Recent years have witnessed transformative changes brought about by artificial intelligence(AI)techniques with billions of parameters for the realization of high accuracy,proposing high demand for the advanced and AI ...Recent years have witnessed transformative changes brought about by artificial intelligence(AI)techniques with billions of parameters for the realization of high accuracy,proposing high demand for the advanced and AI chip to solve these AI tasks efficiently and powerfully.Rapid progress has been made in the field of advanced chips recently,such as the development of photonic computing,the advancement of the quantum processors,the boost of the biomimetic chips,and so on.Designs tactics of the advanced chips can be conducted with elaborated consideration of materials,algorithms,models,architectures,and so on.Though a few reviews present the development of the chips from their unique aspects,reviews in the view of the latest design for advanced and AI chips are few.Here,the newest development is systematically reviewed in the field of advanced chips.First,background and mechanisms are summarized,and subsequently most important considerations for co-design of the software and hardware are illustrated.Next,strategies are summed up to obtain advanced and AI chips with high excellent performance by taking the important information processing steps into consideration,after which the design thought for the advanced chips in the future is proposed.Finally,some perspectives are put forward.展开更多
We are sorry for the mistakes of Affiliation,"a State Key Laboratory of Advanced Fiber Materials,Center for Advanced Low-Dimension Materials,Donghua University,Shanghai 201620,China"should be replaced by&quo...We are sorry for the mistakes of Affiliation,"a State Key Laboratory of Advanced Fiber Materials,Center for Advanced Low-Dimension Materials,Donghua University,Shanghai 201620,China"should be replaced by"a State Key Laboratory of Advanced Fiber Materials,Center for Advanced Low-Dimension Materials,College of Materials Science and Engineering,Donghua University,Shanghai 201620,China".We apologized for the inconvenience caused by this error.展开更多
在托卡马克等离子体的辅助加热系统中,随着高压电源设备的不断增加,对高压电源控制系统的实时传输和数据可靠性有了更严苛的要求。为保障HL-3装置高压电源控制系统的稳定性及数据准确性,研究选用了Chisel语言,设计并开发了一款高压电源...在托卡马克等离子体的辅助加热系统中,随着高压电源设备的不断增加,对高压电源控制系统的实时传输和数据可靠性有了更严苛的要求。为保障HL-3装置高压电源控制系统的稳定性及数据准确性,研究选用了Chisel语言,设计并开发了一款高压电源控制系统的RISC-V微控制器。设计的实现采用了诸如数据前推和分支预测等流水线相关技术以创建一个单发射五级流水线微控制器。这款微控制器已成功通过预定的测试集,并能在高云Tang Primer 25K开发板上顺利运行RT-Thread嵌入式实时操作系统的运行测试。在功耗受限的条件下,此RISC-V微控制器能够高效地执行数据处理,满足可控核聚变领域对嵌入式微控制器开源可定制化设计及可靠性的需求。展开更多
We present a compact cold atom platform where an optical grating chip and planar coil chip are placed inside a compact vacuum chamber to create a magneto-optical trap.This approach significantly reduces the system vol...We present a compact cold atom platform where an optical grating chip and planar coil chip are placed inside a compact vacuum chamber to create a magneto-optical trap.This approach significantly reduces the system volume to about 20×20×20 cm^(3) compared to conventional vacuum systems and offers greater flexibility in accessing the trapped atoms.We demonstrate the trapping of 3×10^(5) cold rubidium atoms at a temperature of 100μK in a vacuum pressure below 10^(−7) mbar.The simplified optical geometry,low power consumption,and high degree of integration make this a promising platform for portable and versatile cold-atom devices in quantum sensing,timing,and information processing.展开更多
Vibration detection using sensors with both wide working frequency range,good sensitivity,and other good performances is a topic of great interest in fields such as inertial navigation,deep-sea fishing boat engines co...Vibration detection using sensors with both wide working frequency range,good sensitivity,and other good performances is a topic of great interest in fields such as inertial navigation,deep-sea fishing boat engines condition monitoring,seismic monitoring,attitude,and heading reference system,etc.This paper investigates two 6H-SIC MEMS diaphragms,one triangular and the other square,used in a fiber optic Fabry–Perot(FP)accelerometer in an experimental scenario.The triangular chip shows a wide working frequency range of 630 Hz–5300 Hz,a natural frequency of 44.3 k Hz,and a mechanical sensitivity of 0.154 nm/g.An optimal structure of the square chip used in a probe such as a fiber optic FP accelerometer also shows a wide working frequency range of 120 Hz–2300 Hz;a good sensitivity of 31.5 m V/g,a resonance frequency of7873 Hz,an accuracy of 0.96%F.S.,a frequency measurement error of 1.15%,and an excellent linearity of 0.9995.展开更多
