Successfully reported this slideshow.
We use your LinkedIn profile and activity data to personalize ads and to show you more relevant ads. You can change your ad preferences anytime.

Mitigating GNSS jamming and spoofing using ML and AI

151 views

Published on

Nir Laufer’s ITSF presentation discussed the threat of GNSS vulnerability to jamming and spoofing and how systems based on machine learning using artificial intelligence could provide the answer.

Published in: Technology
  • Be the first to comment

  • Be the first to like this

Mitigating GNSS jamming and spoofing using ML and AI

  1. 1. ITSF, Nov 7, 2019 Centralized GNSS monitoring and assurance Mitigating GNSS jamming and spoofing utilizing AI and ML Nir Laufer, senior director, product line management, Oscilloquartz
  2. 2. © 2019 ADVA Optical Networking. All rights reserved.22 GNSS vulnerability – known issues
  3. 3. © 2019 ADVA Optical Networking. All rights reserved.33 Identifying the scale of impact can help us classify the cause Am I the only one affected? Cause Impact Segment error Global/regional Adjacent band transmissions Local/regional Spoofing Local/regional Jamming Local/regional Environmental Local/regional Equipment failure Local
  4. 4. © 2019 ADVA Optical Networking. All rights reserved.44 • Each GNSS receiver is working independently • Detection of GNSS outages are done locally • Disadvantages and weaknesses: • Limited/no data from external sources • Decision can’t be made based on the “bigger picture” • Limited memory and compute power Limited and sub-optimal detection and decision making Existing, distributed local (limited) approach
  5. 5. © 2019 ADVA Optical Networking. All rights reserved.55 The centralized-global approch • Relevant data is collected from all the GNSS receivers in the network • External data is collected from independent sources • Advantages and strengths: • Multiple sources of relevant data are available for analysis • Decision is based on the “bigger picture” • Unlimited memory and compute power • Utilizes artificial intelligence (AI) and machine learning (ML) • Can be done remotely – no dedicated HW is needed on site Optimal detection and decision making Can we do better?
  6. 6. © 2019 ADVA Optical Networking. All rights reserved.66 Centralized GNSS monitoring and assurance GNSS receiver GNSS receiver GNSS receiver GNSS receiver GNSS monitoring and assurance NASA servers Weather
  7. 7. © 2019 ADVA Optical Networking. All rights reserved.77 Centralized GNSS monitoring and assurance GNSS receiver GNSS receiver GNSS receiver GNSS receiver GNSS monitoring and assurance NASA servers Weather Use ML and AI Instructions are sent back to the relevant devices (e.g., go to holdover or switch to backup)
  8. 8. © 2019 ADVA Optical Networking. All rights reserved.88 The following data is available from most of the commercial GNSS receivers via API It can be collected remotely over secured interfaces (e.g., CLI-SSH/SNMPv3) • Location: • Latitude, longitude, altitude • Satellites-related data • SV, carrier to noise, health, azimuth, elevation, AGC What data can we get for GNSS receivers?
  9. 9. © 2019 ADVA Optical Networking. All rights reserved.99 • GNSS-related data: • Satellite ephemerides • GNSS broadcast ephemeris files • GNSS maintenance-related info: • Satellite decommission • Satellite health • New satellite • Special events (e.g., leap second and GPS rollover) • Weather and space weather data : • Atmospheric parameters • Tropospheric path delay • Ionospheric conditions Better decision making based on comprehensive, meaningful data What external data is available online?
  10. 10. © 2019 ADVA Optical Networking. All rights reserved.1010 GNSS health – live network summary status Help identify regional and global outages (such as regional jamming or segment error)
  11. 11. © 2019 ADVA Optical Networking. All rights reserved.1111 GNSS health – network historical status Help identify periodic patterns which affect entire network or subnetwork
  12. 12. © 2019 ADVA Optical Networking. All rights reserved.1212 C/N spectrograms (heatmap)
  13. 13. © 2019 ADVA Optical Networking. All rights reserved.1313 Things might not stay as they were … How can we automatically detect changes in site conditions?
  14. 14. © 2019 ADVA Optical Networking. All rights reserved.1414 Blind spot example – customer site Poor signal from the tower direction High tower
  15. 15. © 2019 ADVA Optical Networking. All rights reserved.1515 Example – poor installation GNSS antenna installed near a wall
  16. 16. © 2019 ADVA Optical Networking. All rights reserved.1616 Site analysis – “bad” site Help identify local reception issue
  17. 17. © 2019 ADVA Optical Networking. All rights reserved.1717 Site analysis – “good” site
  18. 18. © 2019 ADVA Optical Networking. All rights reserved.1818 GNSS antenna – signal reception Artificial intelligence and applied machine learning | GNSS use case ML can assist here …
  19. 19. © 2019 ADVA Optical Networking. All rights reserved.1919 • ML and AI can help us automatically detect changes in patterns • Safeguard GNSS signal reception in timing and synchronization networks • Predictive maintenance through long-term satellite telemetry trending and analysis • Detect GNSS anomalies such as signal obstruction caused by trees, walls and signal interference (jamming and spoofing) etc. • Allow early detection of protentional problems • Can be scaled to very large networks • Reduces OPEX AI and ML allow automatic detection of a fault The next level – AI and ML
  20. 20. © 2019 ADVA Optical Networking. All rights reserved.2020 GNSS monitoring and assurance using AI and ML NASA servers Weather AI server Automatic detection of abnormal patterns (patent pending) AlertReact
  21. 21. © 2019 ADVA Optical Networking. All rights reserved.2121 • Our solution is based on centralized GNSS monitoring and assurance • The solution can be applicable to any GNSS receivers which are capable of providing the minimal related data • No need for additional hardware or to physically attend the site • Provides network view of GNSS receiver health • Allows remote in-depth analysis • Utilizes ML and AI for optimal and automated detection of potential GNSS service degradation Summary
  22. 22. Thank you IMPORTANT NOTICE The content of this presentation is strictly confidential. ADVA Optical Networking is the exclusive owner or licensee of the content, material, and information in this presentation. Any reproduction, publication or reprint, in whole or in part, is strictly prohibited. The information in this presentation may not be accurate, complete or up to date, and is provided without warranties or representations of any kind, either express or implied. ADVA Optical Networking shall not be responsible for and disclaims any liability for any loss or damages, including without limitation, direct, indirect, incidental, consequential and special damages, alleged to have been caused by or in connection with using and/or relying on the information contained in this presentation. Copyright © for the entire content of this presentation: ADVA Optical Networking. nlaufer@adva.com

×