Abstract:A music generation method based on variable neighborhood search and masking effect was proposed to overcome the shortcomings of traditional composing methods, such as high threshold and repeatable work. Compared with the existing music generation model, the proposed method is intuitive to operate and can excavate the consistency of music scores over a long period of time. In the training stage, the time domain waveform files were chosen as the input, which were easy to acquire. The music score was obtained by frequency extraction based on pitch significance and time domain block based on masking effect, and the music parameters were pre-set for training. In the generation stage, the method of combining variable neighborhood search with manual operation was adopted. By adding, deleting, and changing the notes in the music score, the artistic music score was generated according to the section, and the music score was manually modified and locked according to the preference of the operator. Then, the music segments were generated in four sections at a time and combined into a long-lasting music segment. Finally, the music score was perfected at different time scales, and the time domain waveform file was output. The idea of visualization and human-computer interaction throughout the whole process could fully mobilize the enthusiasm of music enthusiasts. Besides, a parametric musical instruments simulation method is proposed to modify the music score or generate a new timbre, in which two parts of parametric simulation of musical instruments and human voice can be carried out through sound source excitation and resonant cavity simulation according to the classification of musical instruments or the process of human voice production.

Abstract:In order to improve the visible light catalytic activity of traditional TiO₂ photocatalyst, Mn, N co-doped TiO₂ photocatalyst was synthesized by sol-gel method, which improved the problem of easy recombination of photogenerated carriers. Further, zeolite was used as a carrier to solve the problem of difficult separation and recovery of the conventional photocatalyst, achieving the purpose of reusable. The crystal phase structure, element composition, microtopography, and photocatalytic activity of TiO₂ photocatalyst were characterized by XRD, SEM, UV-vis, and FTIR. Results show that Mn, N co-doped TiO₂/zeolite composites had higher activity than pristine TiO₂. The degradation rate of malachite green in the TiO₂ photocatalyst obtained under optimal doping conditions could reach 97% within 60 min in visible light, which was because the manganese ion doping could inhibit the recombination of TiO₂ photogenerated carriers and promote charge separation. Meanwhile, nitrogen doping could effectively broaden the photoresponse range of TiO₂. Moreover, after five usage cycles of Mn, N co-doped TiO₂/zeolite composites, the degradation rate of malachite green was not greatly reduced, maintaining above 88%, which indicates that the Mn, N co-doped TiO₂/zeolite composites have good photocatalytic cycle stability.

Abstract:When evaluating 2-way constrained covering arrays, the minimum sizes given by the existing methods are too large because the constraints are ignored. In order to obtain more accurate minimum sizes of 2-way constrained covering arrays and evaluate the 2-way constrained covering arrays generated by existing algorithms, a forbidden edge decomposing method was proposed to promote lower bounds of the minimum sizes of 2-way constrained covering arrays in this study. The graphs that describe the input configurations of the systems under test were decomposed into two subgraphs. Through calculating the sizes of the sub constrained covering arrays which cover all the vertices in the two subgraphs and the number of the remaining value combinations to be covered, the minimum sizes of the 2-way constrained covering arrays were promoted compared with the number of the value combinations to be covered. The lower bounds of the minimum sizes obtained from the proposed method were closer to the real values. If the sizes of the 2-way constrained covering arrays are less than the lower bounds of the minimum sizes, then it is impossible for them to exist. The proposed experimental method applied the forbidden edge decomposing method to the existing systems under test, then obtained the lower bounds of the minimum sizes of 2-way constrained covering arrays, and compared the lower bounds of the minimum sizes with the sizes of the 2-way constrained covering arrays given by the generation algorithms. The experimental results show that the lower bounds of the minimum sizes given by the forbidden edge decomposing method can be used to evaluate the 2-way constrained covering arrays generated by the existing algorithms, and are helpful to determine their existence.

