Evaluation of ML-based classification algorithms for GNSS signals in ocean environment
-
Jyothsna S. R. S. Koiloth
and
Padma Raju Koppireddy
Abstract
In the maritime environment, multipath interference exhibits a significantly pronounced influence, resulting in GNSS system performance degradation. Enhancing system performance involves the identification and elimination of NLOS signals. This study focuses on the analysis of multipath data induced by sea waves, collected off the coast of Kakinada Sea (16.98° N, 82.29° E), to categorize signals as Line-of-Sight (LOS), Non-Line-of-Sight (NLOS) and Multipath (MP). A machine learning (ML) approach is employed to identify the presence of LOS, NLOS and MP signals in a coastal environment, both before and after the advancement of tidal waves. In the proposed approach, ML algorithms are trained using 3 key parameters namely elevation angle, signal strength and pseudorange residuals. This study involves the implementation of 14 prominent supervised classification algorithms and their accuracies and computational times are compared. The results due to GPS (L1) and IRNSS (L5 and S1) are considered. Decision Tree and its ensemble function AdaBoost, exhibited exceptional performance of accuracy (99.99 %) and computational time (0.45 s).
Acknowledgments
The research presented in this paper was conducted as part of the project titled “A New Model for Short Term Forecasting of Scintillations using Machine Learning Approach and Generation of Regional Scintillation Maps,” which is supported by the Department of Science and Technology under the SERB-CRG scheme. This support is documented in the sanction letter with reference number CRG/2021/001660, dated February 11, 2022. The authors express gratitude to the Space Applications Centre (SAC), ISRO Ahmedabad, for generously providing the AIGS Receiver for the field trials as stipulated in the Memorandum of Understanding (MoU).
-
Research ethics: Not applicable.
-
Author contributions: All the authors have accepted responsibility for the entire content of this submitted manuscript and approved submission.
-
Competing interests: The authors declare they have no financial interests that could have appeared to influence the work reported in this paper.
-
Research funding: The research presented in this paper is carried out under the sponsored project funded by the Department of Science and Technology (DST), New Delhi, Govt. of India. However, there is no funding for publication processing fee under this project.
-
Data availability: SAC, ISRO sponsored IGS receiver data was acquired in ocean environment. Receiver data are available from the corresponding author based on reasonable requests.
The hardware setup utilized in this experiment is an IRNSS-GPS-SBAS (IGS) receiver and its antenna provided by SAC, ISRO, Ahmedabad, India. It is a multi-constellation and multi-frequency receiver. Table A1 presents the key specifications of the receiver and antenna [25].
Key specifications of IGS receiver and antenna.
| Parameter | Performance specifications | |
|---|---|---|
| Position accuracy | Hybrid positioning | 2 m (rms) |
| Velocity accuracy | GPS/IRNSS | 0.1 m/s |
| Time accuracy | GPS/IRNSS | 25 ns (rms) |
| Signal dynamics | Velocity | 515 m/s |
| Acceleration | 4 g | |
| Jerk | 5 m/s2 | |
| Acquisition sensitivity | GPS/IRNSS | −139 dBm |
| SBAS | −136 dBm | |
| Tracking sensitivity | GPS/IRNSS | −142 dBm |
| SBAS | −138 dBm | |
| Time to first fix | GPS | 40 s |
| IRNSS | 80 s | |
| Re-acquisition | GPS/IRNSS | 1 s |
| Data rate | Ethernet/NMEA | 1 Hz or 5 Hz configurable |
| Antenna details | Type | Patch antenna |
| Polarization | RHCP | |
| Shape | Hemi-spherical | |
| Frequency bands | L and S bands | |
The present work incorporates 14 prominent machine learning algorithms, utilizing signal strength, elevation angle, and pseudorange residuals as input parameters.
The following is the brief explanation of the inputs:
Carrier-to-Noise Ratio (C/N 0 ): It is a crucial parameter in navigation systems, representing the strength of the signal relative to the background noise. Different signal conditions, such as Line-of-Sight (LOS), Multipath (MP), and Non-Line-of-Sight (NLOS), exhibit variations in C/N0 values.
Elevation angle: It refers to the angle between the line of sight (LOS) to the satellite and the local horizontal plane. In general, a high elevation angle increases the likelihood of a clear LOS signal.
Pseudorange residual: It is the difference between the true range and computed pseudorange. The analysis of pseudorange residuals is crucial in Global Navigation Satellite System (GNSS) applications, as it can provide insights into the quality of the received signals and the performance of the positioning solution.
All these parameters can be indicative of various factors, including Line-of-Sight (LOS), Multipath (MP), and Non-Line-of-Sight (NLOS) conditions. The input parameters and the corresponding ranges are given below in Table A2.
