security status of SDN. First, we introduce the network architecture and data flow of SDN. Second, we analyze security threats of the six types given in the STRIDE model, aiming to reveal the vulnerability mechanisms and assess the attack surface. Then, we briefly describe the corresponding defense technologies. Finally, we summarize the work of this paper and discuss the trends of SDN security research.

SIMON and SPECK are two families of lightweight block ciphers that have excellent performance on hardware and software platforms. At CRYPTO 2019, Gohr first introduces the differential cryptanalysis based deep learning on round-reduced SPECK32/64, and finally reduces the remaining security of 11-round SPECK32/64 to roughly 38 bits. In this paper, we are committed to evaluating the safety of SIMON cipher under the neural differential cryptanalysis. We firstly prove theoretically that SIMON is a non-Markov cipher, which means that the results based on conventional differential cryptanalysis may be inaccurate. Then we train a residual neural network to get the 7-, 8-, 9-round neural distinguishers for SIMON32/64. To prove the effectiveness for our distinguishers, we perform the distinguishing attack and key-recovery attack against 15-round SIMON32/64. The results show that the real ciphertexts can be distinguished from random ciphertexts with a probability close to 1 only by 2^8.7 chosen-plaintext pairs. For the key-recovery attack, the correct key was recovered with a success rate of 23%, and the data complexity and computation complexity are as low as 2⁸ and 2^20.1 respectively. All the results are better than the existing literature. Furthermore, we briefly discussed the effect of different residual network structures on the training results of neural distinguishers. It is hoped that our findings will provide some reference for future research.

This paper presents a privacy-preserving data aggregation scheme deals with the multidimensional data. It is essential that the multidimensional data is rarely mentioned in all researches on smart grid. We use the Paillier Cryptosystem and blinding factor technique to encrypt the multidimensional data as a whole and take advantage of the homomorphic property of the Paillier Cryptosystem to achieve data aggregation. Signature and efficient batch verification have also been applied into our scheme for data integrity and quick verification. And the efficient batch verification only requires 2 pairing operations. Our scheme also supports fault tolerance which means that even some smart meters don't work, our scheme can still work well. In addition, we give two extensions of our scheme. One is that our scheme can be used to compute a fixed user's time-of-use electricity bill. The other is that our scheme is able to effectively and quickly deal with the dynamic user situation. In security analysis, we prove the detailed unforgeability and security of batch verification, and briefly introduce other security features. Performance analysis shows that our scheme has lower computational complexity and communication overhead than existing schemes.

While the conventional vehicle's Head-Units played relatively simple roles (e.g., control of heating ventilation and air conditioning, the radio reception), they have been evolving into vehicle-driver interface with the advent of the concept of Connected Car on top of a rapid development of ICT technology. The Head-Unit is now successfully extended as an IVI (In Vehicle Infotainment) that can operate various functions on multimedia, navigation, information with regards to vehicle's parts (e.g. air pressure, oil gauge, etc.). In this paper, we propose a platform architecture for IVI devices required to achieve the goal as a connected car. Connected car platform (CoCaP) consists of vehicle selective gateway (VSG) for receiving and controlling data from major components of a vehicle, application framework including native and web APIs required to request VSG functionality from outside, and service framework for driver assistance. CoCaP is implemented using Tizen IVI and Android on hardware platforms manufactured for IVI such as Nexcom's VTC1010 and Freescale's i.MX6q/dl, respectively. For more practical verification, CoCaP platform was applied to an real-world finished vehicle. And it was confirmed the vehicle's main components could be controlled using various devices. In addition, by deriving several services for driver assistance and developing them based on CoCaP, this platform is expected to be available in various ways in connected car and ITS environments.