Abstract:The quaternion MVDR (Q-MVDR) algorithm has better performance than the traditional complex-domain MVDR algorithm due to the inherent orthogonality between the dipole element components in quaternion signal model of the polarization sensitive array. However, when it comes to the situation where strong desired signal and steering vector are mismatched, the performance of Q-MVDR will be degraded significantly, which can even aggravate the signal self-nulling effect. To address this problem, a robust beamforming algorithm based on quaternion matrix reconstruction was proposed. First, a quaternion signal model of polarization sensitive array was established, and the covariance matrix reconstruction method was extended to the quaternion domain. The steering vector and power of interference signals were obtained by subspace method, and the Capon spectral estimator was adopted to reconstruct the interference-plus-noise covariance matrix. Then, according to the orthogonality of signal subspace and noise subspace, as well as the characteristics that the desired signal vector and signal subspace belong to the same subspace, an improved version of steering vector mismatch correction method was introduced, which utilized the projection technique of weight vector to eigenspace signal subspace. Finally, the effectiveness of the algorithm was verified by numerical simulations. Simulation results show that the proposed technique had better robustness than traditional Q-MVDR algorithm in effectively avoiding the performance degradation caused by signal self-nulling effect, and it is able to provide similar SINR close to the optimal value in both strong desired signal and steering vector mismatch cases.

Abstract:In order to suppress interference caused by the dense deployment of small cell base stations in an ultra-dense network (UDN) and improve system throughput, this paper investigates the power allocation strategy for a spectrum sharing UDN. Firstly, the noncooperative game was adopted to transform the non-convex system sum-rate maximization problem into several convex subproblems that maximize the utility function of each user. By designing a dynamic pricing, each Nash equilibrium point (NE) of the game was a stationary point of the original optimization problem. Secondly, an interference power constraint was applied to guarantee the quality-of-service (QoS) of the macrocell. Finally, under the game theory framework, an iterative global information-based power allocation algorithm was designed. The optimal transmit power of each user could be obtained by solving KKT condition during each iteration, and the proposed iterative algorithm was proved to be convergent to the NE of the game model. In addition, in order to reduce the signaling overhead and improve the resource utilization, a power allocation algorithm based on local information was also proposed. Simulation results show that the proposed global information-based power allocation algorithm achieved a better transmission performance than benchmark methods and the proposed local information-based power allocation algorithm effectively reduced signaling overhead while guaranteed good transmission performance.

Abstract:Traditional communication system fails to utilize the energy of radio frequency signals effectively, which can result in high transmission energy consumption. Therefore, resources must be allocated reasonably to achieve a compromise between information and energy transmission. That is to ensure sufficient transmission energy and improve the information transmission performance of the system as much as possible. In order to save the energy of the communication system, this paper proposes that the communication system simultaneously transmits information and energy in multiple time slots to realize Simultaneous Wireless Information and Power Transfer (SWIPT). First, the transmission time was divided into a number of time slots, and the transmitter transmitted information or energy in different time slots as required. Then, the receiver collected information or harvested energy in the corresponding time slot, and the obtained energy was used to provide circuit energy consumption. Finally, by jointly optimizing the time slot and power allocation, the system transmission rate was maximized based on the guaranteed system energy requirements. Simulation results show that compared with the equal-time power allocation algorithm, the transmission efficiency of the proposed algorithm was improved by about 40 bps, and the performance was significantly improved with the increase of the number of the time slots. When the transmission energy increased, the distributable information power was greater. When the energy demand increased, the information power decreased, and more power was used for energy transmission. Hence, an algorithm was proposed to maximize the performance of information transmission by reasonably allocating time slots and power for information transmission and energy harvesting to ensure sufficient acquisition energy. Therefore, the proposed algorithm could improve the transmission performance effectively via harvesting the wireless power.

