Highlights

    Please wait a minute...
    For Selected: Toggle Thumbnails
    Research Progress on Key Technologies in the Cooperative Vehicle Infrastructure System
    LIN Hongyi, LIU Yang, LI Shen, et al
    Journal of South China University of Technology(Natural Science Edition)    2023, 51 (10): 46-67.   DOI: 10.12141/j.issn.1000-565X.230200
    Abstract4917)   HTML30)    PDF(pc) (1996KB)(415)       Save

    With the steady growth of urban car ownership, the issue of traffic congestion is becoming increasingly prominent, bringing great pressure to urban development. To respond effectively to this challenge, it is critical to develop methods that can improve transport efficiency and reduce energy consumption. In current context, the Cooperative Vehicle Infrastructure System (CVIS), an ideal solution for realizing green and intelligent transportation systems, has become an important direction in both transportation research and practice. By integrating and optimizing various traffic resources, CVIS not only enhances traffic efficiency and reduces energy consumption but also provides key technical support for achieving “dual carbon” goals. This paper thoroughly analyzed the fundamental concepts, research methodologies and application scenarios of CVIS, and delved into its four core technological modules: fusion perception, driving cognition, autonomous decision-making, and cooperative control. The paper reviewed and summarized research achievements within these modules, ranging from traditional methods to the latest in deep reinforcement learning techniques. It also explored the potential applications of these technologies and methods for enhancing traffic efficiency, reducing energy consumption, and improving road safety. Finally, the paper scrutinized numerous challenges that CVIS may encounter in practical applications, including the security of information transmission, system stability, and environmental complexity. To overcome these challenges, the paper looked forward to the future development in four areas: developing datasets that integrate vehicle-side and roadside information, enhancing the fusion accuracy of multi-source perception information, improving the real-time performance and safety of CVIS, and optimizing multi-vehicle cooperative decision-making control methods under complex conditions. As a result, this paper not only has important reference value for the advancement of CVIS technology, but also provides important guidance for the future planning and construction of urban transportation systems.

    Table and Figures | Reference | Related Articles | Metrics | Comments0
    Conflict-Free Path Planning For Multi-AGVs in Automated Terminals Considering Road Load Balancing
    WEN Huiying, YUAN Yuqing, LIN Yifeng
    Journal of South China University of Technology(Natural Science Edition)    2023, 51 (10): 1-10.   DOI: 10.12141/j.issn.1000-565X.230227
    Abstract1806)   HTML22)    PDF(pc) (3304KB)(326)       Save

    With the increasing demand for container transportation and the widespread application of new information technologies, the automation of container terminal operations has become the main development trend in domestic and international ports. It can not only effectively improve the efficiency and safety of terminal operations, but also significantly reduce the demand for human resources and the operational costs. The horizontal transportation system is an essential part of the container terminal handling system and an important link enabling the highly efficient container transportation between the quayside and the storage yard, so its operational reliability and the reasonableness of the scheduling directly affect the operational efficiency of the automated container handling system. The mostly used horizontal transportation equipment in container terminals is the automated guided vehicles (AGVs), which is responsible for horizontal transportation from the front quay crane to the rear yard in automated container terminals. In actual operation process, conflicts and congestion is inevitable when multiple AGVs operate simultaneously. On this basis, this paper used conflict-based search (CBS) to solve the conflict problem arising from the cooperative operation of multi-AGVs at the terminal. The upper layer algorithm searched for conflicts among AGVs, while the lower layer algorithm used the A* algorithm for path planning of AGVs. A load factor was introduced into the heuristic function of the A* algorithm in order to avoid congestion in the path planning and achieve load balancing on terminal roads. Further, a sliding time window conflict resolution (STWCR) based on CBS was adopted to improve computational efficiency for multiple AGVs path planning in the continuous operation scenario of multiple task points at the terminal. Simulation experiments verified that the proposed algorithm in this paper can effectively solve the conflict problem of multiple AGVs path planning at the terminal, while balancing the road network load, alleviating local road congestion, and improving the utilization of road resources. The research results of this paper provide a reference for the optimization of the horizontal transportation system in automated container terminals.

