DE102020119348A1 - BROADBAND SPREADING CODE AUTHENTICATION USING TIME-DEPENDENT FREQUENCY VARIATIONS FOR GNSS - Google Patents

Abstract

A method of operating a global navigation satellite system, GNSS, is disclosed. A spreading code is provided. A carrier wave having a carrier frequency is provided. A frequency function applied to the carrier wave. The frequency function defines a time-dependent course in frequency space. The frequency function changes periodically. The spreading code is then modulated onto the carrier wave.

Description

TECHNICAL AREA

The invention relates to the operation of a global navigation satellite system (GNSS). In particular, the invention relates to the generation, transmission and processing of GNSS signals to protect against manipulation by unauthorized persons.

TECHNICAL BACKGROUND

A global navigation satellite system (GNSS) refers to a system for determining position and navigating on earth, in water, in the air and/or in space by receiving signals from satellites (navigation satellites) and pseudolites such as the ATC (Ancillary Terrestrial Component) or the CGC (Complementary Ground Component). It uses radio signals broadcast by satellites to determine a user's position, velocity and time (PVT). Examples of GNSS are the US GPS, the European Galileo, the Russian GLONASS or the Chinese Beidou.

The importance of GNSS is constantly increasing in various areas of society, business, industry and research. The provision and use of a particular GNSS can be collectively referred to as a GNSS service. A distinction can be made between public GNSS services, which are publicly available for civilian use, and closed/regulated GNSS services, in particular for military use. GNSS services are used for safety-related applications such as autonomous driving, power grid synchronization or time stamps for stock transactions.

A radio signal broadcast by a satellite of a GNSS can be referred to as a GNSS signal. A GNSS receiver can be subject to a so-called spoofing attack. An unauthorized person can imitate a real radio signal transmitted by a satellite, i.e. manipulate it at least partially and transmit the manipulated radio signal (imitation GNSS signal) in such a way that the real radio signal is overlaid by the manipulated radio signal. In particular, the manipulated radio signal is transmitted with an increased transmission power compared to the real GNSS satellite, making it more difficult for an unsuspecting receiver to differentiate the manipulated radio signal from the real radio signal. As a result, navigation systems can be rendered inoperable and/or unsuspecting recipients can receive corrupted GNSS signals in which, for example, navigation, position and/or time data are corrupted. It is also conceivable that such attacks are used to conceal certain positions.

A so-called SCER attack (Secure Code Estimation and Replay attack) is a special type of spoofing attack. Here, an attacker receives a first part of a GNSS signal and estimates the remaining part of the GNSS signal in real time based on the first part received Part. The estimated remaining part of the GNSS signal is manipulated (mimicked) and transmitted, possibly with an increased transmit power. A receiver that may not be able to distinguish the manipulated GNSS signal from the uncorrupted GNSS signal will use the manipulated GNSS signal for positioning and navigation, leading to incorrect results.

It is therefore desirable to improve the protection of GNSS services from a spoofing attack and in particular from a SCER attack in order to ensure the operability of the GNSS and to increase the security for the GNSS. In particular, the protection at chip level (spreading code level) is to be improved. In addition, the independence of the GNSS services from external communication channels should be guaranteed. On the recipient side, the technical requirements for implementing the proposed technology should not be excessively high.

SHORT SUMMARY

The above objects are at least partly solved by the subject matter of claim 1 and the independent claims. Particularly noteworthy embodiments of the claimed subject matter are given by the dependent claims.

Disclosed herein is a method of operating a Global Navigation Satellite System, GNSS. According to the method, a spreading code is provided. A carrier wave having a carrier frequency is provided. A frequency function is provided that defines a time-dependent course in frequency space and changes periodically. The frequency function is applied to the carrier wave. The spreading code is modulated onto the carrier wave.

In particular, the use of the frequency function means that unauthorized persons initially remain unaware of the frequency on which the GNSS signal was transmitted and how this frequency changed over time. The frequency function remains unknown, at least for a while, to the general public and thus to unauthorized persons, including a potential attacker or a user of GNSS services trying to spoof their own position. This makes it difficult for unauthorized persons to imitate a GNSS signal. In addition, the frequency spectrum (frequency band) used for transmission is expanded (broadened) so that the spectral power density spectrum of the signal is distributed more homogeneously over a larger frequency range. This can also make it more difficult to estimate the spread code.

Furthermore, autonomous spreading code authentication is provided by the claimed subject matter. At the same time, compared to a GNSS service without the protection disclosed herein, the increase in the technical requirements for the receiver is relatively small, because the changes in the carrier frequency according to the frequency function can be easily integrated into the conventional phase tracking of the carrier wave. Furthermore, the claimed subject matter enables more efficient use of the available frequency ranges.

Herein, the terms signal, radio signal, and GNSS signal may be used interchangeably unless otherwise noted or unless a signal obviously means something other than a GNSS signal or a radio signal of a GNSS. The subject matter described herein can be implemented for regulated GNSS services, with or without spreading code encryption, as well as for public GNSS services, with or without spreading code authentication.

The spreading code is used herein in accordance with industry standard frequency spreading techniques, particularly Direct Sequence Spread Spectrum (DSSS) and/or Code Division Multiple Access (CDMA). In particular, the spreading code includes a sequence of individual modulation states, which can be binary and are referred to as chips. Accordingly, the spreading code can comprise a binary chip sequence, with a chip of the spreading code being able to denote an individual modulation state in a manner comparable to a bit. For example, a data bit (particularly from a navigation message) is modulated (e.g., superimposed and/or multiplied) with the chips of the spreading code, allowing the spectral power to be distributed (spread) over an increased frequency band. The navigation message can be superimposed with the spread code in that the navigation message and the spread code are each modulated onto the carrier wave. Alternatively or additionally, the navigation message and the spreading code can be superimposed on one another and a signal superimposed in this way can be modulated onto the carrier wave. The navigation message can be modulated onto the spread code, or the spread code can be modulated onto the navigation message. In further examples, a signal generator can be provided which generates such a superimposed signal of the navigation message and the spreading code.

In this case, a rate of the data bit can be lower by a multiple (for example by 5 to 7 orders of magnitude) than a chip rate of the spread code. For example, the rate of the data bit is 1 to 3000 bits per second, in particular 50 to 250 bits per second. The chip rate of the spreading code can be, for example, 1×10 ⁵ to 1×10 ¹² chips per second, in particular 1×1.023 MHz to 10×1.023 MHz.

In particular, the spread code and its chip rate can be set up in such a way that a spectral power of the carrier wave modulated with the spread code drops below a normal level of thermal noise. In this way, a message that was transmitted via the carrier wave modulated with the spreading code can initially remain undetected in the thermal noise of a received GNSS signal. A receiver that knows the spreading code can use the spreading code to extract the message from the received GNSS signal. In addition, each satellite of a GNSS can have its own spread code, which allows the satellite to be identified as a transmitter. Unless otherwise specified, the term transmitter is used generically herein and includes both a satellite of a GNSS and any pseudolites and other transmitter devices that can participate in the GNSS as transmitters.

The carrier wave refers to an electromagnetic wave with a carrier frequency. The carrier wave may be provided as a radio wave (herein synonymous with radio wave or Hertzian wave). The carrier frequency can be between 30 kHz and 3000 GHz, in particular between 3 MHz and 300 GHz, and more precisely between 300 MHz and 3 GHz. The carrier frequency can be specified for the respective GNSS and/or for the respective GNSS service. In specific examples, the carrier frequency may be around 1575.42 MHz (so-called L1/E1 frequency band), 1191.795 MHz, 1278.75 MHz, 1227.60 MHz, 1176.45 MHz and/or around these frequencies within a deviation of vary ± 100MHz. Furthermore, a central frequency can be provided for the transmitter, which can assume one of the aforementioned values, for example, and the carrier frequency can be provided within a frequency band of ±100 MHz around the central frequency.

In specific examples, multiple discrete selection frequencies may be provided and the carrier frequency may assume one of the selection frequencies. The selection frequencies may be spaced a distance apart, in frequency space. This distance in frequency space can be equivalent to a product of a factor and a satellite's fundamental clock, where the factor is 0.01 to 100. Additionally or alternatively, the selection frequencies can be distributed within a frequency range whose width is equivalent to a product of a further factor and the fundamental clock of a satellite, the further factor being 0.01 to 10 ⁴ . For example, the base clock of a satellite is 1.023 MHz.