Abstract:In order to solve the problems that it is difficult to integrate the nonlinear Mises specific plastic power and that the corresponding total rolling power is hard to be calculated analytically, this paper establishes the expression of an linear specific plastic power to analyze the energy of the proposed elliptical velocity field and obtains an analytical solution of rolling force and energy parameters. In this paper, a new yield criterion linearly combined of principle stress components is established by calculating the integral mean value of the variable angle yield function. Its locus on the π-plane is a dodecagonal shape with equal sides and unequal angles, and its Lode parameter expression result was in good agreement with the experimental data. Meanwhile, according to the characteristic that the metal flow velocity increases gradually from the entrance to the exit of heavy plate, a velocity field whose horizontal velocity component satisfies the elliptic equation was proposed, which meets the kinematic admission condition. With the rolling energy analysis, the internal deformation power based on the proposed linear yield criterion, as well as the friction power and the shear power based on the strain vector internal product method were obtained. On this basis, the analytical solutions of rolling torque, rolling force, and stress state coefficient were obtained by the extreme variation of functional, which were compared with the measured data. Results show that the rolling torque and the rolling force obtained by using the yield criterion and the velocity field proposed in this paper were in good agreement with the measured values, where the rolling force error was less than 5.3% and the rolling torque error was about 6%.

Abstract:In metal additive manufacturing (AM), rapid material temperature changes and complicated material behavior inevitably lead to various types of surface and internal defects, which are detrimental to the performance of final parts, and hinder the development and application of metal AM in key areas. Acquiring defect information in time is beneficial for adjusting process parameters, improving manufacturing process, and elevating parts quality. Meanwhile, the defect information can be utilized to conduct subsequent machining and processing. Therefore, it is of significance to conduct research on defect detection technologies to improve AM technology and expand its applications. To resolve this issue, this paper introduces the common defects in AM parts and summarizes their origins. Then typical studies on AM defect detection technologies in recent years are reviewed, with a focus on the monitoring of process signatures and the online non-destructive testing methods. The application scope of different methods is also presented. Finally, the development trend of defect detection technology is forecasted. A conclusion is drawn that defects in AM are complicated and the parts quality needs to be assured by detection methods and technologies. However, methods proposed so far still cannot satisfy the requirement of AM, and the technology should be developed towards integration and intelligence in the future.

Abstract:A new phase-to-height mapping algorithm with loose constraints based on phase measuring profilometry is proposed to improve the measurement accuracy of PMP and relax the constraints of its system. The algorithm requires neither the linking line between the optic centers of the projection system and the imaging system to be parallel to the reference plane, nor the projecting optic axis and the imaging optic axis to be on the same plane. Just by making sure that the imaging optic axis is vertical to the reference plane can the phase-to-height mapping be successfully realized. The mathematic model of the algorithm was deduced in detail, and it was found that the phase-to-height mapping was not only related to the structure parameters of the 3D measurement system but also to the phase distribution of reference plane. In actual measurement, it is difficult to guarantee that the linking line between the optical centers of the projection system and the imaging system is parallel to the reference plane, but the new phase-to-height mapping algorithm avoids this challenge, thus providing a true reflection of the actual measurement system. It also makes the experimental system easier to build. Experimental results show that the mean absolute errors and the root mean square errors of the proposed method were no more than 0.031mm and 0.040 mm, respectively. The algorithm could relax the constraints when building the measurement system and make the beam path easier to realize. The effectiveness of the proposed algorithm was verified by the experiment, indicating that the proposed algorithm is superior to the traditional algorithm in terms of accuracy and repeatability.

Abstract:Botnets choose blockchain networks (such as Bitcoin and Ethereum) as the communication channel for their command-and-control (C&C) mechanism because blockchain networks are anonymous and hard to shut down. Recent research focuses on this mechanism, but the research methods have such defects that the scalability is restricted, and the upstream channel is vulnerable to existing tracing techniques. To solve these problems, D-BitBot, a peer-to-peer (P2P)-based duplex botnet model which utilizes the Bitcoin testnet as the upstream channel is proposed in this paper. The C&C channel used in this model is hard to trace and reduces the cost of data recovery and network scalability. To avoid single point of failure in traditional botnet bootstrap procedure, a Bitcoin blockchain based bootstrap mechanism is presented. Further, to defend against direct routing table poisoning and P2P botnet crawling, a novel peer list exchange algorithm based on the sorted hash values of IP addresses and random salt values is proposed. According to the result of P2P simulation, D-BitBot provided robust network connectivity with an online rate of 100%. In the node request and node crawling algorithm, the proposed algorithm was effective against direct routing table poisoning and reduced the node detection rate of the current crawling algorithm. Lastly, possible countermeasures and the robustness of the proposed C&C channel were discussed at the end of this paper.