    Table and Figures | Reference | Related Articles | Metrics | Comments0
    A Bandwidth Allocation Method of AVB Traffic Based on Link Load Balancing in TSN
    LU Yiqin, XIONG Xin, WANG Meng, et al.
    Journal of South China University of Technology(Natural Science Edition)    2023, 51 (11): 1-9.   DOI: 10.12141/j.issn.1000-565X.220769
    Abstract228)   HTML17)    PDF(pc) (2536KB)(231)       Save

    As a new generation technology of Ethernet, time sensitive networking (TSN) plays an increasingly important role in industrial control, vehicle network and other fields, as it guarantees low-delay and low-jitter transmission of time sensitive traffic. As one of the key shaping technologies of TSN, credit-based shaping (CBS) guarantees the deterministic transmission of audio video bridging (AVB) traffic by reserving bandwidth. The existing bandwidth allocation methods based on network calculus mostly do not consider the impact of routing on the schedulability of AVB traffic, and the bandwidth allocation results are poor and the solution time is long when the network is in a large scale. Therefore, this paper proposed a bandwidth allocation method based on link load balancing. Firstly, the link load balancing routing algorithm was used to calculate the optimal path for each AVB traffic. Then, based on the flow path and network calculus, the arrival and service curves of each switch’s outbound port traffic were analyzed to obtain the worst-case forwarding delay. Finally, the optimization objectives and constraints for bandwidth allocation were established, a heuristic algorithm was used to solve bandwidth allocation, and the bandwidth parameter configuration was optimized during the solving process. The experimental results show that the AVB traffic bandwidth allocation method based on link load balancing can improve the schedulability of AVB traffic by 15 to 45 percentage points, compared with existing bandwidth allocation methods. Optimizing parameter configuration can increase the bandwidth solving speed by more than twice and obtain better bandwidth allocation results, which can effectively cope with large-scale dynamic changes in TSN networks.

    Table and Figures | Reference | Related Articles | Metrics | Comments0
    A Self-Supervised Pre-Training Method for Chinese Spelling Correction
    SU Jindian, YU Shanshan, HONG Xiaobin
    Journal of South China University of Technology(Natural Science Edition)    2023, 51 (9): 90-98.   DOI: 10.12141/j.issn.1000-565X.230031
    Abstract219)   HTML20)    PDF(pc) (946KB)(158)       Save

    Although the pre-trained language models like BERT/RoBERTa/MacBERT can learn the grammatical, semantic and contextual features of characters and words well through the language mask model MLM pre-training task, they lack the ability to detect and correct spelling errors. What’s more, they faces the problem of inconsistency between the pre-training and downstream fine-tuning stages in Chinese spelling correction CSC task. In order to further improve BERT/RoBERTa/MacBERT’s ability of spelling error detection and correction, this paper proposed a self-supervised pre-training method MASC for CSC, which converts the prediction of masked words into recognition and correction of misspelled words on the basis of MLM. First of all, MASC expands the normal word-masking in MLM to whole word masking, aiming to improve BERT’s ability of learning semantic representation at word-level. Then, the masked words are replaced with candidate words from the aspects of the same tone, similar tone and similar shape with the help of external confusion set, and the training target is changed to recognize the correct words, thus enhancing BERT’s ability of detecting and correcting spelling errors. Finally, the experimental results on three open CSC corpora, sighan13, sighan14 and sighan15, show that MASC can further improve the effect of the pre-training language model, i.e. BERT/RoBERTA/MacBERT, in downstream CSC tasks without changing their structures. Ablation experiments also confirm the importance of whole word masking, phonetic and glyph information.

    Reference | Related Articles | Metrics | Comments0
    Operational Reliability Optimization Strategies of Multi-type Bus Lines
    ZHAO Xiaomei, ZHU Xiangyuan, WANG Qin, et al
    Journal of South China University of Technology(Natural Science Edition)    2023, 51 (8): 32-39.   DOI: 10.12141/j.issn.1000-565X.220493
    Abstract1461)   HTML32)    PDF(pc) (894KB)(1351)       Save

    The unstable running time and the difficulty to accurately estimate the arrival time reduce passengers’ satisfaction with bus traveling. With the diversification of cities, bus line types are gradually becoming more diverse, and the variations in bus operating time of the various bus line types cause trouble in bus scheduling and inconvenience for passengers. To solve above problems, firstly, this paper used K-means ++ algorithm to cluster the stations with the sum of squares of error (SSE) as the measurement index. The operation characteristics and reliability influencing factors of different types of bus lines were taken into consideration to determine the alternative station sets of limited-stop bus service. Then, a joint optimization strategy model of limited-stop bus service and speed regulation was proposed to determine the limited-stop bus station, the departure time, and the running speed of all buses. This model took the minimization of bus operation cost, passenger travel cost, and reliability cost as the objectives, and took the constraints of operation process and headway of limited-stop bus service and all-stop bus service into consideration. Finally, genetic algorithm was used to solve the optimization model and Beijing bus system was selected for case analysis. The results show that the limited-stop bus service can reduce the operation cost of public transport, the speed regulation strategy can better lower the passenger travel cost and reliability cost, and the joint optimization strategy of limited-stop bus service and speed regulation can effectively reduce the total cost of the bus system.

