In this paper,an improved error-rate sliding window decoder is proposed for spatially coupled low-density parity-check(SC-LDPC)codes.For the conventional sliding window decoder,the message retention mechanism causes u...In this paper,an improved error-rate sliding window decoder is proposed for spatially coupled low-density parity-check(SC-LDPC)codes.For the conventional sliding window decoder,the message retention mechanism causes unreliable messages along the edges of belief propagation(BP)decoding in the current window to be kept for subsequent window decoding.To improve the reliability of the retained messages during the window transition,a reliable termination method is embedded,where the retained messages undergo more reliable parity checks.Additionally,decoding failure is unavoidable and even causes error propagation when the number of errors exceeds the error-correcting capability of the window.To mitigate this problem,a channel value reuse mechanism is designed,where the received channel values are utilized to reinitialize the window.Furthermore,considering the complexity and performance of decoding,a feasible sliding optimized window decoding(SOWD)scheme is introduced.Finally,simulation results confirm the superior performance of the proposed SOWD scheme in both the waterfall and error floor regions.This work has great potential in the applications of wireless optical communication and fiber optic communication.展开更多
Persistent flows are defined as network flows that persist over multiple time intervals and continue to exhibit activity over extended periods,which are critical for identifying long-term behaviors and subtle security...Persistent flows are defined as network flows that persist over multiple time intervals and continue to exhibit activity over extended periods,which are critical for identifying long-term behaviors and subtle security threats.Programmable switches provide line-rate packet processing to meet the requirements of high-speed network environments,yet they are fundamentally limited in computational and memory resources.Accurate and memoryefficient persistent flow detection on programmable switches is therefore essential.However,existing approaches often rely on fixed-window sketches or multiple sketches instances,which either suffer from insufficient temporal precision or incur substantial memory overhead,making them ineffective on programmable switches.To address these challenges,we propose SP-Sketch,an innovative sliding-window-based sketch that leverages a probabilistic update mechanism to emulate slot expiration without maintaining multiple sketch instances.This innovative design significantly reduces memory consumption while preserving high detection accuracy across multiple time intervals.We provide rigorous theoretical analyses of the estimation errors,deriving precise error bounds for the proposed method,and validate our approach through comprehensive implementations on both P4 hardware switches(with Intel Tofino ASIC)and software switches(i.e.,BMv2).Experimental evaluations using real-world traffic traces demonstrate that SP-Sketch outperforms traditional methods,improving accuracy by up to 20%over baseline sliding window approaches and enhancing recall by 5%compared to non-sliding alternatives.Furthermore,SP-Sketch achieves a significant reduction in memory utilization,reducing memory consumption by up to 65%compared to traditional methods,while maintaining a robust capability to accurately track persistent flow behavior over extended time periods.展开更多
