检索结果-内蒙古大学图书馆

A Scalable BEV Perception Processor for Image/Point Cloud Fusion applications Using CAM-Based Universal Mapping Unit

引用

IEEE JOURNAL OF solid-state CIRCUITS 2025年第3期60卷 1002-1013页

作者： Feng, Xiaoyu Lin, Xinyuan Yang, Huazhong Liu, Yongpan Sun, Wenyu Tsinghua Univ Dept Elect Engn Beijing 100084 Peoples R China

The integration of multi-sensor data like image and point cloud for information complementarity is crucial for 3-D perception scenarios like autonomous driving. Recently, bird's eye view (BEV)-based sensor fusion is attracting more and more attention but the significant computational overhead constrains their widespread application at the edge. First, there are numerous irregular memory access operations in BEV fusion networks. For example, sparse convolutions (SCONVs) in the point cloud branch and irregular BEV plane mapping result in significant memory addressing and mapping overhead. Furthermore, multi-sensor fusion leads to rapid expansion of model size, making it difficult and expensive for single-chip solutions to meet the demands. Based on the above challenges, this work proposes an image and point cloud fusion processor with two highlights: a content addressable memory (CAM)-based deep fusion core to accelerate a variety of irregular BEV operations and chip-level parallelism design supporting flexible interconnect topology. The proposed chip is fabricated in 28-nm CMOS technology. Compared with existing image or point cloud accelerators, the proposed chip achieves higher frequency, 2x higher area efficiency, and 2.61x higher energy efficiency for sparse point cloud processing. To the best of authors' knowledge, this work is the first accelerator for BEV-based multi-modal fusion networks.

关键词： Point cloud compression arrays Three-dimensional displays Semantics Parallel processing Topology Feature extraction Cloud computing Autonomous vehicles Network topology Bird's eye view (BEV) chip-level parallelism content addressable memory (CAM) diverse memory access multi-modal fusion

来源：评论

学校读者我要写书评

暂无评论

solid-state dewetting sputtered ultra-thin Au films for LSPR sensing chip toward African swine fever virus detection

引用

sensorS AND ACTUATORS A-PHYSICAL 2025年 382卷

作者： Lamtha, Thomanai Waiwijit, Uraiwan Duangkanya, Kamonchanok Lertvanithphol, Tossaporn Amarit, Ratthasart Tantiwanichapan, Khwanchai Sathukarn, Asmar Chananonnawathorn, Chanunthorn Hincheeranan, Wantanee Dhanasiwawong, Kittidhaj Choowongkomon, Kiattawee Sritrakul, Tepyuda Boodde, Orawan Sukmak, Manakorn Horprathum, Mati Natl Elect & Comp Technol Ctr Spectroscop & Sensing Devices Res Grp Pathum Thani 12120 Thailand Kasetsart Univ Fac Sci Dept Biochem Bangkok 10900 Thailand Kasetsart Univ Fac Vet Med Dept Vet Publ Hlth Nakhon Pathom 73140 Thailand Kasetsart Univ Fac Vet Med Dept Farm Resources & Prod Med Nakhon Pathom 73140 Thailand

