A method of digitally high pass filtering in frequency domain is proposed to eliminate the background noise of the decoded image in Fresnel zone plate scanning holography. The high pass filter is designed as a circula...A method of digitally high pass filtering in frequency domain is proposed to eliminate the background noise of the decoded image in Fresnel zone plate scanning holography. The high pass filter is designed as a circular stop, which should be suitable to suppressing the background noise significantly and remain much low frequency information of the object. The principle of high pass filtering is that the Fourier transform of the decoded image is multiplied with the high pass filter. Thus the frequency spectrum of the decoded image without the background noise is achieved. By inverse Fourier transform of the spectrum of the decoded image after multiplying operation, the decoded image without the background noise is obtained. Both of the computer simulations and the experimental results show that the contrast and the signal-to-noise ratio of the decoded image are significantly improved with digital filtering.展开更多
Children’s failure to develop simple word decoding skills in early years is linked to future poor reading, school dropout, and poor health [1] [2]. Letter-sound knowledge is needed for word decoding development;howev...Children’s failure to develop simple word decoding skills in early years is linked to future poor reading, school dropout, and poor health [1] [2]. Letter-sound knowledge is needed for word decoding development;however questions remain on what types of letter-sound knowledge help children decode simple words [3]. This study investigated the differences in mean number of consonant-vowel-consonant (CVC) words decoded between two groups of children, a letter-sound reading group and non letter-sound reading group. Children aged 4 to 6 in both groups, attempted to decode a variety of simple words such as tan, sit, hen, pig, dot, and fun. Analysis determined word decoding differences existed between the two groups. The alternate hypothesis was accepted;the letter-sound reading group had a significantly higher mean in number of consonant-vowel-consonant words decoded compared to the non letter-sound reading group. The study informs the teaching approaches needed to improve early decoding skills showing letter-sound reading ability is an important step for learning to decode simple consonant-vowel-consonant words.展开更多
The symbol error rate （SER） performance of a multipulse pulse-position modulation （MPPM） free space optical （FSO） system under the combined effect of turbulence-induced fading modeled by exponentiated Weibull ...The symbol error rate （SER） performance of a multipulse pulse-position modulation （MPPM） free space optical （FSO） system under the combined effect of turbulence-induced fading modeled by exponentiated Weibull （EW） distribution and pointing errors with a soft-decision detector is investigated systematically. Particularly, the theoretical conditional SER （CSER） of soft-decision decoded MPPM is derived. The corresponding closed-form CSER is obtained via curve fitting with the Levenberg-Marquardt method. The analytical SER expression over the aggregated fading channels is then achieved in terms of Laguerre integration. Monte Carlo simulation results are also offered to corroborate the validity of the proposed SER model.展开更多
Aiming at the problem that the bit error rate(BER)of asymmetrically clipped optical orthogonal frequency division multiplexing(ACO-OFDM)space optical communication system is significantly affected by different turbule...Aiming at the problem that the bit error rate(BER)of asymmetrically clipped optical orthogonal frequency division multiplexing(ACO-OFDM)space optical communication system is significantly affected by different turbulence intensities,the deep learning technique is proposed to the polarization code decoding in ACO-OFDM space optical communication system.Moreover,this system realizes the polarization code decoding and signal demodulation without frequency conduction with superior performance and robustness compared with the performance of traditional decoder.Simulations under different turbulence intensities as well as different mapping orders show that the convolutional neural network(CNN)decoder trained under weak-medium-strong turbulence atmospheric channels achieves a performance improvement of about 10^(2)compared to the conventional decoder at 4-quadrature amplitude modulation(4QAM),and the BERs for both 16QAM and 64QAM are in between those of the conventional decoder.展开更多
This paper investigates the uplink spectral efficiency of distributed cell-free(CF)massive multiple-input multiple-output(mMIMO)networks with correlated Rayleigh fading channels based on three different channel estima...This paper investigates the uplink spectral efficiency of distributed cell-free(CF)massive multiple-input multiple-output(mMIMO)networks with correlated Rayleigh fading channels based on three different channel estimation schemes.Specifically,each access point(AP)first uses embedded pilots to estimate the channels of all users based on minimum mean-squared error(MMSE)estimation.Given the high computational cost of MMSE estimation,the low-complexity element-wise MMSE(EW-MMSE)channel estimator and the least-squares(LS)channel estimator without prior statistical information are also analyzed.To reduce non-coherent and coherent interference during uplink payload data transmission,simple centralized decoding(SCD)and large-scale fading decoding(LSFD)are examined.Then,the closedform expressions for uplink spectral efficiency(SE)using MMSE,EW-MMSE,and LS estimators are developed for maximum ratio(MR)combining under LSFD,where each AP may have any number of antennas.The sum SE maximization problem with uplink power control is formulated.Since the maximization problem is non-convex and challenging,a block coordinate descent approach based on the weighted MMSE method is used to get the optimal local solution.Numerical studies demonstrate that LSFD and efficient uplink power control can considerably increase SE in distributed CF m MIMO networks.展开更多