    Table and Figures | Reference | Related Articles | Metrics | Comments0
    Real-Time Template Matching Method for Edge Features
    WANG Shiyong, QIAN Guokang, LI Di, et al.
    Journal of South China University of Technology(Natural Science Edition)    2023, 51 (9): 1-10.   DOI: 10.12141/j.issn.1000-565X.220745
    Abstract655)   HTML21)    PDF(pc) (5488KB)(2058)       Save

    Template matching is a common key technology in the field of machine vision. Currently, edge feature-based template matching methods are facing challenges such as time-consuming searching and low matching accuracy in a complex environment. In order to ensure the robustness while improving the real-time performance, this paper proposed a real-time edge feature-based template matching method. Firstly, in the stage of template creation, a new edge sparse method was proposed, and it can screen out the strong invariant edge points from the template image. It reduces the redundancy of template information while retaining the key template features to ensure the stability and improve the computing efficiency. Secondly, in the stage of pyramid search-based image-matching, a top-level pre-screening method was proposed. Normalized Manhattan distance was used as a constraint to exclude incorrect target poses from the top search results to speed up the search in subsequent layers. Five datasets with different working conditions were constructed, and the proposed template matching method was compared and applied to the fast visual dispensing process for free plane pose. The experimental results show that the proposed matching method can significantly improve the matching speed while ensuring high accuracy. And it can overcome interference factors such as illumination change, rotation, defects, multiple targets, and occlusion, enabling practical applications that require both high robustness and real-time performance.

    Table and Figures | Reference | Related Articles | Metrics | Comments0
    Reconfigurable GNSS RF Receiver for High-Precision Positioning and Orientation
    LI Bin, WANG Riyan, CHEN Zhijian, et al
    Journal of South China University of Technology(Natural Science Edition)    2023, 51 (8): 89-97.   DOI: 10.12141/j.issn.1000-565X.220177
    Abstract778)   HTML30)    PDF(pc) (3689KB)(451)       Save

    Aiming at the problem of multiple types of requirements and large volume, this paper introduced a highly integrated and reconfigurable Global Navigation Satellite System (GNSS) Radio Frequency (RF) receiver for high-precision positioning and orientation of satellite navigation. By adopting four reconfigurable reception channels and receiving full frequency band GNSS signals in parallel, it achieved a single chip supporting high-precision positioning or orientation applications, significantly reducing the volume and cost of navigation terminals. To improve broadband signal reception, the paper proposed a new type of inductorless high linear low noise transconductance amplifier (LNTA). It can eliminate the use of source and load inductors, reduce gain and noise fluctuations when operating in navigation signals at different frequency points, and it is conducive to the reconfiguration of multimode and multi frequency reception and reduces the power consumption of LNTA. A novel IQ phase compensation method was proposed to address the issue of IQ phase imbalance. A programmable switch array with variable impedance was directly designed on the clock controlled latch path of the binary frequency division circuit. By changing the delay time of the 25% duty cycle orthogonal LO, the corresponding branch output LO phase adjustment was realized, achieving calibration of IQ imbalance and improving the image rejection rate (IRR). Testing data shows that the RF receiver achieves full band signal coverage of GNSS from 1.15 to 1.65 GHz, a minimum noise figure of 2.7 dB, and an output third-order intermodulation point power of 34.7 dBm. Adopting a low intermediate frequency and zero intermediate frequency reconfigurable architecture, it can flexibly receive multimode GNSS signals with a bandwidth of 0.8~80 MHz. By compensating for IQ imbalance and improving channel layout, 58.1 dB IRR and 57 dB channel isolation can be achieved, effectively reducing the impact of image interference and inter channel interference. Under a 1.2 V power supply, the power consumption of the receiving channel is only 24.7 mW, which can meet the high integration and diversified application requirements of high-precision positioning and orientation GNSS RF receivers.

