Distributed GNSS Positioning-as-a-Service: Death of a Traditional GNSS Receiver?

Sep 2023 | No Comment

A traditional perception of a black-box GNSS receiver served for the GNSS PNT has dramatically evolved into a completely novel concept of distributed GNSS Positioning-as-a-Service, utilising the latest developments in computer science, mathematics, and statisticsA traditional perception of a black-box GNSS receiver served for the GNSS PNT has dramatically evolved into a completely novel concept of distributed GNSS Positioning-as-a-Service, utilising the latest developments in computer science, mathematics, and statistics

Renato Filjar

Electrical engineer, a Titular Professor of Electronics Engineering (Tenure) with Faculty of Engineering, University of Rijeka, Rijeka, Croatia, and a Professor of Electrical Engineering & Head of Laboratory for Spatial Intelligence, Hrvatsko Zagorje Krapina University of Applied Sciences, Krapina, Croatia

Filip Šklebar

Electrical engineer, and an external associate to Laboratory for Spatial Intelligence, Hrvatsko Zagorje Krapina University of Applied Sciences, Krapina, Croatia

Ivica Zubic

Traffic Engineer, and a Lecturer with Laboratory for Spatial Intelligence, Hrvatsko Zagorje Krapina University of Applied Sciences, Krapina, Croatia


Satellite navigation has become an essential cornerstone of modern civilisation, a fundamental component of national infrastructure, and a public goods, enabling and driving a vast range of technology, and socio-economic systems and services. A prevalent view of GNSS utilisation still depicts an image of a user with a device in his or her hand magically producing position and velocity estimates and perfect timing synchronisation using some marvels of mathematics and statistics, signal processing, radio communications, electrical and mechanical engineering. While such an image may be found in diminishing number of occasions, such a description of GNSS Positioning, Navigation, and Timing (PNT) service is not only changing, but has become obsolete. Recent developments in computer science, mathematics and statistics, and signal processing have revolutionised a traditional view of a black-box GNSS receiver performing the complete GNSS position estimation process [10]. Introduction of the Software-Defined Radio (SDR) concept, Machine Learning (ML) methods and algorithms, fast internet connections and advanced computational techniques, such as Cloud Computing, renders the GNSS position estimation process more resilient to external (environmental) adversarial and detrimental effects, distributed and, therefore, transparent [6, 7, 10]. Transparency and resiliency of GNSS position estimation process has opened an enormous room for developments, enhancements, and improvements which leads to en entirely new concept of GNSS Positioning-as-a-Service, described in this manuscript. The GNSS Positioning-as-a-Service concept is defined as a mathematics-established and software-based GNSS position estimation process set on a distributed computing platform.

The traditional concept of GNSS positioning has equalised the GNSS position estimation process with a GNSS receivers, rendering it a black-box delivering position, velocity, and time estimates [7, 10]. Such a concept was suitable for initially anticipated military applications of satellite navigation. It was later transferred into commercial domain through the process of wider acceptance of satellite navigation [1, 7].

The GNSS PNT service is conducted through the GNSS position estimation process, comprising three essential domains [3, 5, 6, 7], as depicted in Figure 1: (i) Radio Frequency Domain (RFD), (ii) Base-Band Domain (BBD), and (iii) Navigation, or Application, Domain (ND) In the reference to GNSS positioning tasks, the RFD deals with the received modulated satellite signals performing conditioning, demodulation, and conversation of the signals. The BBD measures raw GNSS pseudoranges and extracts related information from the Navigation Message. The ND applies detrimental effects counter measures on raw GNSS pseudorange measurements, such as the ionospheric and tropospheric delays corrections, and performs the GNSS position and positioning errors estimation based on the Navigation Message information and the mitigated GNSS pseudoranges. Selection and applications of methods and techniques for the GNSS position estimation process are governed by the characteristics of the system architecture selected for the GNSS position estimation process implementation [6, 7].