Unless otherwise specified, the term carrier wave is used herein to denote an electromagnetic propagation medium for transmitting a signal, such as a GNSS signal. Unless otherwise specified, the term signal is used herein to denote a character or sequence of characters carried by the carrier wave. Alternatively or additionally, the term signal is understood as a wave packet or a temporally coherent section of the carrier wave, which is modified according to a character or a character string of information to be transmitted. The character or string can be binary. The modification takes place, for example, by a modulation method, in particular a phase modulation. For example, the character or character string of information to be transmitted is modulated onto the carrier wave to form a signal.

The information (message) transmitted by means of a signal (or as a signal) may contain or provide an indication of a navigation message. In particular, the navigation message can contain information about a position (so-called ephemeris) and/or a time of the respective transmitter, for example the respective satellite. Furthermore, the information to be transmitted can contain a code for authenticating the signal. Furthermore, the information to be transmitted can contain a key for decrypting encrypted information.

The frequency function comprises a mathematical function that defines a time-dependent course in frequency space. For example, the frequency function completely describes a change in a frequency over time according to the respective mathematical function. Furthermore, an initial value and an end value for the frequency can be determined for each time interval [t(i), t(i+1)] according to the respective frequency function.

The frequency function may include one or more mathematical terms, which may include at least one or any combination of the following: a constant, an arbitrary degree polynomial, an exponential function, a Gaussian function, a logarithmic function, and a trigonometric function . In particular, the frequency function may contain a constant term that corresponds to the initial value (of the selection frequencies). The final value at the end of the respective time interval (ie at t=t(i+1) for the time interval [t(i), t(i+1)]) can be any frequency, depending on the length of the time interval and the respective frequency function or one of the foregoing multiple discrete selection frequencies (as described below).

The frequency function can be changed periodically. For example, the frequency function is periodically redefined for each time interval [t(i), t(i+1)] of consecutive time intervals i = 1, 2, ..., Z, where Z is a positive integer and a length of the time interval Δt = t(i+1) - t(i) is constant. For example, Δt may be 0.001 second to 10 ³ seconds, such as 1 second to 10 ² seconds, or 2 to 50 seconds. In specific examples, Δt may be 10 seconds, 20 seconds, or 30 seconds, or slightly different by up to 5 seconds. Furthermore, Δt can change over time, so that the time length of each time interval can vary over time. For example, Δt may vary in time by an amount from 000.1 seconds to 10 ³ seconds. In particular, the frequency radio tion be designed in such a way that the frequency function is continuous and differentiable in each individual time interval [t(i), t(i+1)]. In the transition between two consecutive time intervals [t(i), t(i+1)] and [t(i+i), t(i+2)], the frequency function can be continuous and/or differentiable, or neither continuous nor differentiable being.

Consequently, the frequency function can be changed periodically for each (each, per) time interval [t(i), t(i+1)]. Thus, for each time interval [t(i), t(i+1)], a respective form of the time-varying frequency function with i = 1, 2, ..., Z is valid. For convenience, the various forms of the time-varying frequency function may be referred to herein as "multiple frequency functions". The frequency functions can have a respective validity period, which designates a period of time in which the respective frequency function is applied to the carrier wave. The period of validity of the frequency function can designate the respective time interval [t(i), t(i+1)] in which the respective frequency function is valid.

The frequency function can be translated into a parameter string. For example, the parameter string can be a string defining a class of frequency function (eg an nth degree polynomial, exponential, logarithmic, or trigonometric function), a string for a constant term (eg zero for the center frequency, and an integer for a corresponding deviation of 1023 kHz from the central frequency) and a character sequence for coefficients of the respective class of the frequency function. As explained below, the parameter string may further include a string for a respective modulation scheme (e.g. BPSK(n), BOC(n,m), altBOC(n,m), etc.) and/or a string for a phase offset.

In the examples in which the frequency function is changed periodically per time interval [t(i), t(i+1)], the parameter string can be changed accordingly for the respective time interval. Thus, the parameter string may have a validity period in which it is valid, i.e. can be used to extract (recover, map) the frequency function within the same validity period.

For example, multiple parameter strings can be generated using a random number generator. The parameter strings can be stored in a memory of a transmitter, in particular a satellite. The stored parameter character strings can be provided with indexes, for example. Using a further random number generator, one of the several stored frequency functions can be selected and applied for each time interval.

wobei s_i das Signal für das Zeitintervall [t(i), t(i+1)], A die Amplitude, c(t) der Spreizcode, f_T die Trägerfrequenz, und f_i(t) die Frequenzfunktion ist. Wie vorstehend beschrieben, kann die Trägerfrequenz f_T auch zeitlich variabel sein.A signal provided in this way can take the following form: $$s_i=Ac\left(t\right)\text{cos}\left(2\pi \left(f_T+f_i\left(t\right)\right)t\right)$$

where s _{i is} the signal for the time interval [t(i), t(i+1)], A is the amplitude, c(t) is the spreading code, f _{T is} the carrier frequency, and f _i (t) is the frequency function. As described above, the carrier frequency f _{T can} also be variable over time.

In some examples, the signal further comprises a character string of a navigation message, so the signal _{si may} have another factor D(t) for the bitwise navigation message. Accordingly, the signal can take the following form: $$s_i=AD\left(t\right)c\left(t\right)\text{cos}\left(2\pi \left(f_T+f_i\left(t\right)\right)t\right).$$

In this case, D(t) can represent a time-dependent binary function. In particular, D(t) represents a bit sequence of the navigation message. Alternatively, the function c(t) in the previous equation can always represent a multiplication of the navigation message data bit function D(t) by the spreading code. As a result, the navigation message can be spread by the higher-frequency spread code.

According to one example, the spreading code is modulated onto the carrier wave according to a modulation method. The modulation method can be changed periodically. In particular, for two consecutive time intervals [t(i), t(i+1)] and [t(i+i), t(i+2)], different modulation methods can be used for the modulation of the spread code on the carrier wave. In this way, it can be even more difficult for an unauthorized person to estimate the spreading code in order to imitate the navigation message. Thus, the protection of the GNSS service against a spoofing attack or SCER attack can be further improved.

wobei w_i(t, Φ_i) eine Verschlüsselungsfunktion des Spreizcodes darstellt, die von dem jeweiligen Phasenmodulationsverfahren Φ_i für das Zeitintervall [t(i), t(i+1)] abhängt. Ferner kann das Phasenmodulationsverfahren Φ_i in eine (binäre) Zeichenfolge übersetzt werden und mit der vorstehend diskutierten Parameterzeichenfolge ausgesendet werden.A signal provided in this way can take the following form: $$s_i=A{w}_i\left(t,{\Phi }_i\right)c\left(t\right)\text{cos}\left(2\pi \left(f_T+f_i\left(t\right)\right)t\right)$$

where w _i (t, Φ _i ) represents an encryption function of the spreading code, which _{depends on the respective phase modulation method Φ i} for the time interval [t(i), t(i+1)]. Furthermore, the phase modulation method Φ _i can be translated into a (binary) character string and transmitted with the parameter character string discussed above.

An example of a modulation method is BPSK(n), where n can be 0.1 to 20 in particular. Another example of the modulation method is BOC(j,k), where j can be 0.1 to 40, for example, and k can be 0.1 to 40, for example. In particular, j can be 1, 5 or 10, while k can be 1, 2.5 or 5. Other examples of the modulation methods include altBOC, CBOC, and TMBOC, each with appropriate parameters. Both the class of the modulation method (BPSK, BOC, altBOC, etc.) and/or the respective parameters of the modulation method can be changed periodically.

According to an example of the method, a phase offset is also introduced when modulating the spreading code onto the carrier wave. The phase shift can be changed periodically. In particular, a chip phase offset is introduced into the spreading code with respect to a public chip phase. In this way, it can be even more difficult for an unauthorized person to estimate the spreading code in order to imitate the navigation message. Thus, the protection of the GNSS service against a spoofing attack or SCER attack can be further improved.