    Table and Figures | Reference | Related Articles | Metrics | Comments0
    Effect of Particle Size on Hydration Kinetics and Microstructure Development of Recycled Brick Powder-Cement Pastes
    YU Qijun, MA Ting, ZHANG Tongsheng, et al.
    Journal of South China University of Technology(Natural Science Edition)    2023, 51 (11): 63-73.   DOI: 10.12141/j.issn.1000-565X.220784
    Abstract382)   HTML12)    PDF(pc) (6244KB)(192)       Save

    With the advancement of urbanization and the transformation of the old city, the production and stock of construction waste in China are ever increasing, with waste clay bricks accounting for 50%~70%. It has been found that recycled brick powder has the potential as a supplementary cementitious material, but it can lead to a significant reduction in the mechanical properties of cement-based materials. In order to explore the effect of particle size on the activity of recycled brick powder and the hydration kinetics of cement, this study prepared recycled brick powder with different particle sizes by high-energy ball milling. The physical and chemical properties and hydration activity of recycled brick powder were characterized. The effect of particle size of recycled brick powder on the hydration process, microstructure and mechanical properties of recycled brick powder-silicate cement system was analyzed. Based on the Krstulovic-Dabic model, the hydration kinetic parameters of the system were obtained to realize the quantitative evaluation of the hydration process. The results show that with the decrease of the particle size of the recycled brick powder, the lattice distortion of the silicon-aluminum mineral becomes larger and the surface binding energy decreases, resulting in an increase in its hydration activity. The early hydration of recycled brick powder mainly plays a physical filling role, which can accelerate the early hydration of recycled brick powder-cement system and improve the hydration degree of crystallization nucleation and crystal growth→phase boundary reaction→diffusion process. With the decrease of particle size of recycled brick powder, the starting time of pozzolanic reaction is advanced and the degree of pozzolanic reaction is higher. Finally, the later strength of cement mixed with 30% fine-grained recycled brick powder exceeds that of pure cement. This paper lays a foundation for the efficient application of recycled brick powder in cement-based materials.

    Table and Figures | Reference | Related Articles | Metrics | Comments0
    Strengthening and Toughening Process Method of 30CrMnSiA Thin-Wall Cup-Shaped Parts
    XIA Qinxiang, XIE Zhangxiong, CHEN Can, et al
    Journal of South China University of Technology(Natural Science Edition)    2023, 51 (1): 1-7.   DOI: 10.12141/j.issn.1000-565X.220075
    Abstract1958)   HTML30)    PDF(pc) (3521KB)(94)       Save

    30CrMnSiA thin-wall cup-shaped part is a key basic component widely used in the flexible gear of harmonic reducer. Aiming at the problem of the low bearing capacity and short service life of harmonic reducer caused by insufficient strength and toughness of 30CrMnSiA thin-wall cup-shaped parts manufactured by traditional turning method, this paper proposed a plastic deformation-heat treatment process which consists of spinning, quenching, tempering, spinning and aging to manufacture 30CrMnSiA thin-wall cup-shaped parts with less or no cutting and excellent mechanical properties. Through the tensile and impact experiment, by comparing the mechanical properties of each process part, the microorganization of each process part was analyzed. The results show that tempered sorbite microstructure with high strength can be obtained by spinning-quenching and tempering process, but the plasticity was reduced significantly. The fibrous microstructure of spun parts can be further refined by aging heat treatment, and the fine carbides precipitated and uniformly distributed on the ferrite matrix. High strength and good plasticity can be obtained by subsequent aging at 300 ℃ for 6 h. As compared with parts obtained by turning forming after quenching and tempering heat treatment, the yield strength and tensile strength of 30CrMnSiA thin-wall cup-shaped parts manufactured by plastic deformation and heat treatment are improved by 93.65% and 47.88%, respectively. The hardness is increased by 26.87%, and the impact strength is increased by 12.01%. Meanwhile, the elongation and percentage reduction of area are 11.60% and 24.64%, respectively. The thin-wall cup-shaped parts with high strength and toughness can be manufactured by the plastic deformation and heat treatment process of spinning-quenching-tempering-spinning-aging, which provides a new method for manufacturing the thin-wall cup-shaped parts with high-strength and toughness.