Traditional concept of a GNSS receiver assumes the prevalent utilisation of dedicated electronic circuits, electrically pre-programmed to perform a dedicated tasks. The concept allows for fitting into a single black-box GNSS receiver unit, but lacks flexibility in modification, upgrade, and multi-usage of the initially pre-defined hardware of a particular purpose [6, 7].In the process of increased utilisation of GNSS as the critical and enabling foundation technology, the traditional view on the user equipment, GNSS receivers, has started to emerge as a growing obstacle to development. As it happens with every technology, satellite navigation is vulnerable to numerous adversarial effects, both natural, such as space weather/ionospheric and multipath causes of satellite signal delays and reception disruptions, and artificial, such as spoofing, meaconing, and jamming [2, 4, 5, 6, 8, 12, 13]. GNSS operators attempt to assist with mitigation of such effects using the globalised correction models, augmentation systems such as EGNOS and WAAS, and the provision of additional information for the internet-connected user equipment. The attempts generally yield a mixed success, as majority of the effects are related to GNSS positioning environment conditions in the immediate vicinity of the unknown position at which the GNSS signal reception takes place [7]. The GNSS operator is prevented in the provision of the suitable correction and mitigation information by the very nature of the GNSS position estimation and the inability to be aware of the user’s GNSS positioning environment at a required scale. Furthermore, the results of the GNSS positioning process serve increasingly a GNSS application, system or a service, and are generally not presented in its raw form to the GNSS end-user [6, 7]. The manner of assessment of the GNSS positioning performance is more related to the core GNSS processes, and, consequently, suits the traditional view of a GNSS receiver. However, it fails in establishing a productive and pragmatic compliance with the needs and requirements of a GNSS application that will utilise GNSS position estimates to produce a service or an information based on GNSS-based position estimate. GNSS applications need some form of GNSS positioning error estimate to determine the Quality of Service (QoS) of a particular GNSS application [1, 7, 10]. For instance, GNSS operators usually express the estimated positioning error in terms of ideal positioning environment conditions: unobstructed view of horizon, averaged ionospheric and multipath effects. The approach yields the range of GNSS positioning errors, in anticipated extremes, which are generally of sparse use for GNSS application developers, operators, and users, since the positioning error estimates may fluctuate over time widely and unexpectedly in and out of the acceptable range for the particular GNSS application.

The shortcomings of the traditional approach in implementation of the GNSS positioning process through a single black-box mobile GNSS receiver may be summarised as follows: (i) GNSS pseudorange error correction using the global models fail in recognition of the real positioning environment conditions, which render them sub-optimal and generally inefficient in the mitigation of various adversarial effects on GNSS PNT. (ii) Specification of the core GNSS PNT QoS do not translate into GNSS-based application QoS needs, which reduces and limits significantly potentials for GNSS application development and operation. (iii) GNSS augmentation and assistance, such as SBAS: WAAS, EGNOS require additional infrastructure, which is generally very complex and expensive for establishment, operation, and maintenance and still provide sub-optimal results in mitigation of the adversarial effects on GNSS PNT. (iv) The raising number of artificial adversarial effects on GNSS performance and operation requires additional infrastructure and effort for mitigation of their causes (GNSS spoofing, jamming, and meaconing) [2], as GNSS cyberattacks which may additionally raise the mitigation costs.

Our international research team assessed the recent related developments, including: (i) Software-Defined Radio concept utilisation in user segment of GNSS that renders the GNSS positioning process transparent and manageable, (ii) abundance of machine learning methods and techniques, suitable for a wide range of classification and regression (predictive modelling) problems solving, (iii) breakthrough developments in the provision of fast, reliable, and ubiquitous mobile internet communications, (iv) available and affordable advanced computing frameworks and methods, such as cloud computing, (v) availability of machine-readable positioning environment-related data, such as space weather and ionospheric conditions observations, (vi) structured classification of GNSS applications and their needs and requirements for GNSS PNT performance. The assessment results in the proposal of a novel concept in definition, description, and deployment of GNSS positioning process, we called the GNSS Positioning-as-a-Service [7]. The new concept is depicted in Figure 2.