Without precise knowledge of the modulation method used in each case and/or of the chip phase offset, it is considerably more difficult for the unauthorized person to estimate. Thus, the technical requirements for a successful spoofing attack or SCER attack increase significantly, making a successful attack less likely. Most importantly, a successful attack requires estimating the spreading code per selection frequency, per discrete chip phase offset, per modulation scheme, and per frequency function (i.e., the time-dependent change in frequency space as a mathematical function). In accordance with the subject matter described herein, the number of selection frequencies, modulation schemes and chip phases to be estimated that must be estimated simultaneously and in parallel by a potential attacker is increasing. Since correct estimates can only be made with a certain probability, the rate of false detections and thus the probability of the attack failing increases.

wonach die Verschlüsselungsfunktion w_i(t, Φ_i, Δφ_i) ferner von einem jeweiligen Phasenversatz Δφ_i des Spreizcodes für das Zeitintervall [t(i), t(i+1)], dem Spreizcode-Phasenversatz, abhängt. Der Phasenversatz Δφ_i kann in eine (binäre) Zeichenfolge übersetzt werden und mit der vorstehend diskutierten Parameterzeichenfolge ausgesendet werden. Alternativ oder zusätzlich kann der Spreizcode mit dem Spreizcode-Phasenversatz überlagert sein, sodass das Signal die folgende Form aufweist: $$s_i=A{w}_i\left(t,{\Phi }_i,\Delta {\mathrm{\phi }}_i\right)c\left(t,\Delta {\mathrm{\phi }}_i\right)\text{cos}\left(2\pi \left(f_T+f_i\left(t\right)\right)t\right)$$

A signal provided in this way can take the following form: $$s_i=A{w}_i\left(t,{\Phi }_i,\Delta {\mathrm{\phi }}_i\right)c\left(t\right)\text{cos}\left(2\pi \left(f_T+f_i\left(t\right)\right)t\right)$$

according to which the encryption function w _i (t, Φ _i , Δφ _i ) also depends on a respective phase offset Δφ _{i of} the spreading code for the time interval [t(i), t(i+1)], the spreading code phase offset. The phase offset Δφ _i can be translated into a (binary) string and sent out with the parameter string discussed above. Alternatively or additionally, the spreading code can be superimposed with the spreading code phase offset, so that the signal has the following form: $$s_i=A{w}_i\left(t,{\Phi }_i,\Delta {\mathrm{\phi }}_i\right)c\left(t,\Delta {\mathrm{\phi }}_i\right)\text{cos}\left(2\pi \left(f_T+f_i\left(t\right)\right)t\right)$$

Phase offset refers to an offset in the phase of the spreading code from the spreading code in a reference signal, for example from a public pilot channel. In particular, the phase offset depends on the respective phase modulation method for the time interval. For example, the phase offset is equivalent to a product of a factor and the phase length of the respective phase modulation scheme, where the factor can be any number between 0.01 and 1. Herein, the term phase offset, unless explicitly stated otherwise, refers to the spreading code phase offset as described above.

As used herein, the term channel may refer to a particular GNSS signal component with its associated encoding, if any. Different channels, ie multiple GNSS signal components, can be used simultaneously or at different times. The channels can be present in a common frequency spectrum and/or at least partially in separate frequency spectrums. For example, a transmitter can have a data channel and a pilot channel. In the pilot channel the spreading code can be modulated onto the carrier wave without a phase offset and without a navigation message being modulated on. In the data channel the spreading code can be modulated onto the navigation message according to the frequency function, with or without phase offset, and with or without the variable phase modulation method. Furthermore, the transmitter can have an authentication channel via which information for authenticating the navigation message and/or the spread code is transmitted.

According to an example, the spreading code is further overlaid with an encryption chip sequence to obtain an encrypted spreading code. The encrypted spreading code can be modulated onto the carrier wave. The encryption chip sequence can be any random or pseudo-random chip sequence, which in particular can have the same length (i.e. number of chips) as the spreading code. The spreading code can be overlaid with the encryption chip sequence by means of a multiplication or an addition, using a suitable circuit in each case.

The respective spread code for each satellite of a GNSS is usually known. By overlaying the spreading code with the encryption chip sequence, the spreading code remains unknown to the receiver, ie without knowledge of the encryption chip sequence. The encryption chip sequence can be generated and defined according to a cryptographic method and only made available to an authorized and/or licensing group of recipients. This can be done in a similar way to an SCE method (Spreading Code Encryption) as is used in regulated GNSS services.

Alternatively or additionally, the encryption chip sequence can be used to encrypt at least part of the pilot channel, at least part of the data channel and/or the spreading code of the authentication channel of a public GNSS service. This can be done in a similar way to an SCA method (Spreading Code Authentication), as is planned for use in some public GNSS services such as the Galileo system. The encryption chip sequence may be injected and transmitted in one or more of the pilot channel, the data channel and the authentication channel, each optionally encrypted in turn. Alternatively or additionally, the encryption chip sequence may be transmitted over a separate channel or GNSS service (e.g. in a separate frequency band) that is fully or partially encrypted. Furthermore, the encryption chip sequence can be transmitted with a time delay, in particular after the transmission of the navigation message and/or the encrypted part of the signal. For example, the encryption chip sequence can be transmitted together with an (authentication) key for navigation message authentication. Additionally or alternatively, the information can be transmitted via a secure second communication channel such as the Internet or a terrestrial wireless network such as LTE, 5G, WiFi, etc.

Encrypting the spreading code with the encryption chip sequence makes it even more difficult for an unauthorized person to estimate the spreading code in order to impersonate the GNSS signal. Thus, the protection of the GNSS service against a spoofing attack or SCER attack can be further improved.

In combination with the frequency function described above, the encryption of the spread code can thus be extended to a broader frequency range. Accordingly, the encrypted spread code can be periodically modulated onto the time-varying carrier frequency, which can make estimating the spread code considerably more difficult.

According to one example, applying the frequency function to the carrier wave causes the carrier frequency to vary within a frequency band. In particular, the frequency band can be specified for each transmitter and/or the respective GNSS service. The frequency range in which the carrier frequency is variable can have a width equivalent to a product of a factor and the fundamental clock of a satellite, where the factor can be, for example, 0.01 to 10 ⁴ and the fundamental clock of a satellite can be 1.023 MHz. This frequency range can also be with the aforementioned breadth around a central frequency, in which case the central frequency can be from 0.1 to 100 GHz, in particular from 1 GHz to 2 GHz. For example, according to the frequency function, the carrier frequency is variable within a frequency range of 1575.42 GHz ± (20 • 1.023 MHz), where 1575.42 GHz is the center frequency and the aforesaid factor is 20.

In some examples, the carrier frequency may vary continuously within the stated frequency band. Alternatively or additionally, the frequency band can be subdivided evenly or irregularly, so that a number of discrete selection frequencies are obtained for the carrier frequency. The number of such selection frequencies can be at least 2 and generally unlimited above. In specific cases play, the number can be 2 to 10 ⁵ , in particular 2 to 10 ⁴ , or 4 to 10 ³ . The selection frequencies may be spaced apart in frequency space by a distance of 10 ² Hz to 10 ^{8 Hz.}

According to one example, multiple discrete selection frequencies are provided. One of the plurality of discrete selection frequencies can be selected and applied as the carrier frequency of the carrier wave for a time interval [t(i), t(i+1)], respectively. Thus, the carrier frequency can vary periodically between the multiple selection frequencies. The selection frequencies can be provided as described above.

wobei A die Amplitude des Signals ist, w_i(t) eine Verschlüsselungsfunktion für den Spreizcode ist, c(t) der Spreizcode ist, f_T die Trägerfrequenz ist, und f_i(t) die Frequenzfunktion ist.According to an example, a GNSS signal s _{i is} provided and transmitted in a time interval te [t(i), t(i+1)] in the following form: $$s_i=A{w}_i\left(t\right)c\left(t\right)\text{cos}\left(2\pi \left(f_T+f_i\left(t\right)\right)t\right),$$

where A is the amplitude of the signal, w _i (t) is an encryption function for the spreading code, c(t) is the spreading code, f _{T is} the carrier frequency, and f _i (t) is the frequency function.

