High-performance terahertz(THz)logic gate devices are crucial components for signal processing and modulation,playing a significant role in the application of THz communication and imaging.Here,we propose a THz broadb...High-performance terahertz(THz)logic gate devices are crucial components for signal processing and modulation,playing a significant role in the application of THz communication and imaging.Here,we propose a THz broadband NOR logic encoder based on a graphene-metal hybrid metasurface.The unit structure consists of two symmetrical dual-gap metal split-ring resonators(DSRRs)arranged in a staggered configuration,with graphene strips embedded in their gaps.The NOR logic gate metadevice is controlled by the bias voltages independently applied to the two electrodes.Experiments show that when the bias voltages are applied to both electrodes,the metadevice achieves the NOR logic gate within a 0.52 THz bandwidth,with an average modulation depth above 80%.The experimental results match well with theoretical simulations.Additionally,the strong near-field coupling induced by the staggered DSRRs causes redshift at both LC resonance and dipole resonance.This phenomenon was demonstrated by coupled mode theory.Besides,we analyze the surface current distribution at resonances and propose four equivalent circuit models to elucidate the physical mechanisms of modulation under distinct loaded voltage conditions.The results not only advance modulation and logic gate designs for THz communication but also demonstrate significant potential applications in 6G networks,THz imaging,and radar systems.展开更多
Nowadays, increased information capacity and transmission processes make information security a difficult problem. As a result, most researchers employ encryption and decryption algorithms to enhance information secur...Nowadays, increased information capacity and transmission processes make information security a difficult problem. As a result, most researchers employ encryption and decryption algorithms to enhance information security domains. As it progresses, new encryption methods are being used for information security. In this paper, a hybrid encryption algorithm that combines the honey encryption algorithm and an advanced DNA encoding scheme in key generation is presented. Deoxyribonucleic Acid (DNA) achieves maximal protection and powerful security with high capacity and low modification rate, it is currently being investigated as a potential carrier for information security. Honey Encryption (HE) is an important encryption method for security systems and can strongly prevent brute force attacks. However, the traditional honeyword encryption has a message space limitation problem in the message distribution process. Therefore, we use an improved honey encryption algorithm in our proposed system. By combining the benefits of the DNA-based encoding algorithm with the improved Honey encryption algorithm, a new hybrid method is created in the proposed system. In this paper, five different lookup tables are created in the DNA encoding scheme in key generation. The improved Honey encryption algorithm based on the DNA encoding scheme in key generation is discussed in detail. The passwords are generated as the keys by using the DNA methods based on five different lookup tables, and the disease names are the input messages that are encoded by using the honey encryption process. This hybrid method can reduce the storage overhead problem in the DNA method by applying the five different lookup tables and can reduce time complexity in the existing honey encryption process.展开更多
