检索结果-内蒙古大学图书馆

Power Efficient Rotation Control of Linear Antenna Polarization

Power Efficient Rotation Control of Linear Antenna Polarizat...

2024 IEEE International Conference on Microwaves, Communications, Antennas, Biomedical engineering and Electronic Systems, COMCAS 2024

作者： Malka, Omer Cohen, Emanuel Technion Faculty of Electrical and Computer Engineering Israel

ISBN: (纸本)9798350348187

Power delivery in form of electro-magnetic (EM) field is heavily dependent on matching between radiating and observing elements. Both need to share the same resonating frequency and orientation of polarization. While resonating are pre-designed, the last's performance is heavily based on application use. This paper proposes simple, yet innovative technique to overcome loss of polarity and obtain of additional polarization degree of freedom while boosting power efficiency by 50% for Class-A power amplifier (PA) and 30% for Class-B. © 2024 IEEE.

关键词： Polarization

来源：评论

学校读者我要写书评

暂无评论

The Perception-Robustness Tradeoff in Deterministic Image Restoration 41

The Perception-Robustness Tradeoff in Deterministic Image Re...

引用

41st International Conference on Machine Learning, ICML 2024

作者： Ohayon, Guy Michaeli, Tomer Elad, Michael Faculty of Computer Science Technion Haifa Israel Faculty of Electrical and Computer Engineering Technion Haifa Israel

We study the behavior of deterministic methods for solving inverse problems in imaging. These methods are commonly designed to achieve two goals: (1) attaining high perceptual quality, and (2) generating reconstructions that are consistent with the measurements. We provide a rigorous proof that the better a predictor satisfies these two requirements, the larger its Lipschitz constant must be, regardless of the nature of the degradation involved. In particular, to approach perfect perceptual quality and perfect consistency, the Lipschitz constant of the model must grow to infinity. This implies that such methods are necessarily more susceptible to adversarial attacks. We demonstrate our theory on single image super-resolution algorithms, addressing both noisy and noiseless settings. We also show how this undesired behavior can be leveraged to explore the posterior distribution, thereby allowing the deterministic model to imitate stochastic methods. Copyright 2024 by the author(s)

关键词： Stochastic systems

来源：评论

学校读者我要写书评

暂无评论

Equivariant Machine Learning on Graphs with Nonlinear Spectral Filters 38

Equivariant Machine Learning on Graphs with Nonlinear Spectr...

引用

38th Conference on Neural Information Processing Systems, NeurIPS 2024

作者： Lin, Ya-Wei Eileen Talmon, Ronen Levie, Ron Viterbi Faculty of Electrical and Computer Engineering Technion Israel Faculty of Mathematics Technion Israel

Equivariant machine learning is an approach for designing deep learning models that respect the symmetries of the problem, with the aim of reducing model complexity and improving generalization. In this paper, we focus on an extension of shift equivariance, which is the basis of convolution networks on images, to general graphs. Unlike images, graphs do not have a natural notion of domain translation. Therefore, we consider the graph functional shifts as the symmetry group: the unitary operators that commute with the graph shift operator. Notably, such symmetries operate in the signal space rather than directly in the spatial space. We remark that each linear filter layer of a standard spectral graph neural network (GNN) commutes with graph functional shifts, but the activation function breaks this symmetry. Instead, we propose nonlinear spectral filters (NLSFs) that are fully equivariant to graph functional shifts and show that they have universal approximation properties. The proposed NLSFs are based on a new form of spectral domain that is transferable between graphs. We demonstrate the superior performance of NLSFs over existing spectral GNNs in node and graph classification benchmarks. © 2024 Neural information processing systems foundation. All rights reserved.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Point spread function modeling and engineering in black-box lens systems

引用

Optics Express 2025年第3期33卷 4211-4224页

作者： Xiao, Dafei Kedem, Reut Orange Shechtman, Yoav Russell Berrie Nanotechnology Institute Technion–Israel Institute of Technology Haifa Israel Faculty of Biomedical Engineering Technion–Israel Institute of Technology Haifa Israel Faculty of Electrical and Computer Engineering Technion–Israel Institute of Technology Haifa Israel

Point spread function (PSF) engineering in an imaging system involves the introduction of additional optical components to efficiently encode object information. It typically relies on a well-established mapping relation between object space, PSF engineering plane, and image plane. However, this reliance can limit its application in imaging systems with unknown ("black box") attributes, e. g. commercial photographic lenses, where a theoretical imaging model is not readily available. To address this limitation, we conduct wave-optics analysis of general lens systems, explicitly derive a general imaging model based on scaled Fourier transform, and develop an overall pipeline for automatic parameter determination of the model. When applied to a photographic lens, this pipeline demonstrates high PSF modeling accuracy, showcasing the capability of "seeing what you design". We then implemented PSF engineering in this lens to find its optimal PSF for 3D localization. This work improves computational imaging by enabling accurate wavefront shaping design in arbitrary imaging systems. © 2025 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement.