The GNSS Positioning-as-a-Service extend the architecture comprising a Mobile Unit (MU) and a GNSS application Framework (GNSS-F), connected via mobile internet service [7]. The GNSS positioning process is distributed here between the two constitutional units. The MU performs tasks related to the initial RF signal processing and the Base-Band Domain (GNSS pseudorange measurement and extraction of related data from Navigation Message), and packs the tasks results (raw GNSS pseudorange measurements observed Navigation Message data, and optional observations of the immediate GNSS positioning environment conditions, such as geomagnetic field density components values, if the MU is equipped with the suitable sensors) into a tailored protocol for transmission to cloud-based unit. The GNSS-AF utilises the observations of the immediate GNSS positioning environment conditions, if received, and trusted thirdparties observations and estimations for development and operation of bespoke machine learning-based correction models for mitigation the adversarial effects contained in raw GNSS pseudoranges [5, 6, 7, 9]. GNSS pseudoranges become corrected optimally in the sense of the actual immediate GNSS positioning environment conditions. Corrected GNSS pseudoranges then serve as inputs to GNSS position estimation algorithm, which software implementation returns the improved position estimates. In that sense, the cloud-based unit performs the tasks of the Navigation Domain of the GNSS positioning process.

The proposed GNSS Positioning-as-a-Service concept recognises a GNSS application closely integrated with the task related to Navigation within the GNSS-AF. This novelty establishes a GNSS application in charge of definition, development and operation of correction models, as well as of selection of the position estimation methods, all of them in relation to the needs and requirements for the GNSS PNT performance for the GNSS application in question. The integration between the Navigation Domain tasks and a GNSS application anticipated to use the results of the GNSS positioning process resolves the lack of compliance with the core GNSS positioning performance and the needs and requirements for the GNSS application QoS [1, 7]. As the result, the GNSS position estimates emerge as optimised to the PNT needs and requirements of a GNSS application [7].

The active involvement of a GNSS application in the GNSS positioning process allows for tailored configuration of GNSS positioning performance as optimised to the needs and requirements of GNSS application. It utilises the situation awareness concept of mitigation the adversarial effects on GNSS PNT in consideration of the actual positioning environments conditions, based either on the mobile unit’s observations, or on data provide by trusted third-party providers (US NOAA, US NASA, EU Copernicus etc.) [5, 7, 9, 10]. The proposed GNSS Positioning-as-a-Service allows for re-definition of the GNSS positioning performance, including accuracy, in terms of scenarios and conditions of actual utilisation. This approach relieves the GNSS operators of the mission-impossible task to guarantee the GNSS PNT performance, determined mostly with the users positioning micro-environment conditions, the scenario of usage, and the meeds and requirements of a GNSS applications which are all unknown to the GNSS operator. Within the GNSS Positioning-as-a-Service concept, the GNSS operator may focus on advanced spectrum management and protection, and signal structure improvement and protection, without the involvement in issues outside its outreach and control.

Our research team has developed a laboratory GNSS Positioning-as-a-Service development and simulation framework, based on the software developed in the open-source R environment for statistical computing. The framework operates in Laboratory for spatial Intelligence of Hrvatsko Zagorje Krapina University of Applied Sciences in Krapina, Croatia. It serves research in mathematics, statistics, signal processing, and computer science in relation to advance the GNSS PNT performance with the novel position estimation methods, and for detection, identification and mitigation of both natural and artificial adversarial effects on GNSS PNT performance. The results accomplished make contribution to scientific developments, while at the same time being exploiting in industry, and serving as important source of knowledge in the academic education process.

The proposed GNSS Positioning-as-a-Service emerges as a groundbreaking novelty, providing numerous benefits and advantages, while opening widely the room for robust and QoS-guaranteed GNSS applications development and operations. Enhanced PNT performance, detection of and resilience against natural and artificial (spoofing) [2] adversarial effects on GNSS PNT, alignments to GNSS application QoS, and quantifying the risk of the GNSS uitilisation for particular GNSS-based application [11, 12] are among the numerous opportunities resulting from the GNSS Positioning-as-a-Service implementation and operation. The novelty also opens a number of challenges to the existing framework of the GNSS PNT performance management, as GNSS positioning enters the field of ICT services, including: redefinitions of regulations and processes, operations, standardisation & certification, regulations of operations and international trade, international co-operation, targeted academic and professional education. While our team continuous in research, development and exploitation of the proposed concept, we believe the GNSS Positioning-as-a-Service is to set the new standard, attract research interest, and open the vast range of developments, advancements, and co-operation in the field of satellite navigation and its applications.


