Model Refinement Learning and an Example on Channel Estimation With Universal Noise Model
The DOI matches its Crossref record; 51 of the 51 references it cites were found in a registry or an index.
Abstract
Model-based method and data-based method are two basic approaches for the design of wireless communication systems. Model-based methods suffer from inaccurate modeling assumptions due to excessively complex environment. Recently, data-based methods have achieved remarkable performances in the communication system design without the knowledge of accurate model but encounter some challenges such as, lack of available labelled training data and explainability. In this paper, we propose a novel hybrid idea to integrate the strengths of both data and model-based methods, named model refinement learning, which is training affordable, theoretically interpretable and self-adapting. To show the idea more concretely, a novel channel estimation algorithm is proposed in the multiple-input single-output (MISO) system in the case where the noise model is unknown. In particular, we utilize a universal mixture of Gaussian (MoG) model, which can adaptively adjust the involved parameters to fit the true noise distribution by using observed data. We propose a novel variational inference framework to achieve automatical noise model refinement and design the corresponding online channel estimator. To reduce the online algorithm overhead, we propose a decoupled variational Bayesian method to achieve linear computational complexity. Simulations show that our proposed method outperforms both the model-based and data-based counterparts.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
In the citing paper
“For anomaly detection in microservices, TDAI is particularly relevant in edge and IoT scenarios, where embedded AI models are deployed close to the data source [237,238].”
From Anomaly detection and root-cause identification in microservices: a survey (2026) · cites this paper as [238]
Citations
Cites
Book Review: Pattern Recognition and Machine Learning
Nasrabadi, N. M. (2007) · Journal of Electronic Imaging · doi:10.1117/1.2819119 · Cited by 22,066 · cited 2×
“MoG distribution has been widely used in signal processing [30] and pattern recognition [31], which is a simple linear superposition of Gaussian components and can provide a richer class of density distributions [32].”
MethodAuto-Encoding Variational Bayes
Kingma, D. P., & Welling, M. (2013) · arXiv · doi:10.48550/arxiv.1312.6114 · Cited by 20,067
Fundamentals of Statistical Signal Processing: Estimation Theory
Sengijpta, S. K. (1995) · Technometrics · doi:10.1080/00401706.1995.10484391 · Cited by 11,076
Fundamentals of statistical signal processing: Estimation theory
Prasad, K. V. (1994) · Control Engineering Practice · doi:10.1016/0967-0661(94)90195-3 · Cited by 7,902
Pattern Recognition and Machine Learning
Neal, R. M. (2007) · Technometrics · doi:10.1198/tech.2007.s518 · Cited by 4,571
Graphical Models, Exponential Families, and Variational Inference
Wainwright, M. J., & Jordan, M. I. (2007) · now publishers, Inc. eBooks · doi:10.1561/9781601981851 · Cited by 2,959
Power of Deep Learning for Channel Estimation and Signal Detection in OFDM Systems
Ye, H., Li, G. Y., & Juang, B.-H. (2018) · IEEE Wireless Communications Letters · doi:10.1109/lwc.2017.2757490 · Cited by 2,000 · cited 4×
“Such a phenomenon is significantly different from the LMMSE estimator or other data-based and model-based DL methods [13], [22], in which the estimators are fixed after the training steps.”
BackgroundMethodInfluentialGraphical Models, Exponential Families, and Variational Inference
Wainwright, M. J., & Jordan, M. I. (2008) · Foundations and Trends® in Machine Learning · doi:10.1561/2200000001 · Cited by 1,820 · cited 1×
“And the distribution q ( θ ) can be factorized as This factorized form of q ( θ ) in the variational inference corresponds to an approximation framework developed in the statistical physics literature called mean field theory [42].”
MethodA Constructive Definition of Dirichlet Priors
Sethuraman, J. (1991) · doi:10.21236/ada238689 · Cited by 1,572
Malware traffic classification using convolutional neural network for representation learning
Wang, W., Zhu, M., Zeng, X., Ye, X., & Sheng, Y. (2017) · 2017 International Conference on Information Networking (ICOIN) · doi:10.1109/icoin.2017.7899588 · Cited by 1,177 · cited 1×
“Due to the powerful mechanism, DL has made great progress in network traffic prediction [1], [2], data traffic classification [3], [4], network routing [5], network scheduling [6] and resource allocation [7], [8], etc.”
