1 INTRODUCTION

he Satellite Navigation System Galileo was launched

in a joint project between the European Union's

Satellite Navigation and the European Space Agency.

It is part of the Trans-European Transport Network,

which aims to solve navigational and geographic

issues. The Galileo project is an ambitious European

venture aimed at creating the most advanced global

positioning satellite system in the world. Its objectives

are to create an autonomous system that provides

guaranteed global positioning services, as well as

interoperable compatibility with other global

positioning systems, such as GPS and GLONASS. In

this work, we will focus on the signals transmitted by

GNSS Galileo. We will briefly describe the individual

signals. We will show and mathematically analyze

signal models. For each signal, we will visualize and

describe the structure of the signal. The signals of

Galileo or other navigation systems have already been

described in some literature. In the literature [3,4],

models of measurement signals for some navigation

systems are derived. The authors used these models

to evaluate the accuracy of navigation systems. At the

beginning of the system Galileo development, the

system structure and signal models were described in

the literature [9], where the author described the

frequency plan and signal structure. The atmosphere

of the Earth also has a great influence on the accuracy

of signals. The author of the article [1,2] describes how

signals behave when passing through the ionosphere.

In the article [7,8], the author describes how it is

possible to use more than three frequencies for

decimeter positioning accuracy using Galileo and

BeiDou signals. Models of measurement signals for

selected communication systems are presented in the

literature [5]. The mentioned models make it possible

to simulate the signal processing of communication

systems under interference conditions. Galileo's

navigation signals are coherent and transmitted across

three frequencies in the L band, namely E1, E5, and

E6. The E6 signal is transmitted at a carrier frequency

Model of the Signal of the Galileo Satellite Navigation

System

. Džunda, S. Čikovský & L. Melníkova

Technical University of Kosice

, Kosice, Slovakia

ABSTRACT: This article presents an analysis of the Galileo E1 signal and its sensitivity to different types of

interference. The research involved modeling white noise, chaotic impulse interference, and narrowband

interference and the effects of these interfering signals on the E1 signal. Based on the available information,

spectral structures were created for the mentioned types of interference, and subsequently, these interferences

were integrated into the E1 signal in the Matlab program environment. A Kallman filter was used to filter out

white noise from the additive mixture of the E1 signal and white noise. The research aimed to analyze the

influence of white noise, chaotic impulse interference, and narrowband interference on the spectral power

density of the E1 signal. The results of this work can be used in the design of robust receivers and signal

structures capable of withstanding these types of interference.

http:

//www.transnav.eu

the International Journal

on Marine Navigation

and Safety of Sea Transporta

tion

Volume 17

Numb

er 1

March 2023

DOI: 10.12716/1001.17.01.

of 1278.75 MHz and consists of three E6 components:

E6-A, E6-B, and E6-C. On the other hand, the E5

signal, which has a center frequency of 1191.795 MHz,

includes two separate signals called E5a and E5b.

These two signals share a carrier frequency of 1176.45

MHz but are modulated independently. The E5a data

and pilot components can be found below 15.345

MHz on the E5 carrier frequency. Meanwhile, the E5b

signals are modulated on two different carrier

frequencies within the E5 band, allowing them to be

monitored separately [6]. Long codes can be useful in

monitoring weak signals, such as those found inside

buildings, but they can be difficult to obtain because

the receiver detects signals by looking for delays in

the received code, and long codes have more options

than short codes. Short codes are good for quick fixes,

but they can lead to incorrect satellite positioning

when the receiver is exchanging signals between two

satellites. This is because the receiver's ability to

distinguish between two different codes is inversely

proportional to the length of the codes. Signal length

may not be suitable for all types of users, with internal

and static users preferring long codes while external

and fast-moving users preferring shortcodes. To

address this issue, alternative codes with different

properties have been provided for different Galileo

signals. This is one of the reasons why Galileo has so

many signals. Another reason for the abundance of

signals is that the receiver can estimate the

ionospheric delay error, which is caused by the delay

experienced by navigation signals passing through

the ionosphere. This delay can cause the receiver-

measured distance from the satellite to the user to

appear larger than it is, resulting in large position

errors if not corrected. However, this delay is

proportional to the frequency of the signal, with low-

frequency signals experiencing longer delays than

high-frequency signals. As a result, by combining

measurements from two different frequencies on the

same satellite, a new measurement can be created that

removes the ionospheric delay error. The greater the

distance between the two frequencies, the more

effective this cancellation will be. This is why Galileo

services are typically implemented using signal pairs

[11].