Space laser communication(SLC)is an emerging technology to support high-throughput data transmissions in space networks.In this paper,to guarantee the reliability of high-speed SLC links,we aim at practical implementa...Space laser communication(SLC)is an emerging technology to support high-throughput data transmissions in space networks.In this paper,to guarantee the reliability of high-speed SLC links,we aim at practical implementation of low-density paritycheck(LDPC)decoding under resource-restricted space platforms.Particularly,due to the supply restriction and cost issues of high-speed on-board devices such as analog-to-digital converters(ADCs),the input of LDPC decoding will be usually constrained by hard-decision channel output.To tackle this challenge,density-evolution-based theoretical analysis is firstly performed to identify the cause of performance degradation in the conventional binaryinitialized iterative decoding(BIID)algorithm.Then,a computation-efficient decoding algorithm named multiary-initialized iterative decoding with early termination(MIID-ET)is proposed,which improves the error-correcting performance and computation efficiency by using a reliability-based initialization method and a threshold-based decoding termination rule.Finally,numerical simulations are conducted on example codes of rates 7/8 and 1/2 to evaluate the performance of different LDPC decoding algorithms,where the proposed MIID-ET outperforms the BIID with a coding gain of 0.38 dB and variable node calculation saving of 37%.With this advantage,the proposed MIID-ET can notably reduce LDPC decoder’s hardware implementation complexity under the same bit error rate performance,which successfully doubles the total throughput to 10 Gbps on a single-chip FPGA.展开更多
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
Neural machine translation(NMT)has advanced with deep learning and large-scale multilingual models,yet translating lowresource languages often lacks sufficient training data and leads to hallucinations.This often resu...Neural machine translation(NMT)has advanced with deep learning and large-scale multilingual models,yet translating lowresource languages often lacks sufficient training data and leads to hallucinations.This often results in translated content that diverges significantly from the source text.This research proposes a refined Contrastive Decoding(CD)algorithm that dynamically adjusts weights of log probabilities from strong expert and weak amateur models to mitigate hallucinations in lowresource NMT and improve translation quality.Advanced large language NMT models,including ChatGLM and LLaMA,are fine-tuned and implemented for their superior contextual understanding and cross-lingual capabilities.The refined CD algorithm evaluates multiple candidate translations using BLEU score,semantic similarity,and Named Entity Recognition accuracy.Extensive experimental results show substantial improvements in translation quality and a significant reduction in hallucination rates.Fine-tuned models achieve higher evaluation metrics compared to baseline models and state-of-the-art models.An ablation study confirms the contributions of each methodological component and highlights the effectiveness of the refined CD algorithm and advanced models in mitigating hallucinations.Notably,the refined methodology increased the BLEU score by approximately 30%compared to baseline models.展开更多
Action recognition,a fundamental task in the field of video understanding,has been extensively researched and applied.In contrast to an image,a video introduces an extra temporal dimension.However,many existing action...Action recognition,a fundamental task in the field of video understanding,has been extensively researched and applied.In contrast to an image,a video introduces an extra temporal dimension.However,many existing action recognition networks either perform simple temporal fusion through averaging or rely on pre-trained models from image recognition,resulting in limited temporal information extraction capabilities.This work proposes a highly efficient temporal decoding module that can be seamlessly integrated into any action recognition backbone network to enhance the focus on temporal relationships between video frames.Firstly,the decoder initializes a set of learnable queries,termed video-level action category prediction queries.Then,they are combined with the video frame features extracted by the backbone network after self-attention learning to extract video context information.Finally,these prediction queries with rich temporal features are used for category prediction.Experimental results on HMDB51,MSRDailyAct3D,Diving48 and Breakfast datasets show that using TokShift-Transformer and VideoMAE as encoders results in a significant improvement in Top-1 accuracy compared to the original models(TokShift-Transformer and VideoMAE),after introducing the proposed temporal decoder.The introduction of the temporal decoder results in an average performance increase exceeding 11%for TokShift-Transformer and nearly 5%for VideoMAE across the four datasets.Furthermore,the work explores the combination of the decoder with various action recognition networks,including Timesformer,as encoders.This results in an average accuracy improvement of more than 3.5%on the HMDB51 dataset.The code is available at https://github.com/huangturbo/TempDecoder.展开更多
Quantum key distribution is increasingly transitioning toward network applications,necessitating advancements in system performance,including photonic integration for compact designs,enhanced stability against environ...Quantum key distribution is increasingly transitioning toward network applications,necessitating advancements in system performance,including photonic integration for compact designs,enhanced stability against environmental disturbances,higher key rates,and improved efficiency.In this letter,we propose an orthogonal polarization exchange reflector Michelson interferometer model to address quantum channel disturbances caused by environmental factors.Based on this model,we designed a Sagnac reflector-Michelson interferometer decoder and verified its performance through an interference system.The interference fringe visibility exceeded 98%across all four coding phases at 625 MHz.These results indicate that the decoder effectively mitigates environmental interference while supporting high-speed modulation frequencies.In addition,the proposed anti-interference decoder,which does not rely on magneto-optical devices,is well-suited for photonic integration,aligning with the development trajectory for next-generation quantum communication devices.展开更多
