With the advancement of wireless communication technology,intelligent antenna technologies such as beam scanning and beamforming have been extensively applied in operators'5G networks,supported by mature technical...With the advancement of wireless communication technology,intelligent antenna technologies such as beam scanning and beamforming have been extensively applied in operators'5G networks,supported by mature technical solutions.However,the unique characteristics of the railway industry—such as the significant spacing between stations covered by wireless private networks,the high speed of train operations,and the necessity for high network reliability—pose elevated requirements for the construction of 5G private networks.An analysis was conducted on the challenges associated with railway 5G private network coverage.The investigation explored the adaptability of smart antenna technologies in various railway scenarios in combination with the principles and advantages of these technologies.This study analyzed the application prospects of smart antenna technologies in railway 5G private networks,taking into account the characteristics of various train operation scenarios.It evaluated the value of these technologies in enhancing the wireless coverage quality of railway 5G private networks in different scenarios.The findings aim to offer new insights and recommendations for the construction and deployment of railway 5G private networks.展开更多
Founded in September 2020,the International SparkLink Alliance(iSLA)now has approximately 1,200 members in diverse sectors including terminals,homes,vehicles,manufacturing,transportation,finance and healthcare.The iSL...Founded in September 2020,the International SparkLink Alliance(iSLA)now has approximately 1,200 members in diverse sectors including terminals,homes,vehicles,manufacturing,transportation,finance and healthcare.The iSLA has established a technical standards system for wireless short-range communication covering full-stack standards such as the end-to-end protocol system.展开更多
Sparse code multiple access(SCMA) is a novel non-orthogonal multiple access scheme proposed to meet the challenging demand of the future 5G communications, especially in support of the massive connections. The coded b...Sparse code multiple access(SCMA) is a novel non-orthogonal multiple access scheme proposed to meet the challenging demand of the future 5G communications, especially in support of the massive connections. The coded bits from each data stream will be directly mapped as multi-dimensional SCMA codeword in complex domain and then spread onto the physical resource elements in a sparse manner. The number of codewords that can be nonorthogonally multiplexed in one SCMA block can be made much larger than the number of orthogonal resource elements therein, resulting in an overloaded system. The sparsity in the spreading pattern and the design in the multidimensional modulator jointly ensure the SCMA codewords can be robustly decoded with low complexity. In this paper, we focus on the low complexity receiver design and verified the superior of an SCMA system via simulations and real-time prototyping. Lab tests and field tests all show that SCMA is a promising candidate for 5G non-orthogonal multiple access which can provide up to 300% overloading that triples the whole system throughput while still enjoying the link performance close to orthogonal transmissions.展开更多
INFORMATION and Comnmnication Technology(ICT),which is in frill swing in the modern era,has been reconstructing human society.Dedi-cated to innovative ICT products,services,and solutions,Huawei is com-mitted to buildi...INFORMATION and Comnmnication Technology(ICT),which is in frill swing in the modern era,has been reconstructing human society.Dedi-cated to innovative ICT products,services,and solutions,Huawei is com-mitted to building a better and well-connected world and so boosting hnman development.It has also become one of the leading ICT manufacturers with sales revenue of RMB 288 million in 2014,up ao percent over the previous year.展开更多
Accurate prediction of molecular properties is crucial for selecting compounds with ideal properties and reducing the costs and risks of trials.Traditional methods based on manually crafted features and graph-based me...Accurate prediction of molecular properties is crucial for selecting compounds with ideal properties and reducing the costs and risks of trials.Traditional methods based on manually crafted features and graph-based methods have shown promising results in molecular property prediction.However,traditional methods rely on expert knowledge and often fail to capture the complex structures and interactions within molecules.Similarly,graph-based methods typically overlook the chemical structure and function hidden in molecular motifs and struggle to effectively integrate global and local molecular information.To address these limitations,we propose a novel fingerprint-enhanced hierarchical graph neural network(FH-GNN)for molecular property prediction that simultaneously learns information from hierarchical molecular graphs and fingerprints.The FH-GNN captures diverse hierarchical chemical information by applying directed message-passing neural networks(D-MPNN)on a hierarchical molecular graph that integrates atomic-level,motif-level,and graph-level information along with their relationships.Addi-tionally,we used an adaptive attention mechanism to balance the importance of hierarchical graphs and fingerprint features,creating a comprehensive molecular embedding that integrated hierarchical mo-lecular structures with domain knowledge.Experiments on eight benchmark datasets from MoleculeNet showed that FH-GNN outperformed the baseline models in both classification and regression tasks for molecular property prediction,validating its capability to comprehensively capture molecular informa-tion.By integrating molecular structure and chemical knowledge,FH-GNN provides a powerful tool for the accurate prediction of molecular properties and aids in the discovery of potential drug candidates.展开更多
Planar lightwave circuit(PLC)splitters have long been foundational components in passive optical communication networks,achieving commercial success since the 1990s.However,their inherent fixed splitting ratios impose...Planar lightwave circuit(PLC)splitters have long been foundational components in passive optical communication networks,achieving commercial success since the 1990s.However,their inherent fixed splitting ratios impose significant limitations on capacity expansion,often requiring physical replacement and causing service disruptions.Thermally tunable optical splitters address this challenge by enabling adjustable splitting ratios,but their operation is contingent upon a continuous power supply and complex driving systems.In this work,we present a novel,non-volatile tunable PLC platform based on Sb_(2)S_(3)phase-change materials.The proposed device,which incor-porates a Mach-Zehnder interferometer(MZI)optical switch structure,offers tunable splitting ratios via laser-direct writing or ohmic heating,providing flexible reconfiguration capabilities.Experimental results demonstrate non-volatile power splitting ranging from 50∶50 to 20∶80,with a modest increase of approximately 1 dB in additional loss.This work highlights the potential of the proposed platform for low-power,high-efficiency,and reconfigurable photonic networks.展开更多