2 GALILEO SIGNAL E1

The Galileo E1 signal uses BOC modulation, which

employs carrier shift modulation to shift the energy

away from the center of the band. This is significant

because it enables multiple GPS systems to use the

same band. BOC modulation utilizes two independent

parameters, namely the carrier frequency of the

auxiliary signal (fs) in MHz and the code rate of the

code shift (fc) in mega chips per second. This gives the

signal two parameters that can be adjusted to

manipulate the signal's power in specific ways to

reduce interference from other signals on the same

band. Furthermore, the redundant upper and lower

sidebands of BOC modulations offer advantages in

signal processing for receiver acquisition, carrier

tracking, code tracking, and data demodulation [12].

The entire transmitted Galileo E1 signal consists of

the following components [13]:

Figure 1. E1 Signal Modulation scheme [13].

Figure 1 shows the modulation scheme of signal

E1. E1 open service data channel e

E1-B(t) is generated

from I/NAV navigation data stream D

E1-B(t) and

measurement code C

E1-B(t), which are then modulated

by subcarriers SC

E1-B, a(t), and SCE1-B,b(t). The open

service pilot channel E1 e

E1-C(t) is generated from the

range code C

E1-C(t), including its secondary code,

which is then modulated by subcarriers SC

E1-C, a(t), and

E1-C,b(t), in antiphase [13].

The Galileo E1 signal is modulated at a medium

frequency such as [11] :

(

) ( )

( ) ( )

( ) ( ) ( ) ( )

(

cos 2

ττ τ

τ τ τ πθ

−

= − − −−

−− − − +

P c IF

s t A C t d t CBOC t

c t S t CBOC t x f t

(1)

where:

(

)

( )

11 10

−

= −CBOC t BOC t BOC t

(2)

( ) ( ) ( )

11 10

= +CBOC t BOC t BOC t

(3)

( )

(

)

sin .2 .1 .023 . .

BOC t sign x e t

(4)

A is the amplitude of the input signal at the input of

the correlator,

P a CD are extended sequences that carry pilot and

data components,

D represents the navigation message symbol of the

I/NAV modulated data component.

c represents the secondary code present on the pilot

component,

is a sequenced delay,

IF is the center frequency,

is the phase shift of the carrier frequency [12].

GPS C / A and Galileo BOC (1,1) share the L1 / E1

spectrum, which is shown in Figure 1. The mean

frequency of the E1 / L1 signal is 1575.42 MHz. It is

important to remember that the current E1 band was

given the name L1 band for a long time, analogous to

GPS, until 2008, when the name of the L1 signal was

changed to the current E1.

Based on the formulas (1-4), the simulation of the

E1 signal was performed in the Matlab programming

environment. The following parameters of the E1

signal were used in the modeling:

1.023 10 ; 6 1.023 10fsa fsb=⋅ =⋅⋅

Define the subcarrier frequencies for BOC(1, 1) and

BOC(6, 1). The value of fsa is set to 1.023 MHz and fsb

is set to 6 · 1.023 MHz.

F = row space (-20, 20, 40000);

Generates a frequency vector for a PSD plot. The

frequency vector ranges from -20 MHz to 20 MHz

with 400000 points.

0:1/ :10t fsb

−

(5)

Generates the time vector for the signal. The time

vector ranges from 0 to 10 ms with a step size of 1/fsb.

(

)

1 sin 2x fsa t

= ⋅⋅

(6)

( )

6 sin 2x fsb t

= ⋅⋅

(6)

They generate signals for BOC(1, 1) and BOC(6, 1).

Signals are generated by multiplying partial carrier

frequencies by a time vector and evaluating a sine

function.

sca = character(x1)

scb = char(x6)

These lines generate secondary codes for BOC(1, 1)

and BOC(6, 1). Secondary codes are generated by

accepting the sign of the BOC signals.

10 / 11alpha =

(8)

1/ 11beta

(9)

scB alpha sca beta scb= ⋅+ ⋅

(10)

scC alpha sca beta scb= ⋅− ⋅

(11)

These lines generate the primary codes for E1-B

and E1-C. Primary codes are generated by mixing

secondary codes with alpha and beta coefficients.

( )

1 cos 2 1575.42 10eE B t

= ⋅ ⋅⋅

(12)

( )

1 sin 2 1575.42 10eE C t

= ⋅ ⋅⋅

(13)

These lines generate the E1-B and E1-C signals.