Input parameters and the corresponding ranges.
| Input parameter | Range of the parameters used for training | ||
|---|---|---|---|
| LOS | MP | NLOS | |
| Signal strength (dB-Hz) | >45 | 25–45 | <25 |
| Elevation angle (deg) | >30° | <30° | <10° |
| Pseudorange residuals (m) | −2 to +2 | <−2 and >2 | >100 |
References
1. Webb, P. Introduction to oceanography. Montreal, Quebec: Roger Williams University. Rebus Community Publishers; 2019.Search in Google Scholar
2. Min, BG, Jeong, H, Ha, TJ, Cha, JS, Kim, NH. Implementation of navigation system available for marine leisure ship. Int J Control Autom 2014;7:29–36. https://doi.org/10.14257/ijca.2014.7.3.04.Search in Google Scholar
3. Misra, P, Enge, P. Global positioning system: signals, measurements & performance. IEEE Aero Electron Syst Mag 2002;17:36–7. https://doi.org/10.1109/maes.2002.1044515.Search in Google Scholar
4. IRNSS Signal in Space ICD for Standard Positioning Service v. 1.0. ISRO IRNSS-ICD-SPS-1.0. Bangalore: Indian Space Research Organization (ISRO); 2014.Search in Google Scholar
5. Mohd, QJ, Achanta, DS, Supraja, RA, Sridher, T, Nalam, VKR, Pant, TK. Multipath and thermal noise free relative TEC estimation using IRNSS L5 and S1 signals. In: Proc. of int. conf. on innovative technologies in engineering, ICITE-2018. Hyderabad, India; 2018.Search in Google Scholar
6. Kamatham, Y, Achanta, D, Kumar, A, Satyanarayana, K. Spectral analysis and mitigation of GPS multipath error using digital filtering for static applications. IETE J Res 2013;59:156. https://doi.org/10.4103/0377-2063.113036.Search in Google Scholar
7. Yozevitch, R, Ben Moshe, B, Weissman, A. A robust GNSS LOS/NLOS signal classifier. Navigation 2016;63:429–42. https://doi.org/10.1002/navi.166.Search in Google Scholar
8. Hsu, LT. GNSS multipath detection using a machine learning approach. In: IEEE 20th international conference on intelligent transportation systems (ITSC); 2017.10.1109/ITSC.2017.8317700Search in Google Scholar
9. Suzuki, T. Mobile robot localization with GNSS multipath detection using pseudorange residuals. In: Advanced robotics symposia. Toyama, Japan; 2018.10.1080/01691864.2019.1619622Search in Google Scholar
10. Egea-Roca, D, Tripiana-Caballero, A, López-Salcedo, JA, Gonzalo, SG, Wim, DW, Bruno, B, et al.. GNSS measurement exclusion and weighting with a dual polarized antenna: the FANTASTIC project. In: 2018 8th international conference on localization and GNSS (ICL-GNSS). Guimaraes: MDPI Sensors; 2018:1–6 pp.10.1109/ICL-GNSS.2018.8440897Search in Google Scholar
11. Areeyapinun, T, Kitani, T. A prototype of a precision forecasting system for real-time navigation with RTK-GNSS. In: Proc. of the 11th international workshop on informatics (IWIN2017). Informatics Society; 2017:193–200 pp.Search in Google Scholar
12. Xu, H, Zhang, G, Xu, B, Hsu, LT. GNSS shadow matching based on intelligent LOS/NLOS classifier. In: The 16th world congress of the international association of institutes of navigation (IAIN). Chiba, Japan; 2018.Search in Google Scholar
13. Roussel, N, Ramillien, G, Frappart, F, Darrozes, J, Gay, A, Biancale, R, et al.. Sea level monitoring and sea state estimate using a single geodetic receiver. Rem Sens Environ 2015;171:261–77. https://doi.org/10.1016/j.rse.2015.10.011.Search in Google Scholar
14. Groves, PD, Jiang, Z, Rudi, M, Strode, P. A portfolio approach to NLOS and multipath mitigation in dense urban areas. In: Proceedings of the ION GNSS+ 2013. Nashville, TN, USA; 2013:3231–47 pp.Search in Google Scholar
15. Ollander, S, Bode, F, Baum, M. Multi-frequency GNSS signal fusion for minimization of multipath and non-line-of-sight errors: a survey. In: 2018 15th workshop on positioning, navigation and communications (WPNC). Bremen; 2018:1–6 pp.10.1109/WPNC.2018.8555856Search in Google Scholar
16. Uaratanawong, V, Satirapod, C, Tsujii, T. Evaluation of multipath mitigation performance using signal-to-noise ratio (SNR) based signal selection methods. J Appl Geodesy 2021;15:75–85. https://doi.org/10.1515/jag-2020-0045.Search in Google Scholar
17. Strode, PRR, Groves, PD. GNSS multipath detection using three-frequency signal-to-noise measurements. GPS Solut 2015;20:399–412. https://doi.org/10.1007/s10291-015-0449-1.Search in Google Scholar
18. Sun, R, Wang, G, Zhang, W, Hsu, LT, Ochieng, WY. A gradient boosting decision tree based GPS signal reception classification algorithm. Appl Soft Comput 2020;86:105942. https://doi.org/10.1016/j.asoc.2019.105942.Search in Google Scholar
19. Gowdayyanadoddi, NS, Curran, JT, Broumandan, A, Lachapelle, G. A ray-tracing technique to characterize GPS multipath in the frequency domain. Int J Navig Obs 2015;2015:983124–16. https://doi.org/10.1155/2015/983124.Search in Google Scholar