Time Division Multiplexing-Passive Optical Networks(TDM-PONs)play a vital role in Fiberto-the-Home(FTTH)deployments.To improve the service quality of home networks,FTTH is expanding to the Fiber-to-the-Room(FTTR)scena...Time Division Multiplexing-Passive Optical Networks(TDM-PONs)play a vital role in Fiberto-the-Home(FTTH)deployments.To improve the service quality of home networks,FTTH is expanding to the Fiber-to-the-Room(FTTR)scenario,where fibers are deployed to connect individual rooms(i.e.,Fiber In-premises Network(FIN)in the ITU-T G.9940 standard).In this scenario,a point-to-multipoint(P2MP)fiber network is deployed as FTTR FIN to offer gigabit access to each room,which forms a two-tier cascaded network together with the FTTH segment.To optimize the capacity utilization of the cascaded network and reduce the overall system cost,a centralized architecture,known as Centralized Fixed Access Network(C-FAN),has been introduced.C-FAN centralizes the medium access control(MAC)modules of both the FTTH and FTTR networks at the FTTH’s Optical Line Terminal(OLT)for unified control and management of the cascaded network.We develop a unified bandwidth scheduling protocol by extending the ITU-T PON standard for both the upstream and downstream directions of C-FAN.We also propose a unified dynamic bandwidth allocation(UDBA)algorithm for efficient bandwidth allocation for multiple traffic flows in the two-tier cascaded network.Simulations are conducted to evaluate the performance of the proposed control protocol and the UDBA algorithm.The results show that,in comparison to the conventional DBA algorithm,the UDBA algorithm can utilize upstream bandwidth more efficiently to reduce packet delay and loss,without adversely impacting downstream transmission performance.展开更多
Terahertz(THz) and millimeter Wave(mmWave) have been considered as potential frequency bands for 6G cellular systems to meet the need of ultra-high data rates. However, indoor communications could be blocked in THz/mm...Terahertz(THz) and millimeter Wave(mmWave) have been considered as potential frequency bands for 6G cellular systems to meet the need of ultra-high data rates. However, indoor communications could be blocked in THz/mmW cellular systems due to the high free-space propagation loss. Deploying a large number of small base stations indoors has been considered as a promising solution for solving indoor coverage problems. However, base station dense deployment leads to a significant increase in system energy consumption. In this paper, we develop a novel ultra-efficient energy-saving mechanism with the aim of reducing energy consumption in 6G distributed indoor base station scenarios. Unlike the existing relevant protocol framework of 3GPP, which operates the cellular system based on constant system signaling messages(including cell ID, cell reselection information, etc.), the proposed mechanism eliminates the need for system messages. The intuition comes from the observation that the probability of having no users within the coverage area of an indoor base station is high, hence continuously sending system messages to guarantee the quality of service is unnecessary in indoor scenarios. Specifically, we design a dedicated beacon signal to detect whether there are users in the coverage area of the base station and switch off the main communication module when there are no active users for energy saving. The beacon frame structure is carefully designed based on the existing 3GPP specifications with minimal protocol modifications, and the protocol parameters involved are optimized. Simulation results show that the proposed mechanism can reduce the system energy from the order of tens of watts to the order of hundreds of milliwatts. Compared to traditional energy-saving schemes, the proposed mechanism achieves an average energy-saving gain of 58%, with a peak energy-saving gain of 90%.展开更多
To achieve a low-complexity nonlinearity compensation(NLC)in high-symbol-rate(HSR)systems,we propose a modified weighted digital backpropagation(M-W-DBP)by jointly shifting the calculated position of nonlinear phase n...To achieve a low-complexity nonlinearity compensation(NLC)in high-symbol-rate(HSR)systems,we propose a modified weighted digital backpropagation(M-W-DBP)by jointly shifting the calculated position of nonlinear phase noise and considering the correlation of neighboring symbols in the NLC section of DBP.Based on this model,with the aid of neural network optimization,a learned version of M-W-DBP(M-W-LDBP)is also proposed and explored.Furthermore,enough technical details are revealed for the first time,including the principle of our proposed M-W-DBP and M-W-LDBP,the training process,and the complexity analysis of different DBPclass NLC algorithms.Evaluated numerically with QPSK,16QAM,and PS-64QAM modulation formats,1-step-per-span(1-StPS)M-W-DBP/LDBP achieves up to 1.29/1.49 dB and 0.63/0.74 dB signal-to-noise ratio improvement compared to chromatic dispersion compensation(CDC)in 90-GBaud and 128-GBaud 1000-km single-channel transmission systems,respectively.Moreover,1-StPS M-W-DBP/LDBP provides a more powerful NLC ability than 2-StPS LDBP but only needs about 60%of the complexity.The effectiveness of the proposed M-W-DBP and M-W-LDBP in the presence of laser phase noise is also verified and the necessity of using the learned version of M-WDBP is also discussed.This work is a comprehensive study of M-W-DBP/LDBP and other DBP-class NLC algorithms in HSR systems.展开更多
Bilingual lexicon induction focuses on learning word translation pairs,also known as bitexts,from monolingual corpora by establishing a mapping between the source and target embedding spaces.Despite recent advancement...Bilingual lexicon induction focuses on learning word translation pairs,also known as bitexts,from monolingual corpora by establishing a mapping between the source and target embedding spaces.Despite recent advancements,bilingual lexicon induction is limited to inducing bitexts consisting of individual words,lacking the ability to handle semantics-rich phrases.To bridge this gap and support downstream cross-lingual tasks,it is practical to develop a method for bilingual phrase induction that extracts bilingual phrase pairs from monolingual corpora without relying on cross-lingual knowledge.In this paper,the authors propose a novel phrase embedding training method based on the skip-gram structure.Specifically,a local hard negative sampling strategy that utilises negative samples of central tokens in sliding windows to enhance phrase embedding learning is introduced.The proposed method achieves competitive or superior performance compared to baseline approaches,with exceptional results recorded for distant languages.Additionally,we develop a phrase representation learning method that leverages multilingual pre-trained language models.These mPLMs-based representations can be combined with the above-mentioned static phrase embeddings to further improve the accuracy of the bilingual phrase induction task.We manually construct a dataset of bilingual phrase pairs and integrate it with MUSE to facilitate the bilingual phrase induction task.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