Globally,approximately 10 million new tuberculosis(TB)cases are reported annually.Delayed diagnosis due to low detection rates is the primary cause of mortality.Although pathological examination is commonly used for d...Globally,approximately 10 million new tuberculosis(TB)cases are reported annually.Delayed diagnosis due to low detection rates is the primary cause of mortality.Although pathological examination is commonly used for diagnosing TB,5%-30%of cases remain undiagnosed,emphasizing the urgent need to establish quality control(QC)standards to reduce rates of misdiagnosis and missed diagnoses.To address this,we introduced a novel QC chip for detecting Mycobacterium tuberculosis(MTB).A quantitative pathological QC model was constructed by precisely and uniformly integrating MTB and HeLa cells into a photocurable hydrogel.This model was then sliced into uniform sections to create QC chips.It demonstrated that the QC chips exhibited no significant differences in intra-batch or inter-batch variation(coefficient of variation<5%),and remained stable at−80°C for one year.Furthermore,these chips were found to be 100%effective when tested with 240 clinical samples(200 with special staining and 40 with polymerase chain reaction).In addition to enhancing TB detection rates,this approach offers visualization,quantification,and sustainable production.Overall,this work provides a novel framework for developing QC chips for pathological testing,offering a reliable solution to enhance clinical diagnostic workflows.展开更多
The current global economic and trade landscape is undergoing profound changes.Since the outbreak of the China-US trade friction in 2018,the United States has systematically promoted the“de-Sinicizat ion”strategy in...The current global economic and trade landscape is undergoing profound changes.Since the outbreak of the China-US trade friction in 2018,the United States has systematically promoted the“de-Sinicizat ion”strategy in the supply chain through measures such as imposing tariffs,exercising technology blockades,and setting up industrial subsidy barriers.Chinese products,including automot ive par ts and elect ronic equipment,have been significantly impacted.展开更多
智能终端正加速向轻量化、低功耗、高安全、定制化方向演进,传统芯片架构存在的授权约束、功耗瓶颈及灵活性不足等问题,已难以匹配多场景创新需求。第五代精简指令集(Reduced Instruction Set Computing-V,RISC-V)架构凭借开源免费、模...智能终端正加速向轻量化、低功耗、高安全、定制化方向演进,传统芯片架构存在的授权约束、功耗瓶颈及灵活性不足等问题,已难以匹配多场景创新需求。第五代精简指令集(Reduced Instruction Set Computing-V,RISC-V)架构凭借开源免费、模块化定制、低功耗、高扩展性等核心优势,打破了传统架构垄断格局,成为智能终端产业转型升级的关键支撑。从技术突破、生态构建、产业协同3个维度发力,提出优化路径,旨在提升RISC-V与智能终端的适配协同能力,助力产业高质量发展。展开更多
Network-on-Chip(NoC)systems are progressively deployed in connecting massively parallel megacore systems in the new computing architecture.As a result,application mapping has become an important aspect of performance ...Network-on-Chip(NoC)systems are progressively deployed in connecting massively parallel megacore systems in the new computing architecture.As a result,application mapping has become an important aspect of performance and scalability,as current trends require the distribution of computation across network nodes/points.In this paper,we survey a large number of mapping and scheduling techniques designed for NoC architectures.This time,we concentrated on 3D systems.We take a systematic literature review approach to analyze existing methods across static,dynamic,hybrid,and machine-learning-based approaches,alongside preliminary AI-based dynamic models in recent works.We classify them into several main aspects covering power-aware mapping,fault tolerance,load-balancing,and adaptive for dynamic workloads.Also,we assess the efficacy of each method against performance parameters,such as latency,throughput,response time,and error rate.Key challenges,including energy efficiency,real-time adaptability,and reinforcement learning integration,are highlighted as well.To the best of our knowledge,this is one of the recent reviews that identifies both traditional and AI-based algorithms for mapping over a modern NoC,and opens research challenges.Finally,we provide directions for future work toward improved adaptability and scalability via lightweight learned models and hierarchical mapping frameworks.展开更多
基金supported by the Hong Kong Polytechnic University(1-WZ1Y,1-W34U,4-YWER).