Since the signal with strong power need be demodulated first for successive interference cancellation (SIC) receiver in non-orthogonal multiple access (NOMA) systems, the base station (BS) need inform the near user terminal (UT), which has allocated higher power, of the far UT's modulation mode. To avoid unnecessary signaling overhead of control channel, a blind detection algorithm of NOMA signal modulation mode is designed in this paper. Taking the joint constellation density diagrams of NOMA signal as the detection features, the deep residual network is built for classification, so as to detect the modulation mode of NOMA signal. In view of the fact that the joint constellation diagrams are easily polluted by high intensity noise and lose their real distribution pattern, the wavelet denoising method is adopted to improve the quality of constellations. The simulation results represent that the proposed algorithm can achieve satisfactory detection accuracy in NOMA systems. In addition, the factors affecting the recognition performance are also verified and analyzed.

In this paper, we intend to study the energy efficiency (EE) optimization for a simultaneous wireless information and power transfer (SWIPT)-based distributed antenna system (DAS). Firstly, a DAS-SWIPT model is formulated, whose goal is to maximize the EE of the system. Next, we propose an optimal resource allocation method by means of the Karush-Kuhn-Tucker condition as well as an ergodic method. Considering the complexity of the ergodic method, a suboptimal scheme with lower complexity is proposed by using an antenna selection scheme. Numerical results illustrate that our suboptimal method is able to achieve satisfactory performance of EE similar to an optimal one while reducing the calculation complexity.

As the use of smart devices is being increased rapidly by the development of internet and IT technology, the contents production and utilization rate are showing higher increase, too. In addition, the type of contents also shows very diverse forms such as education, game, video, UCC, etc. In the meantime, the contents are reproduced in diverse forms by reprocessing the original contents, and they are being serviced through the contents service platform. Therefore, the platform to make the contents reprocessing easy and fast is needed. As the diverse contents distribution channels such as YouTube, SNS, App Service, etc, easier contents distribution platform is needed, and the development of the relevant area is expected. In addition, as the selective consumption of the contents having easy accessibility through diverse smart devices is distinguished, the demand for the platform and service that can identify the contents consumption propensity by individual is being increased. Therefore, in this study, to vitalize the online contents distribution, the contents reproduction and publishing platform, was designed and materialized, which can reproduce and distribute the contents based on the real-time contents editing technology in URL unit and the consumer propensity analysis technology using the data management-based broadcasting contents distribution metadata technology and the edited image contents streaming technology. In addition, in the results of comparing with other platforms through the experiment, the performance superiority of the suggested platform was verified. If the suggested platform is applied to the areas of education, broadcasting, press, etc, the multi-media contents can be reproduced and distributed easily, through which the vitalization of contents-related industry is expected.

This paper proposes an augmented reality technique to generate acrobatic scenes from hitting motion videos. After a user shoots a motion that mimics hitting an object with hands or feet, their pose is analyzed using motion tracking with deep learning to track hand or foot movement while hitting the object. Hitting position and time are then extracted to generate the object's moving trajectory using physics optimization and synchronized with the video. The proposed method can create videos for hitting objects with feet, e.g. soccer ball lifting; fists, e.g. tap ball, etc. and is suitable for augmented reality applications to include virtual objects.

Vocal detection is one of the fundamental steps in musical information retrieval. Typically, the detection process consists of feature extraction and classification steps. Recently, neural networks are shown to outperform traditional classifiers. In this paper, we report our study on how to improve detection accuracy further by carefully choosing the parameters of the deep network model. Through experiments, we conclude that a feature-classifier model is still better than an end-to-end model. The recommended model uses a spectrogram as the input plane and the classifier is an 18-layer convolutional neural network (CNN). With this arrangement, when compared with existing literature, the proposed model improves the accuracy from 91.8% to 94.1% in Jamendo dataset. As the dataset has an accuracy of more than 90%, the improvement of 2.3% is difficult and valuable. If even higher accuracy is required, the ensemble learning may be used. The recommend setting is a majority vote with seven proposed models. Doing so, the accuracy increases by about 1.1% in Jamendo dataset.