The African swine fever virus (ASFV) has caused enormous economic losses in the swine industry, with no effective vaccine or specific drug treatment currently available. The development of highly specific and sensitive detection for ASFV is crucial to better controlling the spread and management of this disease. In this context, localized surface plasmon resonance (LSPR) based gold nanoparticle (Au NPs) arrays chip has proven great potential for detecting small molecules in label-free biosensor applications. Here, we developed a facile method for the fabrication of an Au NPs-LSPR sensor chip by depositing ultra-thin Au films onto glass slide substrates through magnetron sputtering at various deposition times, ranging from 9 to 18 sec. The subsequent formation of Au NPs was achieved by post-annealing treatment at 500 degrees C for 2 h. The formation of the Au NPs, including diameter, density, and inter-particle spacing, was found to be dependent on the deposition time, allowing tunability of the LSPR peak from 527 to 563 nm. Additionally, the prepared Au NPs exhibited refractive index (RI) sensitivity ranging from 36.118 to 92.165 nm/RIU. To evaluate the LSPR biosensor performance, the optimized Au NPs-LSPR sensor chips were employed in the detection of ASFV antibodies in swine blood serum. The LSPR response was investigated across a dilution range of 1:640-1:10240, achieving a limit of detection (LOD) with high specificity at the 1:5120 dilution level, which demonstrated twice the sensitivity compared to the enzyme-linked immunosorbent assay (ELISA) method. Furthermore, these Au NPs-LSPR sensor chips show high regenerability, long-term stability, and good reproducibility, which are potentially valuable for on-site detection testing to prevent the spread of ASFV. Additionally, our proposed Au NPs-LSPR sensor platform holds great potential as a point-of-field (POF) device for detecting chemicals and biomolecules in diverse applications.

关键词： LSPR AuNPs chip Thin film Sputtering Dewetting ASFV

来源：评论

学校读者我要写书评

暂无评论

A 3.5×3.5mm21.47mW/ch 16-Channel MSS-CMOS Heterogeneous Multi-Modal-Gas-sensor Chip Stack 72

A 3.5×3.5mm21.47mW/ch 16-Channel MSS-CMOS Heterogeneous Mul...

引用

72nd IEEE International solid-state Circuits conference, ISSCC 2025

作者： Naruse, Kotaro Kato, Naru Matsumori, Takuma Shlomi, Jun Midoh, Yoshihiro Hirose, Tetsuya Imamura, Gaku Yoshikawa, Genki Sideris, Constantine Miura, Noriyuki Osaka University Suita Japan National Institute for Materials Science Tsukuba Japan University of Tsukuba Tsukuba Japan University of Southern California Los AngelsCA United States

ISBN: (纸本)9798331541019

Miniaturizing and reducing the power consumption of multi-modal gas sensors with multiple channels can enable a wide array of applications, including integrating with wearable devices for daily breath analysis, encapsulation into ingestible pills for internal body measurements, and embedding into smart tags for food freshness management (Fig. 20.1.1). Traditional gas sensors typically have the transducer and CMOS chip connected via wiring on a substrate, leading to a large-form factor that requires complicated and expensive packaging and cannot readily scale to many-sensor arrays [1]. Recent studies have proposed gas sensors utilizing a MEMS-CMOS stack;however, these have yet to demonstrate responses to different gas types, and there are also challenges regarding increasing the number of channels and reducing the power consumption [2]. In this paper, we propose a heterogeneous multi-modal gas sensor that employs a Membrane-type Surface-stress sensor (MSS) as the transducer stacked on top of CMOS for analog frontend and digital readout. © 2025 IEEE.

关键词： CMOS integrated circuits

来源：评论

学校读者我要写书评

暂无评论

Optoelectronic in-sensor computing based on PD-RRAM array

Optoelectronic in-sensor computing based on PD-RRAM array

引用

Smart Photonic and Optoelectronic Integrated Circuits 2025

作者： Pan, Wen Hao, Zhibiao Sun, Changzheng Xiong, Bing Wang, Jian Han, Yanjun Li, Hongtao Gan, Lin Luo, Yi Wang, Lai Beijing100084 China

ISBN: (纸本)9781510684881

The rapid expansion of the Internet of Things (IoT) necessitates the development of low-power edge devices capable of efficient processing volumes of data. In-sensor computing provides a transformative solution by enabling in-situ data processing within sensor arrays. This study introduces a highly integrated PD-RRAM array for real-time in-sensor computing, which integrates photodiodes (PDs) with resistive random access memories (RRAMs). By consolidating sensing, storage, and computing, the PD-RRAM array achieve excellent efficiency and low power consumption, positioning it as a potential technology for future IoT applications. © 2025 SPIE.