文摘Recent years have witnessed transformative changes brought about by artificial intelligence(AI)techniques with billions of parameters for the realization of high accuracy,proposing high demand for the advanced and AI chip to solve these AI tasks efficiently and powerfully.Rapid progress has been made in the field of advanced chips recently,such as the development of photonic computing,the advancement of the quantum processors,the boost of the biomimetic chips,and so on.Designs tactics of the advanced chips can be conducted with elaborated consideration of materials,algorithms,models,architectures,and so on.Though a few reviews present the development of the chips from their unique aspects,reviews in the view of the latest design for advanced and AI chips are few.Here,the newest development is systematically reviewed in the field of advanced chips.First,background and mechanisms are summarized,and subsequently most important considerations for co-design of the software and hardware are illustrated.Next,strategies are summed up to obtain advanced and AI chips with high excellent performance by taking the important information processing steps into consideration,after which the design thought for the advanced chips in the future is proposed.Finally,some perspectives are put forward.
文摘We are sorry for the mistakes of Affiliation,"a State Key Laboratory of Advanced Fiber Materials,Center for Advanced Low-Dimension Materials,Donghua University,Shanghai 201620,China"should be replaced by"a State Key Laboratory of Advanced Fiber Materials,Center for Advanced Low-Dimension Materials,College of Materials Science and Engineering,Donghua University,Shanghai 201620,China".We apologized for the inconvenience caused by this error.
文摘在托卡马克等离子体的辅助加热系统中,随着高压电源设备的不断增加,对高压电源控制系统的实时传输和数据可靠性有了更严苛的要求。为保障HL-3装置高压电源控制系统的稳定性及数据准确性,研究选用了Chisel语言,设计并开发了一款高压电源控制系统的RISC-V微控制器。设计的实现采用了诸如数据前推和分支预测等流水线相关技术以创建一个单发射五级流水线微控制器。这款微控制器已成功通过预定的测试集,并能在高云Tang Primer 25K开发板上顺利运行RT-Thread嵌入式实时操作系统的运行测试。在功耗受限的条件下,此RISC-V微控制器能够高效地执行数据处理,满足可控核聚变领域对嵌入式微控制器开源可定制化设计及可靠性的需求。
基金supported by the National Key R&D Program(Grant Nos.2021YFA1402004 and 2021YFF0603701)the National Natural Science Foundation of China(Grant Nos.12134014,U21A20433,U21A6006,and 92265108)+1 种基金the Fundamental Research Funds for the Central Universitiesthe University of Science and Technology of China(USTC)Research Funds of the Double First-Class Initiative。
文摘We present a compact cold atom platform where an optical grating chip and planar coil chip are placed inside a compact vacuum chamber to create a magneto-optical trap.This approach significantly reduces the system volume to about 20×20×20 cm^(3) compared to conventional vacuum systems and offers greater flexibility in accessing the trapped atoms.We demonstrate the trapping of 3×10^(5) cold rubidium atoms at a temperature of 100μK in a vacuum pressure below 10^(−7) mbar.The simplified optical geometry,low power consumption,and high degree of integration make this a promising platform for portable and versatile cold-atom devices in quantum sensing,timing,and information processing.
基金Project supported by the National Natural Science Foundation of China(Grant No.32473216)Ningbo Youth Science and Technology Innovation Leading Talent Project(Grant No.2023QL004)。
文摘Vibration detection using sensors with both wide working frequency range,good sensitivity,and other good performances is a topic of great interest in fields such as inertial navigation,deep-sea fishing boat engines condition monitoring,seismic monitoring,attitude,and heading reference system,etc.This paper investigates two 6H-SIC MEMS diaphragms,one triangular and the other square,used in a fiber optic Fabry–Perot(FP)accelerometer in an experimental scenario.The triangular chip shows a wide working frequency range of 630 Hz–5300 Hz,a natural frequency of 44.3 k Hz,and a mechanical sensitivity of 0.154 nm/g.An optimal structure of the square chip used in a probe such as a fiber optic FP accelerometer also shows a wide working frequency range of 120 Hz–2300 Hz;a good sensitivity of 31.5 m V/g,a resonance frequency of7873 Hz,an accuracy of 0.96%F.S.,a frequency measurement error of 1.15%,and an excellent linearity of 0.9995.
