IP Library › Granted Patent US 12,748,956
Granted Patent B2
US 12,748,956 · App. 18/333,063 · Granted Sep 29, 2026

Enhancing the quality of simulated network data using generative adversarial networks

Inventors: Ajit Krishna Patankar (Fremont, CA); Aman Gaurav (Lucknow, IN); Harshavardhan V S Choudary Battula (Saint Paul, MN); Yash Verma (Agra, IN)
Assignee: Hewlett Packard Enterprise Development LP
G06N3/0475G06N3/044G06N3/045G06N3/094
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 12,748,956
App. No.
18/333,063
Granted
Sep 29, 2026
Kind
B2
Abstract

A device may receive real network data associated with a network, and may receive a random latent vector and a random process sample. The device may utilize the random latent vector with a generative adversarial network (GAN) model to generate synthetic network data, and may train the GAN model with the real network data and the synthetic network data to generate a trained GAN model. The device may utilize the random process sample with a random process to generate simulated network data, and may apply weights to the real network data, the synthetic network data, and the simulated network data to generate weighted real network data, weighted synthetic network data, and weighted simulated network data. The device may combine the weighted real network data, the weighted synthetic network data, and the weighted simulated network data to generate interpolated network data, and may perform actions based on the interpolated network data.

Claims (75)

1 . A method, comprising:

receiving, by a device, real network data associated with a network;

receiving, by the device, a random latent vector and a random process sample;

utilizing, by the device, the random latent vector with a generative adversarial network (GAN) model to generate synthetic network data;

training, by the device, the GAN model with the real network data and the synthetic network data to generate a trained GAN model;

utilizing, by the device, the random process sample with a random process to generate simulated network data,

wherein the simulated data network includes network traffic data at one frequency;

applying, by the device and based on the trained GAN model, weights to the real network data, the synthetic network data, and the simulated network data to generate weighted real network data, weighted synthetic network data, and weighted simulated network data;

combining, by the device, the weighted real network data, the weighted synthetic network data, and the weighted simulated network data to generate interpolated network data; and

performing, by the device, one or more actions based on the interpolated network data.

2 . The method of claim 1 , wherein the real network data includes a multivariate dataset.

3 . The method of claim 1 , wherein the GAN model is a Wasserstein recurrent GAN model.

4 . The method of claim 1 , wherein training the GAN model with the real network data and the synthetic network data to generate the trained GAN model comprises:

training the GAN model with the real network data and the synthetic network data to generate the weights.

5 . The method of claim 1 , wherein applying the weights to the real network data, the synthetic network data, and the simulated network data to generate the weighted real network data, the weighted synthetic network data, and the weighted simulated network data comprises:

applying a first weight to the real network data to generate the weighted real network data;

applying a second weight to the synthetic network data to generate the weighted synthetic network data; and

applying a third weight to the simulated network data to generate the weighted simulated network data,

wherein a sum of the first weight, the second weight, and the third weight is equal to one.

6 . The method of claim 5 , wherein a value of the first weight determines a masked quantity of the real network data.

7 . The method of claim 1 , wherein utilizing the random process sample with the random process to generate the simulated network data comprises:

utilizing the random process sample and the real network data to generate two Poisson distributions; and

superposing the two Poisson distributions to generate the simulated network data.

8 . A device, comprising:

one or more memories; and

one or more processors to:

receive real network data associated with a network,

wherein the real network data include a multivariate dataset,

wherein the multivariate dataset set includes multivariate temporal behavior of network traffic throughput;

receive a random latent vector and a random process sample;

utilize the random latent vector with a generative adversarial network (GAN) model to generate synthetic network data;

train the GAN model with the real network data and the synthetic network data to generate a trained GAN model;

utilize the random process sample with a random process to generate simulated network data;

apply weights to the real network data, the synthetic network data, and the simulated network data to generate weighted real network data, weighted synthetic network data, and weighted simulated network data;

combine the weighted real network data, the weighted synthetic network data, and the weighted simulated network data to generate interpolated network data; and

perform one or more actions based on the interpolated network data.