BackgroundEnd-to-end encrypted traffic classification with one-dimensional convolution neural networks
Wang, W., Zhu, M., Wang, J., Zeng, X., & Yang, Z. (2017) · 2017 IEEE International Conference on Intelligence and Security Informatics (ISI) · doi:10.1109/isi.2017.8004872 · Cited by 991
Deep Reinforcement Learning Based Resource Allocation for V2V Communications
Ye, H., Li, G. Y., & Juang, B.-H. F. (2019) · IEEE Transactions on Vehicular Technology · doi:10.1109/tvt.2019.2897134 · Cited by 870
Deep Learning-Based Channel Estimation for Beamspace mmWave Massive MIMO Systems
He, H., Wen, C.-K., Jin, S., & Li, G. Y. (2018) · IEEE Wireless Communications Letters · doi:10.1109/lwc.2018.2832128 · Cited by 823
Deep Learning-based Channel Estimation for Beamspace mmWave Massive MIMO Systems
He, H., Wen, C.-K., Jin, S., & Li, G. Y. (2018) · arXiv · doi:10.48550/arxiv.1802.01290 · Cited by 823
Deep Learning-Based Channel Estimation
Soltani, M., Pourahmadi, V., Mirzaei, A., & Sheikhzadeh, H. (2019) · IEEE Communications Letters · doi:10.1109/lcomm.2019.2898944 · Cited by 780
Middleton, D. (1999) · IEEE Transactions on Information Theory · doi:10.1109/18.761256 · Cited by 695 · cited 1×
“Specifi-cally, MoG distribution provides accurate approximation of the Middleton Class A noise model [34], which is widely used to measure the impulse noise in wireless communication systems [35], [36] with the Gaussian-to-impulse noise power ratio controlled by the MoG parameters.”
MethodLearning to decode linear codes using deep learning
Nachmani, E., Be'ery, Y., & Burshtein, D. (2016) · 2016 54th Annual Allerton Conference on Communication, Control, and Computing (Allerton) · doi:10.1109/allerton.2016.7852251 · Cited by 561 · cited 1×
“Several researches have been proposed to utilize DNNs to substitute a certain module in the physical layer of a wireless transceiver system, such as data detection [9], [10], channel estimation [11]–[14], and channel decoding [15], [16], etc.”
MethodExpectation-Maximization Gaussian-Mixture Approximate Message Passing
Vila, J. P., & Schniter, P. (2013) · IEEE Transactions on Signal Processing · doi:10.1109/tsp.2013.2272287 · Cited by 501 · cited 1×
“MoG distribution has been widely used in signal processing [30] and pattern recognition [31], which is a simple linear superposition of Gaussian components and can provide a richer class of density distributions [32].”
MethodModel-Driven Deep Learning for Physical Layer Communications
He, H., Jin, S., Wen, C.-K., Gao, F., Li, G. Y., & Xu, Z. (2019) · IEEE Wireless Communications · doi:10.1109/mwc.2019.1800447 · Cited by 481
Samuel, N., Diskin, T., & Wiesel, A. (2017) · 2017 IEEE 18th International Workshop on Signal Processing Advances in Wireless Communications (SPAWC) · doi:10.1109/spawc.2017.8227772 · Cited by 429 · cited 1×
“Several researches have been proposed to utilize DNNs to substitute a certain module in the physical layer of a wireless transceiver system, such as data detection [9], [10], channel estimation [11]–[14], and channel decoding [15], [16], etc.”
MethodBayes-Optimal Joint Channel-and-Data Estimation for Massive MIMO With Low-Precision ADCs
Wen, C.-K., Wang, C.-J., Jin, S., Wong, K.-K., & Ting, P. (2016) · IEEE Transactions on Signal Processing · doi:10.1109/tsp.2015.2508786 · Cited by 360 · cited 1×
“For example, if more complex parametric generative models are considered, such as the hardware impairment of the receiver [45], Low-Precision ADCs [46], which bring some non-linearities into the underlying models, how to learn the model parameters under such cases.”