关键词： RRAM

来源：评论

学校读者我要写书评

暂无评论

A Sub-1V 14b 5.8nW/Hz BW/Power-Scalable CT sensor Interface with a Frequency-Controlled Current Source Achieving a 225× Scalable Range 72

A Sub-1V 14b 5.8nW/Hz BW/Power-Scalable CT Sensor Interface ...

引用

72nd IEEE International solid-state Circuits conference, ISSCC 2025

作者： Wu, Xinjie Liu, Yuyan Yu, Xiaopeng Tan, Nick Nianxiong Tang, Zhong Zhejiang University Hangzhou China Vango Technologies Hangzhou China

ISBN: (纸本)9798331541019

Precision and energy-efficient sensor interfaces have always been needed in loT applications. To digitize weak signals (tens of mVs), such as shunt-based current and biomedical sensors, the interface should have low noise [1]-[4]. Moreover, to accommodate different use cases, the bandwidth (BW) and power dissipation of sensor interfaces should be highly scalable [5]-[7]. sensor interfaces based on discrete-time (DT) ADCs, such as SAR ADCs [8], [9] and FIA-based ΔΣ ADCs [5]-[7] are BW/power scalable, but they usually suffer from kT/C noise (Fig. 27.3.1 top, left). Using a continuous-time (CT) first stage, such as an instrumentation amplifier (IA) [1], [3], [4] (Fig. 27.3.1 top, right), or an active RC integrator [10], [11], can suppress the sampling noise of the following stages, however, they usually consume fixed static power, limiting their BW/power scalability. © 2025 IEEE.

关键词： Continuous time systems

来源：评论

学校读者我要写书评

暂无评论

High-Capacity and High-Density FBG Array Sensing Demodulation System Based on High-Speed Wavelength-Swept REC-DFB Laser Array 4

High-Capacity and High-Density FBG Array Sensing Demodulatio...

引用

4th International Computational Imaging conference, CITA 2024

作者： Yang, Shijie Dai, Pan Ban, Qilu Wang, Yu Wang, Jiacheng Xu, Kaichuan Fan, Yaqiang Dong, Junwei He, Zilong Zhang, Yixin Wang, Feng Chen, Xiangfei Key Laboratory of Intelligent Optical Sensing and Manipulation The Ministry of Education & Engineering Research Center of Precision Photonics Integration and System Application The Ministry of Education National Laboratory of Solid State Microstructures College of Engineering and Applied Sciences Institute of Optical Communication Engineering Nanjing University Tongding Joint Lab for large-scale photonic integrated circuits Nanjing University Jiangsu Nanjing210093 China

ISBN: (纸本)9781510688834

Fiber Bragg grating (FBG) sensing technology, with the advantages of precise real-time monitoring and strong anti-interference capabilities, has been widely applied in various fields, including aerospace, petrochemical, and power transmission. This study presents an FBG sensing demodulation system applicable to high-capacity, high-density fiber sensing through quasi-distributed networking using multipoint FBG sensor arrays, capable of dense monitoring through multi-channel parallel operation and wavelength division multiplexing (WDM). The FBG sensing demodulation system primarily consists of a wavelength-swept semiconductor laser source and its parallel driving control module, a multiplexed FBG sensor array section, and a signal acquisition and demodulation module. The laser source is a distributed feedback (DFB) laser array, integrating 20 individual laser units at high density, based on the reconstruction-equivalent-chirp (REC) technique. A continuous coverage of wavelength tuning range, exceeding 40 nm in the C-band range, is achieved, with a maximum sweeping speed of 2.5 kHz and a tuning speed of 100,000 nm/s. The FBG sensor array section can deploy multiple non-identical WDM fiber grating arrays. The system can operate at high sweeping speeds of 1 kHz and 2.5 kHz, yielding great sensing results in temperature experiments, strain experiment and vibration experiment. The proposed FBG sensing system meets the application demands for high-capacity and dense quasi-distributed fiber sensing, exhibiting significant advantages in sensor load capacity, measurement precision, and especially measurement speed, with promising future applications. © 2025 SPIE.