Accurate estimation on the state of health(SOH)is essential for ensuring the safe and reliable operation of batteries.Traditional assessment methods primarily focus on electrical attributes for capacity decay,often ov...Accurate estimation on the state of health(SOH)is essential for ensuring the safe and reliable operation of batteries.Traditional assessment methods primarily focus on electrical attributes for capacity decay,often overlooking the impact of thermal distribution on battery aging.However,thermal effect is a critical factor for degradation process and associated risks throughout their service life.In this paper,we introduce a novel deep learning framework specially designed to estimate the capacity and thermal risks of lithium-ion batteries(LIBs).This model consists of two main components that leverage computer vision technology.One predicts battery capacity by integrating the advantages of thermal and electrical features using a temporal pattern attention(TPA)mechanism,while the other assesses thermal risk by incorporating temperature variation to provide early warnings of potential hazards.An infrared camera is deployed to record temperature evolution of LIBs during the electrochemical process.The thermal heterogeneities are recorded by infrared camera,and the corresponding temperature evolutions are extracted as representative features for analysis.The proposed model demonstrates high accuracy and stability,with an average root mean square error(RMSE)of 0.67% for capacity estimation and accuracy exceeding 93.9% for risk prediction,underscoring the importance of integrating spatial temperature distribution into battery health assessments.This work offers valuable insights for the development of intelligent and robust battery management systems.展开更多
Traceability and trustiness are two critical issues in the logistics sector.Blockchain provides a potential way for logistics tracking systems due to its traits of tamper resistance.However,it is non-trivial to apply ...Traceability and trustiness are two critical issues in the logistics sector.Blockchain provides a potential way for logistics tracking systems due to its traits of tamper resistance.However,it is non-trivial to apply blockchain on logistics because of firstly,the binding relationship between virtue data and physical location cannot be guaranteed so that frauds may exist.Secondly,it is neither practical to upload complete data on the blockchain due to the limited storage resources nor convincing to trust the digest of the data.This paper proposes a traceable and trustable consortium blockchain for logistics T^(2)L to provide an efficient solution to the mentioned problems.Specifically,the authenticated geocoding data from telecom operators’base stations are adopted to ensure the location credibility of the data before being uploaded to the blockchain for the purpose of reliable traceability of the logistics.Moreover,we propose a scheme based on Zero Knowledge Proof of Retrievability(ZK BLS-PoR)to ensure the trustiness of the data digest and the proofs to the blockchain.Any user in the system can check the data completeness by verifying the proofs instead of downloading and examining the whole data based on the proposed ZK BLS-PoR scheme,which can provide solid theoretical verification.In all,the proposed T^(2)L framework is a traceable and trustable logistics system with a high level of security.展开更多
Robotics plays an increasingly important role in all areas of human activity.Teleoperation robots can effectively ensure the safety of operators when operating in difficult and high‐risk industrial scenarios,which ob...Robotics plays an increasingly important role in all areas of human activity.Teleoperation robots can effectively ensure the safety of operators when operating in difficult and high‐risk industrial scenarios,which obviously requires instant and efficient signal compression and transmission in the system.However,most of the existing algorithms cannot fully explore the correlation within the signal,which mostly limits the compression efficiency.In this paper,a novel prediction‐aided kinaestheticsignal compression framework is proposed,which uses semantic communication methods to explore the temporal and spatial correlations of signals and employs neural network predictions to uncover their internal correlations.Specifically,the signal is first divided into two groups:the base part and the predictable part,and then a series of transformation matrices are introduced to establish the correlation between the two groups of the signal,which can be automatically optimised by a well‐designed neural network.This strategy of using learnable transformation matrices for prediction can not only accurately construct the correlation within the signal through massive data mining but also efficiently execute inference in a simple matrix multiplication computing form.Experimental results demonstrate that the proposed method outperforms the existing traditional tactile codecs and the latest tactile semantic communication methods.展开更多
In offshore maritime communication sys-tems,base stations(BSs)are employed along the coastline to provide high-speed data service for ves-sels in coastal sea areas.To ensure the line-of-sight propagation of BS-vessel ...In offshore maritime communication sys-tems,base stations(BSs)are employed along the coastline to provide high-speed data service for ves-sels in coastal sea areas.To ensure the line-of-sight propagation of BS-vessel links,high transceiver an-tenna height is required,which limits the number of geographically available sites for BS deployment,and imposes a high cost for realizing effective wide-area coverage.In this paper,the joint user association and power allocation(JUAPA)problem is investigated to enhance the coverage of offshore maritime systems.By exploiting the characteristics of network topology as well as vessels’motion in offshore communica-tions,a multi-period JUAPA problem is formulated to maximize the number of ships that can be simultane-ously served by the network.This JUAPA problem is intrinsically non-convex and subject to mixed-integer constraints,which is difficult to solve either analyt-ically or numerically.Hence,we propose an iterative augmentation based framework to efficiently select the active vessels,where the JUAPA scheme is iteratively optimized by the network for increasing the number of the selected vessels.More specifically,in each itera-tion,the user association variables and power alloca-tion variables are determined by solving two separate subproblems,so that the JUAPA strategy can be up-dated in a low-complexity manner.The performance of the proposed JUAPA method is evaluated by exten-sive simulation,and numerical results indicate that it can effectively increase the number of vessels served by the network,and thus enhances the coverage of off-shore systems.展开更多