9 . The device of claim 8 , wherein the one or more processors, to utilize the random process sample with the random process to generate the simulated network data, are to:

utilize the random process sample and the real network data to generate two Poisson distributions; and

superpose the two Poisson distributions to generate the simulated network data.

10 . The device of claim 8 , wherein the GAN model includes a generator component and a discriminator component.

11 . The device of claim 8 , wherein the one or more processors, to perform the one or more actions, are to one or more of:

provide the interpolated network data for display; or

retrain the GAN model based on the interpolated network data.

12 . The device of claim 8 , wherein the one or more processors, to perform the one or more actions, are to one or more of:

train a network anomaly detection model with the interpolated network data; or

train a network forecasting model with the interpolated network data.

13 . The device of claim 8 , wherein the one or more processors, to perform the one or more actions, are to:

perform initial training of a network anomaly detection model with the interpolated network data; and

perform fine tune training of the network anomaly detection model with the real network data.

14 . The device of claim 8 , wherein the one or more processors, to perform the one or more actions, are to:

deploy a network forecasting model or a network anomaly detection model, trained with the interpolated network data, in the network.

15 . A non-transitory computer-readable medium storing a set of instructions, the set of instructions comprising:

one or more instructions that, when executed by one or more processors of a device, cause the device to:

receive real network data associated with a network;

receive a random latent vector and a random process sample;

utilize the random latent vector with a generative adversarial network (GAN) model to generate synthetic network data,

wherein the GAN model is a Wasserstein recurrent GAN model;

train the GAN model with the real network data and the synthetic network data to generate a trained GAN model;

utilize the random process sample with a random process to generate simulated network data,

wherein the simulated data network includes network traffic data at one frequency;

apply, based on the trained GAN, weights to the real network data, the synthetic network data, and the simulated network data to generate weighted real network data, weighted synthetic network data, and weighted simulated network data;

combine the weighted real network data, the weighted synthetic network data, and the weighted simulated network data to generate interpolated network data; and

perform one or more actions based on the interpolated network data.

16 . The non-transitory computer-readable medium of claim 15 , wherein the one or more instructions, that cause the device to train the GAN model with the real network data and the synthetic network data to generate the trained GAN model, cause the device to:

train the GAN model with the real network data and the synthetic network data to generate the weights.

17 . The non-transitory computer-readable medium of claim 15 , wherein the one or more instructions, that cause the device to apply the weights to the real network data, the synthetic network data, and the simulated network data to generate the weighted real network data, the weighted synthetic network data, and the weighted simulated network data, cause the device to:

apply a first weight to the real network data to generate the weighted real network data;

apply a second weight to the synthetic network data to generate the weighted synthetic network data; and

apply a third weight to the simulated network data to generate the weighted simulated network data,

wherein a sum of the first weight, the second weight, and the third weight is equal to one.

18 . The non-transitory computer-readable medium of claim 17 , wherein a value of the first weight determines a masked quantity of the real network data.

19 . The non-transitory computer-readable medium of claim 15 , wherein the one or more instructions, that cause the device to utilize the random process sample with the random process to generate the simulated network data, cause the device to:

utilize the random process sample and the real network data to generate two Poisson distributions; and

superpose the two Poisson distributions to generate the simulated network data.

20 . The non-transitory computer-readable medium of claim 15 , wherein the GAN model includes a generator component and a discriminator component.