With the rapid advancement of Voice over Internet Protocol(VoIP)technology,speech steganography techniques such as Quantization Index Modulation(QIM)and Pitch Modulation Steganography(PMS)have emerged as significant c...With the rapid advancement of Voice over Internet Protocol(VoIP)technology,speech steganography techniques such as Quantization Index Modulation(QIM)and Pitch Modulation Steganography(PMS)have emerged as significant challenges to information security.These techniques embed hidden information into speech streams,making detection increasingly difficult,particularly under conditions of low embedding rates and short speech durations.Existing steganalysis methods often struggle to balance detection accuracy and computational efficiency due to their limited ability to effectively capture both temporal and spatial features of speech signals.To address these challenges,this paper proposes an Efficient Sliding Window Analysis Network(E-SWAN),a novel deep learning model specifically designed for real-time speech steganalysis.E-SWAN integrates two core modules:the LSTM Temporal Feature Miner(LTFM)and the Convolutional Key Feature Miner(CKFM).LTFM captures long-range temporal dependencies using Long Short-Term Memory networks,while CKFM identifies local spatial variations caused by steganographic embedding through convolutional operations.These modules operate within a sliding window framework,enabling efficient extraction of temporal and spatial features.Experimental results on the Chinese CNV and PMS datasets demonstrate the superior performance of E-SWAN.Under conditions of a ten-second sample duration and an embedding rate of 10%,E-SWAN achieves a detection accuracy of 62.09%on the PMS dataset,surpassing existing methods by 4.57%,and an accuracy of 82.28%on the CNV dataset,outperforming state-of-the-art methods by 7.29%.These findings validate the robustness and efficiency of E-SWAN under low embedding rates and short durations,offering a promising solution for real-time VoIP steganalysis.This work provides significant contributions to enhancing information security in digital communications.展开更多
The reliable,rapid,and accurate Remaining Useful Life(RUL)prognostics of aircraft power supply and distribution system are essential for enhancing the reliability and stability of system and reducing the life-cycle co...The reliable,rapid,and accurate Remaining Useful Life(RUL)prognostics of aircraft power supply and distribution system are essential for enhancing the reliability and stability of system and reducing the life-cycle costs.To achieve the reliable,rapid,and accurate RUL prognostics,the balance between accuracy and computational burden deserves more attention.In addition,the uncertainty is intrinsically present in RUL prognostic process.Due to the limitation of the uncertainty quantification,the point-wise prognostics strategy is not trustworthy.A Dual Adaptive Sliding-window Hybrid(DASH)RUL probabilistic prognostics strategy is proposed to tackle these deficiencies.The DASH strategy contains two adaptive mechanisms,the adaptive Long Short-Term Memory-Polynomial Regression(LSTM-PR)hybrid prognostics mechanism and the adaptive sliding-window Kernel Density Estimation(KDE)probabilistic prognostics mechanism.Owing to the dual adaptive mechanisms,the DASH strategy can achieve the balance between accuracy and computational burden and obtain the trustworthy probabilistic prognostics.Based on the degradation dataset of aircraft electromagnetic contactors,the superiority of DASH strategy is validated.In terms of probabilistic,point-wise and integrated prognostics performance,the proposed strategy increases by 66.89%,81.73% and 25.84%on average compared with the baseline methods and their variants.展开更多