Abstract:The traditional flash translation layer (FTL) algorithm of spaceborne storage system adopts page-level FTL mapping scheme and fixed partition file management strategy, which has many problems, such as high host occupancy, long system response time, and no consideration of FLASH wear balance. Therefore, a data-driven Adaptive Superblock FTL (DASFTL) algorithm was proposed based on the analysis of the working principle of traditional spaceborne storage system and spaceborne solid state storage system. DASFTL adopts the hierarchical address mapping scheme of adaptive superblock, in which the super block mapping table is the first level mapping and the page mapping table is the second level mapping for improving response time. The superblock was taken as the smallest unit of FLASH address management to reduce the dependence of storage system on host. Dynamic block recovery weight was introduced as the criterion of superblock grouping and target recovery block selection to balance the wear degree of the physical blocks in FLASH chip and prolong its service life. In the end, a hardware test platform was built to verify the DASFTL algorithm. Results show that compared with the traditional spaceborne FTL algorithm, the host occupancy rate and the system response time of the DASFTL algorithm were increased by 51.7% and 46.1%, respectively. The balanced wear performance of the FLASH chip was significantly improved, which can prolong the service life of the FLASH chip.

Abstract:In the application of image compression, to generate effective common bitmap and reduce the distortion risk of color image compressed by single bitmap block truncation coding while remaining compression ratio, a single bitmap block truncation coding method based on binary fireworks algorithm is proposed. First, the color image was divided into non-overlapping blocks, and the initial bitmap of each block was generated by the weight plane method. Then, the positions that need to be optimized in the initial bitmap of each sub-image block were determined by two different strategies, and the values of these positions were used as the initial values of the fireworks algorithm. Next, the fireworks algorithm was changed to binary form and optimized to generate a common bitmap and six quantization values for each block. Finally, each block was restored according to the common bitmap and the quantization values, and the color image was reconstructed by the restored blocks. Through the experiments on the test images, the proposed method was compared with three reference methods from the aspects of the detailed visual effects of the compressed images, the mean square error, and the structural similarity between the compressed images and the original images. Results show that the common bitmap generated by the proposed method was effective, and the global optimization strategy was better than the local optimization strategy. The mean values of the mean square errors between the compressed images generated by the global optimization strategy and the original images were 56.939 7 and 106.317 4 when the block size was 4×4 and 8×8, which were lower than those of the three reference methods. The mean values of the structural similarity index values between the compressed images generated by the global optimization strategy and the original images were 0.968 2 and 0.943 1 when the block size was 4×4 and 8×8, which were higher than those of the three reference methods. It indicates that the similarity between the compressed images generated by the proposed method and the original images is higher, and the accuracy of the compressed images is effectively improved while maintaining the compression ratio.

Abstract:Most of the physical phenomena in the nature exist in the form of fractional order, and integer order equation is just a special case of fractional order equation. Compared with the integer order model, the fractional order model is closer to the real world and has a more attractive development prospect. In order to enrich the types of multi-wing chaotic attractors that coexist in fractional chaotic systems, a novel 3D fractional chaotic system is proposed in this paper. The most significant feature of the system is that there is a coexistence of multiple types of multi-wings chaotic attractors, namely the coexistence of two-wing, three-wing, and four-wing chaotic attractors. The dynamic characteristics of the system were analyzed via numerical simulations of phase diagram, Lyapunov exponent spectrum, and bifurcation diagram. A necessary condition for the existence of chaotic attractors was given, which is q>0.822 4. For fixed system parameters, when q=0.98, there is a coexistence of two-wing, two-wing, and four-wing chaotic attractors in the system; when q=0.83, there is a coexistence of two-wing, three-wing, and four-wing chaotic attractors, which indicates that the chaotic characteristics of the system are complex. The simulation results of an analog circuit of the system by the Multisim were consistent with those by the numerical analysis, which further verified the chaotic behaviors of the system. Based on the fractional Lyapunov stability theory and Theorem 1, an adaptive synchronous controller was designed. The simulation shows that the response system and the drive system reached the synchronization within the range of 0.2 s, and the identification of unknown parameters was completed within the range of 0.2 s, thus the controller is effective.