Multimodal image fusion plays an important role in image analysis and applications.Multimodal medical image fusion helps to combine contrast features from two or more input imaging modalities to represent fused inform...Multimodal image fusion plays an important role in image analysis and applications.Multimodal medical image fusion helps to combine contrast features from two or more input imaging modalities to represent fused information in a single image.One of the critical clinical applications of medical image fusion is to fuse anatomical and functional modalities for rapid diagnosis of malignant tissues.This paper proposes a multimodal medical image fusion network(MMIF-Net)based on multiscale hybrid attention.The method first decomposes the original image to obtain the low-rank and significant parts.Then,to utilize the features at different scales,we add amultiscalemechanism that uses three filters of different sizes to extract the features in the encoded network.Also,a hybrid attention module is introduced to obtain more image details.Finally,the fused images are reconstructed by decoding the network.We conducted experiments with clinical images from brain computed tomography/magnetic resonance.The experimental results show that the multimodal medical image fusion network method based on multiscale hybrid attention works better than other advanced fusion methods.展开更多
Fountain codes are considered to be a promising coding technique in underwater acoustic communication（UAC） which is challenged with the unique propagation features of the underwater acoustic channel and the harsh ma...Fountain codes are considered to be a promising coding technique in underwater acoustic communication（UAC） which is challenged with the unique propagation features of the underwater acoustic channel and the harsh marine environment. And Luby transform（LT） codes are the first codes fully realizing the digital fountain concept. However, in conventional LT encoding/decoding algorithms, due to the imperfect coverage（IC） of input symbols and short cycles in the generator matrix, stopping sets would occur and terminate the decoding. Thus, the recovery probability is reduced,high coding overhead is required and decoding delay is increased.These issues would be disadvantages while applying LT codes in underwater acoustic communication. Aimed at solving those issues, novel encoding/decoding algorithms are proposed. First,a doping and non-uniform selecting（DNS） encoding algorithm is proposed to solve the IC and the generation of short cycles problems. And this can reduce the probability of stopping sets occur during decoding. Second, a hybrid on the fly Gaussian elimination and belief propagation（OFG-BP） decoding algorithm is designed to reduce the decoding delay and efficiently utilize the information of stopping sets. Comparisons via Monte Carlo simulation confirm that the proposed schemes could achieve better overall decoding performances in comparison with conventional schemes.展开更多
语篇要素识别在自动作文评分中发挥着重要作用,提高语篇要素识别的准确率有助于增强自动作文评分的效果以及可解释性。然而,语篇要素识别任务面临着上下文依赖和句子歧义性等挑战。传统的基于规则和特征工程的方法难以捕捉文本中复杂的...语篇要素识别在自动作文评分中发挥着重要作用,提高语篇要素识别的准确率有助于增强自动作文评分的效果以及可解释性。然而,语篇要素识别任务面临着上下文依赖和句子歧义性等挑战。传统的基于规则和特征工程的方法难以捕捉文本中复杂的语义信息和长距离依赖关系,而深度学习方法虽然能够自动学习文本特征,但仍然存在对关键位置信息利用不足的问题。针对上述问题,提出了一种混合位置编码的语篇要素识别模型,即HPE-BiLSTM(Hybrid Position Encoding Bidirectional Long Short-Term Memory)。该模型首先基于预训练的词向量获取句子表示,然后通过双向长短期记忆网络提取句子级特征。在句子级特征的基础上,采用混合的位置编码方案以确保关键位置信息的有效传递。最后,使用线性层和激活函数实现语篇要素识别。该模型在议论文数据集进行实验,并与Feature-based、BERT、BiLSTM、DiSA和DCRGNN五个模型进行比较。实验结果表明,HPE-BiLSTM模型的准确率达到了0.693,在语篇要素识别方面的F 1分数为0.684,优于其他模型。展开更多
高精度的海上船舶轨迹预测是降低船舶碰撞风险、提升船舶搜救效率的重要基础.海上航行环境的多变性使船舶轨迹数据在时间和空间上具有高度复杂性,现有方法对船舶轨迹数据的质量及运动信息关注度不足,难以充分捕捉轨迹中的时空特征和关...高精度的海上船舶轨迹预测是降低船舶碰撞风险、提升船舶搜救效率的重要基础.海上航行环境的多变性使船舶轨迹数据在时间和空间上具有高度复杂性,现有方法对船舶轨迹数据的质量及运动信息关注度不足,难以充分捕捉轨迹中的时空特征和关联信息.因此,文中提出融合数据质量增强和时空信息编码网络的船舶海上轨迹预测方法(Ship Maritime Trajectory Prediction Method Integrating Data Quality Enhancement and Spatio-Temporal Information Encoding Network,DQE-STIEN).首先,基于船舶轨迹数据的特征,设计结合哈希映射分类及局部离群哈希值异常检测的数据质量增强算法,对问题数据进行质量增强.然后,针对多属性的船舶轨迹数据,设计具有双编码通道的时空信息编码网络,充分提取并整合船舶轨迹数据中的位置信息与运动特征.最后,基于时空信息编码提取数据中的时空关联信息,并经解码生成完整的轨迹预测结果.在5个不同区域的AIS数据集上的实验表明DQE-STIEN性能较优.同时,DQE-STIEN具有一定的泛化性,也能有效分析能源、销售、环境和金融等领域的时序数据.展开更多