Assignments (1)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Jun 12, 2023
From: PATANKAR, AJIT KRISHNA; GAURAV, AMAN; BATTULA, HARSHAVARDHAN V S CHOUDARY; VERMA, YASH
To: JUNIPER NETWORKS, INC.
Reel/Frame 063923/0918 →
Continuity (1)
Related Publication 20240412047A1 · Dec 12, 2024
References Cited (27)
US 10702169B2 · Gaurav et al. · 2020 [cited by applicant]
US 11893056B2 · Kontkanen · 2024 [cited by examiner]
US 20190139641A1 · Itu · 2019 [cited by examiner]
US 20200311482A1 · Soni · 2020 [cited by examiner]
US 20210037044A1 · Achanta et al. · 2021 [cited by applicant]
US 20220172050A1 · Dalli · 2022 [cited by examiner]
US 20230155704A1 · Orekondy et al. · 2023 [cited by applicant]
US 20240119199A1 · Balasubramanian · 2024 [cited by examiner]
US 20250078216A1 · Van Aert · 2025 [cited by examiner]
EP 3982298A1 · 2022 [cited by applicant]
Extended European Search Report for Application No. EP23190299 mailed on Feb. 2, 2024, 11 pages. [cited by applicant]
Arjovsky M. et al., Wasserstein Generative Adversarial Networks, ICML'17: Proceedings of the 34th International Conference on Machine Learning, Aug. 6, 2017, 214-223 pages, vol. 70, ACM. [cited by applicant]
Bengio et al., Scheduled Sampling for Sequence Prediction with Recurrent Neural Networks, NIPS'15: Proceedings of the 28th International Conference on Neural Information Processing Systems, Dec. 7, 2015, 1171-1179 pages… [cited by applicant]
Hyland et al., Real-Valued (Medical) Time Series Generation with Recurrent Conditional GANS, CoRR, Jun. 8, 2017, 01-11 Pages, University of Stuttgart, Zurich, Switzerland. [cited by applicant]
Gaurav A. et al., Cuff-Less ppg Based Continuous Blood Pressure Monitoring—A Smartphone Based Approach, 38th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), Aug. 1, 2016, … [cited by applicant]
Goodfellow I. et al., Generative Adversarial Networks, NIPS'14: Proceedings of the 27th International Conference on Neural Information Processing Systems, Dec. 8, 2014, 2672-2680 pages, vol. 2, ACM, Montreal, Canada. [cited by applicant]
Gulrajani I. et al., Improved Training of Wasserstein Gans, NIPS'17: Proceedings of the 31st International Conference on Neural Information Processing Systems, Dec. 4, 2017, 5769-5779 pages, ACM. [cited by applicant]
Lin Z. et al., Using Gans for Sharing Networked Time Series Data: Challenges, Initial Promise, And Open Questions., IMC '20: ACM Internet Measurement Conference, Oct. 27, 2020, 464-483 pages, vol. 70, ACM, Virtual Event… [cited by applicant]
Maheedhar M. et al., Stayfit: A Wearable Application for Gym Based Power Training, 38th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), Aug. 16-20, 2016, 6290-6293 pages, … [cited by applicant]
Mogren et al., C-RNN-GAN:Continuous Recurrent Neural Networks With Adversarial Training, Constructive Machine Learning Workshop (CML) at NIPS 2016, Nov. 29, 2016, Barcelona, Sweden. [cited by applicant]
Pei H. et al., Towards Generating Real-World Time Series Data, 21th IEEE International Conference on Data Mining (ICDM 2021), Dec. 7-10, 2021, 469-478 pages, IEEE. [cited by applicant]
Smith et al., Conditional Gan for Timeseries Generation, Jul. 1, 2020. [cited by applicant]
Smith et al., Time Series Generation Using a One Dimensional Wasserstein Gan, ITISE 2019: International Conference on Time Series and Forecasting, Sep. 25-27, 2019, 771-781 pages, vol. 52, Issue No. 22, Granada, Spain. [cited by applicant]
Van den Oord A. et al., Pixel Recurrent Neural Networks, ICML'16: Proceedings of the 33rd International Conference on International Conference on Machine Learning, Jun. 19, 2016, 1747-1756 pages, vol. 48, ACM. [cited by applicant]
Wiese M. et al., Quant GANs: Deep Generation of Financial Time Series, Quantitative Finance, Apr. 6, 2020, 1419-1440 pages, vol. 20, Issue No. 9, Taylor & Francis Online. [cited by applicant]
Xu T. et al., Cot-Gan: Generating Sequential Data via Causal Optimal Transport, NIPS'20: Proceedings of the 34th International Conference on Neural Information Processing Systems, Dec. 6, 2020, 8798-8809 pages, Article … [cited by applicant]
Yoon J. et al., Time-Series Generative Adversarial Networks, NIPS'19: Proceedings of the 33rd International Conference on Neural Information Processing Systems, Dec. 8, 2019, 5508-5518 pages, Article No. 494, ACM, Los A… [cited by applicant]