In this work,a frame work for time-varying channel modeling and simulation is proposed by using neural network(NN)to overcome the shortcomings in geometry based stochastic model(GBSM)and simulation approach.Two NN mod...In this work,a frame work for time-varying channel modeling and simulation is proposed by using neural network(NN)to overcome the shortcomings in geometry based stochastic model(GBSM)and simulation approach.Two NN models are developed for modeling of path loss together with shadow fading(SF)and joint small scale channel parameters.The NN models can predict path loss plus SF and small scale channel parameters accurately compared with measurement at 26 GHz performed in an outdoor microcell.The time-varying path loss and small scale channel parameters generated by the NN models are proposed to replace the empirical path loss and channel parameter random numbers in GBSM-based framework to playback the measured channel and match with its environment.Moreover,the sparse feature of clusters,delay and angular spread,channel capacity are investigated by a virtual array measurement at 28 GHz in a large waiting hall.展开更多
Internet of Things(IoT) is one of the targeted application scenarios of fifth generation(5 G) wireless communication.IoT brings a large amount of data transported on the network.Considering those data,machine learning...Internet of Things(IoT) is one of the targeted application scenarios of fifth generation(5 G) wireless communication.IoT brings a large amount of data transported on the network.Considering those data,machine learning(ML) algorithms can be naturally utilized to make network efficiently and reliably.However,how to fully apply ML to IoT driven wireless network is still open.The fundamental reason is that wireless communication pursuits the high capacity and quality facing the challenges from the varying and fading wireless channel.So in this paper,we explore feasible combination for ML and IoT driven wireless network from wireless channel perspective.Firstly,a three-level structure of wireless channel fading features is defined in order to classify the versatile propagation environments.This three-layer structure includes scenario,meter and wavelength levels.Based on this structure,there are different tasks like service prediction and pushing,self-organization networking,self adapting largescale fading modeling and so on,which can be abstracted into problems like regression,classification,clustering,etc.Then,we introduce corresponding ML methods to different levelsfrom channel perspective,which makes their interdisciplinary research promisingly.展开更多
Recently DC relay has been concerned as a key component in DC power distribution,management and control systems like aircraft,new energy vehicle,IT and communication industries.Ordinarily,magnetic force and contact mo...Recently DC relay has been concerned as a key component in DC power distribution,management and control systems like aircraft,new energy vehicle,IT and communication industries.Ordinarily,magnetic force and contact moving speed have great influence on arc behaviours in the breaking process.This paper focuses on the numerical investigation of arc during the contact opening process in a real 400V/20 A DC relay product coupling with an inductive load circuit.A 3D air arc model based on the magneto-hydrodynamic theory was built and calculated.A method coupling different computational software was used to take the nonlinear permanent magnet and contact opening process into consideration simultaneously.Arc behaviours under different magnetic field and contact opening speed were presented and discussed carefully.It has been found that the increase of the magnetic field is beneficial to the quick build-up of arc length and voltage.Arc breaking duration becomes shorter with the increase in contact opening speed from 63.5 rad s^-1 to 94.5 rad s^-1,such reduction is less significant with an increase of opening speed from 94.5 rad s^-l to 118.5 rad s^-1.展开更多
Atomic layer deposition(ALD)has become an indispensable thin-film technology in the contemporary microelectronics industry.The unique self-limited layer-by-layer growth feature of ALD has outstood this technology to d...Atomic layer deposition(ALD)has become an indispensable thin-film technology in the contemporary microelectronics industry.The unique self-limited layer-by-layer growth feature of ALD has outstood this technology to deposit highly uniform conformal pinhole-free thin films with angstrom-level thickness control,particularly on 3D topologies.Over the years,the ALD technology has enabled not only the successful downscaling of the microelectronic devices but also numerous novel 3D device structures.As ALD is essentially a variant of chemical vapor deposition,a comprehensive understanding of the involved chemistry is of crucial importance to further develop and utilize this technology.To this end,we,in this review,focus on the surface chemistry and precursor chemistry aspects of ALD.We first review the surface chemistry of the gas–solid ALD reactions and elaborately discuss the associated mechanisms for the film growth;then,we review the ALD precursor chemistry by comparatively discussing the precursors that have been commonly used in the ALD processes;and finally,we selectively present a few newly-emerged applications of ALD in microelectronics,followed by our perspective on the future of the ALD technology.展开更多
文摘With the advancement of wireless communication technology,intelligent antenna technologies such as beam scanning and beamforming have been extensively applied in operators'5G networks,supported by mature technical solutions.However,the unique characteristics of the railway industry—such as the significant spacing between stations covered by wireless private networks,the high speed of train operations,and the necessity for high network reliability—pose elevated requirements for the construction of 5G private networks.An analysis was conducted on the challenges associated with railway 5G private network coverage.The investigation explored the adaptability of smart antenna technologies in various railway scenarios in combination with the principles and advantages of these technologies.This study analyzed the application prospects of smart antenna technologies in railway 5G private networks,taking into account the characteristics of various train operation scenarios.It evaluated the value of these technologies in enhancing the wireless coverage quality of railway 5G private networks in different scenarios.The findings aim to offer new insights and recommendations for the construction and deployment of railway 5G private networks.
