IP Library › Granted Patent US 12,481,862
Granted Patent B2
US 12,481,862 · App. 17/453,258 · Granted Nov 25, 2025

Deep learning for rain fade prediction in satellite communications

Inventors: Aidin Ferdowsi Khosrowshahi (Germantown, MD); David Whitefield (Germantown, MD); Rob Torres (New Market, MD)
Assignee: Hughes Network Systems, LLC
G06N3/04H04B7/18513H04L41/147H04W4/021
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 12,481,862
App. No.
17/453,258
Granted
Nov 25, 2025
Kind
B2
Abstract

A system and method for predicting rain fade. The method may include: training a Neural Network (NN) by importing into the NN a training set of image information and beacon information, wherein the image information includes image datasets including of a cloud view of an Area of Interest (AoI), a geolocation and a timestamp, and the beacon information includes beacon datasets including a beacon strength, a current rain fade state, a geolocation and a timestamp; pre-processing to homogenize and to extract spatially and temporally matching data for the AoI from a live image information and a live beacon information; and forecasting a rain fade based on the data in a near-future. The geolocation of the beacon datasets is located within the AoI, and the periodicity of the live beacon information and the live image information is less than or equal to five (5) minutes.

Claims (39)

1 . A method for predicting rain fade for a rain zone using a deep learning system comprising a computer processor, the method comprising:

training a Neural Network (NN) by importing into the NN a training set of image information and beacon information, wherein the image information comprises image datasets comprising of a cloud view of an Area of Interest (AoI), a geolocation and a timestamp, and the beacon information comprises beacon datasets comprising a beacon strength, a current rain fade state, a geolocation and a timestamp;

receiving, at a receiver, a live beacon;

pre-processing to homogenize and to extract spatially and temporally matching data for the AoI from a live image information and a live beacon information comprising power measurements of the live beacon at the receiver; and

forecasting a rain fade based on the data in a near-future,

wherein the geolocation of one or more of the beacon datasets is located within the AoI,

a beacon periodicity of the live beacon information is less than or equal to five (5) minutes,

the pre-processing flattens a portion of the live image information from 3-dimensional or 2-dimensional information to 1-dimensional information, and

an image periodicity of the live image information is less than or equal to five (5) minutes.

2 . The method of claim 1 , wherein the near-future is less than or equal to sixty-five (65) minutes.

3 . The method of claim 1 , wherein the image periodicity is different than the beacon periodicity, and the method further comprises using a previous copy of the live beacon information or the live image information.

4 . The method of claim 1 , wherein the image periodicity is different than the beacon periodicity, and the method further comprises skipping a previous copy of the live beacon information or the live image information as necessary for the matching.

5 . The method of claim 1 , wherein the beacon information is collected at a satellite transceiver and a geolocation of the satellite transceiver is located within the AoI.

6 . The method of claim 1 , wherein the cloud view comprises a top-view from a satellite of the AoI or a bottom view from a radar of the AoI or a combination thereof.

7 . The method of claim 1 , wherein the live image information comprises a radar image of the AoI and a ground truth for the AoI.

8 . The method of claim 7 , wherein the ground truth comprises a rain label and the pre-processing harmonizes the rain labels with current rain fade states of the beacon information.

9 . The method of claim 1 , wherein the live image information comprises an image of the AoI from a high-altitude platform or satellite and the image comprises images at various spectra.

10 . The method of claim 1 , wherein the pre-processing harmonizes the live image information to an image resolution.

11 . The method of claim 1 , wherein the pre-processing harmonizes a coordinate system of the live image information and the live beacon information.

12 . The method of claim 1 , wherein the training set balances a quantity of clear sky events as compared to a quantity of rain fade events.

13 . The method of claim 1 , wherein the NN processes the data using a 3D convolution neural network.

14 . The method of claim 1 , wherein the NN successively processes the data using a 3D convolution NN, a max pool, a flattening NN and a softmax NN.

15 . The method of claim 1 , wherein the AoI covers a ground area of at least 32 km×32 km.

16 . The method of claim 1 , wherein the AoI is centered over the geolocation of one or more of the beacon datasets.

17 . The method of claim 1 , wherein the AoI comprises a plurality of AoI, the plurality of AoI are located within in a rain zone and the evaluating predicts the rain fade for the plurality of AoI.

18 . The method of claim 1 , further comprising proactively managing gateway diversity based on the forecasting.