关键词： Optical transfer function

来源：评论

学校读者我要写书评

暂无评论

Quantum Coordination Rates in Multi-User Networks

引用

IEEE Transactions on Information Theory 2025年第6期71卷 4428-4453页

作者： Natur, Husein Pereg, Uzi Technion-Israel Institute of Technology Faculty of Electrical and Computer Engineering Helen Diller Quantum Center Haifa3200003 Israel

Quantum coordination is considered in networks with classical and quantum links. We begin with networks with classical links, and characterize the generation of separable and classical-quantum correlations in three primary models: 1) a two-node network with limited common randomness (CR), 2) a no-communication network, and 3) a broadcast network, which consists of a single sender and two receivers. We establish the optimal tradeoff between the classical communication and CR rates in each setting, thus characterizing the minimal resources for simulating classical-quantum correlations. Next, we consider coordination in networks with quantum links. We study the following models: 1) a cascade network with limited entanglement, 2) a broadcast network, and 3) a multiple-access network with two senders and a single receiver. We establish the optimal tradeoff between quantum communication and entanglement rates in each setting, characterizing the minimal resources for entanglement coordination. The examples demonstrate that coordination of entanglement and coordination of separable correlations behave differently. At last, we show the implications of our results on nonlocal games with quantum strategies. © 1963-2012 IEEE.

关键词： Quantum communication

来源：评论

学校读者我要写书评

暂无评论

Comparative Analysis of N-Path Mixer NF vs Blocker Tolerance for Different CMOS Processes

Comparative Analysis of N-Path Mixer NF vs Blocker Tolerance...

引用

2024 IEEE International Conference on Microwaves, Communications, Antennas, Biomedical engineering and Electronic Systems, COMCAS 2024

作者： Jiang, Yan Cohen, Emanuel Technion Faculty of Electrical and Computer Engineering Haifa Israel

ISBN: (纸本)9798350348187

This paper investigates the constraints on N-path mixer design in different CMOS processes. Utilizing a linear time-invariant (LTI) model, we conduct a comparative analysis within the context of mixer-first receiver (MFRX) design considerations. As the process scales down, smaller $C{o f f}$ and gate capacitance $\left(C{g}\right)$ can be obtained for the same switched-on resistance ($R{o n}$). Consequently, a lower noise factor (NF) can be achieved by reducing the clock constraints and maintaining good performance at higher frequencies. Besides, for the same $C{o f f}$, a smaller node enables a better out-of-band (OOB) blocker 1 dB compression point $\left(B{1 d B}\right)$. We investigate the influence of process characteristics $\left(R{o n} C{o f f}, V{t h}, C{g}\right)$ on N-Path circuits performance, with a non-ideal 25% duty cycle. The analysis is based on a 4-Path mixer and with a voltage negative feedback amplifier at its baseband (BB) to form an MFRX topology. Results are verified in TSMC 28nm CMOS and TSMC 65nm CMOS technology. © 2024 IEEE.

关键词： CMOS integrated circuits

来源：评论

学校读者我要写书评

暂无评论

Hyperbolic Distance Based on EMD and Diffusion for Hyperspectral Imaging

Hyperbolic Distance Based on EMD and Diffusion for Hyperspec...

引用

2025 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2025

作者： Lavi, Elad Bourvine, Amir Lin, Ya-Wei Eileen Talmon, Ronen Taub Faculty of Computer Science Technion China Viterbi Faculty of Electrical and Computer Engineering Technion Israel

ISBN: (纸本)9798350368741

In this paper, we introduce EMD-Based Hyperbolic Diffusion Distance (EMD-HDD), a new method for constructing a meaningful distance metric for hierarchical data with latent hierarchical structure. Our method relies on hyperbolic geometry, diffusion geometry, and the Earth Mover's Distance (EMD). Specifically, our method embeds data points into a product manifold of hyperbolic spaces, allowing us to recover the hidden hierarchical structure encoded by the mutual relationships between features. We demonstrate the effectiveness of EMD-HDD through experiments on five hyperspectral imaging datasets, showcasing its capability to capture and reveal the intrinsic hierarchical structures inherent in such data. © 2025 IEEE.

