The Impact of Signal Interference on Static GNSS Measurements

Abstract

Global navigation satellite systems (GNSSs) are an integral part of modern society and are used in various industries, providing users with positioning, navigation, and timing (PNT). However, their effectiveness is vulnerable to signal interference, since GNSSs are based on received satellite signals from space, and that can severely impact applications that rely on continuous and accurate data. Interference can pose significant risks to sectors dependent on GNSSs, including transportation, telecommunications, finance, geodesy, and others. For this reason, in parallel with the development of GNSSs, various interference protection techniques are being developed to enable users to receive GNSS signals without the risk of interference, which can cause various effects, such as reducing the accuracy of positioning, as well as completely blocking signal reception and making it impossible to obtain positioning. There are various sources and methods of interfering with GNSS signals, and the greatest consequences are caused by intentional interference, which includes jamming, spoofing, and meaconing. This study investigates the effects of jamming devices on static GNSS observations using high-accuracy devices through multiple controlled experiments using both single-frequency (SF) and multi-frequency (MF) jammers. The aim was to identify the distances within which signal interference devices disrupt GNSS signal reception and position accuracy. The research conducted herein was divided into several phases where zones within which the jammer completely blocked the reception of the GNSS signal were determined (blackout zones), as were zones within which it was possible to obtain the position (but the influence of the jammer was present) and the influence of the jammer from different directions/azimuths in relation to the GNSS receiver. Various statistical indicators of the jammer’s influence, such as DOP (dilution of precision), SNR (signal-to-noise-ratio), RMS (root mean square), and others, were obtained through research. The results of this study indicate that commercially available, low-cost jamming devices, when operated within manufacturer-specified distances, completely disrupt the reception of GNSS signals. Their impact is also evident at greater distances, where they significantly reduce SNR values, increase DOP, and decrease the number of visible satellites, leading to reduced measurement reliability and integrity. These results underline the necessity of developing effective protection mechanisms against GNSS interference and strategies to ensure reliable signal reception in GNSS-dependent applications, particularly as the use of jamming devices becomes more prevalent.

1. Introduction

GNSSs (global navigation satellite systems) have become a key technology that enables positioning, navigation, and timing (PNT) anywhere and at any time on and near the Earth’s surface. In use today are four global (GPS, GLONASS, BeiDou, Galileo) and two regional (NavIC, QZSS) satellite systems that provide users with different levels of service. The main advantages of GNSSs are global use, obtaining real-time data, high precision, weather independence, compatibility with modern communications, and interoperability with other measuring sensors and systems [1]. Due to technological development, (especially microprocessors, which have made it possible to combine GNSSs with a variety of other sensors), GNSSs have entered many everyday human applications, such as mobile navigation and other location-based services. Today, the use of GNSSs is growing rapidly in all areas of life, and due to their wide distribution, the availability and constancy of the signal are often taken for granted.

Along with the development of GNSSs, technologies for interference with signals from space have also been developed, which partially or completely make it impossible to determine the position. The basic principle of GNSSs is based on the transmission of radio signals, which results in certain limitations when using them. In addition to the fact that GNSSs are not the most accurate measurement systems and are not available always and everywhere, the signal transmitted by satellites is subject to interference that can be natural, intentional, or accidental [2]. Natural interference (caused, for example, by ionospheric scintillation) and accidental interference (caused by, for example, in-band emission or out-of-band emission) are often predictable. GNSSs have been designed to withstand a certain level of radio frequency interference (RFI), which is made possible through the DSSS (direct sequence spread spectrum) modulation used in GNSSs [3,4]. Bigger problems are caused by intentional signal interference, such as jamming and spoofing. Although intentional signal interference is illegal in most countries around the world, access to interference devices is available via several webpages, even at a very low cost, and the techniques of making such devices are publicly available [5,6,7,8,9]. The most common way of interfering with GNSS signals is jamming. Nearly all commercially available jammers transmit chirp signals, a continuous-wave (CW) tone with constantly changing frequency over time (commercial off-the-shelf) [10]. Today, we can find many examples of interference for various reasons, such as blocking the tracking of one’s own location or in cyberattacks. One of the most well-documented examples of interference is when an engineering firm worker in New Jersey had a GPS jammer so that his boss was not able to track him all the time. However, his route took him close to Newark airport, and his jammer affected its satellite systems and caused problems and delays in the operation of the airport [11,12]. Another example is when the US Navy accidentally jammed a GPS system in San Diego while testing an interference device, which caused many problems in the functioning of the city [12]. The motivation for jamming GNSS signals can be an attack on GNSS users to prevent them from positioning or, for personal reasons, using PPDs (personal privacy devices) to prevent people and vehicles from being tracked [7,13]. Jamming devices like PPDs are small, portable, and inexpensive, and they can present a serious threat to safety-of-life (SoL) applications and critical sectors like law enforcement, transportation, communication, and finance [13,14].

The biggest problem with GNSSs is that the signals received from satellites on de-vices are extremely weak, meaning they are very susceptible to interference. Today, the use of interference devices is increasingly common in civilian and military activities. GNSS receivers used for civil applications are especially vulnerable to interference, since the signal characteristics of GNSS civil signals—such as the carrier frequency, polarization, and modulation parameters—are open and always made known to the public [3,15].

Regarding intentional interference, two of the most common techniques are jam-ming and spoofing. Jammers are devices that transmit a high-power radio frequency signal equal to or very close to the frequency of the device whose operation is to be prevented. They aim to prevent the receiver from collecting and tracking GNSS signals. Jammers overwhelm (block) the GNSS signal and disable navigation, so the GNSS is ineffective or degraded for users. On the other hand, spoofing devices emit fake GNSS signals that are similar to authentic signals to the receiver. Fake signals are transmitted at higher power than authentic ones to trick the receiver into picking up those signals and subsequently begin tracking the fake satellites. The result of spoofing is the incorrect user location [16,17,18]. An overview of spoofing and anti-spoofing technology has been presented in Meng et al. [19], where spoofing modes are classified according to signal generation, the implementation stage, and the deployment strategy.

Because of the vulnerabilities of GNSS signals, it is necessary to develop effective protection techniques against interference. The importance of developing protection techniques against GNSS interference is also demonstrated by the GPSJAM project [19], which generates maps of likely GPS interference based on aircraft reports of their navigation system accuracy. GPSJAM aggregates the ADS-B (Automatic Dependent Surveillance–Broadcast) messages broadcast from aircraft containing information about GPS positioning accuracy over the last 24 h. Collected data are displayed in the form of a hexagonal map indicating the level of GPS interference. According to GPSJAM, GPS interference is frequently registered in active war zones (e.g., the Black Sea, the Middle East, and the Baltic region). Moreover, the importance of the development of these techniques was demonstrated by the Jammertest organization in Norway, which has convened for 4 years in a row and enables manufacturers of GNSS devices to test their solutions for GNSS interference in controlled conditions to improve defense against intentional jamming [20].

The threats caused by GNSS interference were recently addressed by three United Nations (UN) organizations (the International Telecommunication Union—ITU, International Civil Aviation Organization—ICAO, and International Maritime Organization—IMO) in a joint declaration on the protection of the satellite radio navigation service from harmful interference [21,22]. To strengthen civil GNSS receivers against jamming and spoofing threats, various technologies and sensors are being developed to make receivers more resilient in contested environments [23,24]. Technological development has produced protective measures against GNSS jamming and spoofing, which are implemented through various combinations of technologies in both receivers and antennas through the use of multiple sources of positioning information and through the authentication of GNSS signals. Protection is achieved through authentication of GNSS signals (e.g., Galileo OSNMA), multiple sources of positioning (e.g., multi-frequency and multi-constellation), better antenna configuration (e.g., Control of Radiation Pattern Antenna (CRPA)), and the implementation of specialized receiver detection and mitigation techniques (e.g., jammer to signal ratio (J/S) level testing and frequency domain filtering). Spoofing detection and signal integrity methods implemented in receivers are usually based on monitoring signal power or carrier-to-noise density ratio (C/N₀) monitoring, time-of-arrival (TOA) discrimination, correlator output distribution testing, and consistency checking of various measurements such as ephemeris data, clock offset changes, or code and carrier Doppler [25].

In the latest GNSS and Secure SATCOM User Technology Report, Issue 1 [25], five families of GNSS receivers were differentiated to better reflect the current technology and capability of the GNSS industry. Survey-grade GNSS receivers commonly used in geomatics and surveying belong to a family of professional, non-regulated receivers providing the highest level of performance in terms of accuracy. The key performance indicators for this family of receivers are accuracy, availability, continuity, integrity and robustness. The measures to mitigate jamming and spoofing threats could be implemented at the level of competent authorities, critical service providers, device manufacturers, and end users in four fields of action: ensuring a clean RF environment, improving antennas and setups, strengthening GNSS (e.g., implementing anti-jamming (A/J) and anti-spoofing (A/S) techniques), and diversifying sources of PNT information [26]. In addition, it has been pointed out that civil GNSS receivers (including surveillance receivers) should be better protected against RFI attacks, as they are designed for benign conditions and therefore even basic anti-jamming (A/J) or anti-spoofing (A/S) measures are not integrated [27].

