Article
Computer Science, Information Systems
Ahmed Aziz, Walid Osamy, Ahmed M. Khedr, Ahmed Salim
Summary: Compressive Sensing (CS) is a widely used sampling theory in signal processing due to its simplicity and efficiency. In this paper, an Adaptive Iterative Forward-Backward Greedy Algorithm (AFB) is proposed to address the challenge of signal reconstruction in CS. AFB algorithm improves the reconstruction performance by solving the least squares problem in the forward phase and refining the selection process. The simulation results demonstrate that AFB outperforms other algorithms in reducing reconstruction error.
JOURNAL OF KING SAUD UNIVERSITY-COMPUTER AND INFORMATION SCIENCES
(2022)
Article
Computer Science, Information Systems
Crisostomo Barajas-Solano, Juan-Marcos Ramirez, Jose Ignacio Martinez Torre, Henry Arguello
Summary: This paper introduces a method for recovering high-quality spectral videos using 4D convolutional sparse representation, without additional optical flow information. Extensive numerical simulations show that this method surpasses the state-of-the-art in terms of quality and border sharpness.
JOURNAL OF VISUAL COMMUNICATION AND IMAGE REPRESENTATION
(2023)
Article
Engineering, Electrical & Electronic
Jue Chen, Tsang-Yi Wang, Jwo-Yuh Wu, Chih-Peng Li, Soon Xin Ng, Robert G. Maunder, Lajos Hanzo
Summary: A new support identification technique based on factor graphs and belief propagation is proposed for compressive sensing aided wireless sensor networks. It achieves a support identification error rate of 10% at a lower SNR compared to the state-of-the-art relative frequency based approach and OMP algorithm, while mitigating signal reconstruction noise by 4 dB.
IEEE SENSORS JOURNAL
(2021)
Article
Engineering, Electrical & Electronic
Ghanbar Azarnia, Abbas Ali Sharifi
Summary: In this study, we propose a distributed and cooperative algorithm based on FOCUSS for signal reconstruction in wireless sensor networks. The algorithm outperforms existing methods in terms of reconstruction accuracy and convergence rate.
INTERNATIONAL JOURNAL OF COMMUNICATION SYSTEMS
(2022)
Article
Acoustics
Zhe Wang, Shen Wang, Qing Wang, Zaifu Zhan, Wei Zhao, Songling Huang
Summary: This study proposes a time-frequency representation (TFR) reconstruction scheme for undersampled Lamb wave signals, incorporating both sparse prior and structural sparse prior, which can effectively recover missing information. Experimental results show that the time-of-flight information extracted from the recovered TFR in noisy environments is accurate with a relative error less than 3%.
Article
Computer Science, Artificial Intelligence
Abhishek Jain, Preety D. Swami, Ashutosh Datar
Summary: Most existing CS-based image acquisition and retrieval algorithms are for gray images only, and recent color image CS methods are computationally complex and have unsatisfactory visual quality. This work proposes a TPGHR model for color image CS, which involves the use of YUV color space and different operations in luminance and chroma channels. Experimental results show that the proposed model outperforms existing CS methods in terms of visual quality, execution time, and PSNR performance.
Article
Engineering, Electrical & Electronic
Qier An, Yuan Shen
Summary: This paper introduces a coverage performance metric considering both perception quality and cover rate, along with a distributed coverage control scheme for mobile anisotropic sensor networks with communication capability. Specific distributed camera coverage control schemes are proposed for anisotropic sensors, focusing on the camera, based on a camera sensing model characterizing the dependence of anisotropic perception trait on relative pose, field of view, and focal length.
IEEE SENSORS JOURNAL
(2021)
Article
Computer Science, Artificial Intelligence
Donghua Jiang, Nestor Tsafack, Wadii Boulila, Jawad Ahmad, J. J. Barba-Franco
Summary: Recent advances in intelligent wearable devices have created opportunities for healthcare monitoring system development. To protect user privacy-related information, a solution based on compression-encryption architecture is proposed.
EXPERT SYSTEMS WITH APPLICATIONS
(2024)
Article
Computer Science, Information Systems
Siwang Zhou, Yi Lian, Daibo Liu, Hongbo Jiang, Yonghe Liu, Keqin Li
Summary: This article addresses the decentralized storage problem in mobile crowdsensing systems and proposes a compressive distributed storage scheme based on compressive sensing. By encoding the sensing data in the local trajectories of participants and exploiting inter-period correlations, the entire information of the area can be stored and recovered with improved accuracy.
ACM TRANSACTIONS ON SENSOR NETWORKS
(2022)
Article
Computer Science, Information Systems
Jin Xu, Zhizhong Fu
Summary: In this paper, an efficient hybrid regularization approach for image CS reconstruction is proposed, which can simultaneously exploit both internal and external image priors. A novel centralized group sparse representation (CGSR) model is designed to effectively exploit internal image sparsity prior, while a state-of-the-art deep image denoiser is plugged into the optimization model to implicitly exploit external deep denoiser prior. Experimental results demonstrate that the proposed method outperforms some state-of-the-art image CS reconstruction methods in terms of both objective quality and visual perception.
JOURNAL OF VISUAL COMMUNICATION AND IMAGE REPRESENTATION
(2023)
Article
Computer Science, Artificial Intelligence
Chengxi Li, Gang Li, Pramod K. Varshney
Summary: This paper explores the distributed detection of sparse signals in energy-limited clustered sensor networks. A new detector is proposed to address the high energy consumption issue of centralized detectors, by combining censoring and locally most powerful test strategies.
INFORMATION FUSION
(2022)
Article
Engineering, Electrical & Electronic
Yuqing Yang, Peng Xiao, Nikos Deligiannis
Summary: This paper presents a new method for underwater source localization by combining the matched field processing method (MFP) with 1-bit compressive sensing (1-bit CS). The Fixed Point Continuation (FPC) method and a deep neural network (DNN) are used to solve the 1-bit recovery problem and evaluate their performance in source localization. Additionally, a simple average technique is proposed to improve the robustness of signal recovery to noise added in the binary measurements.
DIGITAL SIGNAL PROCESSING
(2023)
Article
Computer Science, Information Systems
Jinho Choi
Summary: This study investigates data-aided sensing for distributed detection in wireless sensor networks, proposing a node selection criterion based on J-divergence to ensure reliable decision-making with minimal decision delay. Simulation results confirm that the J-divergence based DAS can provide reliable decisions with fewer sensors compared to other approaches.
IEEE WIRELESS COMMUNICATIONS LETTERS
(2021)
Article
Computer Science, Information Systems
Joannes Sam Mertens, Laura Galluccio, Giacomo Morabito
Summary: This work introduces the i-WSN League framework, which is a comprehensive hardware/software framework for distributed training and inference. The framework combines gossiping and clustering to adapt the operations executed by each node to its capabilities, aiming to minimize energy consumption in resource-limited nodes while preserving accuracy.
IEEE INTERNET OF THINGS JOURNAL
(2023)
Article
Neurosciences
Iiro P. Jaeaeskelaeinen, Enrico Glerean, Vasily Klucharev, Anna Shestakova, Jyrki Ahveninen
Summary: Accumulated findings from MVPA suggest that fingerprint patterns of activations and deactivations in brain activity may reflect neuronal-population level sparse code, which interacts across multiple levels of brain hierarchy and gives rise to perception, emotions, and cognition.