Digital twin(DT)modelling is a prerequisite for the successful application of DT technology in the power industry.However,traditional scene modelling methods are costly,time-consuming,focus on overall features and lac...Digital twin(DT)modelling is a prerequisite for the successful application of DT technology in the power industry.However,traditional scene modelling methods are costly,time-consuming,focus on overall features and lack real-time updates,hindering the interaction between DT models and physical power equipment scenes.Therefore,a scene DT modelling technique focusing on local features in risk areas and real-time updates is urgently needed.Herein,real-time modelling of the±800 kV converter transformer is achieved by improving the neural radiation field based on a hybrid attention mechanism and multiresolution hash encoding.Compared to traditional methods,modelling time is reduced from hours to 1 min without professional equipment or manual intervention.The model quality is more concerned with local features of risk areas in transformers while ensuring the overall scene,and the accuracy is improved by about 6%,realising the real-time modelling of transformers and the DT of scenes.展开更多
Due to the widespread availability of implicit feedback(e.g., clicks and purchases), some researchers have endeavored to design recommender systems based on implicit feedback. However, unlike explicit feedback,implici...Due to the widespread availability of implicit feedback(e.g., clicks and purchases), some researchers have endeavored to design recommender systems based on implicit feedback. However, unlike explicit feedback,implicit feedback cannot directly reflect user preferences. Therefore, although more challenging, it is also more practical to use implicit feedback for recommender systems. Traditional collaborative filtering methods such as matrix factorization, which regards user preferences as a linear combination of user and item latent vectors, have limited learning capacities and suffer from data sparsity and the cold-start problem. To tackle these problems,some authors have considered the integration of a deep neural network to learn user and item features with traditional collaborative filtering. However, there is as yet no research combining collaborative filtering and contentbased recommendation with deep learning. In this paper, we propose a novel deep hybrid recommender system framework based on auto-encoders(DHA-RS) by integrating user and item side information to construct a hybrid recommender system and enhance performance. DHA-RS combines stacked denoising auto-encoders with neural collaborative filtering, which corresponds to the process of learning user and item features from auxiliary information to predict user preferences. Experiments performed on the real-world dataset reveal that DHA-RS performs better than state-of-the-art methods.展开更多
基金supported by the National Natural Science Foundation of China(Grant Nos.62005058 and 62365006)the Natural Science Foundation of Guangxi,China(Grant No.2020GXNSFBA238012)+2 种基金the China Postdoctoral Science Foundation(Grant No.2020M683726)the Innovation Project of Guangxi Graduate Education(Grant Nos.YCSW2024345 and YCBZ2025157)the Guangxi Key Laboratory of Automatic Detecting Technology and Instruments(Grant No.YQ24101).
文摘High-performance terahertz(THz)logic gate devices are crucial components for signal processing and modulation,playing a significant role in the application of THz communication and imaging.Here,we propose a THz broadband NOR logic encoder based on a graphene-metal hybrid metasurface.The unit structure consists of two symmetrical dual-gap metal split-ring resonators(DSRRs)arranged in a staggered configuration,with graphene strips embedded in their gaps.The NOR logic gate metadevice is controlled by the bias voltages independently applied to the two electrodes.Experiments show that when the bias voltages are applied to both electrodes,the metadevice achieves the NOR logic gate within a 0.52 THz bandwidth,with an average modulation depth above 80%.The experimental results match well with theoretical simulations.Additionally,the strong near-field coupling induced by the staggered DSRRs causes redshift at both LC resonance and dipole resonance.This phenomenon was demonstrated by coupled mode theory.Besides,we analyze the surface current distribution at resonances and propose four equivalent circuit models to elucidate the physical mechanisms of modulation under distinct loaded voltage conditions.The results not only advance modulation and logic gate designs for THz communication but also demonstrate significant potential applications in 6G networks,THz imaging,and radar systems.