[1] EUSPA. (2023). User Consultation Platform (UCP) User Needs and Requirements 2022 (Library contains Reports on: (1) Aviation and Drones, (2) Consumer Solutions, (3) Emergency Management and Humanitarian Aid, (4) Energy and Raw Materials, (5) Infrastructure, (6) Insurance and Finance, (7) Maritime, Inland Waterways, Fisheries and Aquaculture). European Agency for Space Programme (EUSPA). Prague, Czechia. Available at:

[2] Filić, M. (2018). Foundations of GNSS Spoofing Detection and Mitigation with Distributed GNSS SDR Receiver. TransNav, 12(4), 649-656. doi: http://dx.doi. org/10.12716/1001.12.04.01

[3] Filić, M, Filjar, R. (2018a). Smartphone GNSS positioning performance improvements through utilisation of Google Location API. Proc of 41 st International Convention MIPRO/CTI, 507-510. Opatija, Croatia. doi: https://doi. org/10.23919/MIPRO.2018.8400087

[4] Filić, M, Filjar, R. (2018b). A South Pacific Cyclone-Caused GPS Positioning Error and Its Effects on Remote Island Communities. TransNav, 12(4), 663-670. doi: http://

[5] Filić, M, Filjar, R. (2018c). Modelling the Relation between GNSS Positioning Performance Degradation, and Space Weather and Ionospheric Conditions using RReliefF Features Selection. Proc of 31st International Technical Meeting ION GNSS+ 2018, 1999-2006. Miami, FL. doi:

[6] Filić, M, Grubišić, L, Filjar, R. (2018). Improvement of standard GPS position estimation algorithm through utilization of Weighted Least-Square approach. Proc of 11th Annual Baška GNSS Conference, 7-19. Baška, Krk Island, Croatia. Available at: https:// zbornici-gnss/2018-GNSS-11.pdf

[7] Filjar, R. (2022a). An application-centred resilient GNSS position estimation algorithm based on positioning environment conditions awareness. Proc ION ITM 2022, 1123 – 1136. Long Beach, CA. doi:

[8] Filjar, R. (2022b). A contribution to short-term rapidly developing geomagnetic storm classification for GNSS ionospheric effects mitigation model development. ICEASE 2021 Conference. Islamabad, Pakistan. doi: ICASE54940.2021.9904168

[9] Filjar, R, Weintrit, A, Iliev, T, Malčić, G, Jukić, O, Sikirica, N. (2020). Predictive Model of Total Electron Content during Moderately Disturbed Geomagnetic Conditions for GNSS Positioning Performance Improvement. Proc USION2020, 256-262. Pretoria, South Africa. doi: FUSION45008.2020.9190264

[10] Filjar, R, Damas, M C, Iliev, T B. (2020). Resilient Satellite Navigation Empowers Modern Science, Economy, and Society. CIEES 2020. IOP Conf. Ser: Mater Sci Eng 1032, 012001 (10 pages). Borovets, Bulgaria. doi: 10.1088/1757-899X/1032/1/012001

[11] Malić, E, Sikirica, N, Špoljar, D, Filjar, R. (2023). A method and a model for risk assessment of GNSS utilisation with a proof-of-principle demonstration for polar GNSS maritime applications. TransNav, 17(1), 43-50. doi: http://dx.doi. org/10.12716/1001.17.01.03

[12] Sikirica, N, Dimc, F, Jukić, O, Iliev, T B, Špoljar, D, Filjar, R. (2021). A Risk Assessment of Geomagnetic Conditions Impact on GPS Positioning Accuracy Degradation in Tropical Regions Using Dst Index. Proc ION ITM 2021, 606- 615. San Diego, CA. doi: https://

[13] Špoljar, D, Jukić, O, Sikirica, N, Lenac, K, Filjar, R. (2021). Modelling GPS Positioning Performance in Northwest Passage during Extreme Space Weather Conditions. TransNav, 15(1), 165-169. doi: http://dx.doi. org/10.12716/1001.15.01.16

1 Star2 Stars3 Stars4 Stars5 Stars (No Ratings Yet)

Leave your response!

You must be logged in to post a comment.