The vulnerability of geodetic GNSS receivers affected by the use of a low-cost jammer in the L1 GPS band and in the E1 Galileo frequency band (L1/E1) was evaluated by the research presented in Bažec et al. [28]. Nine geodetic GNSS receivers from different manufacturers (Trimble, Javad, Leica) and generations were subjected to L1/E1 interference signals in static mode and at different distances from the receiver. Kinematic tests were carried out with a jammer placed in a vehicle travelling through the test area at various speeds. The analysis showed that the use of L1/E1 jammers leads to an increased carrier-to-noise ratio (C/N₀) or, in the worst case, to a loss of signal. When analyzing the effects of a jammer, the authors focused on the changes in the observed values of the carrier-to-noise density (C/N₀) of the received signals and the differences in the coordinates determined during GNSS processing during jamming.

The response of some mass-market and professional receivers to intentional interference from different 3D jammer positions was investigated in Pavlovčič-Prešeren et al. [29]. The results have shown that low-cost receivers are much more susceptible to jamming, while the latest generation of geodetic GNSS receivers are much more resilient. In configurations where the position of a jammer is above a GNSS receiver, the vulnerability and poor performance of the receiver is much more pronounced than in situations where the position of a jammer is below the same type of receiver. This is even more pronounced with low-cost receivers which, unlike geodetic receivers which have some of these algorithms, do not have any interference mitigation implemented.

The response of four smartphones with dual-frequency, and multi-constellation reception capability (Xiaomi Mi8, Xiaomi 11T, Samsung Galaxy S20, and Huawei P40) to various single- and multi-frequency jammers is analyzed in Pavlovič-Prešern et al. [30]. In addition, the research involved four geodetic GNSS receivers (two Leica GS18 and two Leica GS15). The study focused on two aspects—the characteristics of the carrier-to-noise density ratio (CNR) and the analysis of the position quality—both for smartphones and professional receivers during a multi-frequency jamming event, with the aim of localizing the jammer based on the raw measurements of the smartphones. It was found that smartphones have an advantage over professional geodetic receivers in jamming localization, as they can receive signals from satellites even under jamming conditions, while geodetic receivers cancel the measurements in such cases to avoid incorrect positioning.

As mentioned earlier, many industries rely on the uninterrupted reception of GNSS signals, including the geodetic profession, where GNSS is today the main tool for positioning on a global level. So, in this work, the effect of jammers on static GNSS observations was tested with two different types of jammers. Various tests were performed (including the initial jammer test, which identified blackout zones, i.e., zones without GNSS signal reception during jammer operation), and zones where the influence of the jammer on the GNSS signal is visible were determined. Detailed tests were performed on the influence of jammers on static GNSS observations, with all jammer positions from the same direction/azimuth with respect to the GNSS receiver. The last test aimed to evaluate the influence of jammers on the GNSS receiver from different positions (directions/azimuth) around the receiver.

The motivation for our research was the need to gain a better insight into the performance of geodetic survey-grade GPS/GNSS receivers under the influence of SF or MF jammers, which are commercially and easily available on the market. The GPS/GNSS receivers used for our research were two older types (one Trimble 4000 SSi and two Trimble R8 receivers), which presumably had no anti-jam algorithm implemented or at best a less sophisticated anti-jam algorithm implemented. The aim was to assess the vulnerability of geodetic GPS/GNSS receivers under different jamming conditions. While numerous studies have addressed GNSS interference in laboratory settings or through limited field experiments, this work fills a recent gap in the empirical characterization of the vulnerability of high-accuracy geodetic GNSS receivers by combining extensive field-based static and RTK measurements under various operational scenarios, including different jammer types, distances, and azimuths. By quantifying both zones of complete signal loss (blackout zones) and zones of partial degradation and linking them to measurable parameters such as SNR, PDOP, and satellite visibility, the study provides operationally relevant data that can directly inform the design and validation of existing and emerging protection measures. In particular, the results provide a valuable contribution to the improvement of transitional solutions aimed at improving the resilience of current GNSS-dependent systems, such as multi-frequency failover, elevation-based satellite weighting, and hybrid GNSS–INS integration, until fully robust architectures and signal structures become widely implemented.

The second chapter of the paper describes the GNSS signal and its behavior in the event of interference, as well as the consequences of interference and protection techniques. This is followed by a description of the research methodology and the tests carried out, followed by the research results and discussions based on which conclusions were drawn about the effects of interference on static GNSS observations.

2. Theoretical Background and Research Methodology

GNSS provides critical positioning, navigation, and timing (PNT) services across various sectors such as transport, telecommunications, emergency response, finance, and defense. These systems are based on low-power signals transmitted by satellites in medium earth orbit and received by GNSS receivers on the ground. The satellites continuously transmit navigation signals on two or more frequencies, mostly in the L-band. These signals contain codes and navigation data that enable users to calculate the travelling time from the satellite to the receiver and to determine the coordinates at any given epoch. The main components of the signal consist of the carrier wave, the pseudorandom noise code (PRN), and the navigation message [20]. The carrier signals are modulated with PRN codes that enable the satellite clock to be read in the receivers and transmission of other information, such as the orbital parameters. These signals resemble noise but have known frequencies and wavelengths. The navigation message provides information about the synchronization mask, epoch, polynomial coefficients for satellite clock correction modelling, broadcast ephemerides of the satellite, satellite health status, and other complementary data [21,22]. Most of the signal power is dispersed on the way from the satellite to the receiver, leaving only a small amount to reach the receiver, together with unwanted noise and interference. The useful signal would be indistinguishable from the noise if it did not have a much wider frequency band than the noise. This not only helps to distinguish noise but also helps to reduce or prevent signal interference, whether intentional or unintentional, as well as errors and uncertainties due to multipath signal reflections [23].

The use of GNSS signals for positioning plays an important role in modern society. Reliable navigation functions are essential for an increasing number of today’s applications on land, at sea, and in the air. Human dependence on GNSS is increasing daily, but with its use comes the need to consider the vulnerability of the system. The European Commission estimates that 6–7% of GDP (gross domestic product) in Western countries already depends on satellite navigation, which when expressed in monetary terms amounts to approximately EUR 800 million in the European Union [24].

The main weaknesses of GNSS are the weak signal power at the earth’s surface, the known structure of civil signals, open public standards (which can be easily imitated), the presence of blockers and interference techniques, and the implicit trust of users in the received signal [31,32].

GNSS signals are highly vulnerable to various forms of interference, including natural effects such as ionospheric disturbances or multipath effects [2] and unintentional sources such as RF spectrum congestion [2,33]. However, the most critical threat comes from intentional interference, particularly jamming and spoofing, which are becoming increasingly accessible and harder to detect [34]. Jamming involves transmitting signals at GNSS frequencies with increased noise to overpower satellite signals, resulting in reduced accuracy or total signal loss in the receivers, which can cause the inability to maintain continuous observation in the receiver [35]. Jamming techniques are generally categorized into broadband, narrowband, and continuous wave jamming. Broadband jamming, which covers the entire GNSS frequency band (±10.23 MHz around L1/L2), is simple and effective but difficult to filter due to its broad spectrum. In contrast, narrowband and continuous wave jamming have a higher spectral power spectral density and have a greater impact on receiver performance but can often be mitigated using advanced signal processing techniques [36].

Spoofing involves the intentional transmission of false GNSS signals, resulting in false information about the position, speed, and time of the GNSS receiver. Spoofing requires sophisticated equipment to mimic satellite signals and is therefore more difficult to accomplish but also more difficult to detect than jamming [37]. Spoofing can cause the receiver to estimate its position at a location other than where it is located or at a location where it is located but at a different time (one determined by the attacker) [38]. GNSS spoofers generate stronger GNSS signals, fooling relatively simple GNSS receivers that are pre-programmed to use the signals with the highest possible strength to calculate time and position [39].

In this paper, the main focus is on jamming, specifically on examining the effects of jamming on static GNSS measurements with high-precision geodetic receivers. As previously mentioned, jamming can completely disable positioning and navigation or degrade the received satellite signal, thus affecting the reliability and accuracy of observations. The main parameters that indicate the possible presence of interference in GNSS measurements are the signal-to-noise ratio (SNR), the dilution of precision (DOP), and a reduced number of visible satellites.