为了提升运动想象脑电(MI-EEG)信号的分类精度,提出多尺度滑窗注意力时序卷积网络(MSWATCN),充分挖掘MI-EEG信号的时空信息.结合多尺度双流分组卷积、滑动窗口多头注意力机制和窗口化时间卷积模块,实现对MI-EEG信号复杂时空特性的精准解...为了提升运动想象脑电(MI-EEG)信号的分类精度,提出多尺度滑窗注意力时序卷积网络(MSWATCN),充分挖掘MI-EEG信号的时空信息.结合多尺度双流分组卷积、滑动窗口多头注意力机制和窗口化时间卷积模块,实现对MI-EEG信号复杂时空特性的精准解码.利用多尺度卷积模块提取信号的底层时空特征,通过滑动窗口注意力机制聚焦局部关键特征,突出对分类任务重要的信息.窗口化时间卷积模块通过建模时间序列中的长期依赖关系,增强模型处理时序信息的能力.实验结果表明,MSWATCN在BCI Competition IV 2a和2b数据集上的分类准确率和一致性优于对比网络和基准模型.展开更多
针对视频异常事件的时空相关性学习对检测性能存在重要影响的问题,提出了基于融合双支特征的带约束损失的视频异常检测方法(Dual-branch Feature Fusion Based Constrained Multi-loss Video Anomaly Detection,DBF-CML-transMIL)。该...针对视频异常事件的时空相关性学习对检测性能存在重要影响的问题,提出了基于融合双支特征的带约束损失的视频异常检测方法(Dual-branch Feature Fusion Based Constrained Multi-loss Video Anomaly Detection,DBF-CML-transMIL)。该方法考虑多示例学习中片段的显著性和相关性,利用多层线性神经网络学习各片段的空间显著性特征,并设计级联Transformer融合模块来学习示例间的多层时序相关性;然后利用多损失模型对融合特征进行多loss监督学习,以丰富预测的多样性;针对现有top-k的离散性问题,提出了带约束机制的滑窗top-k强化异常事件的相关性。在ShanghaiTech和UCF-Crime数据集上的对比实验与消融实验表明,DBF-CML-transMIL的异常检测曲线下面积(Area Under Curve,AUC)分别达到97.33%和83.82%;各模块都能有效提升视频异常事件检测的性能。展开更多
【目的】青藏高原作为中国最大,世界海拔最高的高原,地表温度垂直递减率(Land Surface Temperature Lapse Rate,LTLR)的时空分布特征对气候变化、生态系统以及水文过程研究具有重要意义。已有研究无法准确表达复杂地形条件下的山区近地...【目的】青藏高原作为中国最大,世界海拔最高的高原,地表温度垂直递减率(Land Surface Temperature Lapse Rate,LTLR)的时空分布特征对气候变化、生态系统以及水文过程研究具有重要意义。已有研究无法准确表达复杂地形条件下的山区近地表气温直减率在时空分布与变化上的精细特征。因此本研究利用地表温度日变化模型估算出青藏高原逐小时地表温度,进而计算出逐小时月均LTLR,以获得青藏高原地区高时空分辨率的LTLR分布。【方法】本研究基于2022年中国西部逐日1 km全天候地表温度数据集TRIMS,利用地表温度日变化模型对青藏高原逐小时地表温度进行估算,进而采用滑动窗口法计算逐小时月均LTLR,分析了研究区LTLR在季节尺度上的时空分布与差异特征。弥补了青藏高原地区缺少高时空分辨率LTLR研究的现状。【结果】(1)4个季节平均LTLR分别为-6.12、-7.63、-5.89和-3.23℃/km,春夏季节整体高于秋冬季节,但横断山脉区域相反,冬季平均LTLR较夏季高出约0.57℃/km;(2)春夏季最大LTLR分别为-14.45℃/km、-13.92℃/km,相对于秋、冬季最大LTLR的-13.60℃/km、-11.61℃/km,更高,因高海拔和干旱晴朗天气影响,羌塘高原区不同季节的最大LTLR差异显著,其中冬季最大LTLR最小,为-13.67℃/km;(3)夏季最小LTLR最为大,高出其他季节约3.05℃/km,其中横断山脉四季最小LTLR均较大,其中春季最小LTLR为-1.16℃/km,比其他3个季节更高,最小的秋季最小LTLR为0.03℃/km,而羌塘高原区四季最小LTLR最小;(4)日变化曲线显示,春秋冬3个季节的LTLR在11:00—14:00最大,春季最小LTLR出现在20:00—23:00,秋季最小LTLR出现时间较春季提前了约1 h,而夏季一天中出现2次最大LTLR,分别在4:00—7:00和15:00—18:00,在21:00—23:00呈现出日最小LTLR特征。【结论】本研究对深入揭示青藏高原地表温度垂直递减率在季节尺度上的时空变化特征与相关影响机制有重要作用。展开更多
基金supported by the National Natural Science Foundation of China (No.62275193)。
文摘In this paper,an improved error-rate sliding window decoder is proposed for spatially coupled low-density parity-check(SC-LDPC)codes.For the conventional sliding window decoder,the message retention mechanism causes unreliable messages along the edges of belief propagation(BP)decoding in the current window to be kept for subsequent window decoding.To improve the reliability of the retained messages during the window transition,a reliable termination method is embedded,where the retained messages undergo more reliable parity checks.Additionally,decoding failure is unavoidable and even causes error propagation when the number of errors exceeds the error-correcting capability of the window.To mitigate this problem,a channel value reuse mechanism is designed,where the received channel values are utilized to reinitialize the window.Furthermore,considering the complexity and performance of decoding,a feasible sliding optimized window decoding(SOWD)scheme is introduced.Finally,simulation results confirm the superior performance of the proposed SOWD scheme in both the waterfall and error floor regions.This work has great potential in the applications of wireless optical communication and fiber optic communication.
基金supported by the National Undergraduate Innovation and Entrepreneurship Training Program of China(Project No.202510559076)at Jinan University,a nationwide initiative administered by the Ministry of Educationthe National Natural Science Foundation of China(NSFC)under Grant No.62172189.
