检索结果-内蒙古大学图书馆

IEEE 31st International Symposium on Industrial Electronics (ISIE)

作者： Baietto, Anthony Boubin, Jayson Farr, Patrick Bihl, Trevor J. Ohio State Univ Dept Comp Sci & Engn Columbus OH 43210 USA Appl Res Solut Beavercreek OH USA US Air Force Res Lab Wright Patterson AFB OH USA

ISBN: (纸本)9781665482400

Spectral notching, a waveform design method used in signal processing and radar, mitigates interference caused by an ever-growing number of technologies which saturate the radio frequency (RF) spectrum. Online waveform design is possible with current technology, but extant techniques can not meet real-time latency requirements on the low size weight and power (SWaP) embedded hardware ideal for contexts where online waveform design is most useful. Current techniques rely on convex optimization, leading to non-trivial execution times and excessive compute and power requirements. Artificial Neural Networks (ANNs) could feasibly be used for spectral notching due to their ability to approximate complex non-linear functions;however, ANNs generally require considerable processing power and result in long inference times. Prior work has shown that ANNs can accurately perform real-time radar waveform design, albeit in very specific contexts. In this paper, we extend that work, demonstrating that lean neural networks, which model successful convex optimization algorithms in a lower dimensional representation, can meet real-time latency constraints for general waveform design without sacrificing accuracy on low SWaP embedded hardware. Our results are exemplified in simulation on real embedded and general purpose hardware. Overall, our lean neural network solution decreases inference time on traditional embedded hardware by 8.5x when compared to the fastest optimization approach, and uses 10.6x less energy. Our system executes quickly on low-power embedded devices, using only 16.6% of the energy a conventional GPU-provisioned system would require.

关键词： signal processing algorithms Radar Interference Artificial neural networks Hardware Real-time systems Convex functions

来源：评论

学校读者我要写书评

暂无评论

Deep Deterministic Policy Gradient Artificial Intelligence for Radar Applications

Deep Deterministic Policy Gradient Artificial Intelligence f...

引用

IEEE Radar Conference (RadarConf)

作者： Reininger, Taylor J. Smith, Graeme E. Johns Hopkins Univ Appl Phys Lab Laurel MD 20723 USA

ISBN: (纸本)9781728153681

The work presented in this paper is intended to build a deeper understanding of reinforcement learning (RL), a branch of artificial intelligence (AI), as it pertains to radar applications. We demonstrate a method for using a deep deterministic policy gradient (DDPG) AI algorithm to solve an example problem of producing a spectral notch in a radar waveform to avoid interfering with other electronic systems operating in the same portion of the spectrum. The DDPG algorithm is able to produce phase-coded waveforms with 80 dB deep notches in their power spectra. As such, it can provide an alternative method to gradient-based optimization and hand-engineered solutions. The radar performance of the resulting notched waveform was not addressed in this effort. Instead, the intent of the paper is to demonstrate that AI can be used in a fundamental radar signal processing task.

关键词： Conferences signal processing algorithms Reinforcement learning Radar signal processing Radar applications Artificial intelligence Task analysis

来源：评论

学校读者我要写书评

暂无评论

An improved event-by-event clustering algorithm for noisy acquisition 8

An improved event-by-event clustering algorithm for noisy ac...

引用

8th International Conference on Event-Based Control, Communication and signal processing (EBCCSP)

作者： Lesage, Xavier Tran, Rosalie Mancini, Stephane Fesquet, Laurent Univ Grenoble Alpes CNRS Grenoble INP TIMA F-38000 Grenoble France

ISBN: (纸本)9781665453493

Event-based image sensors are a new class of sensors developed thanks to non-uniform sampling and asynchronous technology, which overcomes many image sensor limitations such as a high throughput or a huge power consumption. As their behavior and outputs are really different from traditional image sensors, the produced data stream imposes to completely rethink image processing. Indeed, dedicated algorithms are mandatory to take advantage of this specific data stream, known as Address Event Representation (AER). This paper presents an improved and dedicated event-by-event clustering algorithm allowing the object detection in a noisy environment which is still performant with a SNR of 1/4. We measure high recall and precision for different simulated scenarios with multiple objects and show an improvement compared to the previous algorithm. The approach especially demonstrates a low computational complexity and a reduced memory footprint, which is perfectly suited for low-cost and low-power embedded image sensing applications.