In GNSS measurements, one of the main indicators of signal quality is the carrier-to-noise-density ratio (C/N₀), which is the ratio between the carrier power and the noise power spectral density. It is expressed in dB-Hz and calculated according to Equation (1), where Ps stands for the signal power (W) and N₀ for the spectral noise power density (W/Hz) [40,41].

$${\displaystyle \frac{C}{N_0}}=10 {\log }_{10}({\displaystyle \frac{Ps}{N_0}})$$

(1)

C/N₀ is used in GNSS receivers to measure the signal strength of individual satellites and to monitor overall signal quality. On the other hand, SNR, which represents the ratio between the total power of the useful signal and the power of the noise within a certain bandwidth, is often used in statistical analyses, such as the assessment of observation quality. The mathematical relationship between these two quantities is given by Equation (2), where Beff denotes the effective bandwidth, expressed in Hz [42].

$$SNR={\displaystyle \frac{C}{N_0}}\ast B_{eff}$$

(2)

In practice, the SNR is most often expressed in dB, according to Equation (3) [40].

$$SNR [\mathrm{d}\mathrm{B}]={\displaystyle \frac{C}{N_0}}\left[\mathrm{d}\mathrm{B}-\mathrm{H}\mathrm{z}\right]-10 {\log }_{10}\left(B_{eff}\right)$$

(3)

When jammer is present, the effective metric becomes the SINR (signal-to-interference-noise ratio), which takes into account both the noise Pn and the jamming power Pi [43].

$$SINR [\mathrm{d}\mathrm{B}]={\displaystyle \frac{Ps}{Pi+Pn}}$$

(4)

Converting C/N₀ under jamming results in Equation (5):

$${\displaystyle \frac{C}{N_0}}\left[jammed\right]=10 {\log }_{10} ({\displaystyle \frac{Ps}{Pi+Pn}})$$

(5)

This ratio determines the receiver’s ability to maintain lock and make accurate measurements. The degradation of C/N₀ in the presence of jamming can be formally linked to the Spectral Separation Coefficient (SSC), which results from the overlap of signal and jammer spectrum, as shown in recent models for linear chirp jammer. According to [44], the SSC depends on the jammer type and the frequency sweep and is ideally calculated using the spectral overlap. If the jamming power of the spectral overlaps increases, C/N₀ decreases significantly. SSC is represented by Equation (6).

$${\displaystyle \frac{C}{N_0, eff}}=SSC\ast {\displaystyle \frac{C}{N_0, nominal}}$$

(6)

In GNSS measurements, it is usually assumed that a C/N₀ value of more than 40 dB-Hz is a strong signal, while values between 30 and 40 are considered usable but degraded, and anything below 30 is considered a degraded signal. Empirical studies show that GNSS receivers typically lose lock when C/N₀ drops below 27 dB-Hz [45].

In addition to the SNR, one of the interference indicators is the DOP value, which represents the geometric strength of the satellite constellation during positioning. It is derived from the covariance matrix Q (7) of the position solution, where A is the geometry matrix based on the satellite elevation and azimuth angles. The DOP values increase in the event of jamming, when fewer satellites are tracked or the geometry becomes poor. A high PDOP value (>6) indicates the low reliability of the position [40].

$$Q=({A^{T}A)}^{-1}, DOP=\sqrt{trace \left(Q\right)}$$

(7)

Since the number of visible satellites during the observation depends on the signal strength (satellites with C/N₀ greater than the reception threshold (e.g., 30 dB-Hz)) and the satellite geometry, it is to be expected that the number of observed satellites will be reduced during jamming, and thus the DOP value will increase.

On this basis, the basic GNSS signal model, which describes what a GNSS receiver “sees” at its input, can be described as follows (8) [40,43].

$$r\left[k\right]=s\left[k\right]+i\left[k\right]+n\left[k\right]$$

(8)

where r[k] is the total received signal at time k, s[k] is the authentic GNSS signal coming from the satellite, i[k] is the interference signal and is often modelled as a sinusoidal function (CW, chirp, etc.), and n[k] is the additive white Gaussian noise (AWGN), i.e., the receiver thermal noise model. This model forms the basis for most algorithms for jamming detection, signal filtering, and modelling of C/N₀ loss and receiver performance.

As GNSS is essential to modern technology, interference poses a serious risk in all areas that rely on accurate PNT. It can lead to incorrect position and velocity or time calculations, reduced precision, or the inability to calculate position altogether, impacting critical applications such as surveying, especially in time-sensitive projects such as tectonic tracking or RTK [46]. Beyond economic impact, GNSS interference also impacts military operations (e.g., navigation and missile guidance), where it can lead to severe failures [47] and affects emerging technologies such as autonomous vehicles, drones, and robotics by disrupting positioning accuracy (which can lead to accidents or misuse) [48].

Given the importance of unobstructed reception of GNSS signals, interference detection systems are currently being developed, such as the UK Sentinel project, a network that has detected dozens of jamming incidents using a network of sensors in roads [49]. Advanced antennas (e.g., CRPA) are being developed that cancel out the direction from which the jammed signal is coming [50], as well as devices such as the CTL-3520 and systems such as JLOC and Gaardian, which allow jamming to be monitored and eliminated [51]. Receivers equipped with interference mitigation techniques (e.g., Septen-tri) have also been developed, further increasing resilience. Measures to protect against jamming are being implemented in the time, frequency, and spatial domains, and GNSS is increasingly being integrated with complementary sensors such as inertial systems or terrestrial eLoran to increase reliability and redundancy in critical applications. These measures aim to ensure GNSS reliability in highly sensitive applications such as aviation, power grids, and telecommunications, where uninterrupted positioning is essential [52,53]. These measures often play a transitional role, enabling increased resilience of existing systems without requiring a complete redesign of the GNSS architecture or changes to the satellite signals. Such transitional solutions include, for example, the use of multi-frequency and multi-constellation tracking, advanced antennas and spatial filtering, as well as the integration of GNSS with inertial systems or terrestrial navigation networks. The results of this study, in particular the data on distances and conditions under which signal loss or degradation of accuracy occurs, can be used to calibrate and evaluate the effectiveness of these measures under real operational conditions.

Methodology and Conducted Tests

In this study, a mixed methodological approach of theoretical and practical research on the influence of jammers on GNSS signals was applied, and their influence on static GNSS positioning was determined using different geodetic GNSS devices. Figure 1 shows the research flowchart.

The practical research involved a series of tests, starting with an initial phase focused on characterizing the operating mechanisms and effects of the jammers to determine their interference patterns and functional parameters. During the first use of the jammer, it was found that the jammer completely blocks the satellite signals close to the receiver (1–5 m), while its influence is still present at a greater distance from the receiver but then decreases, as assessed by the number of visible satellites observed by the receiver. The first test was used to determine the blackout zone, i.e., the zone within which the jammer completely blocks the reception of GNSS signals on the receiver. In this test, the influence of the SF (single frequency) jammer on the GPS receiver Trimble 4000 SSi was tested, and the test was performed at a distance of up to 55 m from the receiver. The test included two scenarios: a 10-min observation during a complete signal blockage (blackout zone) and a 15-min observation when satellite availability was reduced. In the 15-min observation, the measurements were divided into pre-interference, interference, and post-interference intervals to systematically analyze the receiver response dynamics under different jamming conditions. All observations were performed with relative static positioning, and processing was performed using the CORS (Continuously Operating Referent Station) point ZAGR (Zagreb), which is part of the Croatian permanent station network CROPOS (CROatian POsitioning System) [54]. Baseline processing was analyzed using observational data collected with the Trimble 4000 SSi and the nearby ZAGR CORS station (baseline length < 200 m). The CROPOS network provides users with three services, and this study used the VPPS (High-Precision Real-Time Positioning Service) [54], which provides a network RTK (NRTK) solution. During the processing, various statistical data on satellite visibility, DOP values, RMS (root mean square) and SNR were analyzed and are presented in the research results.

In the second test, the influence of the same SF jammer was tested at a distance of up to 120 m with three receivers, (a Trimble 4000 SSi and two Trimble R8s) to determine impact zone of influence of the jammer (the distance outside the blackout zone where the influence of the jammer is still visible). Relative static observations were performed with two receivers, while the influence on RTK (real-time kinematic) measurements was analyzed with a Trimble R8 receiver. As with the first test, various statistical analyses were performed, and the aim of this test was to determine the zone of influence (the zone in which the influence is noticeable) of the jammer on the GNSS observations. The knowledge gained from these two tests was taken into account in the design of the main test.

In the third test, a multi-frequency (MF) jammer was included in addition to the SF jammer, and the test was performed with the same receivers as in the second test at distances up to 400 m, as the first tests showed that the influence of the SF jammer was noticeable at distances of 120 m. The tests were carried out in the same way as in the second test (relative static positioning and RTK in static mode), and the same analyses and statistical indicators were carried out for both jammers.

Since in the first tests, the effects of the jammer were always analyzed from the same direction (azimuth), the last test was performed to investigate whether the direction/azimuth from which the jammed signal comes has an influence on the GNSS observations. The test was performed in 17 sessions investigating the simultaneous influence of SF and MF jammers placed in different azimuths with respect to the receiver. Figure 2 shows the locations of the tests and the distances of the jammers from the receiver. All tests were conducted in the city of Zagreb, HR.

The relative static method of positioning and RTK using the CROPOS network of stations was used for data collection. The GPS receivers Trimble 4000 SSi (with Compact L1/L2 with ground plane antenna) and GNSS (GPS + GLO) Trimble R8 (multi-frequency with integrated antenna) were used, and data processing was carried out using various software and applications. Trimble Business Center (TBC) v.5.0 was used for processing the relative static observations, and the Trimble Planning Tool (toll under the TBC software) was used to obtain data on satellite elevations and number of visible satellites for all sessions, while the rest of the processing was performed in Excel 365 and MATLAB v.9.14 software, in which various numerical and statistical analyses were performed using the RINEX observation files.