    Table and Figures | Reference | Related Articles | Metrics | Comments0
    Study on CMC/CS/Ca2+ Composite Hydrogel for Transdermal Delivery of SPE in vitro and Its Antioxidant Activity
    YE Jun, LIU Luying, HUANG Zhuosheng, et al
    Journal of South China University of Technology(Natural Science Edition)    2023, 51 (1): 51-60.   DOI: 10.12141/j.issn.1000-565X.220388
    Abstract566)   HTML23)    PDF(pc) (3419KB)(235)       Save

    Peptides, such as soybean peptide (SPE), play an important part in drugs targeted therapy and diagnosis. However, their intolerance for the extremely low pH and ionic barriers of the stomach makes them hard to be delivered orally. Transdermal delivery system can deliver drugs to the blood avoiding the obstacles of the gastric. In this paper, carboxymethyl cellulose/cationic starch/Ca2+ (CMC/CS/Ca2+) composite hydrogel was successfully prepared by freeze-thaw cycle, and it can overcome the barrier of penetration of water-soluble SPE in the stratum corneum during transdermal delivery. The results of particle charge detector (PCD), Fourier transform infrared spectroscopy (FT-IR) and rheology show that chemical and physical cross-linking occurs among CMC, CS and Ca2+, forming a stable hydrogel. Moreover, the interaction among SPE, CS and Ca2+ makes its mechanical properties meet the requirements of transdermal delivery. The results of scanning electron microscope (SEM) and specific surface area measurement (BET) illustrate that there is a porous 3D network structure in those composite hydrogels. The moisture content of the composite hydrogel is higher than 95.63% and remains above 97.34% after 25 min at 25 ℃. All composite hydrogels swell rapidly within 120 min, and the equilibrium swelling fraction can reach up to 636.23% at 720 min. Moreover, it is also indicated that the composite hydrogels have high affinity for water, which is the fundamental reason for improving the transdermal delivery efficiency of SPE. The results of in vitro release show that the release type of SPE releasing from hydrogel belongs to swelling controlled release type, and the maximum scavenging rates of DPPH radical and ABTS radical are 47.07% and 72.16%, respectively. The cumulative permeation per unit area of SPE is 0.55, 0.70 and 0.60 mg/cm2, separately, and the cumulative penetration percentage is 84.13%, 92.93% and 92.57%, respectively. The appearance and pH value, thermostability and cold endurance of the hydrogel all meet the standards of QB/T 2872—2017. Therefore, these composite hydrogels are expected to be applied not only in transdermal delivery of peptides and protein drugs, but also in skin antioxidant care.

    Table and Figures | Reference | Related Articles | Metrics | Comments0
    Smart Contract Vulnerability Detection Method Based on Capsule Network and Attention Mechanism
    LU Lu, LAI Jinxiong
    Journal of South China University of Technology(Natural Science Edition)    2023, 51 (5): 36-44.   DOI: 10.12141/j.issn.1000-565X.220167
    Abstract503)   HTML20)    PDF(pc) (1557KB)(233)       Save

    In recent years, with the increasing number of smart contracts and the increasing economic losses caused by contract loopholes, the security of smart contracts has attracted more and more attention. The vulnerability detection method based on deep learning can solve the problems of low detection efficiency and insufficient accuracy of the early traditional smart contract vulnerability detection method. However, most of the existing deep learning-based vulnerability detection methods directly use smart contract source code, opcode sequence or bytecode sequence as the input of the deep learning model. This fact will weaken the effective information due to the introduction of too much invalid information. To solve this problem, this paper proposed a smart contract vulnerability detection method based on capsule network and attention mechanism. Considering the execution timing information of the program, the study extracted key operation code sequence of the smart contract as the source code feature. Then a hybrid network structure of capsule network and attention mechanism was used for training. The capsule network extracts the context information of the smart contract and the connection between the part and the whole; while the attention mechanism is used to assign different weights to different opcodes according to their importance. The experimental results show that the F1 score and accuracy of the algorithm proposed in this paper in the smart contract data set are 94.48% and 97.15%, indicating that this algorithm is superior to other detection methods in performance.

    Table and Figures | Reference | Related Articles | Metrics | Comments0
    Knowledge Graph Completion Method Based on Interactively Connected Graph Attention Network
    LU Yiqin, PAN Zhoushuang, ZHANG Yang, et al
    Journal of South China University of Technology(Natural Science Edition)    2022, 50 (12): 13-19.   DOI: 10.12141/j.issn.1000-565X.220384
    Abstract1657)   HTML20)    PDF(pc) (1179KB)(362)       Save