The mathematical multiplication of the encryption function w _i (t) by the spreading code c(t) in the above formula may involve applying the encryption chips discussed above to the spreading code. Accordingly, the encryption function w _i (t) can contain a component that corresponds to that of the encryption chips, which can be firmly defined, such as in an SCE method or SCA method.

According to one example, the scrambling function w _i (t) or w _i (t, Φ _i ) further comprises a chip phase offset Δφ _i when modulating the spreading code onto the carrier wave. The phase offset can be implemented as discussed above. In particular, the phase offset can vary between discrete phase offset steps, for example between 0.01 and 1 times the phase length of a specific modulation method. Furthermore, the phase offset can be changed periodically. For example, the phase offset is changed in each time interval. Accordingly, the encryption function can depend on the respective phase modulation method, represented by Φ _i , for the respective time interval and/or on a spreading code phase offset Δφ i as described above. Additionally or alternatively, the spread code can be designed according to the spread code phase offset Δφi. The resulting signal can look like this: $$s_i=A{w}_i\left(t,{\Phi }_i,\Delta {\mathrm{\phi }}_i\right)c\left(t,\Delta {\mathrm{\phi }}_i\right)\text{cos}\left(2\pi \left(f_T+f_i\left(t\right)\right)t\right)$$

Alternatively or additionally, the phase modulation _{method Φ i} can be changed periodically as described above. In these examples, the encryption function w _i can comprise a further variable which depends on the phase modulation method applicable for the respective time interval [t(i), t(i+1)].

The periodically variable spread code phase offset and periodically variable modulation method of the spread code can further complicate the estimation of the spread code and thus the imitation of a GNSS signal and thus contribute to increasing protection against attacks.

According to an example, the carrier wave modulated with the spreading code, scrambling function and frequency function is transmitted. After that, a parameter string is sent out for frequency function extraction and scrambling function. For example, the parameter string is initiated upon completion of transmission of the carrier wave modulated with the spreading code, scrambling function, and frequency function. The transmission of the parameter character string takes place, for example, only after a complete transmission of the GNSS signal using the encryption function and frequency function and possibly including an authentication code for the navigation message. Thus, the parameter character string can be transmitted with a time delay compared to the navigation message. The time delay can be 1 millisecond to 300 seconds, in particular 1 to 10 seconds, or 10 to 30 seconds.

The parameter string can be introduced and transmitted in a data channel and/or a pilot channel of the GNSS. Alternatively or additionally, the parameter string may be transmitted over a separate channel that is fully or partially encrypted. The parameter string can be transmitted with or in the navigation message, for example together with an authentication key for a navigation message authentication method, as explained below. Additionally or alternatively, the parameter string can be transmitted via a secure second communication channel (such as via the Internet or a terrestrial wireless network such as LTE, 5G, WiFi, etc.).

The parameter string can be sent periodically. For example, once per time interval [t(i); t(i+i)] in which the parameter string is valid. In particular, the parameter character string can be transmitted in a subsequent time section of each time interval, for example in a sub-interval [t(i)+p*Δt, t(i+i)] of the time interval [t(i), t(i+i)]. , where the factor p can be 0.4 to 0.99, in particular 0.45 to 0.9. Preferably, the parameter character string is not transmitted until towards the end of a period of time in which the associated frequency function for the respective parameter character string is applied to the carrier wave, in order to effectively conceal the frequency function.

Alternatively or additionally, the parameter character string can be divided into several sections and transmitted at specific time intervals. Preferably, the parameter character string is not fully transmitted until towards the end of a period of time in which the associated frequency function for the respective parameter character string is applied to the carrier wave, in order to effectively hide the frequency function.

According to one example, a navigation message is provided. The navigation message is modulated onto the carrier wave together with the spreading code. Herein the term navigation message can also denote in general a message that is transmitted via a GNSS signal. In particular, the navigation message can have one or more bit sequences that contain information about a position (ephemeris) and/or the current time of an associated satellite. In particular, the message can contain information about an error in the associated satellite clock. For example, the message may have a bit rate of 50 and 250 bits per second (bps). In a specific example, the message may be encoded according to Galileo's Ei-B/NAV.

Thus, a GNSS signal can have at least three components: a navigation message, a spreading code, and a carrier wave. As explained above, the navigation message can be multiplied by a spreading code, the bit rate of the navigation message being about 1 to 3000 Hz, in particular 50 to 250 Hz, and the chip rate of the spreading code being 10 ⁵ Hz to 10 ¹² Hz, in particular 10 ⁶ Hz to 10 can be ⁷ Hz. This spreads the frequency band for transmission of the navigation message. Furthermore, this can cause the level (received power) of the received signal to fall below the thermal noise level, so that the navigation message can be recovered (only) by the application of the spreading code. This allows multiple transmitters or satellites to operate in the same frequency band while maintaining manageable mutual disruption and interference.

The navigation message superimposed with the spreading code can be used to determine a distance between the associated transmitter and a receiver. Based on the navigation message received from a single transmitter, a so-called pseudo range can be determined, with some inaccuracy factors such as a delay time error or a clock error of the transmitter and/or the receiver initially being disregarded. After receiving navigation messages from several (e.g. four) transmitters, the respective runtime and clock error can be compensated and an exact position can be determined. The navigation message can also be used to determine the recipient's speed and clock error. As previously explained, the term transmitter is used herein generically for all transmitter devices, including satellites and pseudolites, that can participate in the GNSS as transmitters. Furthermore, the term receiver is used generically for all receiving devices, including user terminals and terrestrial navigation devices, that can participate as receivers in the GNSS.

The energy (intensity, received power) of a received signal is added up (added up, integrated) for a certain period of time, for example. The time span for the integration can correspond, for example, to a (temporal) length of a bit of the message. Depending on the length and the bit rate of the navigation message, the integration period of the signal can be 0.1 ms to 5000 ms, in particular 4 ms to 20 ms. The summation result can be used for detection and spectral analysis of the signal. For this purpose, for example, a correlation analysis of the received signal with a signal from the pilot channel, in which the spreading code on the carrier wave without the navigation message is modulated can be performed. A sign of the navigation message can also be determined from the correlation analysis.

According to one example, an authentication key is also applied to the navigation message to obtain an authentication code. The authentication code may be a MAC code or may correspond to a MAC code as is commonly used in the industry in the GNSS field. The authentication code can be sent in addition to or instead of the navigation message. The associated authentication key is only sent out, for example, after the authentication code has been sent out in full. In addition or as an alternative, the authentication key can be broken down into a number of sections and sent out at specific time intervals. In examples where the authentication code is sent in addition to the navigation message, the authentication code may be sent before, with, after the navigation message. The authentication code can be transmitted via the data channel and/or the authentication channel.

The authentication key can be a random or quasi-random character string, in particular a bit sequence. The authentication code, which corresponds to an overlay of the navigation message with the authentication key, can also be understood as an encrypted navigation message. Thus the content of the navigation message can remain unknown until the authentication key is present. These processes may be similar to a navigation message authentication (NMA) process.

For example, a chain of authentication keys is generated using a one-way hash function. Starting with a private key, which can be randomly or cryptographically generated, the one-way hash function can be applied repeatedly to the previous key to obtain the chain of authentication keys. The one-way hash function can denote a mathematical transformation that is easy to compute but difficult to virtually impossible to invert. The chain's authentication keys thus obtained can be applied to successive navigation messages in reverse order to the order of generation. It can thus be easy to calculate a previous authentication key based on a currently valid authentication key and thus to verify the authenticity of the key obtained, while it is virtually impossible to determine the next authentication key. In this way, the authenticity of the navigation message can be checked.

In combination with the frequency function described herein, the use of navigation message authentication can result in increased protection against spoofing such as SCER attacks at both the spreading code level and the navigation message level. In this way, possible gaps in circumventing the respective individual protective effect can be closed as a precautionary measure.

In one example, the frequency function extraction parameter string is broadcast over a communication channel external to the GNSS. Such a communication channel outside of the GNSS can in particular include the wired Internet or other terrestrial data networks such as WiFi, LTE, 5G, etc.