Intrinsic decomposition,the process of decomposing an image into reflectance and shading,is widely used in virtual and augmented reality tasks.Reflectance and shading often exhibit large gradients at the object edges,...Intrinsic decomposition,the process of decomposing an image into reflectance and shading,is widely used in virtual and augmented reality tasks.Reflectance and shading often exhibit large gradients at the object edges,and the intrinsic properties on the same object tend to be similar.This spatial coherence is closely related to semantic consistency because objects within the same semantic category often exhibit similar intrinsic properties.Therefore,incorporating semantic segmentation into a deep intrinsic decomposition framework helps the network distinguish between different object instances and understand high-level scene structures.To this end,we design an intrinsic decomposition network jointly trained with a dedicated semantic segmentation module,allowing semantic cues to enhance the decomposition of reflectance and shading.The semantic module provides guidance during training but is removed during inference,improving performance without increasing the inference cost.Additionally,to capture the global contextual dependencies critical for intrinsic decomposition,we adopt a Transformer-based backbone.The proposed backbone enables the model to associate distant regions with similar material properties,thereby maintaining consistency in reflectance and learning smooth illumination patterns across a scene.A convolutional decoder is also designed to output predictions with improved details.Experiments demonstrate that our approach achieves state-of-the-art performance in the quantitative evaluations on the Intrinsic Images in the Wild(IIW)and Shading Annotations in the wild(SAW)datasets.展开更多
Constituted by BCH component codes and its ordered statistics decoding(OSD),the successive cancellation list(SCL)decoding of U-UV structural codes can provide competent error-correction performance in the short-to-med...Constituted by BCH component codes and its ordered statistics decoding(OSD),the successive cancellation list(SCL)decoding of U-UV structural codes can provide competent error-correction performance in the short-to-medium length regime.However,this list decoding complexity becomes formidable as the decoding output list size increases.This is primarily incurred by the OSD.Addressing this challenge,this paper proposes the low complexity SCL decoding through reducing the complexity of component code decoding,and pruning the redundant SCL decoding paths.For the former,an efficient skipping rule is introduced for the OSD so that the higher order decoding can be skipped when they are not possible to provide a more likely codeword candidate.It is further extended to the OSD variant,the box-andmatch algorithm(BMA),in facilitating the component code decoding.Moreover,through estimating the correlation distance lower bounds(CDLBs)of the component code decoding outputs,a path pruning(PP)-SCL decoding is proposed to further facilitate the decoding of U-UV codes.In particular,its integration with the improved OSD and BMA is discussed.Simulation results show that significant complexity reduction can be achieved.Consequently,the U-UV codes can outperform the cyclic redundancy check(CRC)-polar codes with a similar decoding complexity.展开更多
In the field of organic synthesis,the core objective of retrosynthetic methods is to deduce possible synthetic routes and precursor molecules for complex target molecules.Traditional retrosynthetic methods,such as tem...In the field of organic synthesis,the core objective of retrosynthetic methods is to deduce possible synthetic routes and precursor molecules for complex target molecules.Traditional retrosynthetic methods,such as template-based retrosynthesis,have high accuracy and interpretability in specific types of reactions but are limited by the scope of the template library,making it difficult to adapt to new or uncommon reaction types.Moreover,sequence-to-sequence retrosynthetic prediction methods,although they enhance the flexibility of prediction,often overlook the complexity of molecular graph structures and the actual interactions between atoms,which limits the accuracy and reliability of the predictions.To address these limitations,this paper proposes a Molecular Retrosynthesis Top-k Prediction based on the Latent Generation Process(MRLGP)that uses latent variables from graph neural networks to model the generation process and produce diverse set of reactants.Utilising an encoding method based on Graphormer,the authors have also introduced topology-aware positional encoding to better capture the interactions between atomic nodes in the molecular graph structure,thereby more accurately simulating the retrosynthetic process.The MRLGP model significantly enhances the accuracy and diversity of predictions by correlating discrete latent variables with the reactant generation process and progressively constructing molecular graphs using a variational autoregressive decoder.Experimental results on benchmark datasets such as USPTO-50k,USPTO-Full,and USPTO-DIVERSE demonstrate that MRLGP outperforms baseline models on multiple Top-k evaluation metrics.Additionally,ablation experiments conducted on the USPTO-50K dataset further validate the effectiveness of the methods used in the encoder and decoder parts of the model.展开更多
Transfer RNAs(tRNAs)adopt a stable L-shaped tertiary structure crucial for their involvement in protein translation.Among various divalent metal ions,magnesium ions play a pivotal role in preserving the tertiary struc...Transfer RNAs(tRNAs)adopt a stable L-shaped tertiary structure crucial for their involvement in protein translation.Among various divalent metal ions,magnesium ions play a pivotal role in preserving the tertiary structure of tRNA.However,the precise location of the Mg^(2+)binding pocket in human tRNA remains elusive.In this investigation,we identified the Mg^(2+)binding site within human tRNAGln using suppressor tRNA^(Gln).This variant of tRNA recognizes premature stop codons(specificlly UAG)and facilitates the expression of fll-length proteis.By mutating sites 8 and C72 in supprssr tRNAcl,we assessed the decoding efficiency of the resulting mutant suppressor tRNAs,which serves as a measure of tRNA's ability to decode genetic information.Our analysis revealed that the U8C mutant suppressor tRNA exhibited a significantly lower Mg^(2+)content compared to the C72U mutant.Furthermore,we observed a notable reduction in decoding efficiency in the U8-mutated suppressor tRNA,as evidenced by GFP fluorescence and Western blotting analysis.Conversely,mutations at the C72 site had a comparatively minor impact on decoding efficiency.These findings underscored the tight binding of Mg^(2+)to the U8 site of human tRNAGln,crucial for maintaining the stability of tRNA tertiary structure and translation efficacy.Additionally,our investigation delved into the influence of glutamine availability on tRNA decoding efficiency at the cellular level.The results indicated that both the concentration of amino acids and the codon context of TAG could modulate tRNA decoding efficiency.This study provided valuable insights into the structure and function of tRNA,laying the groundwork for further exploration in this field.展开更多