关键词： Fiber Bragg gratings

来源：评论

学校读者我要写书评

暂无评论

A 94.8nW Battery-Free Intelligent Silicon Platform Enabling Distributed, Adaptive, and Event-Driven Multimodal Sensing at the Edge 72

A 94.8nW Battery-Free Intelligent Silicon Platform Enabling ...

引用

72nd IEEE International solid-state Circuits conference, ISSCC 2025

作者： Zhang, Haochen Yu, Wei-Han Zhao, Zhongyu Yang, Zhizhan Un, Ka-Fai Yin, Jun Martins, Rui P. Mak, Pui-In University of Macau China Instituto Superior Tecnico University of Lisboa Lisbon Portugal

ISBN: (纸本)9798331541019

sensor nodes with machine learning (ML) are adept at analyzing intricate environmental and physiological data patterns at the edge [1]-[9]. The design of such ultra-low-power (ULP) devices strives to reduce power consumption, which ensures continuous and energy-harvested operation even with fluctuating ambient available energy levels down to 100nW [10]-[11]. Consequently, ML capabilities on such ULP devices are constrained to perform lightweight detection for events such as voice activity [3] [6], arrhythmia [4] [8], and bearing anomalies [9]. Yet, these isolated, monomodal sensing paradigms suffer from low task complexity and accuracy for overlooking fused information from spatially distributed sensor networks. Additionally, real-world applications subject to data distribution drifts necessitate model adaptability to maintain accuracy in ever-changing environments [12]. Also, current ULP neural network (NN) accelerators consume significant power, limiting their ability to expand the network size, while impeding the overall ML performance. © 2025 IEEE.

关键词： sensor nodes

来源：评论

学校读者我要写书评

暂无评论

A Large-Scale Multi-sensor Dataset for Mapping 5

A Large-Scale Multi-Sensor Dataset for Mapping

引用

5th IEEE International conference on Power, Electronics and Computer applications, ICPECA 2025

作者： Zou, Yujie Zhang, Hanwen Yan, Zhewen Cheng, Hui Sun Yat-sen University Guangzhou China

ISBN: (纸本)9798331533694

In recent years, advancements and applications of computer vision techniques such as neural radiance fields (NeRF) and Gaussian splatting have set new expectations of mapping and 3D reconstruction tasks. Achieving high-accuracy outdoor mapping using radiance field methods requires the integration of robust localization method and carefully designed multi-sensor perception systems. In this paper, we introduce a multimodal dataset tailored for high-precision mapping of large-scale outdoor scenes. To achieve this, we developed an aerial data acquisition platform equipped with a GPS-synchronized solid-state LiDAR, industrial cameras, and inertial measurement units. After comprehensive intrinsic and extrinsic calibration, we collected 3D reconstruction datasets focusing on building exteriors across scenes within a university campus. The raw data, along with documentation on synchronization, calibration procedures, and platform design, will be progressively released on the project's GitHub repository: https://***/lilrookie76/radiance-dataset. © 2025 IEEE.

关键词： Mapping

来源：评论

学校读者我要写书评

暂无评论

MEMS-Free 4096-Pixel CMOS E-Nose Gas-sensor Array with Molecular-Selective Metal-Organic-Framework Sensing and In-Pixel Thermodynamic Modulation for Fast sensor Regeneration 72

MEMS-Free 4096-Pixel CMOS E-Nose Gas-Sensor Array with Molec...