Alternatively or additionally, the parameter string may be broadcast over a data channel of a public GNSS service, with the parameter string being overlaid with a broadcast string. Alternatively or additionally, the parameter string can be sent out in an encrypted part of the pilot channel. In particular, the transmission character sequence can be a chip sequence which is designed in such a way that the frequency spectrum (frequency band) for transmission of the parameter character sequence is spread and as a result the receiving line can be below the level of normal thermal noise. Thus, the parameter string can only be filtered out of the noise by applying the transmission string.

In this way, a secure transmission of the parameter character string can be guaranteed. In particular, the information about the frequency function can thus be transmitted in a secure manner.

According to one example, a GNSS signal is received, where the GNSS signal is carried on the carrier wave, the spreading code is modulated onto the carrier wave, and the frequency function is applied to the carrier wave. The received GNSS signal becomes valid for a period of time equivalent to a duration of the frequency function is saved. The frequency function is mapped using the parameter string. The spreading code can now be mapped using the mapped frequency function. These method steps are carried out at the receiver end.

According to the method described above, the spread code and the carrier wave, and possibly the navigation message, are initially unknown. Therefore, the received GNSS signal is first stored in order to apply the subsequently received parameter string to it, thus recovering information from the GNSS signal.

For example, the received GNSS signal is stored for each time interval [t(i), t(i+1)]. The parameter string can be transmitted in parallel with and simultaneously with the signal component with the navigation message (encrypted or not), for example via a separate channel. Additionally or alternatively, the parameter character string can be transmitted with the navigation message following the signal component.

In other examples, the parameter string is sent out in several separate chunks within a time interval. Thus, the message period and the parameter period may overlap or alternate.

When receiving a GNSS signal, a Doppler shift in the frequency of the carrier wave may have to be considered since the transmitter and receiver are in relative motion. Furthermore, when receiving the GNSS signal, the carrier frequency can be modulated down to an intermediate frequency in order to reduce the technical requirements at the receiver end. Furthermore, a receiver-side phase offset of the carrier wave can be determined and eliminated. If necessary, a frequency offset can be present, which can also be taken into account when processing the GNSS signal. Furthermore, the thermal noise from the GNSS signal may have to be eliminated.

wobei 1 für die Abtastung steht mit 1 = 1, 2, ..., L, wobei L eine positive ganze Zahl ist. Demnach steht t_l für die Abtastungszeit. Ferner ist A_l die Amplitude des empfangenen Signals, f_IF die Zwischenfrequenz, f_o das Frequenz-Offset, f_D die Frequenzverschiebung durch den Doppler-Effekt, θ der empfängerseitige Phasenversatz der Trägerwelle und N_l das thermische Rauschen. Unbefugten ohne Kenntnis der vorgenannten Parameterzeichenfolge bleiben zumindest w_i(t_l) und f_i(t_l) unbekannt. Wie vorstehend diskutiert kann die Frequenzfunktion f_i(t_l) mit einer zeitlich veränderlichen Trägerfrequenz einhergehen, sodass eine Nachverfolgung des GNSS-Signals ohne Kenntnis der Frequenzfunktion hochgradig aufwendig ist. Ferner kann die Verschlüsselungsfunktion w_i(t_l) insbesondere von dem jeweiligen Phasenmodulationsverfahren und/oder einem Phasenversatz für den Spreizcode abhängen. Dies erschwert eine Schätzung des GNSS-Signals durch Unbefugte noch weiter.In particular, the GNSS signal can be received at a specific sampling rate. Accordingly, the received signal can have the following form: $$s_i=A_i{w}_i\left(t_i\right)c\left(t_i\right)\text{cos}\left(2\pi \left(f_{If}+f_0+f_D\left(t_l\right)+f_i\left(t_l\right)\right)t_l+\mathrm{\theta }\right)+N_l,$$

where 1 represents the sample with 1 = 1, 2, ..., L, where L is a positive integer. Accordingly, t _l stands for the sampling time. Furthermore, A _{l is} the amplitude of the received signal, f _{IF is} the intermediate frequency, f _{o is} the frequency offset, f _{D is} the frequency shift due to the Doppler effect, θ is the receiver-side phase offset of the carrier wave, and N _{l is} the thermal noise. _{At least w i} (t _l ) and f _i (t _l ) remain unknown to unauthorized persons without knowledge of the aforementioned parameter character string. As discussed above, the frequency function f _i (t _l ) can be associated with a time-varying carrier frequency, such that tracking the GNSS signal without knowledge of the frequency function is highly expensive. Furthermore, the encryption function w _i (t _l ) can depend in particular on the respective phase modulation method and/or a phase offset for the spread code. This makes it even more difficult for unauthorized persons to estimate the GNSS signal.

According to a further aspect, an apparatus for use in a GNSS is disclosed, the apparatus being arranged to carry out the method described herein or any examples and modifications thereof.

In particular, the device may be a transmitter (as discussed above), comprise a transmitter, or be part of a transmitter. For example, the device is a satellite or a pseudolite. The device can be integrated into a satellite or pseudolite. The device can have a transmitter device such as a parabolic antenna, phased array antenna or other antenna in order to transmit signals. The apparatus may include processor means to perform the method steps and operations described herein. Furthermore, the device may have a memory to store or cache the instructions for execution by the processor means and optionally a private key, a chain of authentication keys, the parameter string, etc. Furthermore, the device can have signal generators for the carrier wave and the spread code, and optionally for the navigation message. Furthermore, the devices can have electrical circuits for carrying out the method steps and operations described herein.

According to a further aspect, a receiver device of a GNSS system is disclosed, which is configured to receive a GNSS signal. The GNSS signal is transmitted on the carrier wave with the spreading code modulated onto the carrier wave and the frequency function applied to the carrier wave. The receiver device is further configured to store the received GNSS signal for a period of time equivalent to a period of validity of the frequency function. The receiver device is configured to map (extract, restore) the frequency function using the parameter string. The receiver device is arranged to map (extract, recover) the spreading code using the mapped frequency function.

The subject matter described herein and its examples have the effect of making it difficult for an unauthorized person to estimate and imitate the GNSS signal, in particular its spreading code. Furthermore, an autonomous spreading code authentication can be provided, which can be independent of external communication channels. Furthermore, the technical requirements for implementing the subject matter described herein remain low in relation to the increased protection. Furthermore, the claimed subject matter enables more efficient use of the available frequency ranges.

character list

• 1 FIG. 12 schematically illustrates a GNSS according to an example.
• 2 12 is a schematic circuit diagram of a circuit according to an example.
• 3 FIG. 12 schematically illustrates time-dependent curves in a frequency space according to some examples.
• 4 schematically illustrates transmitted signals in a time-dependent diagram according to an example.
• 5 12 is a schematic circuit diagram of a circuit according to another example.
• 6 12 is a schematic circuit diagram of a circuit according to another example.
• 7 12 schematically illustrates periodically changing chip phase offset and periodically changing phase modulation method according to an example.
• 8th shows schematically and by way of example spectral power diagrams of a broadband spreading code authentication method and of various phase modulation methods.

DETAILED DESCRIPTION

Exemplary embodiments are described below in conjunction with the drawings. The same reference symbols are used to identify the same or at least similar, analogous or transferrable features.

1 FIG. 12 schematically shows a global navigation satellite system 10, GNSS, according to an example. The GNSS 10 includes at least four satellites 12, 14, 16 and 18 orbiting the earth in respective orbits. As part of the GNSS 10, the satellites 12-18 provide information about their respective identities, their respective positions (ephemeris), and their respective times. The information is broadcast in the form of a GNSS signal via a carrier wave from the respective satellite 12-18, as illustrated by dashed lines 22,32.

In this manner, satellites 12-18 provide a GNSS service that may be public, i.e., without access restrictions, or regulated/military, i.e., inaccessible to the public through encryption. Depending on the type of GNSS service, at least the structure of the GNSS signal (i.e. coding), the information to be transmitted and the frequency band used for the respective GNSS service can vary.

A receiver 20, which may be terrestrial, at sea, in the air, or in space, includes receiver apparatus to receive the GNSS signals, as shown schematically by dashed lines 22. FIG. In particular, the receiver 20 may be in motion relative to the satellites, for example as a navigation device in a moving vehicle.

Furthermore, an unauthorized person 30, referred to simply as an attacker 30 below, can receive the GNSS signals transmitted by the satellites 12-18, as illustrated by the dashed lines 32. For example, the attacker 30 also has a receiver device.