19 . A method for predicting rain fade for a rain zone using a deep learning system comprising a computer processor, the method comprising:

training a Neural Network (NN) by importing into the NN a training set of image information and beacon information, wherein the image information comprises image datasets comprising of a cloud view of an Area of Interest (AoI), a geolocation and a timestamp, and the beacon information comprises beacon datasets comprising a beacon strength, a current rain fade state, a geolocation and a timestamp;

receiving, at a receiver, a live beacon;

pre-processing to homogenize and to extract spatially and temporally matching data for the AoI from a live image information and a live beacon information comprising power measurements of the live beacon at the receiver; and

pre-processing to homogenize and to extract spatially and temporally matching data for the AoI from a live image information and the live beacon information; and

forecasting a rain fade based on the data in a near-future,

wherein the geolocation of one or more of the beacon datasets is located within the AoI,

the near-future is less than or equal to sixty-five (65) minutes,

the beacon information is collected at a satellite transceiver and a geolocation of the satellite transceiver is located within the AoI,

the live image information comprises an image of the AoI from a satellite, a radar image of the AoI and a ground truth for the AoI,

the pre-processing flattens a portion of the live image information from 3-dimensional or 2-dimensional information to 1-dimensional information, and

the NN processes the data using a 3D convolution neural network.

20 . The method of claim 19 , wherein the AoI comprises a plurality of AoI, the plurality of AoI are located within a rain zone and the evaluating predicts the rain fade for the plurality of AoI.