BackgroundNeural Network Detection of Data Sequences in Communication Systems
Farsad, N., & Goldsmith, A. (2018) · IEEE Transactions on Signal Processing · doi:10.1109/tsp.2018.2868322 · Cited by 358 · cited 1×
“Several researches have been proposed to utilize DNNs to substitute a certain module in the physical layer of a wireless transceiver system, such as data detection [9], [10], channel estimation [11]–[14], and channel decoding [15], [16], etc.”
MethodVariational Relevance Vector Machines
Bishop, C., & Tipping, M. E. (2013) · arXiv (Cornell University) · doi:10.48550/arxiv.1301.3838 · Cited by 352 · cited 1×
“In order to reduce the computational complexity, we alternatively to update parameters λ every time after one iteration in the VBE step, which will not effect the performance of the algorithm [43].”
BackgroundMao, B., Fadlullah, Z. M., Tang, F., Kato, N., Akashi, O., Inoue, T., & Mizutani, K. (2017) · IEEE Transactions on Computers · doi:10.1109/tc.2017.2709742 · Cited by 350 · cited 1×
“Due to the powerful mechanism, DL has made great progress in network traffic prediction [1], [2], data traffic classification [3], [4], network routing [5], network scheduling [6] and resource allocation [7], [8], etc.”
BackgroundChannel Estimation for Massive MIMO Using Gaussian-Mixture Bayesian Learning
Wen, C.-K., Jin, S., Wong, K.-K., Chen, J.-C., & Ting, P. (2015) · IEEE Transactions on Wireless Communications · doi:10.1109/twc.2014.2365813 · Cited by 310
Learning to optimize: Training deep neural networks for wireless resource management
Sun, H., Chen, X., Shi, Q., Hong, M., Fu, X., & Sidiropoulos, N. D. (2017) · 2017 IEEE 18th International Workshop on Signal Processing Advances in Wireless Communications (SPAWC) · doi:10.1109/spawc.2017.8227766 · Cited by 308 · cited 1×
“Due to the powerful mechanism, DL has made great progress in network traffic prediction [1], [2], data traffic classification [3], [4], network routing [5], network scheduling [6] and resource allocation [7], [8], etc.”
BackgroundLearning the MMSE Channel Estimator
Neumann, D., Wiese, T., & Utschick, W. (2018) · IEEE Transactions on Signal Processing · doi:10.1109/tsp.2018.2799164 · Cited by 306 · cited 3×
“Several researches have been proposed to utilize DNNs to substitute a certain module in the physical layer of a wireless transceiver system, such as data detection [9], [10], channel estimation [11]–[14], and channel decoding [15], [16], etc.”
MethodA Model-Driven Deep Learning Network for MIMO Detection
He, H., Wen, C.-K., Jin, S., & Li, G. Y. (2018) · 2018 IEEE Global Conference on Signal and Information Processing (GlobalSIP) · doi:10.1109/globalsip.2018.8646357 · Cited by 284
ComNet: Combination of Deep Learning and Expert Knowledge in OFDM Receivers
Gao, X., Jin, S., Wen, C.-K., & Li, G. Y. (2018) · IEEE Communications Letters · doi:10.1109/lcomm.2018.2877965 · Cited by 272
Robust Bayesian mixture modelling
Svensén, M., & Bishop, C. M. (2005) · Neurocomputing · doi:10.1016/j.neucom.2004.11.018 · Cited by 254 · cited 2×
“Based on the general framework of the variational inference in Chapter 10 of [32], [40], [41], we find that the variational inference framework can be used to simultaneously estimate the parameters in two categories, i.e., the random variables and the unknown quantities.”
BackgroundMethodChannel Agnostic End-to-End Learning Based Communication Systems with Conditional GAN
Ye, H., Li, G. Y., Juang, B.-H. F., & Sivanesan, K. (2018) · 2018 IEEE Globecom Workshops (GC Wkshps) · doi:10.1109/glocomw.2018.8644250 · Cited by 244
Gunduz, D., de Kerret, P., Sidiropoulos, N. D., Gesbert, D., Murthy, C. R., & van der Schaar, M. (2019) · IEEE Journal on Selected Areas in Communications · doi:10.1109/jsac.2019.2933969 · Cited by 214 · cited 1×
“On the contrary, we consider that we have no knowledge of the PDF of noise as well as its power spectral density, so that it is not feasible to directly establish an accurate noise model, which hinders the traditional model-based approaches [29].”