In this paper,Wuzhou City of Guangxi was taken as the research object.Through the design of a climatic data warehousing system,the decoding methods of surface meteorological data and their application in the managemen...In this paper,Wuzhou City of Guangxi was taken as the research object.Through the design of a climatic data warehousing system,the decoding methods of surface meteorological data and their application in the management of climatic data were explored.Based on the parsing technology of the monthly report of surface meteorological records(A-file),the design of Wuzhou climatic data warehousing system was realized,completing the precise extraction and database construction of observational elements such as regional temperature,wind direction,and weather phenomena.Based on this system,the meteorological data in 2024 were analyzed,and the probabilistic characteristics of dominant wind direction in Wuzhou(northeast wind accounting for the largest proportion),the spatiotemporal distribution patterns of extreme temperatures(annual extreme high temperature of 37.1℃in August and extreme low temperature of 1.9℃in January),and the general climatic overview of Wuzhou City(annual precipitation 3.2%higher than the standard value)were revealed.The research shows that climate change has a significant impact on agricultural production and economic development in Wuzhou City,and the construction of a reasonable climatic data database is of great significance for enhancing professional meteorological service capabilities in the context of climate change.This study not only provides a scientific basis for the economic development of Wuzhou region,but also offers reference ideas for other regions to cope with regional climate adaptation planning.展开更多
Among the four candidate algorithms in the fourth round of NIST standardization,the BIKE(Bit Flipping Key Encapsulation)scheme has a small key size and high efficiency,showing good prospects for application.However,th...Among the four candidate algorithms in the fourth round of NIST standardization,the BIKE(Bit Flipping Key Encapsulation)scheme has a small key size and high efficiency,showing good prospects for application.However,the BIKE scheme based on QC-MDPC(Quasi Cyclic Medium Density Parity Check)codes still faces challenges such as the GJS attack and weak key attacks targeting the decoding failure rate(DFR).This paper analyzes the BGF decoding algorithm of the BIKE scheme,revealing two deep factors that lead to DFR,and proposes a weak key optimization attack method for the BGF decoding algorithm based on these two factors.The proposed method constructs a new weak key set,and experiment results eventually indicate that,considering BIKE’s parameter set targeting 128-bit security,the average decryption failure rate is lowerly bounded by.This result not only highlights a significant vulnerability in the BIKE scheme but also provides valuable insights for future improvements in its design.By addressing these weaknesses,the robustness of QC-MDPC code-based cryptographic systems can be enhanced,paving the way for more secure post-quantum cryptographic solutions.展开更多
Linear programming(LP)decoding is a classic decoding method for linear block codes,and has attracted recent researches because its potential in joint channel processing.However,for polar codes,LP decoders has long bee...Linear programming(LP)decoding is a classic decoding method for linear block codes,and has attracted recent researches because its potential in joint channel processing.However,for polar codes,LP decoders has long been outperformed by CRCaided successive cancellation list(CA-SCL)decoders.To increase the competitiveness of 5G NR LP polar decoding,it is possible to gain performance improvements by exploiting the cyclic redundancy check(CRC)setup.In this paper,we propose a combined scheme of reduced sparsified factor graph-sparsified CRC(RSFG-SCRC)and augmented generator matrix-CRC(AGM-CRC),for polytope generation in adaptive linear programming(ALP)decoder for 5G polar codes.Augmented generator matrix(AGM)polytope and improved maximum cycle strategy-auxiliary node pairs 4(MCS-ANP-4)algorithm are proposed,to make efficient use of CRC constraints and minimize the constraint size for the decoder.Numerical simulations show that adaptive linear programming decoders with our proposed RSFG-SCRC and AGM-CRC polytopes can achieve significantly better block error rate(BLER)performance than a benchmark CA-SCL-8 decoder especially in harsh low-to-medium SNR regions.展开更多
The convolutional neural network(CNN)method based on DeepLabv3+has some problems in the semantic segmentation task of high-resolution remote sensing images,such as fixed receiving field size of feature extraction,lack...The convolutional neural network(CNN)method based on DeepLabv3+has some problems in the semantic segmentation task of high-resolution remote sensing images,such as fixed receiving field size of feature extraction,lack of semantic information,high decoder magnification,and insufficient detail retention ability.A hierarchical feature fusion network(HFFNet)was proposed.Firstly,a combination of transformer and CNN architectures was employed for feature extraction from images of varying resolutions.The extracted features were processed independently.Subsequently,the features from the transformer and CNN were fused under the guidance of features from different sources.This fusion process assisted in restoring information more comprehensively during the decoding stage.Furthermore,a spatial channel attention module was designed in the final stage of decoding to refine features and reduce the semantic gap between shallow CNN features and deep decoder features.The experimental results showed that HFFNet had superior performance on UAVid,LoveDA,Potsdam,and Vaihingen datasets,and its cross-linking index was better than DeepLabv3+and other competing methods,showing strong generalization ability.展开更多
文摘A method of digitally high pass filtering in frequency domain is proposed to eliminate the background noise of the decoded image in Fresnel zone plate scanning holography. The high pass filter is designed as a circular stop, which should be suitable to suppressing the background noise significantly and remain much low frequency information of the object. The principle of high pass filtering is that the Fourier transform of the decoded image is multiplied with the high pass filter. Thus the frequency spectrum of the decoded image without the background noise is achieved. By inverse Fourier transform of the spectrum of the decoded image after multiplying operation, the decoded image without the background noise is obtained. Both of the computer simulations and the experimental results show that the contrast and the signal-to-noise ratio of the decoded image are significantly improved with digital filtering.