In tests, two types of jammers were used, single-frequency (SF) and multi-frequency (MF) jammers. The SF jammer is designed to be connected to the car’s cigarette socket (Figure 3). This jammer blocks the signals of the L1 frequency band of GPS, GLONASS, QZSS, and the E1 signal of GALILEO and BeiDou-2 so that it has no effect on mobile devices and other electronic devices. According to the technical specifications, the range of the jammer is between 5 and 15 m, but the manufacturer points out that the actual jamming range may vary depending on signal strength, battery capacity, and place of use. The power of the jammer is 0.3 W. The MF jammer has a built-in battery for power supply and is connected to five external antennas (Figure 3). According to the specification, the range of the jammer is 5–30 m, depending on the strength and location of the signal. The built-in battery has a capacity of 8 Ah, which is said to last up to three hours. The device can also be used while charging. It not only interferes with GPS signals but also affects GLONASS, LoJack, and WiFi signals. The total output power is 4.2 W, and the interference frequency range is 173 MHz, 1220–1260 MHz, 1370–1380 MHz, 1570–1620 MHz, 1170–1180 MHz, 2400–2500 MHz.

3. Results

3.1. Blackout Zone and Influence Zone of the Jammer

As described above, the use of jammers can completely block the GNSS signal and degrade observed satellite signals. In the tests carried out, the first two aimed to analyze at what distance from the receiver the use of the jammer has an impact on the GNSS signal received. The first test was used to determine the blackout zone of the jammer, and the second test was used to determine the jammer’s influence zone (the distance up to which the influence of the jammer on the GNSS signal is noticeable).

The blackout zone of the SF jammer was determined in the first test. This test also examined whether the jammer works within the given specifications at 2 to 15 m. The results outside the blackout zone provided information on whether and to what extent the influence of jammers was present in the existing solutions. During the tests, the Trimble 4000 SSi receiver was always placed in the same position, while the distance of the jammer was varied. To determine the blackout zone, the jammer was placed at a distance of 1 m from the receiver until the receiver started to register the satellite signal. The jammer was placed at a distance of 1–15 m, 20 m, 30 m, 40 m, and 55 m, respectively. The observations at a distance of 1–10 m lasted a total of 10 min, half of which was observed with the jammer off and the rest with the jammer on. At greater distances, the observation lasted 15 min, with the jammer off at the beginning of the session, then on for 5–10 min, and then off again. The measurements were carried out in static mode, and the subsequent processing of the baseline was carried out using the CORS ZAGR observation data. The processing was performed using TBC software, and the statistical data for the vector (ZAGR—receiver) of each session was read from the baseline processing reports and the RINEX files, as indicated in

Table 1. Statistical data of testing blackout and the influence zone of the SF jammer.

Session	Jammer Distance (m)	Max Visible Satellites (Jammer Off)	Max Visible Satellites (Jammer On)	Entire Session			With Jammer Only
				Max PDOP	RMS	Solution	Max PDOP	Solution	RMS
S_0	0	7	/	2663	0.011	Fixed	/	/	/
S_1	1	7	0	2759	0.007	Fixed	/	/	/
S_2	2	7	0	2280	0.005	Fixed	/	/	/
S_3	3	7	0	2107	0.005	Fixed	/	/	/
S_4	4	8	0	2296	0.006	Fixed	/	/	/
S_5	5	7	0	2401	0.005	Fixed	/	/	/
S_6	6	7	0	2296	0.005	Fixed	/	/	/
S_7	7	7	0	2146	0.006	Fixed	/	/	/
S_8	8	6	0	2396	0.006	Fixed	/	/	/
S_9	10	6	0	2332	0.006	Fixed	/	/	/
S_10	11–14	8	5	10,670	0.006	Fixed	10,670	Fixed	0.007 m
S_11	15	7	5	6659	0.007	Fixed	6425	Fixed	0.005 m
S_12	20	8	3	10,908	0.007	Fixed	/	Float	/
S_13	30	8	3	30,925	0.007	Fixed	20,249	Fixed	0.007 m
S_14	40	7	5	7046	0.005	Fixed	7046	Fixed	0.004 m
S_15	55	7	5	6000	0.004	Fixed	6000	Fixed	0.004 m

. The blackout zone, as determined in this test, was 13 m. Although no data was recorded during the jamming period, a solution for the jammed session was derived from the non-jammed segments of the session. The TBC software provided a fixed solution calculated based on the parts of the session where the jammer was off. The results of the test, as shown in Table 1, are categorized based on the processing of the entire session (including the jammed and non-jammed intervals) and the processing of only the jammed part of the session.

As can be seen in Table 1, the observation window was shortened for sessions S_10–S_15, where the receiver started to observe the satellite signal (outside the blackout zone) under the influence of the jammer, and processing was performed only for the part of the session under the influence of the jammer. A fixed solution was obtained in all sessions except in session S_12, when the jammer was 20 m away from the receiver. When comparing the PDOP values, no significant changes were observed during the jammed sessions. The maximum PDOP values in the sessions outside the blackout zone were high, ranging from 6 to 30.925. By blocking the satellite signals, the jammer significantly affected the DOP value, and high-quality measurements require low values in addition to a clear horizon. The number of visible satellites for each session was determined by analyzing the observations stored in the RINEX files. Each session within the blackout zone when the jammer was switched on had no visible satellites, while in sessions outside the blackout zone, the number of satellites decreased. Figure 4 shows a comparison of the number of visible satellites per session, depending on whether the jammer was on or not, and comparisons were made with almanac data (on the maximum expected number of visible satellites) obtained using Trimble Planning software. Figure 4 shows that at distances of 15 to 55 m, the number of satellites was lower than expected according to the almanac data. The planning (calculation of the number of visible satellites) was calculated assuming a completely unobstructed horizon above a 10° elevation mask, while the number of satellites actually observed was determined by analyzing the data from the RINEX file.

In addition to the number of satellites, the SNR values were also analyzed during the registration of the satellite signals from the RINEX file. In the sessions within the blackout zone, during the jammer influence, no data was stored, so there were no recorded SNR values (Figure 5—left). Figure 5 (middle and right) shows the SNR values at a distance of 15 m between the jammer and the receiver. In all sessions outside the blackout zone, the results were the same; when the jammer was on, the SNR values were degraded. Figure 5 (center) shows the SNR value on the L1 carrier, where it is noticeable that the recorded SNR values dropped sharply (degradation) when the jammer was turned on and rose again to the initial value when the jammer was switched on. The picture on the right shows the same situation, but on the L2 carrier. The figures refer to the Trimble 4000SSi receiver and show that the average SNR drop for the L1 carrier is 20 dB-Hz, while the average drop for the L2 carrier is slightly lower, at 15 dB-Hz.

In the tests carried out, the blackout zone of the SF jammer’s influence (13 m) was determined, within which no satellite signal data was stored. Furthermore, we conclude that although the receiver stored data outside the blackout zone up to the tested 55 m, the influence of the jammer was recognizable by a reduced number of visible satellites, a consequent increase in PDOP, and/or a degradation in the SNR value.

After the initial tests, further tests were carried out with the SF jammer. Three receivers were used, Trimble 4000 SSi and Trimble R8 (internal designation BR) for static observations and a Trimble R8 receiver (internal designation R1) for static determination of coordinates using the VPPS (High-Precision Real-Time Positioning Service) CROPOS service. In this test, two different types of positioning were tested with the SF jammer at a distance of up to 120 m. The test started with a jammer at 120 m from the receiver, and the jammer was moved closer to the receiver after each session. The sessions lasted 10 min each, with an additional session at the beginning and end of the test in which the jammer was not used. Unlike the first test, all 10 min of each session were fully jammed. The statistical indicators of data processing are shown in

Table 2. Statistical data from testing Trimble 4000 SSi and Trimble R8 receivers with an SF jammer up to 120 m using static relative positioning.

Session	Jammer Distance (m)	Visible SAT (4000 SSi)	Visible SAT (R8)	Max PDOP (4000 SSi)	Max PDOP (R8)	Solution (4000 SSi)	Solution (R8)	RMS (4000 SSi)	RMS (R8)
122_0	/	8	8	5.523	2.162	Fixed	Fixed	0.006 m	0.009 m
122_1	120 m	7	6	4.211	9.313	Fixed	Fixed	0.006 m	0.008 m
122_3	100 m	6	6	6.257	8.765	Fixed	Fixed	0.006 m	0.008 m
122_5	80 m	6	6	5.347	5.347	Fixed	Fixed	0.005 m	0.007 m
122_7	60 m	1	1	/	/	Failed	/	/	/
122_8	50 m	1	0	/	/	Failed	/	/	/
122_9	40 m	3	2	/	/	Failed	/	/	/
122_A	30 m	6	2	/	/	Failed	/	/	/
122_B	20 m	0	1	/	/	/	/	/	/
122_C	15 m	0	0	/	/	/	/	/	/
122_D	10 m	0	0	/	/	/	/	/	/
122_E	5 m	0	0	/	/	/	/	/	/
122_F	/	7	9	2.404	2.086	Fixed	Fixed	0.008 m	0.008 m

. As can be seen from the table, a fixed baseline solution was achieved for static relative positioning at distances greater than 60 m on both devices, and there was no solution at jammer distances up to 60 m. The only difference from the first test is that in this test, the sessions were jammed for their entire duration. For this reason, the receiver was not able to store enough observations to obtain a solution. Comparing the solutions for the different receivers, there are no major differences, and it can be concluded that the receivers are equally sensitive to interference. Table 2 also shows that the PDOP values were relatively high in the sessions in which a fixed solution was obtained compared to the sessions in which the jammer was not used. The RMS values were not significantly degraded in these sessions.