Abstract:Visual attention modeling is a key technique for predicting the distribution of human attention when people are observing scenes, which is widely used in the fields of computer vision. Traditional visual attention models focus on the human eyes fixation points to reflect the eye movement information by calculating saliency maps, while they cannot reflect the perceived semantic information of the brain. To solve this problem, a visual attention model was proposed based on extracting semantic features. First of all, the eye tracking database VOC2012-E was established to study and record the eye movement data of human while observing natural scenes. Then, inspired by image semantic segmentation, the Fully Convolutional Networks(FCN) was used to extract the semantic object features. In order to extract the semantic object features more effectively, the FCN8s network was improved by activation function PReLu and optimization function Adam to mimic the brain’s perception of semantic object features. Next, 28 low-level features such as direction, color, and intensity characteristics were extracted, which attract attention in the human subconscious layer. Finally, Support Vector Machine(SVM) was used to map the previously extracted semantic object features and the low-level features into the human visual space. The real eye movement data was introduced for supervised training, and a visual attention model was obtained which can predict the human visual saliency map. Experimental results showed that the visual attention model proposed in this paper had better performance and biological advantages over the other eight classical models and four advanced models on the VOC2012-E and MIT300 databases.

Abstract:To ensure the rationality of urban taxi stands layout and improve the situation of low utilization rate, a location-decision method of taxi stands based on GPS trajectory data was proposed. The demand for taxi trips and locations of candidate stands were estimated under the background of data. Considering the factors such as the coverage of passenger travel demand, the maximum acceptable distance for passengers, and the service capacity of stand and parking space, the constraint conditions for taxi stands layout were constructed. A bi-objective location optimization model was proposed to minimize construction costs and maximize public satisfaction from the perspectives of planners and users, respectively. The forbearing stratified sequencing method based on genetic algorithm was applied to solve the model. Taking the Chunxi business district of Chengdu as an example, the taxi stands in the area were re-located using the proposed model, and the impact of different expected travel distances of passengers on location decision problem was investigated. Results show that when the expected travel distance of passenger was less than 300 m, the new layout scheme could cover all travel demands in the area with a reasonable number of stands and ensure the passengers’ travel satisfaction at a high level, which verifies the validity of the bi-objective location model.

Abstract:Emoticons have become an important component of network language and is one of the main characteristics of the analysis of social media sentiment. The current social media sentiment analysis methods most focus on Emoji, while there is no study on the sentiment trend of kinesics. In order to obtain the multi-dimensional sentiment polarity of Chinese social media and analyze the group sentiment trend on hot topics, this paper proposes a new multi-dimensional sentiment classification method based on deep learning, which combines Emoticons with short texts. In this framework, the text and Emoji combination and the kinesics in microblog sentences were analyzed using deep learning model, and seven sentiment intensities of the two parts were obtained to explore the correlation between each part and sentiment labels. Then, a computational model was designed to reflect the multi-dimensional sentiment polarity contained in microblog sentences, which can realize the detection of the multi-dimensional sentiment intensity of sentences. The experiment utilized the NLPCC2014 dataset and the crawled microblog dataset containing kinesics for verification. Results show that when the proportion of the text and Emoji combination and the kinesics were 0.6 and 0.4, the effect of sentiment classification was the best. The sentiment classification performance indicator of the sentences containing kinesics was always higher than that without kinesics, which indicates that the combination of Emoticon and short texts can effectively improve the accuracy of microblog sentiment detection. The experiment provides a more fine-grained analysis for group sentiment trend and a new idea for Chinese social media sentiment analysis.