文摘Persistent flows are defined as network flows that persist over multiple time intervals and continue to exhibit activity over extended periods,which are critical for identifying long-term behaviors and subtle security threats.Programmable switches provide line-rate packet processing to meet the requirements of high-speed network environments,yet they are fundamentally limited in computational and memory resources.Accurate and memoryefficient persistent flow detection on programmable switches is therefore essential.However,existing approaches often rely on fixed-window sketches or multiple sketches instances,which either suffer from insufficient temporal precision or incur substantial memory overhead,making them ineffective on programmable switches.To address these challenges,we propose SP-Sketch,an innovative sliding-window-based sketch that leverages a probabilistic update mechanism to emulate slot expiration without maintaining multiple sketch instances.This innovative design significantly reduces memory consumption while preserving high detection accuracy across multiple time intervals.We provide rigorous theoretical analyses of the estimation errors,deriving precise error bounds for the proposed method,and validate our approach through comprehensive implementations on both P4 hardware switches(with Intel Tofino ASIC)and software switches(i.e.,BMv2).Experimental evaluations using real-world traffic traces demonstrate that SP-Sketch outperforms traditional methods,improving accuracy by up to 20%over baseline sliding window approaches and enhancing recall by 5%compared to non-sliding alternatives.Furthermore,SP-Sketch achieves a significant reduction in memory utilization,reducing memory consumption by up to 65%compared to traditional methods,while maintaining a robust capability to accurately track persistent flow behavior over extended time periods.
基金supported in part by the Zhejiang Provincial Natural Science Foundation of China under Grant LQ20F020004in part by the National College Student Innovation and Research Training Program under Grant 202313283002.
文摘With the rapid advancement of Voice over Internet Protocol(VoIP)technology,speech steganography techniques such as Quantization Index Modulation(QIM)and Pitch Modulation Steganography(PMS)have emerged as significant challenges to information security.These techniques embed hidden information into speech streams,making detection increasingly difficult,particularly under conditions of low embedding rates and short speech durations.Existing steganalysis methods often struggle to balance detection accuracy and computational efficiency due to their limited ability to effectively capture both temporal and spatial features of speech signals.To address these challenges,this paper proposes an Efficient Sliding Window Analysis Network(E-SWAN),a novel deep learning model specifically designed for real-time speech steganalysis.E-SWAN integrates two core modules:the LSTM Temporal Feature Miner(LTFM)and the Convolutional Key Feature Miner(CKFM).LTFM captures long-range temporal dependencies using Long Short-Term Memory networks,while CKFM identifies local spatial variations caused by steganographic embedding through convolutional operations.These modules operate within a sliding window framework,enabling efficient extraction of temporal and spatial features.Experimental results on the Chinese CNV and PMS datasets demonstrate the superior performance of E-SWAN.Under conditions of a ten-second sample duration and an embedding rate of 10%,E-SWAN achieves a detection accuracy of 62.09%on the PMS dataset,surpassing existing methods by 4.57%,and an accuracy of 82.28%on the CNV dataset,outperforming state-of-the-art methods by 7.29%.These findings validate the robustness and efficiency of E-SWAN under low embedding rates and short durations,offering a promising solution for real-time VoIP steganalysis.This work provides significant contributions to enhancing information security in digital communications.
基金co-supported by the National Natural Science Foundation of China(Nos.52272403,52402506)Natural Science Basic Research Program of Shaanxi,China(Nos.2022JC-27,2023-JC-QN-0599)。
文摘The reliable,rapid,and accurate Remaining Useful Life(RUL)prognostics of aircraft power supply and distribution system are essential for enhancing the reliability and stability of system and reducing the life-cycle costs.To achieve the reliable,rapid,and accurate RUL prognostics,the balance between accuracy and computational burden deserves more attention.In addition,the uncertainty is intrinsically present in RUL prognostic process.Due to the limitation of the uncertainty quantification,the point-wise prognostics strategy is not trustworthy.A Dual Adaptive Sliding-window Hybrid(DASH)RUL probabilistic prognostics strategy is proposed to tackle these deficiencies.The DASH strategy contains two adaptive mechanisms,the adaptive Long Short-Term Memory-Polynomial Regression(LSTM-PR)hybrid prognostics mechanism and the adaptive sliding-window Kernel Density Estimation(KDE)probabilistic prognostics mechanism.Owing to the dual adaptive mechanisms,the DASH strategy can achieve the balance between accuracy and computational burden and obtain the trustworthy probabilistic prognostics.Based on the degradation dataset of aircraft electromagnetic contactors,the superiority of DASH strategy is validated.In terms of probabilistic,point-wise and integrated prognostics performance,the proposed strategy increases by 66.89%,81.73% and 25.84%on average compared with the baseline methods and their variants.