Signals are generated by multiplying the carrier

frequency by the time vector and evaluating the

cosine and sine functions.

( )

10 log

X fft sE length f

PSD

length f





= ⋅





These lines calculate the FFT of the Galileo E1

signal and convert it to the PSD. The PSD is calculated

as above, where X is the FFT of the signal and f is the

frequency vector.

The results of the E1 signal simulation are shown

in Figure 2.

Figure 2. Structure of signal Galileo E1.

Figure 2 shows the power spectral density (PSD)

for the E1 signal, which is part of the Galileo

navigation system. On the x-axis is the frequency

range from -20 MHz to 20 MHz, while on the y-axis is

the calculated PSD in dBm. Frequency 0 on the x-axis

represents the center frequency of the signal.

The source signal E1 is created by modulating the

carrier wave using PRN (Pseudo Random Noise)

coding. Therefore, the PSD for the E1 signal shows a

characteristic frequency structure that consists of

several layers. The highest layer represents the signal

power in the area of the main carrier frequency of

1575.42 MHz, which occurs at frequency 0 on the x-

axis. This peak has a calculated value of about -162

dBm/Hz.

Additional bands of power are visible around the

main carrier frequency, 1.023 MHz apart, known as

"sidebands" and "adjacent bands". These bands are

part of the Galileo signal modulation and are

responsible for the design of the PRN coding and

other signal parameters.

The overall PSD waveform for the E1 signal shows

how the signal power is distributed over the entire

frequency axis. Since this signal occurs close to other

satellites and transmission channels, it is important to

know its power spectral density and its characteristic

structure to avoid interference and guarantee the

reliable use of Galileo.

3 SIMULATION OF INTERFERING SIGNALS

Research has confirmed that the measurement signals

of the Galileo system are degraded by interfering

signals. These signals can significantly affect the

results of navigational measurements by the Galileo

system. Therefore, we performed the simulation of

interfering signals such as white noise, chaotic

impulse interference, and narrowband interference.

White noise, chaotic impulse interference, and

narrowband interference models were created in the

Matlab programming environment. We proceeded in

the following way when creating interference models.

First, we created an interference model, and then an

additive mixture of the E1 signal and interference. The

simulation results are shown in images no. 2 - 11.

3.1 White noise

We created a white noise model and then simulated

this noise in the Matlab program environment. The

following algorithm was used to generate white noise:

( )

1, _Noise randn N noise level= ⋅

(14)

This line generates a Gaussian noise signal with a

mean of 0, and a standard deviation is given by the

noise level vector.

Where:

fs = 40,000

This line sets the sample rate of the signal to 40000

MHz.

T = 10

-3

; %

This line sets the signal duration time to 10ms.

N = 400

This line sets the number of samples in the signal

to 400.

f = row space(-20·10

, 20·10

, N);

This line creates a vector of N evenly spaced

frequencies ranging from -20 MHz to 20 MHz.

x = rowspace(-T/2, T/2, N);

This line creates a vector of N evenly spaced time

values ranging from -T/2 to T/2.

noise_level = linspace(-1, 6, N);

This line creates a vector of N evenly spaced noise

levels ranging from -1 to 6.

( )

fftshift fft noise

PSD

fs N

⋅

(15)

This line calculates the PSD of the noise signal

using the FFT, taking the absolute value and squaring

it, and dividing it by the product of the sample rate

and the number of samples.

( )

10 log 10 11

dBm

PSD PSD=⋅ ⋅−

(16)

This line converts the PSD from V

/Hz to dBm

units using a reference power of 1 milliwatt and

subtracts 11 to adjust for noise level.

The results of the white noise simulation are

shown in Figure 3.

Figure 3. Structure of White noise.

Figure 3 shows the PSD graph of the Gaussian

noise signal with the specified noise level. PSD is

calculated using the Fast Fourier Transform (FFT) and

converted from V

/Hz units to dBm units. The

resulting PSD is then plotted as a function of

frequency in MHz. The plot shows a symmetrical bell-

shaped curve centered at zero frequency, which

represents the PSD of the Gaussian noise signal. The

curve has a maximum value at zero frequency,

indicating that the noise signal has the highest power

at low frequencies. The curve drops off rapidly with

increasing frequency, indicating that the power of the

noise signal decreases at higher frequencies. The x-

axis of the graph represents frequency in MHz, while

the y-axis represents PSD in dBm. The graph is

labeled with the appropriate axis labels and the title

"White Noise Suppression Structure", which describes

the nature of the analyzed signal. The plot provides a

visual representation of the frequency content of a

Gaussian noise signal, which is useful in a variety of

signal processing and communications applications.