Abstract:To effectively reduce the effects of cluttered background on traditional shape-based object detection methods, a novel object detection algorithm based on contour matching was introduced, which combines the methods of saliency detection and template matching. First, the input image was preprocessed at the super pixel level to obtain the saliency region map without background by saliency feature detection. Then, the edge detection algorithm was applied to get the edge image in the saliency region, and the shape descriptor was used for contour matching after optimizing the edge image. Finally, a depth-first search strategy was applied to identify the hypothetical location of the object and perform hypothesis verification to determine the final location of the object. The experimental result in the ETHZ shape dataset proved the feasibility of this algorithm. Compared with other shape-based methods under the 50%-IoU and 20%-IoU evaluation criteria, according to the data results, the average detection rate of different categories was 96% when the false positive per image (FPPI) was 0.3. The detection rate was 99% when the FPPI was 0.4, and the detection rate would reach 100% if higher FPPI is accepted, which were all higher than the other algorithms. The experimental and comparative analysis results show that the proposed method could improve the detection accuracy and had obvious performance advantages, providing a new solution for object detection in cluttered background.

Abstract:The lever arm effect superimposed on the accelerometer sensitive axis of the attitude and heading reference system (AHRS) under vibration environment will greatly affect the accuracy of the attitude measurement based on the accelerometer. This paper analyzes the low-pass filtering method and the mechanical compensation method in the lever arm effect processing algorithms. The low pass filtering method will have rectification error residual. The compensation of the centripetal acceleration term in the lever arm effect by the mechanical compensation method is limited by the gyro measurement accuracy. Meanwhile, the difference operation of angular velocity will greatly enlarge the compensation error of tangential acceleration term in lever arm effect. Aiming at the shortcomings of the low-pass filtering method and the mechanical compensation method, this paper proposes an improved compensation method of lever-arm effect. For the centripetal acceleration term, the angular frequency threshold is set as decision value. When the angular frequency is lower than the given angular frequency threshold, the low-pass filtering method is adopted; and when the angular frequency is higher than the given angular frequency threshold, the mechanical compensation+low-pass filtering method is adopted. For the tangential acceleration term, the angular acceleration can be solved analytically by multiple accelerometers, which can avoid the error amplification caused by differential operation. Through the theoretical derivation, analysis and simulation, the improved compensation method significantly improves the compensation precision of the lever arm effect, compared with the low-pass filtering method and the mechanical compensation method.

Abstract:The aim of multi-focus image fusion technology is to produce an all-in-focus image in which the clear parts of different source images are integrated to a single image. Most of the existing fusion methods still suffer from the problems such as block artifacts, artificial edges, halo effects, ringing effects, and contrast reduction. To address these problems, a multi-focus image fusion method using the energy of Laplacian CNN is proposed in this paper. The focus information of source images can be extracted effectively by using Laplacian energy operator, and the focus feature extracted from focus information maps by the trained convolutional neural network model can effectively distinguish focused sub-blocks from defocused sub-blocks. The trained convolutional neural network model not only has a good ability to extract the relative focus degree of active windows, but also can obtain an accurate segmentation boundary. After multiple rounds of training, the convolutional neural network model can well establish an effective mapping between source images and a score map, which is essential to generate an accurate focus map. Then, the focus map is further modified using binary segmentation and small region filtering, and the final decision map for fusion is obtained. Finally, according to the weights provided by the final decision map, the final fusion image will be formed by fusing multiple source images. The experimental results show that the proposed method is superior to other existing fusion methods in terms of visual effects and quantitative evaluation.

Abstract:Titanium alloys forming at room temperature has attracted increasing attention due to its low processing cost. As is known, titanium alloys with different microstructures present different forming properties. The forming limit diagram is the basis to study the material failure and forming ability in the numerical simulation of titanium alloy plate forming at room temperature. However, to obtain forming limit requires numerous experiments, which are expensive and time-consuming. In order to reduce the experimental cost and shorten the development cycle, a feasible progression-dependent method based on the finite element (FE) simulation was adopted to determine the process from localized necking to fracture, which was also used to predict the forming limit diagrams of TA17 titanium alloy plate with different microstructures. The FE simulation results of the forming limit were compared with those of uniaxial tensile test, punch bulging test, and round cup drawing test. The results verified the correctness of the FE simulation results. Moreover, the forming limits of three TA17 titanium alloys with different microstructures were compared, and the effect of microstructure on their forming limits at room temperature was analyzed. Results show that the TA17 titanium alloy with the equiaxed microstructure provided the best performance on the forming limit, which was also supported by the fractographs of the samples. This work sheds light on the research on the forming properties of titanium alloys microstructures produced by different heat treatment processes, and establishes a foundation for the subsequent work on the stamping of TA17 titanium alloys.