基金supported by the National Natural Science Foundation of China(Nos.52325504 and 52235007)the Basic Public Welfare Research Project of Zhejiang Province(No.LGY23H160089)the Science and Technology Plan Project of Taizhou City(No.24ywa08).
文摘Globally,approximately 10 million new tuberculosis(TB)cases are reported annually.Delayed diagnosis due to low detection rates is the primary cause of mortality.Although pathological examination is commonly used for diagnosing TB,5%-30%of cases remain undiagnosed,emphasizing the urgent need to establish quality control(QC)standards to reduce rates of misdiagnosis and missed diagnoses.To address this,we introduced a novel QC chip for detecting Mycobacterium tuberculosis(MTB).A quantitative pathological QC model was constructed by precisely and uniformly integrating MTB and HeLa cells into a photocurable hydrogel.This model was then sliced into uniform sections to create QC chips.It demonstrated that the QC chips exhibited no significant differences in intra-batch or inter-batch variation(coefficient of variation<5%),and remained stable at−80°C for one year.Furthermore,these chips were found to be 100%effective when tested with 240 clinical samples(200 with special staining and 40 with polymerase chain reaction).In addition to enhancing TB detection rates,this approach offers visualization,quantification,and sustainable production.Overall,this work provides a novel framework for developing QC chips for pathological testing,offering a reliable solution to enhance clinical diagnostic workflows.
文摘The current global economic and trade landscape is undergoing profound changes.Since the outbreak of the China-US trade friction in 2018,the United States has systematically promoted the“de-Sinicizat ion”strategy in the supply chain through measures such as imposing tariffs,exercising technology blockades,and setting up industrial subsidy barriers.Chinese products,including automot ive par ts and elect ronic equipment,have been significantly impacted.
文摘智能终端正加速向轻量化、低功耗、高安全、定制化方向演进,传统芯片架构存在的授权约束、功耗瓶颈及灵活性不足等问题,已难以匹配多场景创新需求。第五代精简指令集(Reduced Instruction Set Computing-V,RISC-V)架构凭借开源免费、模块化定制、低功耗、高扩展性等核心优势,打破了传统架构垄断格局,成为智能终端产业转型升级的关键支撑。从技术突破、生态构建、产业协同3个维度发力,提出优化路径,旨在提升RISC-V与智能终端的适配协同能力,助力产业高质量发展。
基金the Deanship of Graduate Studies and Scientific Research at University of Bisha for supporting this work through the Fast-Track Research Support Programthe Deanship of Scientific Research at Northern Border University,Arar,KSA for funding this research work through the project number“NBU-FFR-2025-2903-09”.
文摘Network-on-Chip(NoC)systems are progressively deployed in connecting massively parallel megacore systems in the new computing architecture.As a result,application mapping has become an important aspect of performance and scalability,as current trends require the distribution of computation across network nodes/points.In this paper,we survey a large number of mapping and scheduling techniques designed for NoC architectures.This time,we concentrated on 3D systems.We take a systematic literature review approach to analyze existing methods across static,dynamic,hybrid,and machine-learning-based approaches,alongside preliminary AI-based dynamic models in recent works.We classify them into several main aspects covering power-aware mapping,fault tolerance,load-balancing,and adaptive for dynamic workloads.Also,we assess the efficacy of each method against performance parameters,such as latency,throughput,response time,and error rate.Key challenges,including energy efficiency,real-time adaptability,and reinforcement learning integration,are highlighted as well.To the best of our knowledge,this is one of the recent reviews that identifies both traditional and AI-based algorithms for mapping over a modern NoC,and opens research challenges.Finally,we provide directions for future work toward improved adaptability and scalability via lightweight learned models and hierarchical mapping frameworks.