As described above, the attacker 30 can receive and analyze a first part of a conventional GNSS signal, and guess (guess) the remaining part of the same GNSS signal. For this purpose, the attacker 30 can take advantage of the fact that a large part of the technical information about the GNSS signal is freely available, such as the carrier frequency of the carrier wave, the spread code of the respective satellite 12-18, the structure of the GNSS signal, the orbit and the speed of the respective satellite (to compensate for Doppler effects). For GNSS services with unknown spreading codes, further information about the GNSS signal can be obtained, e.g. using spectral analysis, such as the modulation method used to modulate the spreading code, the number of chips in the spreading code, the length of a chip, etc.

For example, based on the estimated and known information about the GNSS signal, the attacker 30 may attempt to arbitrarily manipulate a GNSS signal (hereinafter simply referred to as “impersonate”) and transmit the impersonated GNSS signal to the receiver 20, as by dashed line 34 illustrated. For this purpose, the attacker 30 can use a signal generator, for example, in order to imitate positions, movements and/or times of one or more satellites 12-18 at will. Furthermore, the attacker 30 can use a powerful transmitter device to transmit the imitation GNSS signal with a high transmission power. Such an attack can be generically referred to as a spoofing attack.

In addition, the attacker can perform a SCER attack as described above. Thus, the attacker 30 estimates a navigation message bit-by-bit, and/or chip-by-chip spreading code in real-time. In a specific example, the attacker 30 receiving the GNSS signal (32 in 1 ), estimates the sign of the respective bit or chip after about half the time length of a bit or a chip by integrating the energy of the GNSS signal during this time span. Based on the estimate, manipulated bits of the navigation message or manipulated chips of the spreading code can be generated and transmitted. For example, the imitation GNSS signal is transmitted with a higher transmission power than the original GNSS signal in order to deceive receivers. A major risk can be that the attacker 30 may not need any knowledge of the structure of the GNSS signal and/or the spreading code for the SCER attack if the attacker 30 is equipped with powerful hardware and sufficient computing power. Therefore, the SCER attack can pose a serious risk to the public as well as the regulated GNSS service.

As in 1 As illustrated, attacker 30 may use a receiving antenna to amplify the received GNSS signal. A computing unit can estimate the bits and/or chips in real time and generate an imitation GNSS signal, also called a spoofing signal. A transmitting antenna can emit the spoofing signal with a high transmission power.

A SCER attack can be divided into two stages, with the bits of the message of the GNSS signal being estimated in a first stage and the spreading code being estimated in a second stage in real time. Due to the different rate of bits (on the order of 1 to 3000 Hz) and chips (on the order of 10 ⁵ to 10 ¹² Hz), bit estimation can be easier to implement than chip estimation. Namely, a bit may have a typical length of time on the order of milliseconds, while a chip may have a length of time on the order of microseconds.

2 12 shows a schematic circuit diagram of a circuit 100 according to an example. The circuit can be used in particular in a GNSS, such as the GNSS 10 in 1 , to be implemented. Furthermore, at least a portion of circuitry 100 may be implemented in a satellite, such as one or more of satellites 12-18.

The circuit 100 includes a carrier wave signal generator 110 which generates _{a carrier wave having a carrier frequency f T .} The carrier wave signal generator 110 can be set up to generate _{the carrier wave with a constant carrier frequency f T .} Alternatively or additionally, the carrier wave signal generator 110 can be set up to change _{the carrier frequency f T over time.} For example, the carrier wave signal generator 110 can be set up to provide a plurality of selection frequencies and to generate the carrier wave with one of the selection frequencies as the carrier frequency.

wobei A die Amplitude ist. Wie vorstehend erwähnt kann die Trägerfrequenz f_T konstant sein oder zeitlich verändert werden. Die Trägerwelle wird einem Modulator 140 zugeführt, wie durch einen Pfeil 112 in 2 veranschaulicht.Thus, the carrier wave signal generator 110 can provide the carrier wave as an output signal. The carrier wave generated by the carrier wave signal generator 110 can be represented as follows: $$A\text{cos}\left(2\pi f_{T}t\right).$$

where A is the amplitude. As mentioned above, the carrier frequency f _{T can be} constant or can be varied over time. The carrier wave is fed to a modulator 140 as indicated by arrow 112 in 2 illustrated.

The carrier frequency f _T can correspond to a central frequency fz of a frequency band of the respective GNSS service. For example, the center frequency fz is 1176.45 MHz, 1191.795 MHz, 1227.60 MHz, 1278.75 MHz or 1575.42 MHz. Furthermore, a plurality of selection frequencies can be provided within a frequency band spaced from the central frequency fz and by regular intervals from the central frequency fz. For example, the frequency spectrum (frequency band) has a width equivalent to an integer multiple of the satellite's fundamental clock, for example, the satellite's fundamental clock being 1.023 MHz. The intervals between the selection frequencies can be achieved by dividing the frequency band by integers.

The circuit 100 also has a spreading code signal generator 120 . The spreading code signal generator 120 generates a sequence of modulation states, a chip sequence, according to a spreading code that is unique to, and therefore characteristic of, each satellite. The spreading code signal generator 120 may be hard-programmed for this purpose (such as using a shift register), or retrieve the spreading code from memory (not shown). For convenience, the chip sequence generated by the spreading code signal generator 120 is referred to as the spreading code c(t).

As in 2 shown, the spreading code can be binary and jump between two states, resulting in a rectangular waveform. The spreading code is applied to the modulator 140 as indicated by an arrow 122 in 2 illustrated.

The circuit 100 includes a frequency function generator 130 . The frequency function generator 130 is set up _{to generate a frequency function f i} (t) for a time interval [t(i), (i+1)]. The frequency function f _i (t) specifies a time profile in a frequency space. The waveform of the frequency function f _i (t) can be represented as cos(2π f _i (t)t).

For example, the frequency function generator 130 receives instructions for generating the frequency function from a control unit 150, which _{can specify a class, an initial value, all coefficients, etc. of the respective frequency function f i} (t). Alternatively or additionally, the frequency function generator 130 can be connected to a memory (not shown) in order to access a selection of frequency functions stored there and _{to select a frequency function f i} (t) for the respective time interval. The frequency function generated by frequency function generator 130 is applied to modulator 140 as indicated by arrow 132 in FIG 2 illustrated.

The circuit 100 may further include the control unit 150 that controls the frequency function generator 130 . In particular, the control unit 150 can specify the frequency function to be generated by the frequency function generator 130 for a respective time interval [t(i), t(i+1)], including the class, initial value and coefficients. The control unit 150 may include processor means to execute operations and instructions to control the frequency function generator 130 and thus to generate the frequency function f _i (t).

In further examples, the circuit 100 may further comprise a communication unit (not shown) to receive instructions regarding the generation of the frequency function f _i (t) from an administrator of the satellite, located for example at a center on earth. In particular, the communication unit can be connected to the control unit 150 in order to control it from the earth.

The modulator 140 is set up to _{apply the frequency function f i} (t) to the carrier wave. Thus, the carrier frequency f _{T of} the carrier wave varies with time as defined by the frequency function f _i (t). In particular, the frequency function f _i (t) is designed in such a way that the carrier frequency f _{T is changed} periodically after a specific time period Δt, where Δt=[t(i), t(i+1)] and i is an integer.

The modulator 140 is also set up to modulate the spreading code onto the carrier wave. Thus, the modulator 140 produces a signal with the frequency function f _i (t) applied to the carrier wave and the spreading code modulated onto the carrier wave. The signal is output from the modulator 140 as a result, as illustrated by arrow 142 . This signal 142 can be fed to a transmitter device and sent out.

In 2 For example, modulator 140 is shown as a single element for simplification and illustration. In reality, the modulator 140 may comprise several separate parts. In particular, the application of the frequency function f _i (t) to the carrier wave and the modulation of the spreading code c(t) to the carrier wave can take place in separate parts of the modulator 140 . For example, the spreading code signal generator 120 and the frequency function generator 130 may each first receive the carrier wave from the carrier wave signal generator 110 and modulate the spreading code c(t) thereto or _{apply the frequency function f i} (t) thereto. The intermediate results (the carrier wave after application of the frequency function f _i (t) and the carrier wave after modulation with the spreading code c(t)) are combined (superimposed, multiplied) and output as the output signal of the modulator 140, as illustrated by an arrow 142 .

