Filters








5,645 Hits in 2.6 sec

Sparse Quantized Spectral Clustering [article]

Zhenyu Liao, Romain Couillet, Michael W. Mahoney
2020 arXiv   pre-print
with very aggressively sparsified or quantized spectral clustering.  ...  We illustrate how these results depend on the nonlinearity, we characterize a phase transition beyond which spectral clustering becomes possible, and we show when such nonlinear transformations can introduce  ...  Most of our technical results hold for rather generic functions f , e.g., those of interest beyond sparse quantized spectral clustering, but we are particularly interested in f with nontrivial numerical  ... 
arXiv:2010.01376v1 fatcat:wvelfmqfczayzgvctslb43ceay

A Multichannel Gray Level Co-Occurrence Matrix for Multi/Hyperspectral Image Texture Representation

Xin Huang, Xiaobo Liu, Liangpei Zhang
2014 Remote Sensing  
Accordingly, in this study, we propose to represent the multichannel textures for multi/hyperspectral imagery by the use of: (1) clustering algorithms; and (2) sparse representation, respectively.  ...  Specifically, K-means and fuzzy c-means methods are adopted to generate the codes of an image from the clustering point of view, while a sparse dictionary learning method based on two coding rules is proposed  ...  Clustering-Based Quantization Clustering aims at partitioning pixels with similar spectral properties to the same class, and thus it can be used to quantize a multi/hyperspectral image into N g levels.  ... 
doi:10.3390/rs6098424 fatcat:7sgrspg5mjbptd5hwo5xuuhwe4

Learning Sparse Convolutional Neural Network via Quantization with Low Rank Regularization

Xin Long, Zongcheng Ben, Xiangrong Zeng, Yan Liu, Maojun Zhang, Dianle Zhou
2019 IEEE Access  
Considering the different peculiarities of weight quantization and sparse regularization, in this paper, we propose a low rank sparse quantization (LRSQ) method to quantize network weights and regularize  ...  INDEX TERMS Convolutional neural network (CNN), weight quantization, spectral regularization, sparsity, visualization, channel pruning.  ...  LOW RANK SPARSE QUANTIZATION In this work, we propose using, as a regularizer for convolutional layers, the regularization constraint of spectral clustering.  ... 
doi:10.1109/access.2019.2911536 fatcat:w2zgpgmhknchxiapn6mbndvxga

Sparse Joint Transmission for Cell-Free Massive MIMO: A Sparse PCA Approach [article]

Deokhwan Han, Jeonghun Park, Namyoon Lee
2021 arXiv   pre-print
By simulations, we show that sparse-JT achieves higher ergodic spectral efficiencies than those attained by multi-cell zero-forcing precoding with the user-centric AP clustering algorithm in all system  ...  Sparse-JT jointly identifies the user-centric cooperative APs sets, precoding vectors for beamforming and compression, and power allocation that maximizes a lower bound of the sum-spectral efficiency under  ...  Fig. 1 . 1 An illustration of cell-free massive MIMO systems applied with sparse-dynamic clustering under the limited backhaul capacity.  ... 
arXiv:1912.05231v2 fatcat:xsgeb3b3ujfhlb4lita56pktoy

Sparse coding of auditory features for machine hearing in interference

Richard F. Lyon, Jay Ponte, Gal Chechik
2011 2011 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)  
We recently extended our testing of this approach using sound mixtures, and found that the sparse-coded auditory-image features degrade less in interference than vector-quantized MFCC sparse features do  ...  For audio-file ranking and retrieval from text queries, based on stabilized auditory images, we took a multi-scale approach, using vector quantization to choose one sparse feature in each of many overlapping  ...  For the vector-quantized SAI features, we tuned one parameter, the number of clusters used for quantization of each local box pattern (we tested 500, 750, 1000, 1500, and 2000 clusters per box).  ... 
doi:10.1109/icassp.2011.5947698 dblp:conf/icassp/LyonPC11 fatcat:tmzhx26olfh6lkgyych67yzwlu