文摘Nowadays, increased information capacity and transmission processes make information security a difficult problem. As a result, most researchers employ encryption and decryption algorithms to enhance information security domains. As it progresses, new encryption methods are being used for information security. In this paper, a hybrid encryption algorithm that combines the honey encryption algorithm and an advanced DNA encoding scheme in key generation is presented. Deoxyribonucleic Acid (DNA) achieves maximal protection and powerful security with high capacity and low modification rate, it is currently being investigated as a potential carrier for information security. Honey Encryption (HE) is an important encryption method for security systems and can strongly prevent brute force attacks. However, the traditional honeyword encryption has a message space limitation problem in the message distribution process. Therefore, we use an improved honey encryption algorithm in our proposed system. By combining the benefits of the DNA-based encoding algorithm with the improved Honey encryption algorithm, a new hybrid method is created in the proposed system. In this paper, five different lookup tables are created in the DNA encoding scheme in key generation. The improved Honey encryption algorithm based on the DNA encoding scheme in key generation is discussed in detail. The passwords are generated as the keys by using the DNA methods based on five different lookup tables, and the disease names are the input messages that are encoded by using the honey encryption process. This hybrid method can reduce the storage overhead problem in the DNA method by applying the five different lookup tables and can reduce time complexity in the existing honey encryption process.
基金supported by Qingdao Huanghai University School-Level ScientificResearch Project(2023KJ14)Undergraduate Teaching Reform Research Project of Shandong Provincial Department of Education(M2022328)+1 种基金National Natural Science Foundation of China under Grant(42472324)Qingdao Postdoctoral Foundation under Grant(QDBSH202402049).
文摘Multimodal image fusion plays an important role in image analysis and applications.Multimodal medical image fusion helps to combine contrast features from two or more input imaging modalities to represent fused information in a single image.One of the critical clinical applications of medical image fusion is to fuse anatomical and functional modalities for rapid diagnosis of malignant tissues.This paper proposes a multimodal medical image fusion network(MMIF-Net)based on multiscale hybrid attention.The method first decomposes the original image to obtain the low-rank and significant parts.Then,to utilize the features at different scales,we add amultiscalemechanism that uses three filters of different sizes to extract the features in the encoded network.Also,a hybrid attention module is introduced to obtain more image details.Finally,the fused images are reconstructed by decoding the network.We conducted experiments with clinical images from brain computed tomography/magnetic resonance.The experimental results show that the multimodal medical image fusion network method based on multiscale hybrid attention works better than other advanced fusion methods.
基金supported by the National Natural Science Foundation of China(61371099)the Fundamental Research Funds for the Central Universities of China(HEUCF150812/150810)
文摘Fountain codes are considered to be a promising coding technique in underwater acoustic communication（UAC） which is challenged with the unique propagation features of the underwater acoustic channel and the harsh marine environment. And Luby transform（LT） codes are the first codes fully realizing the digital fountain concept. However, in conventional LT encoding/decoding algorithms, due to the imperfect coverage（IC） of input symbols and short cycles in the generator matrix, stopping sets would occur and terminate the decoding. Thus, the recovery probability is reduced,high coding overhead is required and decoding delay is increased.These issues would be disadvantages while applying LT codes in underwater acoustic communication. Aimed at solving those issues, novel encoding/decoding algorithms are proposed. First,a doping and non-uniform selecting（DNS） encoding algorithm is proposed to solve the IC and the generation of short cycles problems. And this can reduce the probability of stopping sets occur during decoding. Second, a hybrid on the fly Gaussian elimination and belief propagation（OFG-BP） decoding algorithm is designed to reduce the decoding delay and efficiently utilize the information of stopping sets. Comparisons via Monte Carlo simulation confirm that the proposed schemes could achieve better overall decoding performances in comparison with conventional schemes.