引用

72nd IEEE International solid-state Circuits conference, ISSCC 2025

作者： Saif, Marco Jiang, Fuze Lo, Shao-Wei Wang, Adam Huang, Zhikai Wang, Jiachen Liu, Hangxing Shih, Chih-Jen Burger, Thomas Wang, Hua ETH Zürich Zürich Switzerland

ISBN: (纸本)9798331541019

Accurate and real-time sensing of gas molecule species and concentrations is critical for a myriad of applications, including environmental protection, industry automation, public health monitoring, and bio-/chemical-security surveillance. For example, gas sensing in large chemical plants is crucial for safety and production efficiency, with ammonia (NH3) being particularly important due to its extensive use [1]. Electronic noses (e-noses) consist of a chemical sensor array with surface recognition modules for capturing specific or multi-species gas molecules followed by electronic signal acquisition and processing [2]. Despite decades of exploration [3]-[5], the development of e-noses is still in its early stages, and existing e-noses continue to face major challenges as: (1) limited selectivity on target molecules (2) poor sensor regeneration due to slow and incomplete gas desorption process, which can result in sensing memory effect, i.e., affecting low-trace gas detection due to prior exposure to high concentrations, (3) a small number of sensing nodes instead of a large array, also lacking in-pixel signal processing and data digitization [6], [7]. To address these issues, we present a monolithically integrated molecular specific metal-organic-frameworks (MOFs) CMOS e-nose sensor array with in-pixel fast capacitance-gas interfacing, readout, and data digitization with local temperature regulation and sensing. For demonstration, NiNi-Pyz MOFs have been employed to achieve NH3 specificity. In particular, the in-pixel temperature regulation can modulate the thermodynamic of NH3 adsorption and desorption at the sensor interfaces, significantly improving the sensitivity and allowing rapid sensor regeneration. © 2025 IEEE.

关键词： CMOS integrated circuits

来源：评论

学校读者我要写书评

暂无评论

A Novel Multiple Modes Resonant sensor Featuring Blue Sideband Excitation 38

A Novel Multiple Modes Resonant Sensor Featuring Blue Sideba...

引用

38th IEEE International conference on Micro Electro Mechanical Systems, MEMS 2025

作者： Xu, Jiao Zheng, Zhuoyue Xi, Jingqian Zhang, Ziqian Liu, Huafeng Zhang, Pan Chen, Jianlin Wang, Chen Kraft, Michael Wang, Yuan Martins, R.P. Mak, Pui-In Huazhong University of Science and Technology Moe Key Laboratory of Fundamental Physical Quantities Measurement Pgmf School of Physics Wuhan430074 China University of Macau Institute of Microelectronics State Key Laboratory of Analog and Mixed-Signal Vlsi 999078 China Peking University National Key Laboratory of Advanced Micro and Nano Manufacture Technology School of Integrated Circuits 100871 China Shanghai University School of Microelectronics 200444 China Ku Leuven Department of Electrical Engineering-MNS Leuven3000 Belgium

ISBN: (纸本)9798331508890

This paper reports a novel MEMS resonant structure functioning as a baseline resonant electrostatic force/temperature sensor, highlighting its substantial potential in sensing applications. The resonator, characterized by a dual-cosine structure, is capable of simultaneous multiple resonant modes actuation within the 300 kHz range, through a sole-source drive approach using the recent emerging Blue Sideband Excitation (BSE) technique. This method enables a distinctive multi-modal response, essential for multi-parameter measurements. The research focuses on the sensor's temperature characteristics, electrostatic perturbation sensitivity, and noise floor. Notably, the study reveals intriguing temperature responses across different modes, where certain modes exhibit a significant frequency rise with increasing temperature, this is counter-intuitive with common MEMS resonator devices, prompting further investigation into the underlying mechanisms. The analysis of electrostatic perturbation sensitivity and noise background further unfolds the resonator's performance under various excitation conditions, laying a foundation for advancing sensor functionality, and providing solid evidence on the viability of the subject. To sum up, this study highlights the resonator's potential as a high-precision sensor, opening new opportunities for MEMS sensor design and multi-parameter measurement applications. © 2025 IEEE.

关键词： Microchannels

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：