20. Butt, J, Wieser, A, Gojcic, Z, Zhou, C. Machine learning and geodesy: a survey. J Appl Geodesy 2021;15:117–33. https://doi.org/10.1515/jag-2020-0043.Search in Google Scholar
21. Shalev-Shwartz, S, Ben-David, S. Understanding machine learning: from theory to algorithms. Cambridge, England: Cambridge University Press; 2014.10.1017/CBO9781107298019Search in Google Scholar
22. Ghosh, U, Sarma, AD, Javeed, MQ, Koteswara Rao, NV. Selective suppression of IRNSS S-band signals for specific applications. In: IEEE international WIE conference on electrical and computer engineering (WIECON-ECE). WIT, Dehradun, India; 2017.10.1109/WIECON-ECE.2017.8468869Search in Google Scholar
23. Kumar, PN, Sarma, AD, Reddy, AS. Modelling of ionospheric time delay of global positioning system (GPS) signals using taylor series expansion for GPS aided geo augmented navigation applications. IET Radar, Sonar Navig 2014;8:1081–90. https://doi.org/10.1049/iet-rsn.2013.0351.Search in Google Scholar
24. Sarma, AD, Sultana, Q, Srinivas, VS. Augmentation of Indian regional navigation satellite system to improve dilution of precision. J Navig 2010;63:313–21. https://doi.org/10.1017/S037346330999035X.Search in Google Scholar
25. IRNSS/GPS/SBAS receiver user manual, Accord software & systems Pvt. Ltd, Issue 1.2; 2016.Search in Google Scholar
© 2024 Walter de Gruyter GmbH, Berlin/Boston
Articles in the same Issue
- Frontmatter
- Original Research Articles
- Ionospheric TEC modeling using COSMIC-2 GNSS radio occultation and artificial neural networks over Egypt
- Regional GPS orbit determination using code-based pseudorange measurement with residual correction model
- Analysis of different combinations of gravity data types in gravimetric geoid determination over Bali
- Assessment of satellite images terrestrial surface temperature and WVP using GNSS radio occultation data
- GNSS positioning accuracy performance assessments on 1st and 2nd generation SBAS signals in Thailand
- Differential synthetic aperture radar (SAR) interferometry for detection land subsidence in Derna City, Libya
- Advanced topographic-geodetic surveys and GNSS methodologies in urban planning
- Detection of GNSS ionospheric scintillations in multiple directions over a low latitude station
- Spatiotemporal postseismic due to the 2018 Lombok earthquake based on insar revealed multi mechanisms with long duration afterslip
- Practical implications in the interpolation methods for constructing the regional mean sea surface model in the eastern Mediterranean Sea
- Validation of a tailored gravity field model for precise quasigeoid modelling over selected sites in Cameroon and South Africa
- Evaluation of ML-based classification algorithms for GNSS signals in ocean environment
- Development of a hybrid geoid model using a global gravity field model over Sri Lanka
- Implementation of GAGAN augmentation on smart mobile devices and development of a cooperative positioning architecture
- On the GPS signal multipath at ASG-EUPOS stations
Articles in the same Issue
- Frontmatter
- Original Research Articles
- Ionospheric TEC modeling using COSMIC-2 GNSS radio occultation and artificial neural networks over Egypt
- Regional GPS orbit determination using code-based pseudorange measurement with residual correction model
- Analysis of different combinations of gravity data types in gravimetric geoid determination over Bali
- Assessment of satellite images terrestrial surface temperature and WVP using GNSS radio occultation data
- GNSS positioning accuracy performance assessments on 1st and 2nd generation SBAS signals in Thailand
- Differential synthetic aperture radar (SAR) interferometry for detection land subsidence in Derna City, Libya
- Advanced topographic-geodetic surveys and GNSS methodologies in urban planning
- Detection of GNSS ionospheric scintillations in multiple directions over a low latitude station
- Spatiotemporal postseismic due to the 2018 Lombok earthquake based on insar revealed multi mechanisms with long duration afterslip
- Practical implications in the interpolation methods for constructing the regional mean sea surface model in the eastern Mediterranean Sea
- Validation of a tailored gravity field model for precise quasigeoid modelling over selected sites in Cameroon and South Africa
- Evaluation of ML-based classification algorithms for GNSS signals in ocean environment
- Development of a hybrid geoid model using a global gravity field model over Sri Lanka
- Implementation of GAGAN augmentation on smart mobile devices and development of a cooperative positioning architecture
- On the GPS signal multipath at ASG-EUPOS stations