3.1.1 The additive mixture of E1 signal and white noise

If in the previous steps, we generated the Galielo

E1 signal and subsequently white noise, the next step

is to generate the noisy E1 signal. E1 signal with the

addition of white noise is possible using the following

relationship:

( ) ( )

( )

11 1

sE eE B alpha scB beta scB eE C alpha scC beta scC noise=⋅⋅⋅+⋅−⋅⋅−⋅+

(17)

Add white Gaussian noise to the signal:

sE1 = sE1 + noise (18)

X = fft(sE1, length(f)) (19)

Calculates the FFT of the sE1 signal with a length

of (f) points

(

)

log

PSD

length f







(20)

Calculates the power spectral density of the signal.

The results of the simulation of the additive

mixture of the useful signal E1 and white noise are

shown in Figure 4.

Figure 4. Signal E1 with added White noise.

The white Gaussian noise added to the sE1 signal

can be seen in Figure no. 4. The effect of adding noise

to a signal is that it increases the power spectral

density (PSD) of the signal at all frequencies,

including those of the original signal and the noise

itself. The PSD is calculated using the fast Fourier

transform (FFT) of the noise signal sE1, which is

represented by the variable X. The PSD is then plotted

against the frequency f in the image. The noise level is

specified using the noise_level variable, which

controls the standard deviation of the white Gaussian

noise added to the signal. A higher level of noise

results in a higher PSD of the noisy signal, which

means that the signal is harder to detect and more

accurately decoded, especially if the signal-to-noise

ratio is low. Adding noise to the signal also introduces

errors in the decoding of the spreading codes and the

original signals, which can affect the overall

performance of the communication system. In

particular, the accuracy of timing and frequency

synchronization can be affected, which can lead to

further degradation of signal quality.

3.2 Chaotic impulse interference

We created a model of chaotic impulse interference

and then simulated this interference in the Matlab

environment. Chaotic impulse interference is created

by combining a Lorentzian chaotic signal with an

impulse response. In relation (23), the impulse

response is defined as a vector of zeros with one 1 at

index 100:

pulse = zero (size (t)) (21)

impulse(100) = 1 (22)

The conv() function is then used to convolve this

impulse response with the first component of the

Lorenz signal (x(:,1)):

interference = conv(x(:,1), impulse ) (23)

The resulting interference signal is a version of the

Lorenz signal with an added pulse at index 100.

This relationship was used to generate the

structure of impulsive chaotic interference, the

spectrum of which is shown in Figure 5.

Parameters for the Lorenz chaotic system: sigma to

10, beta to 8/3, rho to 28, and the initial state vector x0

to [1;1;1]. The ode45 function is then used to solve the

Lorenz system for the given time range t and initial

conditions x0.

(

) (

)

(

)

( )

(

)

(

)

(

)

( )

2 1;1 3 2;1 2 3

sigma x x x rhod

x x x x beta

xt xd

⋅ − ⋅− − ⋅ ⋅=

−

(24)

A function called "Lorenz" defines the differential

equations for the Lorenz chaotic system. It takes in the

input arguments "~" and "x", which are not used in the

function. It also takes in the parameters "sigma",

"beta", and "rho", which are used to define the Lorenz

system equations.

The output "dxdt" is a vector of the same size as

"x", which defines the rate of change of each state

variable in the Lorenz system at a given time. The first

element of "dxdt" is the rate of change of x(1), the

second element is the rate of change of x(2), and the

third element is the rate of change of x(3). The

equations used in this function are the classic Lorenz

equations, which are commonly used in chaos theory

to study the behavior of nonlinear dynamical systems.

Parameters for the Lorenz chaotic system: sigma to

10, beta to 8/3, rho to 28, and the initial state vector x0

to [1;1;1]. The ode45 function is then used to solve the

Lorenz system for the given time range t and initial

conditions x0.

The results of the chaotic impulse interference

simulation are shown in Figure 5.

Figure 5. Structure of Chaotic Impulse Interference.

Figure 5 shows the power spectral density (PSD) of

chaotic impulse interference as a function of

frequency. The horizontal axis shows the frequency in

MHz, while the vertical axis shows the PSD in dBm.