关键词： Image sensors signal processing algorithms Clustering algorithms Streaming media signal processing Throughput Robustness

来源：评论

学校读者我要写书评

暂无评论

RF-SITL: A Software-in-the-loop Channel Emulator for UAV Swarm Networks 23

RF-SITL: A Software-in-the-loop Channel Emulator for UAV Swa...

引用

23rd IEEE International Symposium on a World of Wireless, Mobile and Multimedia Networks (IEEE WoWMoM)

作者： Mastronarde, Nicholas Russell, Daniel Guan, Zhangyu Sklivanitis, George Pados, Dimitris Bentley, Elizabeth Serena Medley, Michael Univ Buffalo Dept Elect Engn Buffalo NY 14260 USA GE Aviat Grand Rapids MI 49512 USA Florida Atlantic Univ Dept Elect Engn & Comp Sci Boca Raton FL 33431 USA Air Force Res Lab AFRI I-13440 Rome NY Italy

ISBN: (数字)9781665408769

ISBN: (纸本)9781665408769

We introduce RF-SITL, a radio frequency (RF) software-in-the-loop (SITL) channel emulator developed with GNU Radio and the University at Buffalo's Airborne Networking and Communications (UB-ANC) emulator to enable integrated simulation of systems comprising multiple unmanned aerial vehicles (UAVs) interacting over a wireless communication channel. RF-SITL could be paired with any multi-robot simulator to enable I/Q sample-level fidelity simulation of communication interactions between the robots by accurately simulating channel effects, including interference, noise, distance-dependent path loss, and packet losses. RF-SITL works as follows: 1) it instantiates a virtual software-defined transceiver in GNU Radio for each UAV simulated in the UB-ANC Emulator;2) it builds an interference channel model in which each network node receives the superposition of signals transmitted from other nodes;and 3) it synchronizes the location of each simulated UAV in the UB-ANC Emulator with the virtualized RF transceivers in RF-SITL, such that the communication channel between nodes can accurately model distance-dependent channel effects, such as path loss. With these capabilities, we can use both off-the-shelf and custom-built signal processing flowgraphs that simulate Gaussian Minimum Shift Keying (GMSK), 802.11-like Orthogonal Frequency Division Multiplexing (OFDM), and direct sequence spread-spectrum (DSSS) links in GNU Radio to simulate swarm UAV networks prior to their deployment in software-defined radios in a swarm UAV network.

关键词： Radio frequency Wireless communication OFDM Atmospheric modeling signal processing algorithms Spread spectrum communication Autonomous aerial vehicles

来源：评论

学校读者我要写书评

暂无评论

Measuring The Fatigue of Persons Based on EOG signal processing Mmethods

Measuring The Fatigue of Persons Based on EOG Signal Process...

引用

International Conference on Engineering Technology and their Applications (IICETA)

作者： Hassanein R. Mahmood Mohammed S. Irshayyid Manaf K. Hussien Basim k. J. Al-Shammari Department of Electrical Engineering College of Engineering University of Wasit Wasit Iraq

ISBN: (数字)9781665472159

ISBN: (纸本)9781665472166

Driver fatigue is one of the major causes of a road accidents. Various driver fatigue detection systems have been designed to detect and warn the driver of impending fatigue. Most available prototypes and ongoing research have focused on the video-based eye-tracking system, which demands high computing power due to real-time video processing. In our research, the use of electrooculography (EOG) as an alternative to video-based systems in detecting eye activities caused by fatigue is evaluated. The cornea retinal potential (CRP) is the source of the EOG signals, which are being generated by the movement of the eyeballs within the conductive environment of the skull. Electrooculogram signals are generated from neurological conditions actions and EOG is generally used to capture these signals. EOG is an easy and low-cost method of recording eye movements. This paper proposed a fatigue detection mechanism which is based on EOG recorder with (Ag- AgC1) electrodes. Five subjects (2M, 3F) were participated in this study. Six features were presented to evaluate the suggested algorithm. The presented approach has managed to provide 89.2% to 96.5% effective blink detection.