文摘Founded in September 2020,the International SparkLink Alliance(iSLA)now has approximately 1,200 members in diverse sectors including terminals,homes,vehicles,manufacturing,transportation,finance and healthcare.The iSLA has established a technical standards system for wireless short-range communication covering full-stack standards such as the end-to-end protocol system.
文摘Sparse code multiple access(SCMA) is a novel non-orthogonal multiple access scheme proposed to meet the challenging demand of the future 5G communications, especially in support of the massive connections. The coded bits from each data stream will be directly mapped as multi-dimensional SCMA codeword in complex domain and then spread onto the physical resource elements in a sparse manner. The number of codewords that can be nonorthogonally multiplexed in one SCMA block can be made much larger than the number of orthogonal resource elements therein, resulting in an overloaded system. The sparsity in the spreading pattern and the design in the multidimensional modulator jointly ensure the SCMA codewords can be robustly decoded with low complexity. In this paper, we focus on the low complexity receiver design and verified the superior of an SCMA system via simulations and real-time prototyping. Lab tests and field tests all show that SCMA is a promising candidate for 5G non-orthogonal multiple access which can provide up to 300% overloading that triples the whole system throughput while still enjoying the link performance close to orthogonal transmissions.
文摘INFORMATION and Comnmnication Technology(ICT),which is in frill swing in the modern era,has been reconstructing human society.Dedi-cated to innovative ICT products,services,and solutions,Huawei is com-mitted to building a better and well-connected world and so boosting hnman development.It has also become one of the leading ICT manufacturers with sales revenue of RMB 288 million in 2014,up ao percent over the previous year.
基金supported by Macao Science and Technology Development Fund,Macao SAR,China(Grant No.:0043/2023/AFJ)the National Natural Science Foundation of China(Grant No.:22173038)Macao Polytechnic University,Macao SAR,China(Grant No.:RP/FCA-01/2022).
文摘Accurate prediction of molecular properties is crucial for selecting compounds with ideal properties and reducing the costs and risks of trials.Traditional methods based on manually crafted features and graph-based methods have shown promising results in molecular property prediction.However,traditional methods rely on expert knowledge and often fail to capture the complex structures and interactions within molecules.Similarly,graph-based methods typically overlook the chemical structure and function hidden in molecular motifs and struggle to effectively integrate global and local molecular information.To address these limitations,we propose a novel fingerprint-enhanced hierarchical graph neural network(FH-GNN)for molecular property prediction that simultaneously learns information from hierarchical molecular graphs and fingerprints.The FH-GNN captures diverse hierarchical chemical information by applying directed message-passing neural networks(D-MPNN)on a hierarchical molecular graph that integrates atomic-level,motif-level,and graph-level information along with their relationships.Addi-tionally,we used an adaptive attention mechanism to balance the importance of hierarchical graphs and fingerprint features,creating a comprehensive molecular embedding that integrated hierarchical mo-lecular structures with domain knowledge.Experiments on eight benchmark datasets from MoleculeNet showed that FH-GNN outperformed the baseline models in both classification and regression tasks for molecular property prediction,validating its capability to comprehensively capture molecular informa-tion.By integrating molecular structure and chemical knowledge,FH-GNN provides a powerful tool for the accurate prediction of molecular properties and aids in the discovery of potential drug candidates.
基金sponsored by the National Key Research and Development Program of China(2020YFA0714504,2019YFA0709100)the program of the National Natural Science Foundation of China(U24A20309,62305043).
文摘Planar lightwave circuit(PLC)splitters have long been foundational components in passive optical communication networks,achieving commercial success since the 1990s.However,their inherent fixed splitting ratios impose significant limitations on capacity expansion,often requiring physical replacement and causing service disruptions.Thermally tunable optical splitters address this challenge by enabling adjustable splitting ratios,but their operation is contingent upon a continuous power supply and complex driving systems.In this work,we present a novel,non-volatile tunable PLC platform based on Sb_(2)S_(3)phase-change materials.The proposed device,which incor-porates a Mach-Zehnder interferometer(MZI)optical switch structure,offers tunable splitting ratios via laser-direct writing or ohmic heating,providing flexible reconfiguration capabilities.Experimental results demonstrate non-volatile power splitting ranging from 50∶50 to 20∶80,with a modest increase of approximately 1 dB in additional loss.This work highlights the potential of the proposed platform for low-power,high-efficiency,and reconfigurable photonic networks.