On the graph, there is a suitable peak near the

frequency of 0 MHz, which represents the DC

component, i.e., alternating current with zero

frequency. Next, we will see peaks that occur in

different frequency bands, which are the frequency

chaotic behavior of the system. Chaotic impulse

interference contains many different frequency

components that are propagated into the environment

and affect other neighboring systems.

3.2.1 The additive mixture of E1 signal and chaotic

impulsive noise

We created a model of chaotic impulse interference

and then simulated this interference in the Matlab

environment. We then created an additive mixture of

the useful signal E1 and the chaotic impulse

interference using the following algorithms:

sE1 = sE1 + interference (25)

The disturbance is generated by the convolution of

the first variable (x(:,1)) of the output of the

Lorentzian impulse response system. The resulting

interference signal is then added to the sE1 signal. The

simulation results of the additive mixture of useful

signal E1 and chaotic impulse interference are shown

in Figure 6.

Figure 6. Structure of Galielo signal E1 with added Chaotic

impulse Interference.

Figure 6 shows the PSD of the Galileo E1 signal,

which is subsequently affected by chaotic interference.

This process can have a major impact on signal

quality and cause interference and communication

disruptions. A Lorenz chaotic system is a differential

equation that describes chaotic behavior. In this case,

the system consists of three equations that describe

the evolution of three variables over time. These

variables represent the state of the system and change

depending on time and other parameters. The chaotic

disturbance in this case is generated by the

convolution of the signal from the Lorentz system

with the impulse response. The result of this

convolution is a time-shifted signal that can cause

interference and communication breakdowns. The

effect of interference on the Galileo E1 signal can be

observed using the power spectral density (PSD),

which is calculated using the Fourier transform. PSD

shows the distribution of signal power as a function of

frequency. In this case, we can see that chaotic

interference causes interference in a wide frequency

band, which can cause disturbances in receiving

navigation information. The result of this process is a

signal that is affected by chaotic interference, which

manifests itself as interference in the entire frequency

band. The resulting signal may be of lower quality

and cause communication problems. Therefore, it is

important to ensure a sufficient level of protection

against interference and to minimize the effect of

chaotic interference on the signal.

3.3 Narrowband interference

Narrowband interference is a type of interference that

occurs when there is a strong signal at a frequency

equal to the frequency of the E1 signal. This

interference signal causes errors in the navigation

measurements of the Galileo system. We generated

the narrowband interference according to the

following algorithms:

( )

1 10 2 0

f phi

x cos f delta t delta

= ⋅ ⋅ + ⋅+

(26)

That relationship generates narrowband

interference by the sum of N sine waves with random

frequency and phase variations around the center

frequency f0.

Where:

delta

f = (rand(1, N) - 0.5) · 2 · 10

(27)

Generates N random values between -1 MHz and 1

MHz. These values represent frequency variations.

delta

phi = rand(1, N) · 2 π (28)

Generates N random values between 0 and 2π.

These values represent phase variations.

f0 = 1575.42 · 10

Sets the center frequency to 1575.42 MHz.

The results of the narrowband interference

simulation are shown in Figure 7.

Figure 7. Narrowband interference structure.

The power spectral density (PSD) of a narrowband

interference signal with a random frequency and

phase variations is shown in Figure 7. The x-axis

represents frequency in MHz and the y-axis

represents the PSD in dBm. The center frequency of

the signal is 1575.42 MHz, and the frequency

variations around this center frequency are randomly

generated with a range of +/- 2 MHz. The phase of the

signal also varies randomly. The resulting PSD plot

shows a main lobe centered at the center frequency

and several smaller lobes on either side due to the

random frequency variations. The height and width of

the lobes depend on the amplitude and duration of

the interference signal. The plot can be used to

identify the frequency and strength of the interference

signal, which is important for mitigating its effects on

the system.

3.3.1 The additive mixture of E1 signal and narrowband

interference

We used the programming environment of the

Matlab program to create an additive mixture of the

E1 signal and narrowband noise. Interference is added

to the original signal (Figure 2) using the relation:

sE1_with_interference = sE1 +x1 (29)

The result of adding narrowband interference to

the E1 signal can be seen in Figure 8.

Figure 8. Structure of Signal E1 with added Narrowband

interference.