Abstract:To improve the property of M-ary chaotic communication system and develop the chaotic synchronization technique of Duffing chaotic system, a quaternary chaotic digital communication system using Duffing oscillator is presented in this paper. The binary chaotic modulation method with Duffing oscillator was combined with Quadrature Amplitude Modulation (QAM) to realize quaternary chaotic modulation with Duffing oscillator. Meanwhile, according to the demodulation theory of Duffing oscillator, the zone partition was simplified, and the chaotic signal demodulator based on Duffing oscillators array as well as the quaternary chaotic digital receiver based on Duffing oscillator was constructed, which realized the quaternary chaotic digital communication system design based on Duffing oscillator. Simulation results show that the quaternary chaotic modulation could be achieved by the quaternary chaotic signal transmitter. The simplified zone partition could recognize the phase trajectory of the Duffing oscillator with easier realization of structure and algorithm. The quaternary chaotic digital receiver based on Duffing oscillator could realize noncoherent demodulation of the quaternary Duffing chaotic signal, and results show that the BER property of the chaotic system was better than that of the classical quaternary chaotic systems. Therefore, the quaternary chaotic digital communication system based on Duffing oscillator introduced in this paper simplifies the chaotic modulation method and noncoherent demodulation with Duffing oscillator, which provides reference for the construction of M-ary chaotic communication system.

Abstract:With the development of the GPS spoofing signal generating technology, the spoofing signal can be similar to the authentic signal in such attributes as arrival angle and Doppler frequency, making it difficult to detect the spoofing signal by only employing one spoofing interference detection method. While if directly combining several methods, the detection performance will be decreased when one detector is degraded. Thus, considering that under different deception jamming scenarios, the detection performance of detectors has real-time changes and the priori information is missing, a detection fusion method is proposed in this paper to detect spoofing signals by evaluating real-time performance of detectors. By establishing the evaluation criteria for performance error, the real-time performance comparison between the proposed method and the single detector method was conducted. With this method, the degraded detector will be rejected if the performance falls down below the threshold, which can ensure that when a detector fails, it will not affect the overall performance of the fusion. This method can be applied to the fusion of any number and any kind of deception detection methods. Based on the detection methods of phase difference and carrier Doppler variance, an engineering implementation method is presented in this paper. Finally, simulation results showed that the spoofing detection fusion algorithm for real-time performance evaluation had better deception performance and wider adaptability than one detector and traditional detection fusion algorithms.

Abstract:In order to study the full-range failure process of bar-wrapped cylinder concrete pressure pipe (BCCP) under internal water pressure, tests on three prototype BCCPs with different diameters were designed. Annular strain gauges were arranged on the cylinder, steel bar, and concrete protective cover of the BCCP with a diameter of 1 800 mm, and the force response of each part under the action of internal water pressure was obtained by increasing the applied internal water pressure to 2.5 MPa step by step. It was found that the load-bearing process of BCCP from prestressing to ultimate failure could be divided into 5 stages. First, the concrete core was subjected to the force of the prestressed steel bar without protective cover. After being wrapped, the core formed by cylinder and concrete was subjected to an initial pre-compression stress. Second, the whole pipe was in the elastic stage before the protective cover cracked. It corresponded to the stage when the internal water pressure in the test was less than 1.5 MPa, and the core was still under pressure, while the prestressed steel bar and the outer concrete protective cover were under tension. Third, the concrete core was in the elastic stage after the protective cover cracked. When the internal water pressure arrived at 1.6 MPa, the cover reached the tensile strength and began to crack, and the concrete core also transformed from the initial compression to tension slowly and remained in elastic state. Fourth, the cylinder and the steel bar were in the elastic stage of tension after the concrete core cracked. As the water pressure arrived at 2.2 MPa, the concrete core cracked radially, but the stress of cylinder and bars still increased steadily with the internal water pressure. Fifth, the pipe was completely destroyed. The cylinder and the steel bars reached the ultimate yield strength, and the whole pipe lost its bearing capacity. The research results could provide basis for the popularization and application of BCCP in water transfer engineering and the formulation of relevant standards.