In the same test, the Trimble R8_R1 receiver was used to determine the coordinates in real time with VPPS CROPOS. Before the RTK observations, it is necessary to establish initialization. On the test polygon, initialization was only possible if no jammer was used. Even if the jammer was 120 m away, no initialization was achieved. After a successful initialization with the jammer being off, the RTK positioning was carried out in three repetitions of 10 epochs (seconds) each. The statistics for the RTK observations are shown in

Table 3. Statistical data of testing the Trimble R8 receiver with an SF jammer up to 120 m using the RTK method.

Jammer DIST (m)	Max SAT Visibility	Max PDOP	Solution	SNR
				L1	L2	L2C
/	9	2.017	Fixed
5 m	0	/	Float	/	/	/
10 m	0	/	Float	/	/	/
15 m	2	/	Float	<36	/	<46
20 m	2	/	Float	<35	/	<46
30 m	6	/	Float	<37	<18	<47
40 m	5	/	Float	<38	<18	<47
50 m	6	/	Float	<37	<18	<47
60 m	4	5.4	Float	<38	<20	<47
80 m	5	5.1	Float	<43	<25	<47
100 m	5		Float	<39	<25	<47
120 m	6		Float	<40	<25

. In terms of jammer distance, the results show the maximum number of visible satellites during the observation in each session; the maximum PDOP; the average SNR value on the L1, L2 and L2C carrier; and the solution type, which is fixed only without the use of jammers. Although the receiver was able to detect satellite signals with jammers at a distance of more than 10 m, initialization was not possible, and a fixed solution was not obtained. In addition, the average SNR values (especially on L2) were degraded, making it impossible for the receiver to determine the fixed solution under the influence of jammers. The SNR data was displayed on the screen of the TSC2 controller running the Trimble Survey Controller software v12.45. In addition, the table shows that the SNR values increase with increasing distance from the jammer.

These tests provided the first insights into the influence of jammers on GNSS observations. A blackout zone was determined, and it was evident that the jammer affected the visibility of the satellites, the degradation of the SNR value, and the increase in PDOP at distances up to 120 m. In addition, during the tests, the direction of the SF jammer antenna was set in two different directions. The first direction was vertical (towards the zenith), while in certain cases, the top of the antenna was orientated towards the receiver’s horizon. These tests showed that the influence of the jammer is the same regardless of the direction of the jammer antenna. The results served as input parameters for further tests at greater distances in order to find out at what distances from the receiver and to what extent the SF jammer no longer affects the GNSS measurements.

3.2. Influence of Single-Frequency and Multi-Frequency Jammers on Static Measurements

The third test aimed to analyze the effects of two different jammers, the SF and the MF, on three receivers; it was carried out in the same way as the previous tests. The Trimble 4000 SSi and Trimble R8_BR receivers were placed at two points for relative static positioning, and the Trimble R8_R1 was placed at one point for RTK observation with VPPS CROPOS, which was also used to determine the coordinates of the points and the distances of the jammer to the receiver. In this test, the MF jammer was also used at distances of 15, 30, 50 and 100 m, as its specifications provide for a range of 5 to 30 m. At each distance, the observation was conducted for 10 min, with an additional session at the beginning and end where no jammer was used. All 10 min of the session were jammed.

3.2.1. Influence of MF Jammer

When using MF jammer on the Trimble 4000 SSi and Trimble R8_BR receivers for relative static positioning, both receivers provided solutions at a distance of 30 m from the jammer that were not significantly different from the first session without the jammer. When investigating the effects on RTK measurements (Trimble R8_R1), initialization had to be performed first. In RTK mode (Trimble R8_R1), the solutions were fixed at distances of 30 m and more, and the MF jammer did not affect the positioning. At a distance of 15 m, three satellites were visible, and the solution was float. The statistics of these observations (RTK and relative static positioning) are shown in

Table 4. Statistical data for sessions up to 100 m using the MF jammer.

DIST (m)	Trimble 4000 SSi					Trimble R8_BR					Trimble R8_R1
	Session	MAX VIS SAT	MAX PDOP	Sol.	RMS (m)	Session	MAX VIS SAT	MAX PDOP	Sol.	RMS (m)	MAX VIS SAT	Sol.
/	MF_1	8	2.037	Fixed	0.010	MF_6	8	2.037	Fixed	0.010	9	Fixed
15	MF_2	3	/	/	/	MF_7	3	/	/	/	3	Float
30	MF_3	8	3.001	Fixed	0.011	MF_8	8	3.001	Fixed	0.011	7	Fixed
50	MF_4	7	2.977	Fixed	0.009	MF_9	7	2.977	Fixed	0.009	8	Fixed
100	MF_5	7	2.379	Fixed	0.013	MF_	7	2.379	Fixed	0.013	9	Fixed

. The number of satellites recorded for relative static positioning was determined from the RINEX file, while the number of satellites recorded for RTK positioning was obtained from the JOB file.

In addition to the statistical indicators listed in Table 4, the number of visible satellites for each session was analyzed for relative static positioning. The satellite visibility information was obtained by comparing the data from the RINEX observation file with the data from the Trimble Planning output (almanac data). Figure 6 shows the number of satellites on the Trimble 4000 SSi receiver in a session without a jammer (MF_1) and in a session where the jammer is 30 m away from the receiver (MF_3). The same sessions were compared for the Trimble R8_BR receiver (MF_6 and MF_8).

Compared to the figures in Figure 6, the visibility of the satellites did not change significantly when the jammer was at a distance of 30 m. In some parts of the session, the number of satellites dropped to seven, but this had no influence on the positioning. However, when comparing the number of visible satellites from RINEX and the almanac data, it was noticeable that a larger number of satellites from the almanac data were displayed throughout the session. In session MF_1, where no jammer was used, only one satellite out of a possible nine was not stored by the receiver. In session MF_3, where an MF jammer was used at a distance of 30 m, 10 satellites were visible in some parts of the session, while the receiver stored the signals of 7 satellites. The satellite signals that the receiver did not store have lower elevations (G19 = 11°, G25 = 58°, and G31 = 11°), which can be seen in Figure 7.

Figure 8 shows the SNR values of the observed satellites in this session, where it can be seen that satellites with low SNR were no longer recorded by the receiver (less than 15 dB-HZ), and the SNR was degraded for the remaining satellites. It should also be noted that the difference between the number of visible satellites based on the almanac data and the number of recorded satellites with the jammer off may be due to the actual horizon. The planning was carried out under the assumption of a completely unobstructed horizon, while the observations were conducted under real horizon conditions. The Trimble R8_BR receiver had the same number of visible satellites in both sessions (with and without a jammer); this is the reason why the investigation of an MF jammer lasted up to 100 m. Furthermore, a comparison of the number of visible satellites from the RINEX file with the almanac data shows no difference; there were nine visible satellites during most of the first session and during all of session MF_6. So, at 30 m, an MF jammer had no effect on the visibility of the satellites in the Trimble R8_BR receiver.

The SNR values were also analyzed. The SNR values of the Trimble 4000 SSi receiver in session MF_1, in which no jammer was used, are shown in Figure 8. The top left image shows the SNR value on the GPS L1 carrier, and the top right image shows that on the GPS L2 carrier. In the session, eight satellites were visible, with SNR values between 7 and 35 dB-HZ. The same figure (below) shows the SNR values when the jammer was 30 m away from the receiver. The figure on the bottom left shows the SNR values on GPS L1, and the right figure shows the values on the GPS L2 carrier. The SNR values range from 1 to 12 dB-HZ, which shows that the SNR value was degraded during the jammed observation. The influence of the jammer on the SNR value was higher on the GPS L2 carrier, where the SNR was recorded for only three satellites with extremely low values and short storage intervals.

A similar analysis was performed for relative static positioning with Trimble R8_BR and is shown in Figure 9. The SNR values during the session without interference are shown on the top left of the L1 carrier, and the top right shows values for the L2 carrier. In the session without a jammer, nine satellites were visible, with SNR values between 35 and 55 dBHz on the L1 carrier and between 20 and 50 dBHz on the L2 carrier (top left and top right figures). In the session where the jammer was used at a distance of 30 m, eight satellites were visible with SNR values between 25 and 45 dBHz on the L1 carrier and between 15 and 30 dBHz on the L2 carrier (bottom left and right images). There was also an imperceptible drop in SNR on both carriers, although this was not as sharp as in the case of the Trimble 4000 SSi receiver.