Scaling Deep Learning-Based Decoding of Polar Codes via Partitioning
Cammerer, S., Gruber, T., Hoydis, J., & ten Brink, S. (2017) · GLOBECOM 2017 - 2017 IEEE Global Communications Conference · doi:10.1109/glocom.2017.8254811 · Cited by 196 · cited 1×
“Several researches have been proposed to utilize DNNs to substitute a certain module in the physical layer of a wireless transceiver system, such as data detection [9], [10], channel estimation [11]–[14], and channel decoding [15], [16], etc.”
MethodVariational approximations in Bayesian model selection for finite mixture distributions
McGrory, C. A., & Titterington, D. M. (2007) · Computational Statistics & Data Analysis · doi:10.1016/j.csda.2006.07.020 · Cited by 169 · cited 1×
“Based on the general framework of the variational inference in Chapter 10 of [32], [40], [41], we find that the variational inference framework can be used to simultaneously estimate the parameters in two categories, i.”
MethodSparse Bayesian Learning for the Time-Varying Massive MIMO Channels: Acquisition and Tracking
Ma, J., Zhang, S., Li, H., Gao, F., & Jin, S. (2019) · IEEE Transactions on Communications · doi:10.1109/tcomm.2018.2855197 · Cited by 168 · cited 1×
“…critical issue for the Bayesian channel estimation and has been considered in multiple typical wireless communication scenarios, such as multi-cell massive MIMO channel [25], time-varying massive MIMO channel [39], group-clustering channel [23], based on the expectation-maximization (EM) algorithm.”
BackgroundXu, Z., & Sun, J. Z. (2017) · National Science Review · doi:10.1093/nsr/nwx099 · Cited by 143 · cited 3×
“The significant difference between our model refine-ment learning and its counterparts, i.e., data-based DL and model-driven DL [11]–[14], [19]–[22] is that in the training procedure, we refine the model by learning the model parameters which have specific physical meanings of underlying communication…”
BackgroundMethodNetwork Traffic Prediction Based on Deep Belief Network in Wireless Mesh Backbone Networks
Nie, L., Jiang, D., Yu, S., & Song, H. (2017) · IEEE Wireless Communications and Networking Conference · doi:10.1109/wcnc.2017.7925498 · Cited by 132 · cited 2×
“Due to the powerful mechanism, DL has made great progress in network traffic prediction [1], [2], data traffic classification [3], [4], network routing [5], network scheduling [6] and resource allocation [7], [8], etc.”
BackgroundBackpropagating Through the Air: Deep Learning at Physical Layer Without Channel Models
Raj, V., & Kalyani, S. (2018) · IEEE Communications Letters · doi:10.1109/lcomm.2018.2868103 · Cited by 131 · cited 1×
“One particularly thorough example is using a paired neural network called autoencoders to implement an end-to-end communication system without having to physically model, design, and explicitly implement an equalizer or an error control mechanism, based only on training data [17], [18].”
MethodPrasad, R., Murthy, C. R., & Rao, B. D. (2015) · IEEE Transactions on Signal Processing · doi:10.1109/tsp.2015.2451071 · Cited by 127 · cited 1×
“…Gaussian distribution, the derivations in the following sections can be directly extended into more complex wireless communication scenarios, such as the cases where the channel distributions are non-Gaussian [23], [25], [28] or the Gaussian channel is controlled by some hyper-parameters [26].”
BackgroundOverlapping Mixtures of Gaussian Processes for the data association problem
Lázaro-Gredilla, M., Vaerenbergh, S. V., & Lawrence, N. D. (2011) · Pattern Recognition · doi:10.1016/j.patcog.2011.10.004 · Cited by 111
Clustering Analysis in the Wireless Propagation Channel with a Variational Gaussian Mixture Model
Li, Y., Zhang, J., Ma, Z., & Zhang, Y. (2018) · IEEE Transactions on Big Data · doi:10.1109/tbdata.2018.2840696 · Cited by 82 · cited 5×
“In wireless communication, MoG distribution has been adopted for the channel modeling, such as in outdoor-to-indoor (O2I) multiple propagation channel [23], [33], multi-cell massive MIMO channel [25] and temporal correlated channel [24].”