    Knowledge graph provides underlying support for many intelligent information service applications, including intelligent search, public safety, finance, medical care and other fields. However, the existing knowledge graph is usually incomplete, and knowledge graph completion has become an urgent problem to be solved. The existing knowledge graph completion method models ignore the important information rich in neighbor nodes and relationships, and often simply splice neighbor nodes and relationships together, ignoring the different importance of different relationships and neighbor nodes to nodes. To solve this problem, this paper proposed a knowledge graph completion method (ICGAT) based on the interactive connection graph attention network. The method firstly finds out the potential relationship by finding two-hop neighbor nodes, and expands the triples of each node. Then it fuses the relationship in each triplet with the features of the node, and adopts the method of interactive connection between nodes and neighbor nodes, using 4 space vectors to represent the interactively connected relationship. Finally, the vector of interactive connection was input into the graph attention network to obtain the weight of relations and neighbor nodes to the node, so as to illustrate its importance. In order to effectively represent triples of complex relationships such as one-to-many, many-to-many, etc., this method used the RotatE model as a pre-training model. The experimental results in the link prediction task show that the performance of the mean rank and HR@10indicators of the ICGAT method in the WN18RR and FB15k-237 datasets have been improved to a certain extent, indicating that ICGAT can improve the accuracy of the link prediction task.

    Table and Figures | Reference | Related Articles | Metrics | Comments0
    Security-Aware Scheduling Method for Time-Sensitive Networking
    LU Yiqin, XIE Wenjing, WANG Haihan, et al
    Journal of South China University of Technology(Natural Science Edition)    2023, 51 (5): 1-12.   DOI: 10.12141/j.issn.1000-565X.220394
    Abstract695)   HTML23)    PDF(pc) (2301KB)(168)       Save

    The authenticity of information is the key security factor of system in time-sensitive networking (TSN). However, the direct introduction of traditional security authentication mechanism will lead to a significant reduction in schedulability of the system. The existing methods still have the problems of few application scenarios and high resource consumption. To address this problem, a security-aware scheduling method for TSN was proposed. Firstly, based on the traffic characteristics of TSN, a time-efficient one-time signature security mechanism was designed to provide efficient multicast source authentication for messages. Secondly, the corresponding security model was proposed to evaluate the mechanism and describe the impact of the security mechanism on tasks and traffic. Finally, the proposed security-aware scheduling method was modeled mathematically. On the basis of traditional scheduling constraints, some constraints related to security mechanisms were added. At the same time, the optimization objective was to minimize the end-to-end delay of applications, and constraint programming was used to solve the problem. Simulation results show that the introduction of the improved one-time signature mechanism can effectively protect the authenticity of key information in TSN, and has limited impact on scheduling. In multiple test cases of different sizes generated based on real industrial scenarios, the average end-to-end delay and bandwidth consumption of the generated applications only increased by 13.3% and 5.8% respectively. Compared with other similar methods, this method consumes less bandwidth, thus more suitable for TSN networks with strict bandwidth restrictions.

    Table and Figures | Reference | Related Articles | Metrics | Comments0
    Beamspace Channel Estimation Algorithm Based on Deep Compressed Sensing
    ZHENG Juanyi, MU Jinyu, XING Lirong, et al
    Journal of South China University of Technology(Natural Science Edition)    2022, 50 (12): 101-108.   DOI: 10.12141/j.issn.1000-565X.220017
    Abstract1996)   HTML25)    PDF(pc) (1825KB)(533)       Save

    In the millimeter-wave (mmWave) massive multiple-input multiple-output (MIMO) system with lens antenna array, because the radio frequency (RF) link is much less than the number of antennas, it is necessary to recover the high-dimensional channel from the low-dimensional effective measurement signal by channel estimation. The current channel estimation methods basically make use of the sparsity of the beamspace channel, transforming the channel estimation into compressed sensing problem and then estimating with different methods. Aiming at the limitation that approximate message passing (AMP) algorithm needs channel prior information in channel estimation, this paper proposed an improved channel estimation algorithm. Firstly, a new noise term was derived based on the AMP algorithm and fitted with a convolutional neural network (CNN). Then the iterative denoising process was expanded into a deep network to solve the linear inverse transformation of the measurement signal to the cha-nnel. Finally, the initially estimated channel was further optimized by a residual noise removal network. In addition, the controllable parameters were introduced to increase the flexibility of the channel estimation process, and the sen-sing matrix was jointly trained with other network parameters to improve the channel estimation accuracy. This paper verified the proposed algorithm from two aspects of channel estimation accuracy and system transmission quality, and carried out the theoretical formula derivation and system simulation analysis on the Saleh-Valenzuela channel model. Simulation results show that the proposed algorithm has less model parameters and computation than the traditional algorithm, and can improve the accuracy of channel estimation and the transmission quality of the communication system.