When comparing the SNR values in Figure 8 and Figure 9, which show the effects of a jammer on two different GNSS receivers (Trimble 4000 SSi—Figure 8 and Trimble R8—Figure 9) at the same distance from the jammer, it is evident that the receivers have different levels of sensitivity to interference. The results clearly show that the Trimble 4000 SSi experiences greater signal degradation. This device is not equipped with integrated anti-jamming technology, adaptive filtering, interference detection, or advanced signal processing and only supports tracking of a single channel per satellite, making it highly vulnerable to jamming. Its protection depends entirely on the quality of the antenna and the measurement environment. In this study, the Trimble 4000 SSi was used with a compact L1/L2 antenna with a ground plane, which only provides passive protection against low-elevation interference. In contrast, the Trimble R8 with integrated antenna offers better tracking dynamics, multiple channels per satellite, and integrated digital filtering. Its primary protection against interference is to monitor the drop in C/N₀ values, which provides moderate resistance to interference detection [55,56]. The difference in performance between the two receivers is clearly visible in the graphs provided. As interference becomes more common in modern GNSS applications, newer receivers are equipped with more advanced protection mechanisms. These include multi-frequency tracking, which allows receivers to compensate for degraded signals on one band by using other bands, advanced digital filtering and signal processing, and CRPA or other specialized antennas that reduce the effects of interference. Most detection techniques are based on sudden drops in C/N₀, satellite losses and anomalies in DOP values. Each manufacturer implements its own algorithms for interference detection and signal suppression [57].

3.2.2. Influence of the SF Jammer

After the tests with an MF jammer, the influence of an SF jammer was tested at distances of up to 400 m. The tests were carried out until a fixed solution was obtained by RTK observation. The observations were processed in the same way as for the MF jammer. The statistical indicators of the relative static observations (Trimble 4000 SSi and Trimble R8_BR) and the RTK observations (Trimble R8_R1) are shown in

Table 5. Statistical data for observation sessions up to 400 m using the SF jammer.

DIST (m)	Trimble 4000 SSi					Trimble R8_BR					Trimble R8_R1
	Session	MAX VIS SAT	MAX PDOP	Sol.	RMS (m)	Session	MAX VIS SAT	MAX PDOP	Sol.	RMS (m)	MAX VIS SAT	Sol.
/	SF_1	8	2.037	Fixed	0.010	SF_10	9	1.981	Fixed	0.010	9	Fixed
15	SF_2	0	/	/	/	SF_11	0	/	/	/	2	Float
30	SF_3	4	9.306	Fixed	0.011	SF_12	2	/	/	/	3	Float
50	SF_4	4	7.252	Fixed	0.011	SF_13	2	/	/	/	4	Float
100	SF_5	5	6.821	Fixed	0.015	SF_14	3	/	/	/	3	Float
150	SF_6	6	8.371	Fixed	0.016	SF_15	8	3.602	Fixed	0.025	4	Float
200	SF_7	6	3.134	Fixed	0.018	SF_16	7	4.543	Fixed	0.021	5	Fixed
300	SF_8	6	3.180	Fixed	0.018	SF_17	7	2.430	Fixed	0.019	6	Fixed
400	SF_9	7	2.354	Fixed	0.020	SF_18	8	2.343	Fixed	0.011	8	Fixed

. With the Trimble 4000SSi receiver, a fixed solution was obtained at 30 m, but at up to 300 m, the jammer affected the number of visible satellites. With the Trimble R8_BR receiver, the solution was fixed at 150 m, while the number of visible satellites was affected at up to 300 m. With RTK positioning, successful initialization was achieved when the jammer was 200 m away from the receiver. The tests were carried out up to a distance of 400 m, as at 200 m, when the initialization was established, only five out of nine possible satellites were visible (according to the almanac data). At a distance of 400 m, the number of satellites with the fixed solution was eight, which is only one less than the almanac data, so the impact of the jammer was reduced. The decrease in the number of visible satellites consequently led to an increase in the PDOP values. In contrast to the MF jammer, the SF jammer had a greater impact on the Trimble R8 than on the Trimble 4000 SSi, as can be seen from the fixed solutions obtained.

The SF jammer was also used to analyze the visible satellites on each receiver in relation to the almanac data. Figure 10 shows the number of visible satellites from the RINEX file of the Trimble 4000SSi receiver compared to the almanac data for the sessions in which the jammer was at 300 m (session SF_8) and 400 m (SF_9). In both sessions, the number of satellites observed by the receiver was one less than the almanac data. The same analysis was performed for the Trimble R8_BR. In session SF_17, when the jammer was 300 m away, and in session SF_18, when the jammer was 400 m away from the receiver, the number of visible satellites observed by the receiver was equal to the almanac data.

The SNR values were evaluated in further analysis. Figure 11 shows the SNR values in session SF_8 when the jammer was 300 m away from the Trimble 4000 SSi receiver. The top left image shows the SNR value on the L1 carrier, and the top right image shows the SNR value on the L2 carrier. In this session, six satellites were visible, with SNR values between 0 and 25 dBHz. The same image shows the SNR values in session SF_9, when the jammer was 400 m away. The bottom left image shows the SNR value on the L1 carrier, and the bottom right shows the same value on the L2 carrier. In this session, seven satellites were visible, with SNR values between 0 and 30 dBHz. As there was no significant influence on the number of visible satellites at an altitude of 400 m, the SNR values were still low. The figures also show some jumps in the SNR values, possibly caused by the chainsaw chirp characteristic of the jammer signal. Satellite elevations in the same sessions are also shown (Figure 12), where some satellites have elevations of less than 20°. The relationship between SNR and the elevations of individual satellites is explored below.

The same analysis was performed for the Trimble R8_BR receiver when the jammer was at 300 m (session SF_17) and 400 m (session SF_18). The SNR values are shown in Figure 13. The top left image shows the SNR value on the L1 carrier, and the top right image shows the L2 carrier in session SF_17. In this session, seven satellites were visible, with SNR values between 30 and 55 dBHz on the L1 carrier and between 15 and 50 dBHz on the L2 carrier. The image on the bottom left shows the SNR values of the jammer at 400 m on the L1 carrier, and the bottom right shows the SNR values on the L2 carrier in the same session. In this session, eight satellites were visible, with SNR values between 15 and 50 dB. There was no significant difference in the SNR values in the analyzed sessions. This is a further indication that the Trimble R8 receiver was not influenced by the SF jammer at these distances.

3.2.3. Correlation of the SNR Values and Satellite Elevation

When processing the results, a correlation was found between the influence of jammers on the SNR values and the elevations of the associated satellites. Satellites with lower elevations had lower SNR values and vice versa. In the sessions in which the jammer did not completely overwhelm the satellite signals, the SNR values of the satellites with lower elevations degraded the most. Moreover, if the satellite was at a lower elevation, its signal was completely overrun in some cases when the jammer was used. The correlations between the SNR values and the satellite elevation were calculated for the sessions in which both jammers (SF and MF) were used and separately by session for the Trimble 4000SSi and the Trimble R8_BR receiver. The correlation coefficients were calculated based on the mean values of SNR and elevations for each satellite in each session.

Table 6. Correlation coefficient of SNR and elevation for the SF and MF jammer using Trimble 4000 SSi and Trimble R8.

Jammer DIST [m]	Correlation Coeff. (SF)		Correlation Coeff. (MF)
	4000SSi	R8_BR	4000SSi	R8_BR
30	0.85	/	0.95	0.93
50	0.91	/	0.94	0.93
100	0.99	/	0.97	0.94
150	0.98	0.93	/	/
200	0.97	0.93	/	/
300	0.98	0.92	/	/
400	0.95	0.89	/	/
/	0.78	0.88	0.95	0.93

shows the correlation coefficients for all sessions. A comparison of the correlations of the sessions under the influence of the jammer and the sessions without a jammer showed no significant difference.

In order to visualize the correlation graphically and numerically more clearly, one session of each receiver under the influence of the MF jammer was taken and compared with the session without the jammer. Figure 14 shows a tabular and graphical representation of the mean SNR values and elevations for individual satellites in sessions MF_1 (a session without a jammer) and MF_3 (a session in which an MF jammer was used at a distance of 30 m from the receiver). The data was obtained with the Trimble 4000SSi receiver.

Figure 15 shows tabular and graphical representations of the mean SNR values and elevations in sessions in which an SF jammer was used. Figure 15 refers to session SF_10, when the jammer was not used, and SF_17, when the jammer was used at 300 m from the receiver. The sessions refer to data obtained with the Trimble R8 receiver.

Finally, the linear regressions between the SNR and the satellite elevations, the equations of the regression lines, and the coefficients of determination (R2) are presented. Figure 16 shows the regression lines between SNR values and elevations for individual satellites in sessions MF_1 (no jammer) and MF_3 (MF jammer at 30 m). These sessions refer to data obtained with the Trimble 4000SSi receiver.

Comparing the correlations between these two sessions in which a jammer was used at a distance of 30 m from the receiver (blue regression line) and where no jammer was used (orange regression line), it is noticeable that the correlation between SNR and satellite elevation is better when the jammer is used, which is also shown by the coefficient of determination, which is 0.37 for the session without a jammer and 0.91 for the session with a jammer.

Figure 17 shows the linear regression in session SF_10 (a session without a jammer) and SF_17 (a session in which an SF jammer was used at 300 m) with the Trimble R8_BR receiver.