关键词： Earth Mover's Distance Hyperbolic geometry Hyperspectral Imaging Manifold Learning

来源：评论

学校读者我要写书评

暂无评论

Direct Learning Neural Network Digital Predistortion Using Backpropagation Through a Memory Power Amplifier Model

Direct Learning Neural Network Digital Predistortion Using B...

引用

2023 IEEE/MTT-S International Microwave Symposium, IMS 2023

作者： Loebl, Erez Ginzberg, Nimrod Cohen, Emanuel Faculty of Electrical and Computer Engineering Technion Israel

ISBN: (纸本)9798350347647

This work presents a new direct learning for a neural network based digital predistortion linearizer using backpropagation through a memory power amplifier. The learning method is compared with indirect learning among other methods. The proposed learning methods are evaluated on a wideband CMOS power amplifier using a 160 MHz 802.11ax WiFi signal with an 11.5 dB peak-to-average power ratio and a center frequency of 1 GHz at 8.2 dBm average transmit power. The direct learning using memory backpropagation shows superior performance to other proposed learning methods, achieving a correction of 10.8 dB to the error vector magnitude to -39 dB EVM. © 2023 IEEE.

关键词： Backpropagation

来源：评论

学校读者我要写书评

暂无评论

Classification Diffusion Models: Revitalizing Density Ratio Estimation 38

Classification Diffusion Models: Revitalizing Density Ratio ...

引用

38th Conference on Neural Information Processing Systems, NeurIPS 2024

作者： Yadin, Shahar Elata, Noam Michaeli, Tomer Faculty of Electrical and Computer Engineering Technion - Israel Institute of Technology Israel

A prominent family of methods for learning data distributions relies on density ratio estimation (DRE), where a model is trained to classify between data samples and samples from some reference distribution. DRE-based models can directly output the likelihood for any given input, a highly desired property that is lacking in most generative techniques. Nevertheless, to date, DRE methods have failed in accurately capturing the distributions of complex high-dimensional data, like images, and have thus been drawing reduced research attention in recent years. In this work we present classification diffusion models (CDMs), a DRE-based generative method that adopts the formalism of denoising diffusion models (DDMs) while making use of a classifier that predicts the level of noise added to a clean signal. Our method is based on an analytical connection that we derive between the MSE-optimal denoiser for removing white Gaussian noise and the cross-entropy-optimal classifier for predicting the noise level. Our method is the first DRE-based technique that can successfully generate images beyond the MNIST dataset. Furthermore, it can output the likelihood of any input in a single forward pass, achieving state-of-the-art negative log likelihood (NLL) among methods with this property. Code is available on the project's webpage. © 2024 Neural information processing systems foundation. All rights reserved.

关键词：

来源：评论

学校读者我要写书评

暂无评论

ON THE POSTERIOR DISTRIBUTION IN DENOISING: APPLICATION TO UNCERTAINTY QUANTIFICATION 12

ON THE POSTERIOR DISTRIBUTION IN DENOISING: APPLICATION TO U...

引用

12th International Conference on Learning Representations, ICLR 2024

作者： Manor, Hila Michaeli, Tomer Faculty of Electrical and Computer Engineering Technion - Israel Institute of Technology Israel

Denoisers play a central role in many applications, from noise suppression in low-grade imaging sensors, to empowering score-based generative models. The latter category of methods makes use of Tweedie's formula, which links the posterior mean in Gaussian denoising (i.e., the minimum MSE denoiser) with the score of the data distribution. Here, we derive a fundamental relation between the higher-order central moments of the posterior distribution, and the higher-order derivatives of the posterior mean. We harness this result for uncertainty quantification of pre-trained denoisers. Particularly, we show how to efficiently compute the principal components of the posterior distribution for any desired region of an image, as well as to approximate the full marginal distribution along those (or any other) one-dimensional directions. Our method is fast and memory-efficient, as it does not explicitly compute or store the high-order moment tensors and it requires no training or fine tuning of the denoiser. Code and examples are available on the project website. © 2024 12th International Conference on Learning Representations, ICLR 2024. All rights reserved.

关键词：

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：