基金supported by National Nature Science Founding of China(62101372)Open Fund of IPOC(BUPT,IPOC2022A07)+1 种基金State Key Laboratory of Advanced Optical Communication Systems and Networks(2023GZKF11)Leading Youth Talents of Innovation and Entrepreneurship of Gusu(ZXL2023162).
文摘Time Division Multiplexing-Passive Optical Networks(TDM-PONs)play a vital role in Fiberto-the-Home(FTTH)deployments.To improve the service quality of home networks,FTTH is expanding to the Fiber-to-the-Room(FTTR)scenario,where fibers are deployed to connect individual rooms(i.e.,Fiber In-premises Network(FIN)in the ITU-T G.9940 standard).In this scenario,a point-to-multipoint(P2MP)fiber network is deployed as FTTR FIN to offer gigabit access to each room,which forms a two-tier cascaded network together with the FTTH segment.To optimize the capacity utilization of the cascaded network and reduce the overall system cost,a centralized architecture,known as Centralized Fixed Access Network(C-FAN),has been introduced.C-FAN centralizes the medium access control(MAC)modules of both the FTTH and FTTR networks at the FTTH’s Optical Line Terminal(OLT)for unified control and management of the cascaded network.We develop a unified bandwidth scheduling protocol by extending the ITU-T PON standard for both the upstream and downstream directions of C-FAN.We also propose a unified dynamic bandwidth allocation(UDBA)algorithm for efficient bandwidth allocation for multiple traffic flows in the two-tier cascaded network.Simulations are conducted to evaluate the performance of the proposed control protocol and the UDBA algorithm.The results show that,in comparison to the conventional DBA algorithm,the UDBA algorithm can utilize upstream bandwidth more efficiently to reduce packet delay and loss,without adversely impacting downstream transmission performance.
基金supported by the National Natural Science Foundation of China under Grant No. 62201121the Fundamental Research Funds for Central Universities under Grant No. ZYGX2024XJ070.
文摘Terahertz(THz) and millimeter Wave(mmWave) have been considered as potential frequency bands for 6G cellular systems to meet the need of ultra-high data rates. However, indoor communications could be blocked in THz/mmW cellular systems due to the high free-space propagation loss. Deploying a large number of small base stations indoors has been considered as a promising solution for solving indoor coverage problems. However, base station dense deployment leads to a significant increase in system energy consumption. In this paper, we develop a novel ultra-efficient energy-saving mechanism with the aim of reducing energy consumption in 6G distributed indoor base station scenarios. Unlike the existing relevant protocol framework of 3GPP, which operates the cellular system based on constant system signaling messages(including cell ID, cell reselection information, etc.), the proposed mechanism eliminates the need for system messages. The intuition comes from the observation that the probability of having no users within the coverage area of an indoor base station is high, hence continuously sending system messages to guarantee the quality of service is unnecessary in indoor scenarios. Specifically, we design a dedicated beacon signal to detect whether there are users in the coverage area of the base station and switch off the main communication module when there are no active users for energy saving. The beacon frame structure is carefully designed based on the existing 3GPP specifications with minimal protocol modifications, and the protocol parameters involved are optimized. Simulation results show that the proposed mechanism can reduce the system energy from the order of tens of watts to the order of hundreds of milliwatts. Compared to traditional energy-saving schemes, the proposed mechanism achieves an average energy-saving gain of 58%, with a peak energy-saving gain of 90%.
基金supported in part by National Natural Science Foundation of China(No.62271080)in part by Fund of State Key Laboratory of IPOC(BUPT)(No.IPOC2022ZT06)in part by BUPT Excellent Ph.D Students Foundation(No.CX2022102).
文摘To achieve a low-complexity nonlinearity compensation(NLC)in high-symbol-rate(HSR)systems,we propose a modified weighted digital backpropagation(M-W-DBP)by jointly shifting the calculated position of nonlinear phase noise and considering the correlation of neighboring symbols in the NLC section of DBP.Based on this model,with the aid of neural network optimization,a learned version of M-W-DBP(M-W-LDBP)is also proposed and explored.Furthermore,enough technical details are revealed for the first time,including the principle of our proposed M-W-DBP and M-W-LDBP,the training process,and the complexity analysis of different DBPclass NLC algorithms.Evaluated numerically with QPSK,16QAM,and PS-64QAM modulation formats,1-step-per-span(1-StPS)M-W-DBP/LDBP achieves up to 1.29/1.49 dB and 0.63/0.74 dB signal-to-noise ratio improvement compared to chromatic dispersion compensation(CDC)in 90-GBaud and 128-GBaud 1000-km single-channel transmission systems,respectively.Moreover,1-StPS M-W-DBP/LDBP provides a more powerful NLC ability than 2-StPS LDBP but only needs about 60%of the complexity.The effectiveness of the proposed M-W-DBP and M-W-LDBP in the presence of laser phase noise is also verified and the necessity of using the learned version of M-WDBP is also discussed.This work is a comprehensive study of M-W-DBP/LDBP and other DBP-class NLC algorithms in HSR systems.