文摘Children’s failure to develop simple word decoding skills in early years is linked to future poor reading, school dropout, and poor health [1] [2]. Letter-sound knowledge is needed for word decoding development;however questions remain on what types of letter-sound knowledge help children decode simple words [3]. This study investigated the differences in mean number of consonant-vowel-consonant (CVC) words decoded between two groups of children, a letter-sound reading group and non letter-sound reading group. Children aged 4 to 6 in both groups, attempted to decode a variety of simple words such as tan, sit, hen, pig, dot, and fun. Analysis determined word decoding differences existed between the two groups. The alternate hypothesis was accepted;the letter-sound reading group had a significantly higher mean in number of consonant-vowel-consonant words decoded compared to the non letter-sound reading group. The study informs the teaching approaches needed to improve early decoding skills showing letter-sound reading ability is an important step for learning to decode simple consonant-vowel-consonant words.
基金supported by the National Natural Science Foundation of China(No.61474090)the Fundamental Research Funds for the Central Universities(No.JB160105)the 111 Project of China(No.B08038)
文摘The symbol error rate （SER） performance of a multipulse pulse-position modulation （MPPM） free space optical （FSO） system under the combined effect of turbulence-induced fading modeled by exponentiated Weibull （EW） distribution and pointing errors with a soft-decision detector is investigated systematically. Particularly, the theoretical conditional SER （CSER） of soft-decision decoded MPPM is derived. The corresponding closed-form CSER is obtained via curve fitting with the Levenberg-Marquardt method. The analytical SER expression over the aggregated fading channels is then achieved in terms of Laguerre integration. Monte Carlo simulation results are also offered to corroborate the validity of the proposed SER model.
基金supported by the National Natural Science Foundation of China(No.12104141).
文摘Aiming at the problem that the bit error rate(BER)of asymmetrically clipped optical orthogonal frequency division multiplexing(ACO-OFDM)space optical communication system is significantly affected by different turbulence intensities,the deep learning technique is proposed to the polarization code decoding in ACO-OFDM space optical communication system.Moreover,this system realizes the polarization code decoding and signal demodulation without frequency conduction with superior performance and robustness compared with the performance of traditional decoder.Simulations under different turbulence intensities as well as different mapping orders show that the convolutional neural network(CNN)decoder trained under weak-medium-strong turbulence atmospheric channels achieves a performance improvement of about 10^(2)compared to the conventional decoder at 4-quadrature amplitude modulation(4QAM),and the BERs for both 16QAM and 64QAM are in between those of the conventional decoder.
基金supported by National Natural Science Foundation of China(NSFC No.62020106001)。
文摘This paper investigates the uplink spectral efficiency of distributed cell-free(CF)massive multiple-input multiple-output(mMIMO)networks with correlated Rayleigh fading channels based on three different channel estimation schemes.Specifically,each access point(AP)first uses embedded pilots to estimate the channels of all users based on minimum mean-squared error(MMSE)estimation.Given the high computational cost of MMSE estimation,the low-complexity element-wise MMSE(EW-MMSE)channel estimator and the least-squares(LS)channel estimator without prior statistical information are also analyzed.To reduce non-coherent and coherent interference during uplink payload data transmission,simple centralized decoding(SCD)and large-scale fading decoding(LSFD)are examined.Then,the closedform expressions for uplink spectral efficiency(SE)using MMSE,EW-MMSE,and LS estimators are developed for maximum ratio(MR)combining under LSFD,where each AP may have any number of antennas.The sum SE maximization problem with uplink power control is formulated.Since the maximization problem is non-convex and challenging,a block coordinate descent approach based on the weighted MMSE method is used to get the optimal local solution.Numerical studies demonstrate that LSFD and efficient uplink power control can considerably increase SE in distributed CF m MIMO networks.
基金supported by the National Key R&D Program of China(Grant No.2022YFA1005000)the National Natural Science Foundation of China(Grant No.62101308 and 62025110).