Spatially Sparse Precoding in Millimeter Wave MIMO Systems

Omar El Ayach, Sridhar Rajagopal, Shadi Abu-Surra, Zhouyue Pi, Robert W. Heath
2014 IEEE Transactions on Wireless Communications  
Beamforming with multiple data streams, known as precoding, can be used to further improve mmWave spectral efficiency.  ...  We exploit the spatial structure of mmWave channels to formulate the precoding/combining problem as a sparse reconstruction problem.  ...  SPATIALLY SPARSE PRECODING FOR THE SINGLE USER MMWAVE CHANNEL We seek to design hybrid mmWave precoders (F RF , F BB ) that maximize the spectral efficiency expression in (3) .  ... 
doi:10.1109/twc.2014.011714.130846 fatcat:x2wqp6fbffhiljqqk5jf5echja

Limited Channel Feedback Scheme for Reconfigurable Intelligent Surface Assisted MU-MIMO Wireless Communication Systems

Beom-Sik Shin, Ji-Hye Oh, Young-Hwan You, Duck-Dong Hwang, Hyoung-Kyu Song
2022 IEEE Access  
By utilizing sparse nature of mmWave propagation environment, the downlink CSI is compressed with the small number of channel vector and the corresponding index is transferred to the BS.  ...  RIS is the state-of-the-art technology that is extensively researched as the solution to solve the problems derived from the sparse propagation environment of the millimeter wave (mmWave) communication  ...  To reduce channel feedback overhead, the codebook generated by clustering algorithm is adopted to reduce rate loss derived from quantization error. • Finally, the BS recovers the sparse vector by the orthogonal  ... 
doi:10.1109/access.2022.3174095 fatcat:kze2wlnx4zhjnemycvgbizhs4i

A New Design of Codebook for Hybrid Precoding in Millimeter-Wave Massive MIMO Systems

Gang Liu, Honggui Deng, Kai Yang, Zaoxing Zhu, Jitai Liu, Hu Dong
2021 Symmetry  
with quantization algorithm under low SNR and different numbers of RF chains.  ...  The simulation results demonstrate that the spectral efficiency of the algorithm is obviously outstanding compared with that of the OMP-based joint codebook algorithm and the hybrid precoding algorithm  ...  with quantization in [1] in the SNR interval [−5, 30].  ... 
doi:10.3390/sym13050743 doaj:54549bb3cb874d00a4b1743576961de6 fatcat:ef6f76b6fjdmtbdrtqxm3zpq7e

Non-Uniform Quantization Codebook-Based Hybrid Precoding to Reduce Feedback Overhead in Millimeter Wave MIMO Systems

Yun Chen, Da Chen, Tao Jiang
2019 IEEE Transactions on Communications  
Specifically, we firstly group the angles of the arrive/departure (AOAs/AODs) of the scattering paths into several spatial lobes by exploiting the sparseness property of the millimeter wave in the angular  ...  Then, we map the quantization bits non-uniformly to different coverage angles and construct NUQ codebooks, where high numbers of quantization bits are employed for the effective coverage angles to quantize  ...  Then, non-uniform quantization codebooks are proposed by exploiting the sparseness property of the millimeter wave in the angular domain.  ... 
doi:10.1109/tcomm.2018.2890227 fatcat:fosigb55cvccndld7gtarhtnby

Automatic clustering based on an information-theoretic approach with application to spectral anomaly detection

Mark J. Carlotto, Ivan Kadar
2006 Signal Processing, Sensor Fusion, and Target Recognition XV  
VQ clusterings are evaluated within an anomaly detector, which detects manmade object/changes as spectral outliers within a set of background clusters.  ...  The method is evaluated using two different clustering algorithms: a mode finder based on scale-space algorithm, and a vector quantizer (VQ).  ...  The cluster map (c) shows the VQ algorithm has partitioned the image roughly into bare soil/concrete and sparse vegetation, with the airplanes assigned to the sparse vegetation cluster.  ... 
doi:10.1117/12.668805 fatcat:s5z2lgn5e5bz7ixpzlaukotagq