Figure 8 shows the resulting power spectral

density (PSD) of the E1 signal and narrowband

interference with random frequency and phase

deviations. PSD indicates the distribution of signal

power concerning the frequency and is expressed in

units of dBm/MHz. The E1 signal is a blue curve and

corresponds to the global navigation satellite system

Galileo, which transmits at a frequency of 1575.42

MHz. The blue curve has a maximum value in the

middle of the spectrum, which is chosen as a reference

value for the entire spectrum and is therefore placed

at zero. The red curve shows narrowband

interference, which is characterized by random

frequencies and phase deviations. This type of

interference can arise, for example, from sources with

an unstable frequency, such as generators with short-

term deviations, or the influence of external

interference. Thus, the figure shows that narrowband

interference has a significant effect on signal quality if

the transmitted signal and the interference are on the

same frequency. Therefore, it is important to protect

the transmitted signals from interference and avoid

overlapping frequency bands of different signals and

interference.

4 USING KALLMAN FILTRE

White noise has a pronounced effect on the original

structure of the E1 signal. It is therefore important to

remove this noise from the signal. In this chapter, we

will describe the effect of using a Kallman filter to

filter out white noise from a noisy signal (Figure 4.)

Defining the system matrices for the Kalman filter:

A = 1; state transition matrix

H = 1; observation matrix

Q = 0.01; process noise covariance

R = 0.1; covariance of measurement noise

PO = 1; initial covariance state

In this block of code, the matrices that define the

Kalman filter are defined. A is the state transition

matrix, H is the observation matrix, Q is the process

noise covariance, R is the measurement noise

covariance, and P0 is the initial state covariance.

The simulation results are shown in Figure 9.

Figure 9. Filtered signal E1 using the Kallman filter.

In this case, white noise was suppressed using a

Kalman filter (Figure 9.) A Kalman filter is a

mathematical algorithm that uses a series of

measurements observed over time to estimate the

state of a linear system. In this case, the state is the

actual power spectral density of the signal, and the

measurements are the power spectral density values

obtained from the noisy signal. The Kalman filter

works by predicting the state of the system at each

time step based on a previous estimate of the system's

state and dynamics, and then updating the state

estimate based on the current measurement. The filter

also estimates the uncertainty in the state estimate,

which is used to weigh the importance of the

prediction and measurement in the overall estimate.

In this implementation, the state transition matrix A

was set to 1, indicating that the system does not

change over time. The observation matrix H was also

set to 1, indicating that the measurement is a direct

observation of the state. The covariance Q of the

process noise and the covariance R of the

measurement noise was set to 0.01 and 0.1,

respectively, which are assumptions about the noise

variance in the system. A filter was applied to each

frequency bin of the power spectral density using a

loop. At each cycle iteration, the state and covariance

estimates were predicted using the state transition

model. Then, the state estimate was updated using the

current measurement and the Kalman gain, which is a

weighting factor that balances the contribution of

prediction and measurement to the overall estimate.

Finally, the filtered value of the power spectral

density was stored in the array. The filtered power

spectral density values were then plotted against the

original power spectral density values to show the

effectiveness of the filter in suppressing white noise.

Figure 10. Detailed Figure 4.

Figure 10 shows a more detailed spectrum of the

additive mixture of the useful signal E1 and white

noise.

Figure 11. Detailed Figure 9.

Figure 11 shows the spectrum of the additive

mixture of signal E1 and white noise after passing the

signal through the Kallman filter. The Kalman filter

suppressed the white Gaussian noise in the additive

mixture of the E1 signal and white noise. It is clear

from the picture that the (orange color) filter

suppressed the effect of noise on the useful E1 signal

from the Galileo system.

5 CONCLUSIONS

In this paper, mathematical models of the

measurement signal E1 of the Galileo system were

created to simulate intentional interference and the

influence of atmospheric conditions on signal

propagation. The study also presents the

determination of the Galileo satellite navigation

system, including the frequency and structure of the

E1 signal and its block diagram for signal generation.