文摘Space laser communication(SLC)is an emerging technology to support high-throughput data transmissions in space networks.In this paper,to guarantee the reliability of high-speed SLC links,we aim at practical implementation of low-density paritycheck(LDPC)decoding under resource-restricted space platforms.Particularly,due to the supply restriction and cost issues of high-speed on-board devices such as analog-to-digital converters(ADCs),the input of LDPC decoding will be usually constrained by hard-decision channel output.To tackle this challenge,density-evolution-based theoretical analysis is firstly performed to identify the cause of performance degradation in the conventional binaryinitialized iterative decoding(BIID)algorithm.Then,a computation-efficient decoding algorithm named multiary-initialized iterative decoding with early termination(MIID-ET)is proposed,which improves the error-correcting performance and computation efficiency by using a reliability-based initialization method and a threshold-based decoding termination rule.Finally,numerical simulations are conducted on example codes of rates 7/8 and 1/2 to evaluate the performance of different LDPC decoding algorithms,where the proposed MIID-ET outperforms the BIID with a coding gain of 0.38 dB and variable node calculation saving of 37%.With this advantage,the proposed MIID-ET can notably reduce LDPC decoder’s hardware implementation complexity under the same bit error rate performance,which successfully doubles the total throughput to 10 Gbps on a single-chip FPGA.
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M.Faheem is supported by VTT Technical Research Center of Finland.
文摘Neural machine translation(NMT)has advanced with deep learning and large-scale multilingual models,yet translating lowresource languages often lacks sufficient training data and leads to hallucinations.This often results in translated content that diverges significantly from the source text.This research proposes a refined Contrastive Decoding(CD)algorithm that dynamically adjusts weights of log probabilities from strong expert and weak amateur models to mitigate hallucinations in lowresource NMT and improve translation quality.Advanced large language NMT models,including ChatGLM and LLaMA,are fine-tuned and implemented for their superior contextual understanding and cross-lingual capabilities.The refined CD algorithm evaluates multiple candidate translations using BLEU score,semantic similarity,and Named Entity Recognition accuracy.Extensive experimental results show substantial improvements in translation quality and a significant reduction in hallucination rates.Fine-tuned models achieve higher evaluation metrics compared to baseline models and state-of-the-art models.An ablation study confirms the contributions of each methodological component and highlights the effectiveness of the refined CD algorithm and advanced models in mitigating hallucinations.Notably,the refined methodology increased the BLEU score by approximately 30%compared to baseline models.
基金Shanghai Municipal Commission of Economy and Information Technology,China (No.202301054)。
文摘Action recognition,a fundamental task in the field of video understanding,has been extensively researched and applied.In contrast to an image,a video introduces an extra temporal dimension.However,many existing action recognition networks either perform simple temporal fusion through averaging or rely on pre-trained models from image recognition,resulting in limited temporal information extraction capabilities.This work proposes a highly efficient temporal decoding module that can be seamlessly integrated into any action recognition backbone network to enhance the focus on temporal relationships between video frames.Firstly,the decoder initializes a set of learnable queries,termed video-level action category prediction queries.Then,they are combined with the video frame features extracted by the backbone network after self-attention learning to extract video context information.Finally,these prediction queries with rich temporal features are used for category prediction.Experimental results on HMDB51,MSRDailyAct3D,Diving48 and Breakfast datasets show that using TokShift-Transformer and VideoMAE as encoders results in a significant improvement in Top-1 accuracy compared to the original models(TokShift-Transformer and VideoMAE),after introducing the proposed temporal decoder.The introduction of the temporal decoder results in an average performance increase exceeding 11%for TokShift-Transformer and nearly 5%for VideoMAE across the four datasets.Furthermore,the work explores the combination of the decoder with various action recognition networks,including Timesformer,as encoders.This results in an average accuracy improvement of more than 3.5%on the HMDB51 dataset.The code is available at https://github.com/huangturbo/TempDecoder.
基金supported by the National Natural Science Foundation of China under Grant No.62001440。
文摘Quantum key distribution is increasingly transitioning toward network applications,necessitating advancements in system performance,including photonic integration for compact designs,enhanced stability against environmental disturbances,higher key rates,and improved efficiency.In this letter,we propose an orthogonal polarization exchange reflector Michelson interferometer model to address quantum channel disturbances caused by environmental factors.Based on this model,we designed a Sagnac reflector-Michelson interferometer decoder and verified its performance through an interference system.The interference fringe visibility exceeded 98%across all four coding phases at 625 MHz.These results indicate that the decoder effectively mitigates environmental interference while supporting high-speed modulation frequencies.In addition,the proposed anti-interference decoder,which does not rely on magneto-optical devices,is well-suited for photonic integration,aligning with the development trajectory for next-generation quantum communication devices.
基金Supported by Science and Technology Innovation 2030:Major Project of“New Generation Artificial Intelligence”(No.2022ZD0115901)the National Natural Science Foundation of China(No.62332003).