关键词： Electrodes Electrooculography Road accidents signal processing algorithms Streaming media Fatigue Retina

来源：评论

学校读者我要写书评

暂无评论

EXPECTATION CONSISTENT PLUG-AND-PLAY FOR MRI 47

EXPECTATION CONSISTENT PLUG-AND-PLAY FOR MRI

引用

47th IEEE International Conference on Acoustics, Speech and signal processing (ICASSP)

作者： Shastri, Saurav K. Ahmad, Rizwan Metzler, Christopher A. Schniter, Philip Ohio State Univ Dept ECE Columbus OH 43210 USA Ohio State Univ Dept BME Columbus OH 43210 USA Univ Maryland Dept CS College Pk MD 20742 USA

ISBN: (纸本)9781665405409

For image recovery problems, plug-and-play (PnP) methods have been developed that replace the proximal step in an optimization algorithm with a call to an application-specific denoiser, often implemented using a deep neural network. Although such methods have been successful, they can be improved. For example, the denoiser is often trained using white Gaussian noise, while PnP's denoiser input error is often far from white and Gaussian, with statistics that are difficult to predict from iteration to iteration. PnP methods based on approximate message passing (AMP) are an exception, but only when the forward operator behaves like a large random matrix. In this work, we design a PnP method using the expectation consistent (EC) approximation algorithm, a generalization of AMP, that offers predictable error statistics at each iteration, from which a deep-net denoiser can be effectively trained.

关键词： Magnetic resonance imaging Message passing Gaussian noise Neural networks signal processing algorithms signal processing Prediction algorithms

来源：评论

学校读者我要写书评

暂无评论

Unsupervised Blind Joint Dereverberation and Room Acoustics Estimation With Diffusion Models

IEEE Transactions on Audio, Speech and Language Processing

引用

IEEE Transactions on Audio, Speech and Language processing 2025年 33卷 2244-2258页

作者： Jean-Marie Lemercier Eloi Moliner Simon Welker Vesa Välimäki Timo Gerkmann Signal Processing group at Universität Hamburg Hamburg Germany Acoustics Lab Department of Information Communications Engineering Aalto University Espoo Finland

This paper presents an unsupervised method for single-channel blind dereverberation and room impulse response (RIR) estimation, called BUDDy. The algorithm is rooted in Bayesian posterior sampling: it combines a likelihood model enforcing fidelity to the reverberant measurement, and an anechoic speech prior implemented by an unconditional diffusion model. We design a parametric filter representing the RIR, with exponential decay for each frequency subband. Room acoustics estimation and speech dereverberation are jointly carried out, as the filter parameters are iteratively estimated and the speech utterance refined along the reverse diffusion trajectory. In a blind scenario where the RIR is unknown, BUDDy successfully performs speech dereverberation in various acoustic scenarios, significantly outperforming other blind unsupervised baselines. Unlike supervised methods, which often struggle to generalize, BUDDy seamlessly adapts to different acoustic conditions. This paper extends our previous work by offering new experimental results and insights into the algorithm’s versatility. We demonstrate the robustness of our proposed method to new acoustic and speaker conditions, as well as its adaptability to high-resolution singing voice dereverberation, using both instrumental metrics and subjective listening evaluation. We study BUDDy’s performance for RIR estimation and observe it surpasses a state-of-the-art supervised DNN-based estimator on mismatched acoustic conditions. Finally, we investigate the sensitivity of informed dereverberation methods to RIR estimation errors, thereby motivating the joint acoustic estimation and dereverberation design.