BackgroundMethodInfluentialDemir, Ö. T., & Björnson, E. (2019) · IEEE Open Journal of the Communications Society · doi:10.1109/ojcoms.2019.2959913 · Cited by 54 · cited 1×
“For example, if more complex parametric generative models are considered, such as the hardware impairment of the receiver [45], Low-Precision ADCs [46], which bring some non-linearities into the underlying models, how to learn the model parameters under such cases.”
BackgroundWei, Y., Zhang, Z., Yu, F. R., & Han, Z. (2018) · IEEE International Conference on Communications workshops · doi:10.1109/iccw.2018.8403711 · Cited by 44 · cited 1×
“Due to the powerful mechanism, DL has made great progress in network traffic prediction [1], [2], data traffic classification [3], [4], network routing [5], network scheduling [6] and resource allocation [7], [8], etc.”
BackgroundA Neural Network for Quality of Experience Estimation in Mobile Communications
Pierucci, L., & Micheli, D. (2016) · IEEE Multimedia · doi:10.1109/mmul.2016.21 · Cited by 35
Low-Complexity Bayesian Estimation of Cluster-Sparse Channels
Ballal, T., Al-Naffouri, T. Y., & Ahmed, S. F. (2015) · IEEE Transactions on Communications · doi:10.1109/tcomm.2015.2480092 · Cited by 18 · cited 1×
“…Gaussian distribution, the derivations in the following sections can be directly extended into more complex wireless communication scenarios, such as the cases where the channel distributions are non-Gaussian [23], [25], [28] or the Gaussian channel is controlled by some hyper-parameters [26].”
BackgroundNonlinear MMSE Equalizer for Impulsive Noise Mitigation in OFDM-Based Communications
Soltanpur, C., Paravi, R., Ghamari, M., & Adebisi, B. (2019) · IEEE Signal Processing Letters · doi:10.1109/lsp.2019.2917018 · Cited by 15 · cited 1×
“Specifi-cally, MoG distribution provides accurate approximation of the Middleton Class A noise model [34], which is widely used to measure the impulse noise in wireless communication systems [35], [36] with the Gaussian-to-impulse noise power ratio controlled by the MoG parameters.”
MethodGaussian Mixture Based Context-Aware Short-Term Characterization of Wireless Channels
Mukherjee, P., Mishra, D., & De, S. K. (2019) · IEEE Transactions on Vehicular Technology · doi:10.1109/tvt.2019.2955512 · Cited by 11 · cited 2×
“channel [25] and temporal correlated channel [24].”
BackgroundOptimal Power Allocation for Parallel Two-State Gaussian Mixture Impulse Noise Channels
Axell, E., Wiklundh, K., & Stenumgaard, P. F. (2015) · IEEE Wireless Communications Letters · doi:10.1109/lwc.2015.2388759 · Cited by 10 · cited 1×
“Specifi-cally, MoG distribution provides accurate approximation of the Middleton Class A noise model [34], which is widely used to measure the impulse noise in wireless communication systems [35], [36] with the Gaussian-to-impulse noise power ratio controlled by the MoG parameters.”
MethodA novel method to obtain CSI based on Gaussian Mixture Model and Expectation Maximization
Li, H., Li, Y., Zhou, S., & Wang, J. (2016) · International Conference on Wireless Communications and Signal Processing (WCSP) · doi:10.1109/wcsp.2016.7752732 · Cited by 10 · cited 3×
“In wireless communication, MoG distribution has been adopted for the channel modeling, such as in outdoor-to-indoor (O2I) multiple propagation channel [23], [33], multi-cell massive MIMO channel [25] and temporal correlated channel [24].”
BackgroundMethodRobust CSI Estimation Under Complex Communication Environment
Zhang, H., Xue, J., Meng, D., Zhao, Q., & Xu, Z. (2019) · International Conference on Communications · doi:10.1109/icc.2019.8761122 · Cited by 5 · cited 2×
“MoG distribution can also be used to measure the noise distributions in complex wireless communication environments and network structures, where the noise distributions are highly non-Gaussian [27], [37].”