The blue regression line refers to the session without a jammer, and its coefficient of determination is R2 = 0.78, while the orange regression line refers to the session where an SF jammer was used at 300 m from the Trimble R8_BR receiver, and its coefficient of determination is higher, at R2 = 0.85.

The calculated correlation coefficients in Table 6 and the linear regression with the coefficients of determination show that the SNR values and the elevations of the individual satellites are closely related. The correlations obtained in almost every session with both jammers for both receivers are close to 1. It can be concluded that in almost every session, a satellite with a lower elevation has a lower SNR value and vice versa, and if the satellite has a lower SNR value initially, it is more likely to be completely overwhelmed by the influence of the jammer.

3.3. Jammer’s Impact in Relation to the Azimuth

This study also tested the influence of the direction/azimuth in which the jammer is placed and how the different directions from which the signal from the jammer comes influence the GNSS observations. In the above tests, the direction of the jammer signal was always in the same azimuth relative to the GNSS receiver. The Trimble R8 device and the same two jammers were used for data collection. The measurements were performed using relative static positioning, and a total of 17 sessions were observed. The first session lasted 15 min and was conducted without a jammer. The next 16 sessions were measured under the simultaneous influence of an SF and MF jammer. The measurements were divided into four parts, depending on the azimuth at which the SF jammer was placed in relation to the receiver. The first part of the measurement refers to seven sessions in which the position of the SF jammer was fixed at an azimuth of 180° with respect to the receiver, while the position of the MF jammer was variable in each session; it was 0° during the first session and increased by 45° in each session until the last session in which the azimuth was 315°. In the second part of the measurement, the SF jammer was placed at 270°, and the MF jammer changed position at 0°, 90°, 180° and 270°. The third part refers to sessions where the SF jammer was at 0° and the MF jammer was at 0°, 90° and 270°. In the fourth part, both the SF and MF jammers were at an azimuth of 90° to the receiver.

The distances between the jammer and the receiver are based on earlier tests. In the first test, the jammers were set up at a distance of 40 m for the SF jammer and 20 m for the MF jammer. With this combination of distances, the receiver was unable to observe data, as the number of visible satellites dropped to zero. The jammers were then moved until a combination was found at which the receiver could store the solution. Accordingly, the SF jammer was placed at 60 m in all sessions, while the MF jammer was placed at 40 m in all sessions except the session in which the jammers were placed at an azimuth of 270°/180°. From analyzing the jammer distance, it can be concluded that the combined effect of the MF and SF jammers extends the range of the interference more than either jammer alone. During the 270°/180° session, a distance of 40 m was not sufficient for the MF jammer as the receiver could not observe enough satellites due to the interference. It was therefore moved a further 5 m and placed 45 m away from the receiver. All observed vectors were calculated in TBC using the generated VRS (Virtual Reference Station) point of the CROPOS grid near the observation location. The GNSS device observed GPS and GLONASS satellites simultaneously in all sessions. The first session, in which no jammer was used, was observed for 15 min, while all other sessions were observed for 6 min with the jammer switched on for the entire duration of the session.

Table 7. Statistics for testing the influence of the azimuth of the SF and MF jammer.

Session	SF Jammer		MF Jammer		MAX VIS SAT	∆Hp	∆Vp	RMS	∆PDOP
	Azimuth	Distance	Azimuth	Distance
S0	/	/	/	/	14	0.0028	0.0037	0.0082	2.728
S1	180°	65 m	0°	40 m	14	+0.0046	+0.0049	+0.0073	+6.139
S2	180°	65 m	45°	40 m	10	+0.0051	+0.0110	+0.0010	+0.738
S3	180°	65 m	90°	40 m	9	+0.0030	+0.0114	+0.0008	+0.118
S4	180°	65 m	135°	40 m	11	+0.0015	+0.0083	+0.0025	+0.948
S5	180°	65 m	180°	40 m	12	+0.0024	+0.0130	+0.0003	+0.122
S6	180°	65 m	225°	40 m	14	+0.0062	+0.0286	+0.0020	+5.510
S7	180°	65 m	270°	40 m	14	+0.0028	+0.0059	+0.0015	+0.231
S8	180°	65 m	315°	40 m	12	+0.0033	+0.0066	+0.0007	+1.021
S9	270°	65 m	180°	45 m	12	+0.0142	+0.0161	+0.0013	+7.013
S10	270°	65 m	270°	40 m	12	+0.0053	+0.0047	+0.0014	+0.954
S11	270°	65 m	0°	40 m	12	+0.0044	+0.0046	+0.0016	+0.946
S12	270°	65 m	90°	40 m	11	+0.0086	+0.0106	+0.0041	+0.286
S13	0°	65 m	270°	40 m	9	+0.0111	+0.0281	+0.0064	+0.733
S14	0°	65 m	0°	40 m	9	+0.0098	+0.0245	+0.0040	+0.522
S15	0°	65 m	90°	40 m	9	+0.0095	+0.0223	+0.0054	+0.402
S16	90°	65 m	90°	40 m	11	+0.0076	+0.1373	0.0015	0.998

provides an insight into all sessions, the distances of the jammers, and the azimuths at which they were located in relation to the receiver, the maximum number of visible satellites in each session, the PDOP, and the RMS in relation to the session without jammers. Also shown are the differences in horizontal (∆Hp) and vertical (∆Vp) accuracy, which represent the change in horizontal and vertical accuracy due to the presence of interference. These parameters are automatically calculated by the receiver and refer to the estimated horizontal and vertical position accuracy at each epoch. These indicators provide a statistical estimate of the reliability of positioning in real time and are influenced by satellite geometry, signal quality, and the presence of interference. All solutions were fixed, given that the measurements were taken at distances at which the receiver could store the observations, as described above.

The statistical data do not show any major deviations in terms of horizontal and vertical accuracy and RMS. The number of visible satellites varied depending on the session from a minimum of 9 to a maximum of 14. The lowest number of satellites was in sessions with the jammer azimuths 180°/90°, 0°/270°, 0°/0°, 0°/90°. There were greater deviations in the PDOP values. Compared to the first session without the jammer, the largest increase in PDOP, being greater than 1, was recorded for the following sessions: 180°/0°, 180°/225°, 180°/315°, 270°/180°. If we look at Figure 18, which shows the sky plot (display of all satellites and their azimuths during the observation), most of the satellites were close to the 270° azimuth. If we consider that the 270°/180° session is the one in which the PDOP increase is the highest and the only session in which the fixed solution at 40 m was not obtained, we can conclude that this combination of azimuths is the most affected by jammers.

As with the previous tests, the SNR value was also analyzed here. The average SNR values on the L1 and L2 carriers were calculated separately for each session. As can be seen in

Table 8. SNR values for the session where the jammers were placed at different azimuths.

Session	SF Jammer		MF Jammer		AVG SNR L1	AVG SNR L2
	Azimuth	Distance	Azimuth	Distance
S1	180°	65 m	0°	40 m	36.03	27.19
S2	180°	65 m	45°	40 m	34.00	17.00
S3	180°	65 m	90°	40 m	35.10	19.52
S4	180°	65 m	135°	40 m	36.75	25.42
S5	180°	65 m	180°	40 m	33.95	21.04
S6	180°	65 m	225°	40 m	35.50	24.60
S7	180°	65 m	270°	40 m	32.70	23.36
S8	180°	65 m	315°	40 m	33.37	19.59
S9	270°	65 m	180°	45 m	32.70	23.36
S10	270°	65 m	270°	40 m	34.70	27.81
S11	270°	65 m	0°	40 m	36.41	29.81
S12	270°	65 m	90°	40 m	34.95	29.81
S13	0°	65 m	270°	40 m	32.80	27.61
S14	0°	65 m	0°	40 m	34.50	30.00
S15	0°	65 m	90°	40 m	32.80	27.40
S16	90°	65 m	90°	40 m	34.33	28.18

, the greater influence of the jammer was on the L2 carrier, where the values were significantly lower compared to the L1 carrier. Of the L1 and L2 carriers together, session S2 had the lowest SNR value, where the jammers were placed at an azimuth of 180°/45°.

When the SNR values are analyzed in more detail and plotted, it can be concluded that the satellites that were at approximately the same azimuth as the jammer had lower SNR values, and in some cases, the satellite signal was completely overwhelmed by the jammer. Session S9 is shown as an example. In this session, the jammers were located at azimuths of 270°/180° and at distances of 65 m and 45 m. In Figure 19, the SNR values of the L2 carrier were lower compared to the L1 carriers of other sessions, but when looking at the SNR values of the L2 carrier, it is clear that the SNR values were extremely low, often dropping to 0. If we compare this with Figure 18, it is clear that the lowest SNR values, i.e., the satellites whose SNR drops to 0, are at azimuth 180° and close to it. The PDOP values are highest in the same session.

4. Discussion

The conducted tests aimed to investigate the effects of signal interference from jammers on static GNSS observations, focusing on the effects of SF and MF jammers at different distances from the receivers. This research provides insights into the vulnerability of GNSS signals and the potential risks posed by intentional signal interference.