3 shows exemplary time curves in the frequency domain. In particular, the chronological sequences result in 3 after applying the frequency function f _i (t) to the carrier frequency as described above.

In 3 several consecutive time intervals [t(i), t(i+1)] to [t(i+4), t(i+5)] are shown along the vertical axis. The time intervals are of uniform length, ie a time period Δt of a respective time interval is constant, where Δt=t(i+1)−t(i) and i is an integer between 1 and 5. For example, Δt can generally be 10 ^-4 to 10 ³ seconds, in particular 0.01 to 1 second.

The horizontal axis of the chart in 3 represents frequency in arbitrary units. A number of discrete frequencies including a center frequency fz and selection frequencies f _-3 to f ₊₃ are represented along the horizontal axis.

According to the subject matter described herein, the center frequency fz can be chosen arbitrarily. For certain GNSS services, the center frequency fz (sometimes also called center frequency in the literature) can be fixed according to a standard frequency band of the respective GNSS (eg Galileo, GPS, Glonass, Beidou, etc.). For example, the center frequency fz can be 1575.42 MHz for the Galileo GNSS Ei frequency band (and also L1 GPS GNSS frequency band), 1278.75 MHz for the Galileo GNSS E6 frequency band, 1191.795 for the E _5- frequency band of the Galileo GNSS, 1176.45 MHz for the _e-5 a frequency band of the Galileo GNSS ₍₅ L and frequency band of the GPS GNSS), 1207.14 MHz for the E5b-frequency band of the Galileo GNSS or 1227.6 MHz for the L2 frequency band of the GPS GNSS.

The selection frequencies f _-3 to f ₊₃ can be determined by setting a distance Δf between the adjacent selection _{frequencies f -3} to f ₊₃ and the central frequency fz and a number of the selection frequencies. in the in 3 In the example shown, the number of selection frequencies is 6, providing 3 selection frequencies on either side of the center frequency fz in the frequency space. The distance Δf between adjacent selection _{frequencies f −3} to f ₊₃ and the central frequency fz, ie between f _a+1 and f _a (where fz=̂ f _o ) can be equivalent to a product of a factor with the basic clock (1.023 MHz). , where the factor is 0.01 to 1000.

Alternatively or additionally, the selection frequencies can be determined by specifying the bandwidth F of the respective frequency range together with the number of selection frequencies. in the in 3 example shown, the number of selection frequencies is 6, with 3 selection frequencies on both sides of the central frequency fz in the frequency domain. The bandwidth F = fmax - fmin can be determined based on the base clock (1.023 MHz) of the satellite, specifically as an integer multiple of the base clock. For example, the bandwidth is equivalent to a product of the basic clock and a factor, where the factor can be an integer from 0.01 to 1000, in particular 0.5 to 2.

In the example of 3 the individual time curves are each shown with a thick line f _T(i) to f _T(i+4) . As explained above, the time curves initially represent the content of the frequency function f _i (t); the time characteristics of the carrier frequency f _T(i) to f _T(i+4) are obtained by applying the frequency function f _i (t) to the carrier frequency f _{T .}

The course over time in the time interval [t(i), t(i+1)] in the frequency domain is, for example, linear and constant. Accordingly, the carrier frequency f _{T (i)} remains constant at the selection _{frequency f -1} , which corresponds to a difference fz - Δf. Accordingly, the frequency function f _i (t) can have a single constant term and be represented by f _i (t) = f _-1 .

The course over time in the time interval [t(i+1), t(i+2)] is linear, for example, and begins at the selection _{frequency f+2} , which corresponds to an addition fz+2Δf. Accordingly, the carrier frequency f _{T(i+1) changes} linearly from the selection _{frequency f +2} to the selection _{frequency f −2} , which corresponds to a subtraction f _{Z −2Δf.} Accordingly, the frequency function f _i+1 (t) can be represented by f _i+1 (t) = c ₁ t + c _o , where c _o and c _{1 are} coefficients of a linear equation with c _o = f ₊₂ and c ₁ = (f _-2 - f ₊₂ )/Δt.

The course over time in the time interval [t(i+2), t(i+3)] is, for example, parabolic, ie a polynomial of the second order. Accordingly, the carrier frequency f _{T(i+2) changes} starting from the selection _{frequency f +1} , which corresponds to an addition fz+Δf, along a parabola to the selection frequency f ₊₂ . Accordingly, the frequency function f _i+2 (t) can be represented by f _i+2 (t) = c ₂ t ² + c ₁ t + c _o , where c _o , c ₁ and c _{2 are} coefficients of a second degree polynomial with c _o = f _+1. c ₁ can be any number and c ₂ can be any positive number. It should be noted that the curve in 3 is illustrative and may not be congruent with a parabola.

The course over time in the time interval [t(i+3), t(i+4)] is exponential, for example. Accordingly, the carrier frequency f _{T(i+3) changes,} starting from the selection _{frequency f −3} , which corresponds to a subtraction f _Z −3Δf, along an exponential curve to the central frequency fz. For example, the frequency function f _i+3 (t) can be represented by f _i+3 (t) = c ₂ exp(c ₁ t) + c _o where c _o , c ₁ and c _{2 are} respective coefficients with c _o = f- ₃ . c ₁ and c ₂ can be any positive numbers. The exponent can have another polynomial instead of a linear equation and itself have a constant term. It should be noted that the curve in 3 is illustrative and may not be congruent with an exponential curve.

The course over time in the time interval [t(i+4), t(i+5)] is linear, for example. Accordingly, the carrier frequency f _{T(i+4) changes,} starting from the selection _{frequency f +3,} according to a linear slope. For example, the frequency function f _i+4 (t) can be represented by f _i+4 (t) = c ₁ t + c _o where c _o = f ₊₃ and c _{1 is} any negative number. In particular, the value at the end of the time interval [t(i+4), t(i+5)] is not one of the selection frequencies. This can also apply to other classes of mathematical functions.

In the examples above, the initial value was always defined by a constant term of the frequency function. Alternatively, as discussed above, the respective constant term in the frequency function may be omitted. Instead, the carrier frequency itself can be provided differently per time interval. This can be done, for example, by the carrier wave signal generator 110 or by an additional unit (not shown) that can modify the carrier wave.

In some examples, the frequency function f _i (t) determines one of the selection _{frequencies f -3} to f ₊₃ as the initial value for a respective time interval [t(i), t(i+i)]. The initial value may be given in the form of a constant (time-independent) mathematical term in the frequency function, as explained above. Alternatively or additionally, the carrier wave signal generator 110 _{can select one of the selection frequencies f -3} to f ₊₃ as the carrier frequency f _T and generates the carrier wave accordingly.

The examples show that the time profile of the carrier frequency can be changed periodically. It should be noted that the in 3 curves shown are exemplary and illustrative and are partly simplified. The order of the gradients and the form of the individual gradients can be changed as desired. In particular, the mathematical function classes shown as an example (exponential, linear, polynomial) can be combined with one another. The individual curves can alternatively or additionally contain a mathematical term that is 3 is not shown, such as logarithmic, trigonometric, etc., as discussed above.

The time courses according to the subject matter described here do not necessarily have to start at one of the selection frequencies, i.e. the initial value for a single time interval can be chosen arbitrarily and outside of the selection frequencies. For example, any starting value that lies within the available frequency band can be determined using a constant term.

4 shows a schematic diagram with time sequences of transmitted signals. In 4 shown are two examples of a GNSS signal 210 and 220. The horizontal axis in 4 represents a time axis t. The time axis t is divided into several time intervals, where in 4 exemplarily the time intervals [t(i-2), t(i-1)] (partly), [t(i-1), t(i)], [t(i), t(i+1)] and [t(i+i), t(i+2)] (partially). Although the time intervals in 4 are shown as having a constant length Δt = t(i) - t(i-1), in other examples (not shown) the length of each time interval may be variable, as discussed above. In other words, the widths of the time intervals can vary over time.