Assignments (2)
SECURITY INTEREST Recorded Feb 10, 2022
From: HUGHES NETWORK SYSTEMS, LLC
To: U.S. BANK GLOBAL CORPORATE TRUST WEST SIDE FLATS, ST. PAUL
Reel/Frame 058971/0319 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Nov 2, 2021
From: KHOSROWSHAHI, AIDEN FERDOWSI; WHITEFIELD, DAVID; TORRES, ROB
To: HUGHES NETWORK SYSTEMS, LLC
Reel/Frame 057995/0749 →
Continuity (2)
Provisional Application 63203351 · Jul 19, 2021
Related Publication 20230019771A1 · Jan 19, 2023
References Cited (37)
US 20200242733A1 · Kumar et al. · 2020 [cited by examiner]
US 20210058293A1 · Whitefield et al. · 2021 [cited by examiner]
Sonderby et al, MetNet: A Neural Weather Model for Precipitation Forecasting, Mar. 30, 2020 (Year: 2020). [cited by examiner]
Chen et al, A Machine Learning System for Precipitation Estimation Using Satellite and Ground Radar Network Observations, Feb. 2020 (Year: 2020). [cited by examiner]
Dahman et al, Rain attenuation prediction model for satellite communications based on the Météo-France ensemble prediction system PEARP, Dec. 18, 2018 (Year: 2018). [cited by examiner]
Higashiyama et al, Feature Extraction of NWP Data for Wind Power Forecasting Using 3D-Convolutional Neural Networks, Nov. 11, 2018 (Year: 2018). [cited by examiner]
Liu, Daytime Rainy Cloud Detection and Convective Precipitation Delineation Based on a Deep Neural Network Method Using GOES-16 ABI Images, Oct. 30, 2019 (Year: 2019). [cited by examiner]
Mishra et al, Deep Rainrate Estimation from Highly Attenuated Downlink Signals of Ground-Based Communications Satellite Terminals, May 4, 2020 (Year: 2020). [cited by examiner]
Nielsen et al, CloudCast: A Satellite-Based Dataset and Baseline for Forecasting Clouds, Mar. 2, 2021 (Year: 2021). [cited by examiner]
Roy et al, Attenuation prediction for fade mitigation using neural network with in situ learning algorithm, Jan. 5, 2011 (Year: 2011). [cited by examiner]
“Nowcasting the Next Hour of Rain”, Sep. 29, 2021, printed on Nov. 2, 2021 from https://deepmind.com/blog/article/nowcasting. [cited by applicant]
A. H. Nielsen, A. Iosifidis, and H. Karstoft, “Cloudcast: A satellite-based dataset and baseline for forecasting clouds,” IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, vol. 14, pp. 34… [cited by applicant]
B. Roy, R. Acharya, and M. Sivaraman, “Attenuation prediction for fade mitigation using neural network with in situ learning algorithm,” Advances in Space Research, vol. 49, No. 2, pp. 336-350, 2012. [cited by applicant]
B. Welford, “Note on a method for calculating corrected sums of squares and products,” Technometrics, vol. 4, No. 3, pp. 419-420, 1962. [cited by applicant]
C. K. Sønderby, L. Espeholt, J. Heek, M. Dehghani, A. Oliver, T. Salimans, S. Agrawal, J. Hickey, and N. Kalchbrenner, “Metnet:A neural weather model for precipitation forecasting,” arXiv preprint arXiv:2003.12140, 2020. [cited by applicant]
C.-S. Lu, Z.-W. Zhao, Z.-S. Wu, L.-K. Lin, P. Thiennviboon, X. Zhang, and Z.-F. Lv, “A new rain attenuation prediction model for the earthspace links,” IEEE Transactions on Antennas and Propagation, vol. 66, No. 10, pp.… [cited by applicant]
D. Das and A. Maitra, “Time series prediction of rain attenuation from rain rate measurement using synthetic storm technique for a tropical location,” AEU—International Journal of Electronics and Communications, vol. 68… [cited by applicant]
E. Matricciani, “Global formulation of the synthetic storm technique to calculate rain attenuation only from rain rate probability distributions,” in 2008 IEEE Antennas and Propagation Society International Symposium, 2… [cited by applicant]
FCC, “Fixed satellite service by hughes network systems, llc,” 2016. [Online]. Available: https://fcc.report/IBFS/SAT-LOA-20160624-00061. [cited by applicant]
FCC, “Fixed satellite service by hughes network systems, llc,” 2017. [Online]. Available: https://fcc.report/IBFS/SAT-LOA-20170621-00092. [cited by applicant]
H. Dao, M. R. Islam, and K. A. S. Al-Khateeb, “Rain fade slope model in satellite path based on data measured in heavy rain zone,” IEEE Antennas and Wireless Propagation Letters, vol. 12, pp. 50-53, 2013. [cited by applicant]
HughesNet, “Echostar xix,” 2017. [Online]. Available: https://www.hughes.com/technologies/hughes-high-throughput-satellite-constellation/echostar-xix. [cited by applicant]
I. Dahman, P. Arbogast, N. Jeannin, and B. Benammar, “Rain attenuation prediction model for satellite communications based on the meteo-france ensemble prediction system pearp,” Natural Hazards and Earth System Sciences… [cited by applicant]
J. X. Yeo, Y. H. Lee, and J. T. Ong, “Rain attenuation prediction model for satellite communications in tropical regions,” IEEE Transactions on Antennas and Propagation, vol. 62, No. 11, pp. 5775-5781, 2014. [cited by applicant]
L. A. R. d. S. Mello and M. S. Pontes, “Improved unified method for the prediction of rain attenuation in terrestrial and earth space links,” in 2009 Sbmo/IEEE MTT-S International Microwave and Optoelectronics Conferenc… [cited by applicant]
M. A. Samad, and D.-Y. Choi, “Learning-Assisted Rain Attenuation Prediction Model,” Appl. Sci. Oct. 2020, 6017; doi:10.3390/app10176017. [cited by applicant]
M. A. Samad, F. D. Diba, and D.-Y. Choi, “A survey of rain fade models for earth-space telecommunication links—taxonomy, methods, and comparative study,” Remote Sensing, vol. 13, No. 10, p. 1965, 2021. [cited by applicant]
M. N. Ahuna, T. J. Afullo, and A. A. Alonge, “Rain attenuation prediction using artificial neural network for dynamic rain jade mitigation,” SAIEE Africa Research Journal, vol. 110, No. 1, pp. 11-18, 2019. [cited by applicant]
P. Series, “Propagation data and prediction methods required for the design of earth-space telecommunication systems,” Recommendation ITU-R, pp. 618-12, 2015. [cited by applicant]
Q. W. Pan, J. E. Allnutt, and C. Tsui, “Evaluation of diversity and power control techniques for satellite communication systems in tropical and equatorial rain climates,” IEEE Transactions on Antennas and Propagation, … [cited by applicant]
R. Prudden, S. Adams, D. Kangin, N. Robinson, S. Ravuri, S. Mohamed, and A. Arribas, “A review of radar-based nowcasting of precipitation and applicable machine learning techniques,” arXiv preprint arXiv:2005.04988, 202… [cited by applicant]
R. T. Union, “Prediction method of fade dynamics on earth-space paths,” Recommendation ITU-R, 2005. [cited by applicant]
S. Gong, Y. Gao, H. Shi, and G. Zhao, “A practical mga-arima model for forecasting real-time dynamic rain-induced attenuation,” Radio Science, vol. 48, No. 3, pp. 208-225, 2013. [cited by applicant]
S. N. Livieratos and P. G. Cottis, “Rain attenuation along terrestrial millimeter wave links: A new prediction method based on supervised machine learning,” IEEE Access, vol. 7, pp. 138 745-138 756, 2019. [cited by applicant]
Chen Haonan et al: “A Machine Learning System for Precipitation Estimation Using Satellite and Ground Radar Network Observations”, IEEE Transactions on Geoscience and Remote Sensing, IEEE, USA, vol. 58, No. 2, Feb. 1, 2… [cited by applicant]
International Search Report for PCT Application No. PCT/US2022/073767. [cited by applicant]
Mishra Kumar Vijay et al: “Deep Rainrate Estimation from Highly Attenuated Downlink Signals of Ground-Based Communications Satellite Terminals”, ICASSP 2020—2020 IEEE International Conference on Acoustics, Speech and Si… [cited by applicant]