文摘为了提升运动想象脑电(MI-EEG)信号的分类精度,提出多尺度滑窗注意力时序卷积网络(MSWATCN),充分挖掘MI-EEG信号的时空信息.结合多尺度双流分组卷积、滑动窗口多头注意力机制和窗口化时间卷积模块,实现对MI-EEG信号复杂时空特性的精准解码.利用多尺度卷积模块提取信号的底层时空特征,通过滑动窗口注意力机制聚焦局部关键特征,突出对分类任务重要的信息.窗口化时间卷积模块通过建模时间序列中的长期依赖关系,增强模型处理时序信息的能力.实验结果表明,MSWATCN在BCI Competition IV 2a和2b数据集上的分类准确率和一致性优于对比网络和基准模型.
文摘针对视频异常事件的时空相关性学习对检测性能存在重要影响的问题,提出了基于融合双支特征的带约束损失的视频异常检测方法(Dual-branch Feature Fusion Based Constrained Multi-loss Video Anomaly Detection,DBF-CML-transMIL)。该方法考虑多示例学习中片段的显著性和相关性,利用多层线性神经网络学习各片段的空间显著性特征,并设计级联Transformer融合模块来学习示例间的多层时序相关性;然后利用多损失模型对融合特征进行多loss监督学习,以丰富预测的多样性;针对现有top-k的离散性问题,提出了带约束机制的滑窗top-k强化异常事件的相关性。在ShanghaiTech和UCF-Crime数据集上的对比实验与消融实验表明,DBF-CML-transMIL的异常检测曲线下面积(Area Under Curve,AUC)分别达到97.33%和83.82%;各模块都能有效提升视频异常事件检测的性能。
文摘【目的】青藏高原作为中国最大,世界海拔最高的高原,地表温度垂直递减率(Land Surface Temperature Lapse Rate,LTLR)的时空分布特征对气候变化、生态系统以及水文过程研究具有重要意义。已有研究无法准确表达复杂地形条件下的山区近地表气温直减率在时空分布与变化上的精细特征。因此本研究利用地表温度日变化模型估算出青藏高原逐小时地表温度,进而计算出逐小时月均LTLR,以获得青藏高原地区高时空分辨率的LTLR分布。【方法】本研究基于2022年中国西部逐日1 km全天候地表温度数据集TRIMS,利用地表温度日变化模型对青藏高原逐小时地表温度进行估算,进而采用滑动窗口法计算逐小时月均LTLR,分析了研究区LTLR在季节尺度上的时空分布与差异特征。弥补了青藏高原地区缺少高时空分辨率LTLR研究的现状。【结果】(1)4个季节平均LTLR分别为-6.12、-7.63、-5.89和-3.23℃/km,春夏季节整体高于秋冬季节,但横断山脉区域相反,冬季平均LTLR较夏季高出约0.57℃/km;(2)春夏季最大LTLR分别为-14.45℃/km、-13.92℃/km,相对于秋、冬季最大LTLR的-13.60℃/km、-11.61℃/km,更高,因高海拔和干旱晴朗天气影响,羌塘高原区不同季节的最大LTLR差异显著,其中冬季最大LTLR最小,为-13.67℃/km;(3)夏季最小LTLR最为大,高出其他季节约3.05℃/km,其中横断山脉四季最小LTLR均较大,其中春季最小LTLR为-1.16℃/km,比其他3个季节更高,最小的秋季最小LTLR为0.03℃/km,而羌塘高原区四季最小LTLR最小;(4)日变化曲线显示,春秋冬3个季节的LTLR在11:00—14:00最大,春季最小LTLR出现在20:00—23:00,秋季最小LTLR出现时间较春季提前了约1 h,而夏季一天中出现2次最大LTLR,分别在4:00—7:00和15:00—18:00,在21:00—23:00呈现出日最小LTLR特征。【结论】本研究对深入揭示青藏高原地表温度垂直递减率在季节尺度上的时空变化特征与相关影响机制有重要作用。