    Table and Figures | Reference | Related Articles | Metrics | Comments0
    Study on Cutting Forces and Machining Performance in Turning Stainless Steel with Variable-Length Restricted Contact Tool
    PANG Xueqin, DENG Wenjun, LI Songqing
    Journal of South China University of Technology(Natural Science Edition)    2022, 50 (8): 49-61.   DOI: 10.12141/j.issn.1000-565X.210741
    Abstract201)   HTML13)    PDF(pc) (8231KB)(72)       Save

    As a typical difficult-to-machine material, AISI 316L austenitic stainless steel has the problems of low machining efficiency and large thermo-mechanical load during machining. The present methods to improve the machining performance of stainless steel such as adopting lubrication, designing tool surface structure, etc., all exist obvious defects. For example, the extensive use of cutting fluid is detrimental to the environment and human health, and tool wear will be aggravated if the tool surface structure is designed unreasonably. Given that the existing methods fail to well deal with the problems like large thermo-mechanical load during machining austenitic stainless steel, a new restricted contact tool with variable contact length was designed aiming at machining AISI 316L auste-nitic stainless steel. Compared with traditional restricted contact tools, the new tool is superior in reducing cutting force and friction coefficient. This paper first developed a semi-analytical model for turning to predict cutting forces for inconstant restricted contact length based on the non-equidistant shear zone model and the principles of unified cutting mechanics. Then, the conventional and variable-length restricted contact pattern (rectangular and trapezoidal restricted contact pattern) was fabricated on the rake surface of uncoated cemented carbide by the W-EDM and micro-EDM process. Subsequently, extensive experiments were performed to validate the proposed model. Results show that the cutting forces predicted by the proposed semi-analytical model are in good agreement with the experimental values. As compared with conventional restricted contact tools, the developed restricted contact tool can significantly reduce cutting forces. Furthermore, the influences of conventional and developed restricted contact tools on cutting performance were compared and analyzed under dry machining conditions. The restricted contact tool developed in this paper can effectively decrease the cutting temperature and tool wear. This study can provide a basis for designing the cutting tools in machining difficult-to-cut materials.