The tests showed that the use of jammers, especially near receivers, can significantly interfere with the reception of GNSS signals. During the test, the first task was to set up a blackout zone within which the jammer completely blocked the reception of satellite signals and no satellites were recorded. This zone was set at 13 m for SF jammers. Within the blackout zone, it is not possible for the receiver to determine a position by relative static observation, as the number of visible satellites drops to zero. Furthermore, the investigation showed that the influence of jammers extends beyond the blackout zone and affects the number of visible satellites, the SNR values, the PDOP values, and the determination of a fixed position. In initial tests, the SF jammer was tested at distances of up to 55 m. Outside the blackout zone, the number of visible satellites was reduced compared to the almanac data; the SNR value of the signals at the receiver were degraded; and the PDOP value increased significantly. Stable and high SNR values and PDOP values of less than 6 are crucial for accurate positioning. By extending the tests to 120 m, it was concluded that at distances up to 60 m, the receiver was not able to achieve a fixed solution for relative static observations when the entire observation session was jammed. At distances up to 120 m, the influence of the jammer was clearly visible in the values of SNR, PDOP, and the number of visible satellites. In other words, the receiver did not record satellite observation data; therefore, this data cannot be used for baseline calculation or relative static positioning. The same test was performed for the RTK observation, which could not initialize the measurement and achieve a fixed position at all distances up to 120 m. Analysis of the data obtained with different receivers, such as the Trimble 4000 SSi and the Trimble R8, showed differences in susceptibility to jamming signals, with the Trimble R8 having a higher sensitivity to jammer interference, which was to be expected as it is a newer device with more advanced technological solutions.

In the tests with MF jammers, both static data collection and RTK measurements were only minimally affected at distances greater than 30 m, with fixed solutions still possible. However, the degradation of SNR values and reduced satellite visibility indicated a decline in signal quality during the jammed sessions, highlighting the need for robust anti-jamming mechanisms to ensure signal integrity and accuracy in positioning applications. On the other hand, the tests with an SF jammer showed a stronger degradation of GNSS observations, especially at greater distances. With the Trimble 4000 SSi, fixed solutions could be achieved from 30 m and with the Trimble R8 from 150 m. In the RTK case, initialization was obtained at 150 m. The tests were carried out up to 400 m. Despite the fixed solutions achieved beyond 150 m, the reduced number of visible satellites and degraded SNR values were visible for all receivers. Looking at the carrier phases of the satellite signal, the interference had a stronger effect on the L2 carrier, where the SNR values dropped more sharply compared to L1.

Statistical analysis has shown that the degradation of the SNR value is greater for satellites at lower elevations. To prove this, the correlation between the SNR value and the elevation of the satellite was analyzed. Correlation coefficients and linear regressions with the coefficients of determination were calculated, indicating that the SNR values and the elevations of the individual satellites are closely related. Correlations were found in almost every session for both the receivers, and the jammers were close to 1, indicating that a satellite with a lower elevation has a lower SNR value, and if the satellite has a lower SNR value, it is more likely to be completely overwhelmed by the influence of the jammer.

As the jammed signals came from the same direction at all tested distances, additional tests were conducted to investigate the effects of jammers coming from different directions or azimuths. In this test, both jammers were used simultaneously. The results show that their combined use increases the overall impact on the observations and that the direction from which the jammed signal comes has no significant impact on the observations. There were some isolated cases where a satellite at lower elevations with an azimuth of +/−30° (with respect to the azimuth of the jammer) had a lower SNR value than other satellites. However, regardless of the azimuth of the jammer, jammers have a stronger effect on satellites with low elevation. If a pattern is to be established to reduce the influence of jammers on satellites, more weight should be given to the elevation of the satellite than to the azimuth of the jamming satellite. It can also be concluded that the antenna of the jammer is omnidirectional.

In the context of related studies, previous work examining GNSS receivers for intentional jamming has utilized controlled laboratory environments or limited field trials with a single-jammer configuration. The results of this work extend existing knowledge by providing distance-based vulnerability, elevation sensitivity analysis, and azimuthal impact assessment under realistic field conditions while recording SNR, PDOP, and visible satellite changes for high-accuracy geodetic receivers. Compared to Bažec et al. [28], who tested nine geodetic receivers under L1/E1 chirp jamming in controlled environments, these experiments extend the analysis range up to 400 m and distinguish between complete blackout and partial influence zones, providing useful data for elevation-based satellite weighting in transitional mitigation strategies. In contrast to Pavlovčič-Prešeren et al. [29], who found stronger effects of vertical interference, our tests with omnidirectional jammers did not show significant azimuthal distortion but confirmed a higher susceptibility of low-elevation satellites, suggesting that elevation filtering and multi-frequency redundance should be prioritized in the mitigation of interference. While Pavlovčič-Prešeren et al. [30] have shown that dual-frequency smartphones can support jammer localization, our discovery of weak residual signals at certain distances suggests the potential for similar hybrid detection–navigation solutions. In line with other related work [25,26,27,28,29,30] on CRPA antennas, spatial filtering, spectral processing, and multi-constellation tracking, the empirical parameters identified here—such as interference distances, C/N₀ degradation rates, and PDOP increases—provide operationally relevant constraints for the calibration and validation of transitional protection measures. By linking these observations to practical mitigation measures, the study contributes both theoretical depth and practical guidance for improving the resilience of GNSS until fully robust architectures are deployed.

These findings emphasize the critical importance of developing effective protection mechanisms against signal interference from jammers, and this research contributes to ongoing efforts to improve the resilience of GNSS-based systems to unintentional interference. The study emphasizes the urgent need to continuously develop interference protection techniques to ensure the integrity and reliability of GNSS signals in the face of the growing threat of intentional signal jamming.

5. Conclusions

This study investigated the effects of intentional GNSS signal interference, through jamming, on static GNSS measurements. Our research focused on evaluating the effects of two types of jammers, single-frequency (SF) and multi-frequency (MF), on GNSS signal reception and provided insights into the vulnerability of GNSS systems to jamming attacks. The results showed the susceptibility of GNSS receivers to jammers, with signal degradation occurring within certain distances from the jammer. Through a series of controlled experiments, critical zones were identified where signal interference significantly disrupts GNSS signal reception and position accuracy. The experiments identified two critical zones: the “blackout” zone where the GNSS signal was completely disrupted, and the “influence” zone where signal degradation and a reduced number of satellites were observed but not completely lost. The SF jammer created a blackout zone of up to 13 m, while its influence extended up to 300 m, depending on the type of GNSS receiver used and the type of positioning. With the MF jammer, the signal interference was observed up to 30 m. The effects included a reduction in the number of visible satellites, increased PDOP values, and degraded signal-to-noise ratio (SNR) values, all of which had a negative impact on positioning accuracy. This study also showed that different GNSS receivers had different sensitivities to jammers and that both static relative positioning and real-time kinematic (RTK) observations were affected. The Trimble 4000 SSi and Trimble R8 receivers used in this study experienced signal degradation and a reduction in the number of satellites stored/recorded during relative static observation, but the effects were more pronounced in the real-time observations, where initialization often failed under jamming conditions. In addition, the analysis revealed a correlation between SNR values and satellite elevation, suggesting that satellites at lower elevations are more susceptible to signal degradation during jamming. When analyzing the SNR values of satellites located at the same azimuth as the line from the jammer to the re-receiver, no regularity was found that would indicate more interference from satellites positioned within 30° of the jammer azimuths than others. However, it should be emphasized that this observation applies specifically to the jammers used in this study, which were omnidirectional devices emitting signals uniformly in all directions. Therefore, this conclusion may not be applicable to directional jammers or other sources of interference. By comparing the SNR values of satellites at jammed azimuths and satellites that are not jammed but are at approximately the same elevations, it was found that regardless of azimuth, the interference is stronger at low satellite altitudes. In addition to the characterization of blackout and influence zones for SF and MF jammers, this study also provides comparative insights into existing transitional protection methods and their possible optimization. By linking our field-based results with previous studies and mitigation concepts, the study shows how empirical distance- and elevation-based vulnerability can inform the design of anti-jamming strategies such as elevation-dependent satellite selection, multi-frequency redundancy, and hybridized navigation. These findings not only fill the current gap in operationally relevant interference testing but also provide a practical basis for validating transitional measures prior to the implementation of fully robust GNSS infrastructures. When establishing a pattern to reduce the influence of signal jammers on satellites, more weight should be given to the elevation of the satellite than the azimuth of the jamming satellite. The research findings herein emphasize the need for improved anti-jamming techniques to protect critical GNSS-dependent applications as intentional signal jamming becomes more common. The findings also highlight the growing need for effective detection and mitigation of jamming threats, given the increasing reliance on GNSS for various critical applications, including transport, telecommunications, and public safety. The insights gained from this research can help in developing better strategies to protect GNSS signals and ensure the integrity and accuracy of positioning data, especially in scenarios where GNSS reliability is of paramount importance.

6. Future Work

This research represents the first in a series of studies on the impact of jamming on the precise GNSS geodetic receivers involved in static relative and RTK positioning. As most geodetic works are performed using RTK positioning relying on national permanent networks (e.g., CROPOS in Croatia), future research will investigate the impact of jamming on reference stations of permanent networks and how this affects the users.