基金National Key Research and Development Program of China,Grant/Award Number:2023YFC3305003National Natural Science Foundation of China,Grant/Award Number:62376076。
文摘Bilingual lexicon induction focuses on learning word translation pairs,also known as bitexts,from monolingual corpora by establishing a mapping between the source and target embedding spaces.Despite recent advancements,bilingual lexicon induction is limited to inducing bitexts consisting of individual words,lacking the ability to handle semantics-rich phrases.To bridge this gap and support downstream cross-lingual tasks,it is practical to develop a method for bilingual phrase induction that extracts bilingual phrase pairs from monolingual corpora without relying on cross-lingual knowledge.In this paper,the authors propose a novel phrase embedding training method based on the skip-gram structure.Specifically,a local hard negative sampling strategy that utilises negative samples of central tokens in sliding windows to enhance phrase embedding learning is introduced.The proposed method achieves competitive or superior performance compared to baseline approaches,with exceptional results recorded for distant languages.Additionally,we develop a phrase representation learning method that leverages multilingual pre-trained language models.These mPLMs-based representations can be combined with the above-mentioned static phrase embeddings to further improve the accuracy of the bilingual phrase induction task.We manually construct a dataset of bilingual phrase pairs and integrate it with MUSE to facilitate the bilingual phrase induction task.
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financial support of the Fundamental Research Funds for the Central Universities(SCU2023HGXY)Special Research Funds for Intelligent Battery Cell Multidimensional Signal Sensing Technology Project from Huawei Technologies Co.Ltd.(24H1117)。
文摘Accurate estimation on the state of health(SOH)is essential for ensuring the safe and reliable operation of batteries.Traditional assessment methods primarily focus on electrical attributes for capacity decay,often overlooking the impact of thermal distribution on battery aging.However,thermal effect is a critical factor for degradation process and associated risks throughout their service life.In this paper,we introduce a novel deep learning framework specially designed to estimate the capacity and thermal risks of lithium-ion batteries(LIBs).This model consists of two main components that leverage computer vision technology.One predicts battery capacity by integrating the advantages of thermal and electrical features using a temporal pattern attention(TPA)mechanism,while the other assesses thermal risk by incorporating temperature variation to provide early warnings of potential hazards.An infrared camera is deployed to record temperature evolution of LIBs during the electrochemical process.The thermal heterogeneities are recorded by infrared camera,and the corresponding temperature evolutions are extracted as representative features for analysis.The proposed model demonstrates high accuracy and stability,with an average root mean square error(RMSE)of 0.67% for capacity estimation and accuracy exceeding 93.9% for risk prediction,underscoring the importance of integrating spatial temperature distribution into battery health assessments.This work offers valuable insights for the development of intelligent and robust battery management systems.
基金sponsored by National Natural Science Foundation of China(No.61571049).
文摘Traceability and trustiness are two critical issues in the logistics sector.Blockchain provides a potential way for logistics tracking systems due to its traits of tamper resistance.However,it is non-trivial to apply blockchain on logistics because of firstly,the binding relationship between virtue data and physical location cannot be guaranteed so that frauds may exist.Secondly,it is neither practical to upload complete data on the blockchain due to the limited storage resources nor convincing to trust the digest of the data.This paper proposes a traceable and trustable consortium blockchain for logistics T^(2)L to provide an efficient solution to the mentioned problems.Specifically,the authenticated geocoding data from telecom operators’base stations are adopted to ensure the location credibility of the data before being uploaded to the blockchain for the purpose of reliable traceability of the logistics.Moreover,we propose a scheme based on Zero Knowledge Proof of Retrievability(ZK BLS-PoR)to ensure the trustiness of the data digest and the proofs to the blockchain.Any user in the system can check the data completeness by verifying the proofs instead of downloading and examining the whole data based on the proposed ZK BLS-PoR scheme,which can provide solid theoretical verification.In all,the proposed T^(2)L framework is a traceable and trustable logistics system with a high level of security.
基金supported in part by the National Natural Science Foundation of China(NSFC)(Grants 62302128 and 624B2049)supported by Shenzhen Science and Technology Innovation Committee(Grant RCBS20231211090749086).
文摘Robotics plays an increasingly important role in all areas of human activity.Teleoperation robots can effectively ensure the safety of operators when operating in difficult and high‐risk industrial scenarios,which obviously requires instant and efficient signal compression and transmission in the system.However,most of the existing algorithms cannot fully explore the correlation within the signal,which mostly limits the compression efficiency.In this paper,a novel prediction‐aided kinaestheticsignal compression framework is proposed,which uses semantic communication methods to explore the temporal and spatial correlations of signals and employs neural network predictions to uncover their internal correlations.Specifically,the signal is first divided into two groups:the base part and the predictable part,and then a series of transformation matrices are introduced to establish the correlation between the two groups of the signal,which can be automatically optimised by a well‐designed neural network.This strategy of using learnable transformation matrices for prediction can not only accurately construct the correlation within the signal through massive data mining but also efficiently execute inference in a simple matrix multiplication computing form.Experimental results demonstrate that the proposed method outperforms the existing traditional tactile codecs and the latest tactile semantic communication methods.
基金supported by the National Key Research and Development Program of China under Grant 2018YFA0701601by the Program of Jiangsu Province under Grant NTACT-2024-Z-001.