文摘Intrinsic decomposition,the process of decomposing an image into reflectance and shading,is widely used in virtual and augmented reality tasks.Reflectance and shading often exhibit large gradients at the object edges,and the intrinsic properties on the same object tend to be similar.This spatial coherence is closely related to semantic consistency because objects within the same semantic category often exhibit similar intrinsic properties.Therefore,incorporating semantic segmentation into a deep intrinsic decomposition framework helps the network distinguish between different object instances and understand high-level scene structures.To this end,we design an intrinsic decomposition network jointly trained with a dedicated semantic segmentation module,allowing semantic cues to enhance the decomposition of reflectance and shading.The semantic module provides guidance during training but is removed during inference,improving performance without increasing the inference cost.Additionally,to capture the global contextual dependencies critical for intrinsic decomposition,we adopt a Transformer-based backbone.The proposed backbone enables the model to associate distant regions with similar material properties,thereby maintaining consistency in reflectance and learning smooth illumination patterns across a scene.A convolutional decoder is also designed to output predictions with improved details.Experiments demonstrate that our approach achieves state-of-the-art performance in the quantitative evaluations on the Intrinsic Images in the Wild(IIW)and Shading Annotations in the wild(SAW)datasets.
基金supported by the National Natural Science Foundation of China(NSFC)with project ID 62071498the Guangdong National Science Foundation(GDNSF)with project ID 2024A1515010213.
文摘Constituted by BCH component codes and its ordered statistics decoding(OSD),the successive cancellation list(SCL)decoding of U-UV structural codes can provide competent error-correction performance in the short-to-medium length regime.However,this list decoding complexity becomes formidable as the decoding output list size increases.This is primarily incurred by the OSD.Addressing this challenge,this paper proposes the low complexity SCL decoding through reducing the complexity of component code decoding,and pruning the redundant SCL decoding paths.For the former,an efficient skipping rule is introduced for the OSD so that the higher order decoding can be skipped when they are not possible to provide a more likely codeword candidate.It is further extended to the OSD variant,the box-andmatch algorithm(BMA),in facilitating the component code decoding.Moreover,through estimating the correlation distance lower bounds(CDLBs)of the component code decoding outputs,a path pruning(PP)-SCL decoding is proposed to further facilitate the decoding of U-UV codes.In particular,its integration with the improved OSD and BMA is discussed.Simulation results show that significant complexity reduction can be achieved.Consequently,the U-UV codes can outperform the cyclic redundancy check(CRC)-polar codes with a similar decoding complexity.
基金supported by the China Postdoctoral Science Foundation(No.2014m561331)Science and Technology Research Project of Heilongjiang Provincial Education Department(No.12521073)National Natural Science Youth Fund(No.61300115).
文摘In the field of organic synthesis,the core objective of retrosynthetic methods is to deduce possible synthetic routes and precursor molecules for complex target molecules.Traditional retrosynthetic methods,such as template-based retrosynthesis,have high accuracy and interpretability in specific types of reactions but are limited by the scope of the template library,making it difficult to adapt to new or uncommon reaction types.Moreover,sequence-to-sequence retrosynthetic prediction methods,although they enhance the flexibility of prediction,often overlook the complexity of molecular graph structures and the actual interactions between atoms,which limits the accuracy and reliability of the predictions.To address these limitations,this paper proposes a Molecular Retrosynthesis Top-k Prediction based on the Latent Generation Process(MRLGP)that uses latent variables from graph neural networks to model the generation process and produce diverse set of reactants.Utilising an encoding method based on Graphormer,the authors have also introduced topology-aware positional encoding to better capture the interactions between atomic nodes in the molecular graph structure,thereby more accurately simulating the retrosynthetic process.The MRLGP model significantly enhances the accuracy and diversity of predictions by correlating discrete latent variables with the reactant generation process and progressively constructing molecular graphs using a variational autoregressive decoder.Experimental results on benchmark datasets such as USPTO-50k,USPTO-Full,and USPTO-DIVERSE demonstrate that MRLGP outperforms baseline models on multiple Top-k evaluation metrics.Additionally,ablation experiments conducted on the USPTO-50K dataset further validate the effectiveness of the methods used in the encoder and decoder parts of the model.
基金National Natural Science Foundation of China(Grant No.U23A20106)National Key Research and Development Program of China(Grant No.91510100MA6CG8UJ4K)。
文摘Transfer RNAs(tRNAs)adopt a stable L-shaped tertiary structure crucial for their involvement in protein translation.Among various divalent metal ions,magnesium ions play a pivotal role in preserving the tertiary structure of tRNA.However,the precise location of the Mg^(2+)binding pocket in human tRNA remains elusive.In this investigation,we identified the Mg^(2+)binding site within human tRNAGln using suppressor tRNA^(Gln).This variant of tRNA recognizes premature stop codons(specificlly UAG)and facilitates the expression of fll-length proteis.By mutating sites 8 and C72 in supprssr tRNAcl,we assessed the decoding efficiency of the resulting mutant suppressor tRNAs,which serves as a measure of tRNA's ability to decode genetic information.Our analysis revealed that the U8C mutant suppressor tRNA exhibited a significantly lower Mg^(2+)content compared to the C72U mutant.Furthermore,we observed a notable reduction in decoding efficiency in the U8-mutated suppressor tRNA,as evidenced by GFP fluorescence and Western blotting analysis.Conversely,mutations at the C72 site had a comparatively minor impact on decoding efficiency.These findings underscored the tight binding of Mg^(2+)to the U8 site of human tRNAGln,crucial for maintaining the stability of tRNA tertiary structure and translation efficacy.Additionally,our investigation delved into the influence of glutamine availability on tRNA decoding efficiency at the cellular level.The results indicated that both the concentration of amino acids and the codon context of TAG could modulate tRNA decoding efficiency.This study provided valuable insights into the structure and function of tRNA,laying the groundwork for further exploration in this field.