关键词： Acoustics Diffusion models signal processing algorithms Reverberation Noise reduction Kernel Training Robustness Inverse problems Indexes

来源：评论

学校读者我要写书评

暂无评论

Robustifying GNN Via Weighted Laplacian 14

Robustifying GNN Via Weighted Laplacian

引用

14th IEEE International Conference on signal processing and Communications (SPCOM)

作者： Runwal, Bharat Vivek Kumar, Sandeep Indian Inst Technol Delhi New Delhi India Samsung Res Bangalore Bengaluru India

ISBN: (数字)9781665482509

ISBN: (纸本)9781665482509

Graph neural network (GNN) is achieving remarkable performances in a variety of application domains. However, GNN is vulnerable to noise and adversarial attacks in input data. Making GNN robust against noises and adversarial attacks is an important problem. The existing defense methods for GNNs are computationally demanding, are not scalable, and are architecture dependent. In this paper, we propose a generic framework for robustifying GNN known as Weighted Laplacian GNN (RWL-GNN). The method combines Weighted Graph Laplacian learning with the GNN implementation. The proposed method benefits from the positive semi-definiteness property of Laplacian matrix, feature smoothness, and latent features via formulating a unified optimization framework, which ensures the adversarial/noisy edges are discarded and connections in the graph are appropriately weighted. For demonstration, the experiments are conducted with Graph convolutional neural network(GCNN) architecture, however, the proposed framework is easily amenable to any existing GNN architecture. The simulation results with benchmark dataset establish the efficacy of the proposed method over the state-of-the-art methods, both in accuracy and computational efficiency.

关键词： Laplace equations Simulation Conferences signal processing algorithms Computer architecture signal processing Graph neural networks

来源：评论

学校读者我要写书评

暂无评论

FPGA Implementation of ANN Using Customized Instruction Set for 1-D signal processing

FPGA Implementation of ANN Using Customized Instruction Set ...

引用

International Conference on Computing Communication Control and Automation (ICCUBEA)

作者： Anudita Bagal Varsha Harpale Electronics and Telecommunication Engineering Pimpri-Chinchwad College of Engineering Pune Pune India

ISBN: (纸本)9781665484527

This work describes the implementation of hardware architecture based on ANN and its hardware realization using hardware description language (Verilog) on FPGA. The objective of this project is “FPGA implementation of ANN using customized Instruction set for 1-D signal processing”. Our aim in this project is on designing small in size and functional processor architecture. The proposed project presents architectural design and execution of artificial neural network processor with a customized instruction set processor for 1 D signal processing. The ANN processor architecture is flexible, supporting random counts of layers and a number of 784 artificial neurons (ANs), The proposed architecture will support the feed-forward algorithm. The proposed architecture is flexible with the counts of input data; output data and signals will represent the fixed-point format and will compare the accuracies of the Tanh function and sigmoid function.

关键词： Instruction sets Computational modeling Time series analysis signal processing algorithms Artificial neural networks Computer architecture signal processing

来源：评论

学校读者我要写书评

暂无评论

A New Outage Probability Bound for IR-HARQ and Its Application to Power Adaptation 23

A New Outage Probability Bound for IR-HARQ and Its Applicati...

引用

23rd International Workshop on signal processing Advances in Wireless Communication (SPAWC)

作者： Wang, Wenyu Shen, Kaiming Chinese Univ Hong Kong Sch Sci & Engn Shenzhen Peoples R China

ISBN: (数字)9781665494557

ISBN: (纸本)9781665494557

This work proposes a novel upper bound on outage probability for the incremental redundancy hybrid automatic repeat request (IR-HARQ) protocol over block fading channels. The new bound is much tighter than the classical bound due to Laneman, especially when the transmit power is limited as is often the case in the Internet of Things (IoT). We demonstrate the practical use of this new bound by showing that it enables a geometric programming approach to power adaptation for IR-HARQ. Because the new bound approximates the outage probability much more exactly, the resulting new power adaptation scheme significantly outperforms those existing methods based on the classical bound.

关键词： Fading channels Wireless communication Upper bound Protocols Redundancy signal processing algorithms Probability

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：