文摘In offshore maritime communication sys-tems,base stations(BSs)are employed along the coastline to provide high-speed data service for ves-sels in coastal sea areas.To ensure the line-of-sight propagation of BS-vessel links,high transceiver an-tenna height is required,which limits the number of geographically available sites for BS deployment,and imposes a high cost for realizing effective wide-area coverage.In this paper,the joint user association and power allocation(JUAPA)problem is investigated to enhance the coverage of offshore maritime systems.By exploiting the characteristics of network topology as well as vessels’motion in offshore communica-tions,a multi-period JUAPA problem is formulated to maximize the number of ships that can be simultane-ously served by the network.This JUAPA problem is intrinsically non-convex and subject to mixed-integer constraints,which is difficult to solve either analyt-ically or numerically.Hence,we propose an iterative augmentation based framework to efficiently select the active vessels,where the JUAPA scheme is iteratively optimized by the network for increasing the number of the selected vessels.More specifically,in each itera-tion,the user association variables and power alloca-tion variables are determined by solving two separate subproblems,so that the JUAPA strategy can be up-dated in a low-complexity manner.The performance of the proposed JUAPA method is evaluated by exten-sive simulation,and numerical results indicate that it can effectively increase the number of vessels served by the network,and thus enhances the coverage of off-shore systems.
基金supported by the National Nature Science Foundation of China(NSFC)under grant No.61771194supported by Key Program of Beijing Municipal Natural Science Foundation with No.17L20052
文摘In this work,a frame work for time-varying channel modeling and simulation is proposed by using neural network(NN)to overcome the shortcomings in geometry based stochastic model(GBSM)and simulation approach.Two NN models are developed for modeling of path loss together with shadow fading(SF)and joint small scale channel parameters.The NN models can predict path loss plus SF and small scale channel parameters accurately compared with measurement at 26 GHz performed in an outdoor microcell.The time-varying path loss and small scale channel parameters generated by the NN models are proposed to replace the empirical path loss and channel parameter random numbers in GBSM-based framework to playback the measured channel and match with its environment.Moreover,the sparse feature of clusters,delay and angular spread,channel capacity are investigated by a virtual array measurement at 28 GHz in a large waiting hall.
基金supported by National Science and Technology Major Program of the Ministry of Science and Technology(No.2018ZX03001031)Key program of Beijing Municipal Natural Science Foundation(No.L172030)+1 种基金Beijing unicipal Science and Technology Commission Project(No.Z181100003218007)National Key Technology Research and Development Program of the Ministry of Science and Technology of China(NO.2012BAF14B01)
文摘Internet of Things(IoT) is one of the targeted application scenarios of fifth generation(5 G) wireless communication.IoT brings a large amount of data transported on the network.Considering those data,machine learning(ML) algorithms can be naturally utilized to make network efficiently and reliably.However,how to fully apply ML to IoT driven wireless network is still open.The fundamental reason is that wireless communication pursuits the high capacity and quality facing the challenges from the varying and fading wireless channel.So in this paper,we explore feasible combination for ML and IoT driven wireless network from wireless channel perspective.Firstly,a three-level structure of wireless channel fading features is defined in order to classify the versatile propagation environments.This three-layer structure includes scenario,meter and wavelength levels.Based on this structure,there are different tasks like service prediction and pushing,self-organization networking,self adapting largescale fading modeling and so on,which can be abstracted into problems like regression,classification,clustering,etc.Then,we introduce corresponding ML methods to different levelsfrom channel perspective,which makes their interdisciplinary research promisingly.
基金National Natural Science Foundation of China(Nos.51707144,51877165 , 51577144)Shaanxi Province Key R&D Program under 2019ZDLGY18-05This manuscript is recommended by international symposium on insulation and discharge computation for power equipment IDCOMFU2019.
文摘Recently DC relay has been concerned as a key component in DC power distribution,management and control systems like aircraft,new energy vehicle,IT and communication industries.Ordinarily,magnetic force and contact moving speed have great influence on arc behaviours in the breaking process.This paper focuses on the numerical investigation of arc during the contact opening process in a real 400V/20 A DC relay product coupling with an inductive load circuit.A 3D air arc model based on the magneto-hydrodynamic theory was built and calculated.A method coupling different computational software was used to take the nonlinear permanent magnet and contact opening process into consideration simultaneously.Arc behaviours under different magnetic field and contact opening speed were presented and discussed carefully.It has been found that the increase of the magnetic field is beneficial to the quick build-up of arc length and voltage.Arc breaking duration becomes shorter with the increase in contact opening speed from 63.5 rad s^-1 to 94.5 rad s^-1,such reduction is less significant with an increase of opening speed from 94.5 rad s^-l to 118.5 rad s^-1.
基金supported by NSFC(22175005)Guangdong Basic and Applied Basic Research Foundation(2020B1515120039)+1 种基金Shenzhen Fundamental Research Program(JCYJ20200109110628172,GXWD20201231165807007-20200802205241003)Guangdong Technology Center for Oxide Semiconductor Devices and ICs。
文摘Atomic layer deposition(ALD)has become an indispensable thin-film technology in the contemporary microelectronics industry.The unique self-limited layer-by-layer growth feature of ALD has outstood this technology to deposit highly uniform conformal pinhole-free thin films with angstrom-level thickness control,particularly on 3D topologies.Over the years,the ALD technology has enabled not only the successful downscaling of the microelectronic devices but also numerous novel 3D device structures.As ALD is essentially a variant of chemical vapor deposition,a comprehensive understanding of the involved chemistry is of crucial importance to further develop and utilize this technology.To this end,we,in this review,focus on the surface chemistry and precursor chemistry aspects of ALD.We first review the surface chemistry of the gas–solid ALD reactions and elaborately discuss the associated mechanisms for the film growth;then,we review the ALD precursor chemistry by comparatively discussing the precursors that have been commonly used in the ALD processes;and finally,we selectively present a few newly-emerged applications of ALD in microelectronics,followed by our perspective on the future of the ALD technology.