Cyber powers around the world are conducting cyber information-gathering activities in cyberspace, a global domain within the Internet-based information environment. Accordingly, it is imperative to obtain the latest information through the cyber intelligence preparation of the battlefield (IPB) process to prepare for future cyber operations. Research utilizing the cyber battlefield visualization method for effective cyber IPB and situation awareness aims to minimize uncertainty in the cyber battlefield and enable command control and determination by commanders. This paper designed architecture by classifying cyberspace into a physical, logical network layer and cyber persona layer to visualize the cyber battlefield using BGP archive data, which is comprised of BGP connection information data of routers around the world. To implement the architecture, BGP archive data was analyzed and pre-processed, and cyberspace was implemented in the form of a Di-Graph. Information products that can be obtained through visualization were classified for each layer of the cyberspace, and a visualization method was proposed for performing cyber IPB. Through this, we analyzed actual North Korea's BGP and OSINT data to implement North Korea's cyber battlefield centered on the Internet network in the form of a prototype. In the future, we will implement a prototype architecture based on Elastic Stack.

With the rapid development of information technology, it is now possible to analyze the spatial patterns of cultivated land and its evolution by combining GIS, geostatistical analysis models and spatiotemporal big data for the dynamic monitoring and management of cultivated land resources. The spatial pattern of cultivated land and its evolutionary patterns in Luoyang City, China from 2009 to 2019 were analyzed using spatial autocorrelation and spatial autoregressive models on the basis of GIS technology. It was found that: (1) the area of cultivated land in Luoyang decreased then increased between 2009 and 2019, with an overall increase of 0.43% in 2019 compared to 2009, with cultivated land being dominant in the overall landscape of Luoyang; (2) cultivated land holdings in Luoyang are highly spatially autocorrelated, with the 'high-high'-type area being concentrated in the border area directly north and northeast of Luoyang, while the 'low-low'-type area is concentrated in the south and in the municipal area of Luoyang, and being heavily influenced by topography and urbanization. The expansion determined during the study period mainly took place in the Luoyang City, with most of it being transferred from the 'high-low'-type area; (3) elevation, slope and industrial output values from analysis of the bivariate spatial autocorrelation and spatial autoregressive models of the drivers all had significant effects on the amount of cultivated land holdings, with elevation having a positive effect, and slope and industrial output having a negative effect.

Road damage detection is important for road maintenance. With the development of deep learning, more and more road damage detection methods have been proposed, such as Fast R-CNN, Faster R-CNN, Mask R-CNN and RetinaNet. However, because shallow and deep layers cannot be extracted at the same time, the existing methods do not perform well in detecting objects with fewer samples. In addition, these methods cannot obtain a highly accurate detecting bounding box. This paper presents a Multi-level Feature Pyramids method based on M2det. Because the feature layer has multi-scale and multi-level architecture, the feature layer containing more information and obvious features can be extracted. Moreover, an attention mechanism is used to improve the accuracy of local boundary boxes in the dataset. Experimental results show that the proposed method is better than the current state-of-the-art methods.

With the advent of 5G communications, internet of vehicles technology has been widely used in vehicles. Then the dynamic spread of information between vehicles began to come into focus with more research. It is well known that the identification of nodes with great spread influence has always been a hot topic in the field of information spreading. Most of the existing work measures the propagation influence by degree centrality, betweenness centrality and closeness centrality. In this paper, we will identify influential vehicle nodes based on the mobility characteristics of vehicles to explore the information spreading between vehicles in VANETs. Different from the above methods, we mainly explore the influence of the radius of gyration and vehicle kilometers of travel on information spreading. We use a real vehicle trajectory data to simulate the information transmission process between vehicles based on the susceptible-infected-recovered SIR model. The experimental results show that the influence of information spreading does not enhance with increasing radius of gyration and vehicle kilometers of travel. The fact is that both the radius of gyration and the distance travelled have a significant influence on information spreading when they are close to the median. When the value of both is large or small, it has little influence on information spreading. In view of this results, we can use the radius of gyration and vehicle kilometers of travel to better facilitate the transmission of information between vehicles.