文摘语篇要素识别在自动作文评分中发挥着重要作用,提高语篇要素识别的准确率有助于增强自动作文评分的效果以及可解释性。然而,语篇要素识别任务面临着上下文依赖和句子歧义性等挑战。传统的基于规则和特征工程的方法难以捕捉文本中复杂的语义信息和长距离依赖关系,而深度学习方法虽然能够自动学习文本特征,但仍然存在对关键位置信息利用不足的问题。针对上述问题,提出了一种混合位置编码的语篇要素识别模型,即HPE-BiLSTM(Hybrid Position Encoding Bidirectional Long Short-Term Memory)。该模型首先基于预训练的词向量获取句子表示,然后通过双向长短期记忆网络提取句子级特征。在句子级特征的基础上,采用混合的位置编码方案以确保关键位置信息的有效传递。最后,使用线性层和激活函数实现语篇要素识别。该模型在议论文数据集进行实验,并与Feature-based、BERT、BiLSTM、DiSA和DCRGNN五个模型进行比较。实验结果表明,HPE-BiLSTM模型的准确率达到了0.693,在语篇要素识别方面的F 1分数为0.684,优于其他模型。
文摘高精度的海上船舶轨迹预测是降低船舶碰撞风险、提升船舶搜救效率的重要基础.海上航行环境的多变性使船舶轨迹数据在时间和空间上具有高度复杂性,现有方法对船舶轨迹数据的质量及运动信息关注度不足,难以充分捕捉轨迹中的时空特征和关联信息.因此,文中提出融合数据质量增强和时空信息编码网络的船舶海上轨迹预测方法(Ship Maritime Trajectory Prediction Method Integrating Data Quality Enhancement and Spatio-Temporal Information Encoding Network,DQE-STIEN).首先,基于船舶轨迹数据的特征,设计结合哈希映射分类及局部离群哈希值异常检测的数据质量增强算法,对问题数据进行质量增强.然后,针对多属性的船舶轨迹数据,设计具有双编码通道的时空信息编码网络,充分提取并整合船舶轨迹数据中的位置信息与运动特征.最后,基于时空信息编码提取数据中的时空关联信息,并经解码生成完整的轨迹预测结果.在5个不同区域的AIS数据集上的实验表明DQE-STIEN性能较优.同时,DQE-STIEN具有一定的泛化性,也能有效分析能源、销售、环境和金融等领域的时序数据.
基金National Key Research and Development Program of China,Grant/Award Number:2021YFB2401700。
文摘Digital twin(DT)modelling is a prerequisite for the successful application of DT technology in the power industry.However,traditional scene modelling methods are costly,time-consuming,focus on overall features and lack real-time updates,hindering the interaction between DT models and physical power equipment scenes.Therefore,a scene DT modelling technique focusing on local features in risk areas and real-time updates is urgently needed.Herein,real-time modelling of the±800 kV converter transformer is achieved by improving the neural radiation field based on a hybrid attention mechanism and multiresolution hash encoding.Compared to traditional methods,modelling time is reduced from hours to 1 min without professional equipment or manual intervention.The model quality is more concerned with local features of risk areas in transformers while ensuring the overall scene,and the accuracy is improved by about 6%,realising the real-time modelling of transformers and the DT of scenes.
基金supported by the National Natural Science Foundation of China (No. 61370077)Collaborative Innovation Center of Novel Software Technology and Industrialization
文摘Due to the widespread availability of implicit feedback(e.g., clicks and purchases), some researchers have endeavored to design recommender systems based on implicit feedback. However, unlike explicit feedback,implicit feedback cannot directly reflect user preferences. Therefore, although more challenging, it is also more practical to use implicit feedback for recommender systems. Traditional collaborative filtering methods such as matrix factorization, which regards user preferences as a linear combination of user and item latent vectors, have limited learning capacities and suffer from data sparsity and the cold-start problem. To tackle these problems,some authors have considered the integration of a deep neural network to learn user and item features with traditional collaborative filtering. However, there is as yet no research combining collaborative filtering and contentbased recommendation with deep learning. In this paper, we propose a novel deep hybrid recommender system framework based on auto-encoders(DHA-RS) by integrating user and item side information to construct a hybrid recommender system and enhance performance. DHA-RS combines stacked denoising auto-encoders with neural collaborative filtering, which corresponds to the process of learning user and item features from auxiliary information to predict user preferences. Experiments performed on the real-world dataset reveal that DHA-RS performs better than state-of-the-art methods.