Jittered random sampling with a successive approximation ADC

Chenchi Eric Luo, Lingchen Zhu
2014 2014 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)  
A reconstruction algorithm called Successive Sine Matching Pursuit (SSMP) is proposed to recover spectrally sparse signals when sampled by the proposed SAR ADC at a sub-Nyquist rate.  ...  Based on a discrete jittered random sampling theory, this paper analyzes the impact of the random jitters and the resulting randomized quantization noise for a class of sparse or compressible signals.  ...  Three factors: aliasing noise, spectral leakage, and amplitude quantization noise make the sampled signal not perfectly sparse in the frequency domain.  ... 
doi:10.1109/icassp.2014.6853908 dblp:conf/icassp/LuoZ14 fatcat:gppjt4fksje6zespkfxy3got24

Multilayer bootstrap networks [article]

Xiao-Lei Zhang
2018 arXiv   pre-print
It consists of a group of k-centroids clusterings.  ...  Each clustering randomly selects data points with randomly selected features as its centroids, and learns a one-hot encoder by one-nearest-neighbor optimization.  ...  Our MBN builds each layer of sub-quantizers on the output sparse feature of its lower layer which is fundamentally different from the hierarchical product quantization method.  ... 
arXiv:1408.0848v8 fatcat:boihnyheffhqhl34ssv5sbglvy

SCENE CLASSFICATION BASED ON THE SEMANTIC-FEATURE FUSION FULLY SPARSE TOPIC MODEL FOR HIGH SPATIAL RESOLUTION REMOTE SENSING IMAGERY

Qiqi Zhu, Yanfei Zhong, Liangpei Zhang
2016 The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences  
In this paper, a semantic-feature fusion fully sparse topic model (SFF-FSTM) is proposed for HSR imagery scene classification.  ...  In SFF-FSTM, three heterogeneous features – the mean and standard deviation based spectral feature, wavelet based texture feature, and dense scale-invariant feature transform (SIFT) based structural  ...  In SFF-FSTM, the spectral, texture, and SIFT features are quantized separately by k-mean clustering algorithm to acquire three distinct 1-D histograms, H 1 , H 2 , and H 3 .  ... 
doi:10.5194/isprs-archives-xli-b7-451-2016 fatcat:htf52a2tlbg2zaw2bgku4ljvu4

SCENE CLASSFICATION BASED ON THE SEMANTIC-FEATURE FUSION FULLY SPARSE TOPIC MODEL FOR HIGH SPATIAL RESOLUTION REMOTE SENSING IMAGERY

Qiqi Zhu, Yanfei Zhong, Liangpei Zhang
2016 The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences  
In this paper, a semantic-feature fusion fully sparse topic model (SFF-FSTM) is proposed for HSR imagery scene classification.  ...  In SFF-FSTM, three heterogeneous features – the mean and standard deviation based spectral feature, wavelet based texture feature, and dense scale-invariant feature transform (SIFT) based structural  ...  In SFF-FSTM, the spectral, texture, and SIFT features are quantized separately by k-mean clustering algorithm to acquire three distinct 1-D histograms, H 1 , H 2 , and H 3 .  ... 
doi:10.5194/isprsarchives-xli-b7-451-2016 fatcat:dj2jmjeyevcunlrl5onrbkttzi

P-BOOST: Parallel Boosting of Optimal Narrow-Band Direction of Arrival Estimators [article]

Elio D. Di Claudio, Raffaele Parisi, Giovanni Jacovitti
2019 arXiv   pre-print
However, because of quantized locations, conventional sparse solvers present some ambiguity problems.  ...  Sparse under-determined solvers are considered as viable solutions to this problem, since they drastically reduce the dimensionality of the search space by exploiting the array model sparseness.  ...  In particular, the DOA quantization bias results in either spurious peaks, or in excess fitting errors and spectral loss of resolution w.r.t. to their ML counterparts.  ... 
arXiv:1909.07185v1 fatcat:arxybyss75ghpi44rdcuv2ftza
« Previous Showing results 1 — 15 out of 5,645 results