BackgroundDetection for OFDM-AF Relay Systems Under Gaussian Mixture Noise With Imperfect CSI
Zhang, J., Li, F., & Chen, J. (2020) · IEEE Transactions on Vehicular Technology · doi:10.1109/tvt.2020.2980140 · Cited by 4 · cited 1×
“MoG distribution can also be used to measure the noise distributions in complex wireless communication environments and network structures, where the noise distributions are highly non-Gaussian [27], [37].”
Cited by
Sadeghi, A., Bellavista, P., Song, W., & Yazdani-Asrami, M. (2024) · IEEE Access · doi:10.1109/access.2024.3371902 · Cited by 100
“…control and reconfiguration commands in the proximity of their actuators; in addition, distributed DT on industrial gateways or MEC nodes can continue the training/learning process also locally, via emerging machine learning techniques such as refinement learning [211] and federated learning [212].”
Design and Implementation of 5G e-Health Systems: Technologies, Use Cases, and Future Challenges
Zhang, D., Rodrigues, J. J. P. C., Zhai, Y., & Sato, T. (2021) · IEEE Communications Magazine · doi:10.1109/mcom.001.2100035 · Cited by 38
Deep Learning Based Efficient Symbol-Level Precoding Design for MU-MISO Systems
Bo, Z., Liu, R., Li, M., & Liu, Q. (2021) · IEEE Transactions on Vehicular Technology · doi:10.1109/tvt.2021.3093079 · Cited by 21
“The authors in [8], [9] exploited learning-based schemes for more accurate channel estimation.”
BackgroundGrouping-Based Joint Active User Detection and Channel Estimation With Massive MIMO
Jiang, J., & Wang, H. (2021) · IEEE Transactions on Wireless Communications · doi:10.1109/twc.2021.3111104 · Cited by 15
“On the other hand, to deal with the liner mixing of sparse vector, some approximations are considered, such as AMP approximation [17], mean field (MF) approximation [33], [42].”
BackgroundSelf-citationRobust 3-D AOA Localization Against Malicious Attacks in Non-Gaussian Noise
Yan, Q., Wang, H., Chen, Y., & Gao, C. (2024) · IEEE Sensors Journal · doi:10.1109/jsen.2024.3373649 · Cited by 12
“With the demand of autonomous robots and vehicle of the Internet of Thing (IoT) in the forthcoming sixth-generation (6G) wireless networks, the localization accuracy faces new challenges under malicious attacks in complex signal propagation environment with channel variations [1], [2], [3], [4].”
BackgroundSelf-citation
Metadata
- Authors
- Wang, H.-M., Jiang, J.-C., & Wang, Y.-N.
- Affiliations
- Xi'an Jiaotong University (CN)
- ORCID
- Wang, Hui-Ming
- Venue
- IEEE Journal on Selected Areas in Communications
- Published
- 2021-01
- Volume
- 39
- Issue
- 1
- Pages
- 31-46
- ISSN
- 0733-8716 (ISSN-L)
- 1558-0008
- OpenAlex
- W3100624467
- Cited by
- 14
- References
- 51
- Citations by year
- 2026: 1 · 2024: 3 · 2023: 2 · 2022: 2 · 2021: 6
- FWCI
- 0.86
- Type
- journal article
- Language
- en
- Open access
- closed
- Locations
- IEEE Journal on Selected Areas in Communications (publishedVersion)
- Funders
- National Natural Science Foundation of China (61941118, 61941105)
- National Key Research and Development Program of China (2019YFE0113200)
- Innovation Team Research Fund of Shaanxi Province (2019TD-013)
- Topics
- Physical Sciences › Engineering › Electrical and Electronic Engineering › Advanced Wireless Communication Techniques
- Physical Sciences › Computer Science › Computer Networks and Communications › Wireless Communication Networks Research
- Physical Sciences › Engineering › Control and Systems Engineering › Control Systems and Identification
- Keywords
- channel estimation, variational inference, MISO systems, wireless communication system design
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.