    Reference | Related Articles | Metrics | Comments0
    A cross-modal face retrieval method based on metric learning
    WO Yan, LIANG Jiyun, HAN Guoqiang
    Journal of South China University of Technology(Natural Science Edition)    2022, 50 (6): 1-9.   DOI: 10.12141/j.issn.1000-565X.210709
    Abstract2767)      PDF(pc) (2022KB)(649)       Save
    Metric learning is an important technique to reduce modal differences. Existing cross-modal retrieval methods based on metric learning for cross-modal face retrieval tasks lack attention to pose differences and domain differences, and there are two problems in the process of metric learning: lack of learning of global information and the existence of a large number of redundant triplets. In this paper, a cross-modal common representation generation algorithm based on metric learning is proposed. Our study uses the yaw angle equivariant module to compensate for yaw angle differences so that we can obtain the image features with robustness, uses the multi-layer attention mechanism to obtain video features with differentiability; combines global triplets and local triplets to jointly train the cross-modal common representation generation network, then accelerates the convergence of the loss function through the screening of semi-hard triplets; combines domain calibration and transfer learning to improve the generalization of common representations. Finally, the results of comparison experiments on three face video datasets: PB, YTC and UMD Faces, demonstrate that our algorithm can improve the accuracy of cross-modal face retrieval, and the results of fine-tuning the cross-modal common representation generation network using different numbers of samples demonstrate that our algorithm can improve the accuracy of cross-modal retrieval of target domain images.
    Related Articles | Metrics | Comments0
    Composition and Optimization Method of Coordination Path Set in Control Subarea
    LU Kai, WU Wei, DENG Xingdong, et al
    Journal of South China University of Technology(Natural Science Edition)    2022, 50 (2): 1-14.   DOI: 10.12141/j.issn.1000-565X.210162
    Abstract1146)      PDF(pc) (4388KB)(242)       Save
    To achieve the refinement of urban traffic signal coordination control, the selection of coordinated object should not just focus on the subarea, but needs to be further refined to the attachments of direction paths. In this paper, first, a concept system reflecting the directional demand of coordination control was established by defining the coordination path, the coordination path chain and the coordination path set, as well as the parent path and sub path which constitute the path chain. Then, the counting method of path chains was analyzed. Moreover, according to the trajectory distribution of vehicles in the road network, the corresponding optimization rules and processes of coordination path set were established from the perspective of the intersections number (namely the length of the path chain) and the traffic flow carried by the path chain. The results of example analysis show that the proposed method can select the main path chains with large flow in the control area according to the number of intersections to be coordinated and the flow of traffic path, which can further enrich the division theory of control sub area and intersection group.
    Related Articles | Metrics | Comments0
    Analysis of Effect Factors in the Synchronization Process of Multi-cone Friction Pair
    WANG Yanzhong, DOU Delong, ZHANG Zhen, et al
    Journal of South China University of Technology(Natural Science Edition)    2022, 50 (2): 58-66,83.   DOI: 10.12141/j.issn.1000-565X.210119
    Abstract1071)      PDF(pc) (3255KB)(234)       Save
    Based on the lubrication mechanics and the friction principle of asperity, a mathematical model describing the synchronization process of multi-cone friction pair is established and numerically solved. Then, the change laws of the relative speed of the driving and driven plates, the thickness of the oil film, the bearing capacity of the oil film, the bearing capacity of the micro peak, the viscous shear torque, the rough friction torque, etc., with the synchronization time during the synchronization process, are analyzed. Moreover, the effects of such factors as the cone angle, the number of friction cones, the surface roughness, the initial relative speed and the loading pressure on the synchronization process are investigated based on the established mathematical model. The results show that (1) the multi-cone friction pair has an obvious wedge-shaped self-energizing effect and can provide large cone contact pressure with small axial loading force| (2) the synchronous torque can be increased and the synchronization time can be shortened by reducing the cone angle and by increasing the number of friction cones, the surface roughness and the axial loading force| and (3) the synchronization time is roughly linear to the initial relative speed.
    Related Articles | Metrics | Comments0
    Effects of Fermentation and Polyphenol Complex on the Digestibility of Rice Starch
    LI Xiaoxi, SHEN Shaodan, LU Ping, et al
    Journal of South China University of Technology (Natural Science Edition)    2022, 50 (1): 1-8.   DOI: 10.12141/j.issn.1000-565X.210103
    Abstract3481)      PDF(pc) (2212KB)(362)       Save
    In order to improve the anti-digestibility of rice starch,the change law of digestibility and multi-scale structures of the fermented rice starch-proanthocyanidins complex,including the particle morphology,lamellar structure,crystalline structure,helical structure and short-range ordered structure,were systematically studied with modern analytical techniques and the method of fermentation and interaction with proanthocyanidins.Meanwhile,the molecular mechanism of rice starch anti-digestibility controlling by fermentation and combination with proanthocyanidins was revealed.The results indicates that the rapidly digestible starch and slow digestible starch contents of the fermented rice starch after interacting with proanthocyanidins are decreased,whereas the resistant starch contents are significantly increased.Moreover,the interaction between proanthocyanidins and fermented rice starch can promote the formation of specific ordered structure.And partial proanthocyanidins in the amorphous region can be released to inhibit the action of amylase during the digestion process,which synergisticly reduces the digestibility of fermented rice starch.The results provides foundations for improving the nutritional functions of fermented rice products.
    Related Articles | Metrics | Comments0
    Obstacle Avoidance Algorithm for Unmanned Aerial Vehicle Vision Based on Deep Learning
    ZHANG Xiangzhu, ZHANG Lijia, SONG Yifan, et al
    Journal of South China University of Technology(Natural Science Edition)    2022, 50 (1): 101-108, 131.   DOI: 10.12141/j.issn.1000-565X.210096
    Abstract4595)      PDF(pc) (3578KB)(977)       Save
    In order to solve the obstacle avoidance problem of unmanned aerial vehicle (UAV) based on monocular vision, a quadrotor autonomous obstacle avoidance method based on monocular depth estimation and object detection was proposed.The monocular depth estimation model provides the pixel-level depth information of the obstacle, and the object detection model provides the location information of the obstacles.The depth map and object detection results of a single Red-Green-Blue image were obtained by convolutional neural network (CNN).The region division of the image was based on the object detection results, and the region depth was calculated based on the depth estimation results.The linear velocity and angular velocity of UAV were calculated by the planning algorithm based on the regional depth and regional division results, so as to realize the autonomous obstacle avoidance of UAV.In order to verify the autonomous obstacle avoidance performance of the algorithm, the Parrot Bebop2 UAV was employed to carry out real flight comparison experiments between the proposed algorithm and the direct flight algorithm.The results show that the proposed algorithm can be used for low speed autonomous obstacle avoi-dance of quadrotor.
    Related Articles | Metrics | Comments0
News
 
Featured Article
Most Read
Most Download
Most Cited