The modeling results can be used to evaluate the

immunity of the Galileo system to interference. The

results of the simulations of the measurement signal

E1 with white noise, chaotic impulsive interference,

and narrowband interference allow a spectral analysis

of the structure of this signal. Through simulation, we

found that the frequency spectrum of the additive

mixture of the E1 signal and interference is

significantly different from the frequency spectrum of

the E1 signal itself. This fact indicates that signal

measurements that are degraded by interference can

cause errors in navigation measurements by the

Galileo system. The simulation results confirmed that

chaotic impulse interference is the most dangerous

type of interference. We can explain it by the fact that

chaotic impulse interference has a wide frequency

spectrum. Its influence on the original signal has

destructive effects, and when processing the

measurement signal E1, it is necessary to eliminate

this interference as much as possible. A Kallman filter

was designed to filter E1 signal interference. This filter

was used to filter out white noise from the additive

mixture of the E1 signal and white noise. The

simulation results confirmed that such a filter is

effective for suppressing white noise and can be used

at the receiver input for E1 signal processing. The

simulation results confirmed that when synthesizing

algorithms for processing the E1 measurement signal

from the Galileo system, it is necessary to pay

attention to the influence of white noise, chaotic

impulse interference, and narrowband interference on

the measurement results. Dual frequency is used to

improve the measurement results of the Galileo

system. Galileo's use of dual frequency positioning

provides an advantage over other satellite systems,

particularly in dense urban areas or forested

environments. The use of dual frequency allows for

more accurate positioning.

REFERENCES

[1] Bałdysz, Z., Szołucha, M., Nykiel, G., & Figurski, M.

(2017, June). Analysis of the impact of Galileo's

observations on the tropospheric delay estimation. In

2017 Baltic Geodetic Congress (BGC Geomatics) (pp. 65-

71). IEEE. DOI: 10.1109/bgc.geomatics.2017.22.

[2] Bidaine, B. (2006). Ionosphere Crossing of GALILEO

Signals.

[3] Dzunda, M; Kotianova, N; Dzurovčin, P., et all.: Selected

Aspects of Using the Telemetry Method in Synthesis of

RelNav System for Air Traffic Control. Jan 2020

INTERNATIONAL JOURNAL OF ENVIRONMENTAL

RESEARCH AND PUBLIC HEALTH 17 (1)

[4] Dzunda, M; Dzurovcin, P; et all. : Selected Aspects of

Navigation System Synthesis for Increased Flight Safety,

Protection of Human Lives, and Health. Mar 2020,

INTERNATIONAL JOURNAL OF ENVIRONMENTAL

RESEARCH AND PUBLIC HEALTH 17 (5)

[5] Dzunda, M.; Kotianova, N.; Holota, K.; et al.: Use of

Passive Surveillance Systems in Aviation. ACTIVITIES

IN NAVIGATION: MARINE NAVIGATION AND

SAFETY OF SEA TRANSPORTATION. Published: 2015.

Pages: 249- 253

[6] Galileo navigation signals and frequencies [online].

Available from:

https://www.esa.int/Applications/Navigation/Galileo/Ga

lileo_navigation_signals_and_frequencies

[7] Geng, J., Guo, J. Beyond three frequencies: an extendable

model for single-epoch decimeter-level point positioning

by exploiting Galileo and BeiDou-3 signals. J Geod 94, 14

(2020). https://doi.org/10.1007/s00190-019-01341-y

[8] Hadas, T., Kazmierski, K., & Sośnica, K. (2019)

Performance of Galileo-only dual-frequency absolute

positioning using the fully serviceable Galileo

constellation. GPS Solutions, 23(4), 108. DOI:

10.1007/s10291-019-0900-9.

[9] Hein, Guenter & Godet, Jeremie & Issler, Jean-Luc &

Martin, Jean-Christophe & Lucas-Rodriguez, Rafael &

Pratt, Tony. (2001). The Galileo Frequency Structure and

Signal Design. Proceedings of the 14th International

Technical Meeting of the Satellite Division of The

Institute of Navigation.

[10] European Commission (2010), European GNSS (Galileo)

Open Service – Signal-In-Space Interface Control

Document Issue 1, February.

[11] EUROPEAN GNSS (GALILEO) OPEN SERVICE

SIGNAL- IN-SPACE INTERFACE CONTROL

DOCUMENT Issue 2.0, January 2021

[12] Olivier Julien, Christophe Macabiau, Emmanuel

Bertrand. Analysis of Galileo E1 OS unbiased

BOC/CBOC tracking techniques for mass market

applications. NAVITEC 2010, 5th ESA Workshop on

Satellite Navigation Technologies and European

Workshop on GNSS Signals, Dec 2010, Noordwijk,

Netherlands. pp 1-8, 10.1109/NAVITEC.2010.5708070 .

hal-01022203

[13] Rodríguez, J.A,A. Galileo Signal Plan, University FAF

Munich, Germany, 2011