Abstract:Collaborative filtering is a simple recommendation method which uses related knowledge, but its performance is poor when the data is highly sparse. Factorization machines (FM) can solve the problem of feature combination in the case of data sparsity. Combining with high-order feature extraction via deep neural networks, a series of deep learning prediction models have been proposed and good results have achieved. However, these models mainly depend on the combination of a large number of labels and the understanding of high-order features, whose performance can be seriously degraded when the label categories of data are scarce. In order to solve the problem of recommendation in sparse data and scarce labels, a novel neural network-based recommendation model was proposed which embeds multi-domain labels into a unified domain. First, labels were divided by domain and transformed into feature vectors through embedding layer. Then, the mapping layer was used to embed the feature vectors from the current domain into the unified domain. Finally, the spatial relations of the unified domain-embedded vectors were calculated and predictions were made. Experiments on several open datasets show that the proposed model had higher accuracy and better performance than the mainstream neural network based predicting models. The model overcomes the training bottleneck when label is scarce, and provides a solution for recommender system with limited original data.

Abstract:Visual fatigue is a fatal traffic safety hazard for drivers in the process of high-speed driving. Therefore, the analysis of the driving visual space of the driver in a quantitative way is an important path to reveal the relationship between driving safety and driving visual behavior. In response to these characteristics, such as a monotonous visual environment and a high proportion of viaducts in the urban expressway, this research adopted the automatic imaging technology of vehicle-mounted isometric and variable-distance for the simulation experiment, and the numerical analysis method is used to analyze the structural characteristics of the driving visual space between the common viaduct section and the zero-interface viaduct section in the urban expressway, to determine the composition and frequency of visual elements in the two typical spaces. Results show that the visual elements of the urban viaduct section are lower than the common urban roads in addition to the sky background. The main visual elements of the viaduct section are the bridge pavement with the sky as the background and the right-side street building, and the distribution of these elements is too balanced and simplification, thereby causing visual fatigue of the driver. In addition, the visual elements of the zero-interface viaduct section are in a state of scarcity for a long time, which is easy to cause negative driving psychology, such as anxiety, panic and instability. The reasonable design of traffic safety facilities to guide the driver's sight at the right time can effectively alleviate the driver's visual fatigue, and increasing the visual elements as the elevation reference is the most direct measure to eliminate the negative driving psychology of the zero-interface viaduct section.

Abstract:Good control handing comfort not only reduces fatigue but also improves efficiency. Aiming at the uncertainty and fuzziness of control handing comfort evaluation, this study builds up a control handing comfort evaluation model for standing posture based on Takagi-Sugeno Fuzzy Neural Network (T-S FNN). Training and testing data were collected during the experiment. Twenty adult test subjects were asked to complete 100 different operation tasks. Test subjects’ joint angle, foot pressure distribution, anthropometric dimensions, target position, and subjective comfort rating were collected during the experiment. The proposed model was trained using 90% of the data obtained from the experiment and was verified by the remaining 10% experiment data. It was then compared with the subjective comfort rating estimated by BP Neural Network. Results show that the proposed model had smaller root mean square error than BP Neural Network (1.2 vs. 4.5). Subsequently, 15 groups of different tasks were randomly selected to further test this model. Results show that the correlation coefficients between the value obtained by this model and the actual value, and those obtained by the Rapid Upper Limb Assessment (RULA) and the Ovako Working Posture Analysing System (OWAS) were 0.962 (P<0.01), 0.833 (P<0.01), and 0.694 (P<0.01), respectively. This study demonstrates that the proposed model is effective in estimating control handing comfort.