基金Supported by the Fifth Batch of Innovation Teams of Wuzhou Meteorological Bureau“Wuzhou Innovation Team for Enhancing the Comprehensive Meteorological Observation Ability through Digitization and Intelligence”Project of Wuzhou Science and Technology Bureau(202402122)Wuzhou Science and Technology Planning Project(202402119).
文摘In this paper,Wuzhou City of Guangxi was taken as the research object.Through the design of a climatic data warehousing system,the decoding methods of surface meteorological data and their application in the management of climatic data were explored.Based on the parsing technology of the monthly report of surface meteorological records(A-file),the design of Wuzhou climatic data warehousing system was realized,completing the precise extraction and database construction of observational elements such as regional temperature,wind direction,and weather phenomena.Based on this system,the meteorological data in 2024 were analyzed,and the probabilistic characteristics of dominant wind direction in Wuzhou(northeast wind accounting for the largest proportion),the spatiotemporal distribution patterns of extreme temperatures(annual extreme high temperature of 37.1℃in August and extreme low temperature of 1.9℃in January),and the general climatic overview of Wuzhou City(annual precipitation 3.2%higher than the standard value)were revealed.The research shows that climate change has a significant impact on agricultural production and economic development in Wuzhou City,and the construction of a reasonable climatic data database is of great significance for enhancing professional meteorological service capabilities in the context of climate change.This study not only provides a scientific basis for the economic development of Wuzhou region,but also offers reference ideas for other regions to cope with regional climate adaptation planning.
基金funded by Beijing Institute of Electronic Science and Technology Postgraduate Excellence Demonstration Course Project(20230002Z0452).
文摘Among the four candidate algorithms in the fourth round of NIST standardization,the BIKE(Bit Flipping Key Encapsulation)scheme has a small key size and high efficiency,showing good prospects for application.However,the BIKE scheme based on QC-MDPC(Quasi Cyclic Medium Density Parity Check)codes still faces challenges such as the GJS attack and weak key attacks targeting the decoding failure rate(DFR).This paper analyzes the BGF decoding algorithm of the BIKE scheme,revealing two deep factors that lead to DFR,and proposes a weak key optimization attack method for the BGF decoding algorithm based on these two factors.The proposed method constructs a new weak key set,and experiment results eventually indicate that,considering BIKE’s parameter set targeting 128-bit security,the average decryption failure rate is lowerly bounded by.This result not only highlights a significant vulnerability in the BIKE scheme but also provides valuable insights for future improvements in its design.By addressing these weaknesses,the robustness of QC-MDPC code-based cryptographic systems can be enhanced,paving the way for more secure post-quantum cryptographic solutions.
基金supported by China Postdoctoral Science Foundation(No.2020M670469)National Key Research and Development Program of China(No.2019YFB1803303,No.2020YFB1806702).
文摘Linear programming(LP)decoding is a classic decoding method for linear block codes,and has attracted recent researches because its potential in joint channel processing.However,for polar codes,LP decoders has long been outperformed by CRCaided successive cancellation list(CA-SCL)decoders.To increase the competitiveness of 5G NR LP polar decoding,it is possible to gain performance improvements by exploiting the cyclic redundancy check(CRC)setup.In this paper,we propose a combined scheme of reduced sparsified factor graph-sparsified CRC(RSFG-SCRC)and augmented generator matrix-CRC(AGM-CRC),for polytope generation in adaptive linear programming(ALP)decoder for 5G polar codes.Augmented generator matrix(AGM)polytope and improved maximum cycle strategy-auxiliary node pairs 4(MCS-ANP-4)algorithm are proposed,to make efficient use of CRC constraints and minimize the constraint size for the decoder.Numerical simulations show that adaptive linear programming decoders with our proposed RSFG-SCRC and AGM-CRC polytopes can achieve significantly better block error rate(BLER)performance than a benchmark CA-SCL-8 decoder especially in harsh low-to-medium SNR regions.
基金supported by National Natural Science Foundation of China(No.52374155)Anhui Provincial Natural Science Foundation(No.2308085 MF218).
文摘The convolutional neural network(CNN)method based on DeepLabv3+has some problems in the semantic segmentation task of high-resolution remote sensing images,such as fixed receiving field size of feature extraction,lack of semantic information,high decoder magnification,and insufficient detail retention ability.A hierarchical feature fusion network(HFFNet)was proposed.Firstly,a combination of transformer and CNN architectures was employed for feature extraction from images of varying resolutions.The extracted features were processed independently.Subsequently,the features from the transformer and CNN were fused under the guidance of features from different sources.This fusion process assisted in restoring information more comprehensively during the decoding stage.Furthermore,a spatial channel attention module was designed in the final stage of decoding to refine features and reduce the semantic gap between shallow CNN features and deep decoder features.The experimental results showed that HFFNet had superior performance on UAVid,LoveDA,Potsdam,and Vaihingen datasets,and its cross-linking index was better than DeepLabv3+and other competing methods,showing strong generalization ability.