The GNSS signal 210 according to an example has a first signal component 212 and a second signal component 214 . The first signal component 212 can correspond to a pilot channel (partially encrypted), authentication channel or data channel or be transmitted on a data channel. The second signal component 214 may be or be transmitted over a data channel separate from the aforementioned data channel and/or an authentication channel. In other examples, the second signal component 214 may be superimposed on the first signal component 212 in the data channel.

In other examples, the second signal component 214 may be transmitted over a separate, external communication channel outside of satellite signaling. The external communication channel can include a wired network connection, a wireless network connection or other radio connections such as the Internet, LTE, Wifi, 5G, etc.

Unless explicitly stated otherwise, the terms data channel, pilot channel, and authentication channel as used herein may be defined according to industry-standard language usage in the field of GNSS. In particular, a data channel can generally denote a channel for a radio signal of the GNSS over which the navigation message is transmitted.

As described above, a navigation message, the spreading code and a frequency function are modulated onto the carrier wave. The respective modulation can be done individually or in combination. The content of the navigation message is as described above and includes in particular position data (ephemeris) and a respective time of the satellite and is updated regularly. The frequency function changes for each time interval [t(i), t(i+1)] and is therefore valid for the respective time interval. The time interval at which the frequency function is changed is independent of or may be linked to an update period of the navigation message.

The navigation message can be encrypted. This can be done using an NMA method. For example, a message authentication code, or MAC code for short, can be generated by applying an authentication key to the navigation message. The MAC code can be sent in addition or as an alternative to the navigation message. In some examples, a chain of authentication keys can be generated, for example using a one-way hash function. The authentication keys of such a chain can be sequentially applied to the navigation message in a reverse order to their generation in order to encrypt it. In such examples, the MAC code may be broadcast in addition to or instead of the navigation message.

The first signal component 212 may represent a signal resulting from the modulation of the navigation message (and/or a MAC code), the spreading code, and the frequency function onto the carrier wave. Thus, the first signal component 212 can transmit an encrypted part of the GNSS signal. In 4 the first signal component 212 is shown continuously and illustrates that some information, in particular the navigation message, can be sent out continuously. However, depending on the implementation, the first signal component may have discontinuities (not shown).

The second component signal 214 transmits information for authentication and/or decryption of the first component signal 212 . In addition to the parameter string, the authentication key for MAC code decryption can be sent out. Thus, the second signal component 214 can transmit a decrypting and/or authenticating part of the GNSS signal.

As in 4 As illustrated, the second signal component 214 is not emitted continuously, but is emitted at regular intervals in a final portion of each time interval. Accordingly, the receiver receives the first signal component 212 continuously and stores it for the entire time interval vall while receiving the second signal component 214 at regular intervals. The parameter string, which is only fully received at the end of the respective time interval, is used to map the frequency function. The frequency function can be used to emulate the spreading code for the respective time interval and thus to decrypt or authenticate the navigation message. If necessary, further keys, which are transmitted via the second signal component, can be used in order to completely decrypt the navigation message and/or to authenticate the received GNSS signal.

The GNSS signal 220 according to another example has a first signal component 222 and a second signal component 224 . The signal components 222, 224 of the GNSS signal 220 are transmitted in a common channel, in particular in a data channel. In this example, the first signal component 222, which transmits the encrypted part of the GNSS signal, and the second signal component 224, which transmits the decrypting and/or authenticating part of the GNSS signal, are transmitted alternately. Accordingly, the transmission of the first signal component 222 is interrupted in order to transmit the second signal component 224. In particular, the first signal component 222 is transmitted in a first, initial portion of a respective time interval. The second signal component 224 is transmitted in a second section, in particular towards the end, of a respective time interval.

The composition of the first signal component 222 and the second signal component 224 may be as described above with respect to the first signal component 212 and the second signal component 214 of the signal 210 . The reception and use of signal components 222, 224 may also be as described above with respect to signal 210. A repetition of the same analogous features is therefore avoided.

In further, in 4 In some examples that are not explicitly shown, the parameter string can be sent in encrypted form via a public GNSS service, with the recipient receiving a key for decrypting the parameter string separately.

5 10 shows a schematic circuit diagram of a circuit 10 ² according to a further example. The circuit 10 ² has all the structural and functional features of the circuit 100 of FIG 2 on, which can be recognized by the same reference numbers.

The circuit 10 ² in 5 further comprises an encryption function generator 160 which _{generates a signal according to an encryption function w i} (t). The encryption function may comprise a chip sequence having the same overall length as the chip sequence of the spreading code c(t). The control unit 150 can control the selection and provision of the encryption function w _i (t) and the corresponding signal.

The signal generated by the encryption function generator 160 according to the encryption function w _i (t) is fed to a second modulator 144 as illustrated by an arrow 162 . The second modulator 144 multiplies the chip sequences of the spreading code c(t) and the encryption function w _i (t) and feeds the result to the modulator 140 .

According to an example, a modulation method for the spreading code can be changed periodically as described above. The encryption function w _i (t) depends on the particular modulation scheme and therefore has the modulation scheme as a variable or parameter. The modulation scheme for the spreading code can include in particular between phase modulation schemes such as BPSK(n), BOC(i,j), altBOC, TMBOC, etc. with time varying parameters as described above.

6 12 shows a schematic circuit diagram of a circuit 104 according to a further example. The circuit 104 of 6 has all the structural and functional features of the circuit 100 of FIG 2 and circuit 102 of FIG 5 on, which can be recognized by the same reference numbers.

The circuit 104 of 6 further includes a chip phase shift generator 170 . The chip phase offset generator 170 generates a chip phase offset (time delay) by which the phase of the spreading code is shifted, in particular cyclically. The phase offset of the spreading code can be related to an associated pilot signal of the transmitter, as described above. The pilot signal can be provided and sent to, among other things, the extraction of the spreading code and / or the to facilitate message. For example, a correlation process between a data signal (ie carrier signal superimposed with the message and the spreading code including respective encryption) and the pilot signal (carrier wave modulated with the spreading code) is performed to extract the message and/or the spreading code.

The offset generated by chip phase offset generator 170 is provided to a third modulator 146 as illustrated by arrow 172 . Thus, signals entering the third modulator 146 include the spreading code c(t), the encryption function W _i (t), a periodically changed modulation scheme, and a chip phase offset for the spreading code. The resulting encrypted and modified spreading code is applied to modulator 140 as indicated by arrow 148 and modulated onto the carrier wave and transmitted as indicated by arrow 142. FIG.

7 Figure 12 is a schematic representation of periodically varying chip phase offset and periodically varying phase modulation scheme. The BPSK(2) modulation method can be used as a reference for the chip phase offset, with the parameter (2) specifying a chip rate, with the digit specifying a multiplier with the basic clock (1.023 MHz).

In the example of 7 a chip phase offset Φ corresponding to 0.5 of the phase length of the BPSK(2) modulation method is shown. This is followed by a BPSK(2) modulation, followed by a BPSK(1) modulation and a BOC(1,1) modulation. For example, that puts in 7 The example shown represents a chip sequence of (-1, +1, -1), the chips being produced by modulation methods which are different from one another.

The example in 7 is purely schematic and is used for illustration purposes. In particular, the modulation method can be changed periodically per time interval [t(i), t(i+1)] and remain unchanged within a time interval.

8th shows schematic and exemplary power spectra of different phase modulation methods and a broadband spreading code authentication method (BSCA). It is in 8th to recognize that the application of a single modulation method leaves behind a characteristic power spectrum with pronounced maxima in predictable frequency ranges. The BSCA according to the invention, on the other hand, can enable a homogeneous, uniform use of the entire available frequency range and disguise the modulation methods used in the power spectrum.

Reference List

10

GNSS

12-18

satellite

20

recipient

22

GNSS signal

30

Unauthorized/attacker

32

GNSS signal

34

mock GNSS signal

100, 102, 104

circuit

110

Carrier Wave Signal Generator

112

carrier wave signal

120

Spreading Code Signal Generator

122

spreading code signal

130

frequency function generator

132

frequency function signal

140, 144, 146

modulator

142

output signal

148

encrypted spreading code signal

150

control unit

160

encryption function generator

162

Encryption Function Signal

170

Chip Phase Shift Generator

172

chip phase shift

210

Channel

212-220

sent signals

230

duration

240

recipient

242-250

received signals

Claims (15)

A method of operating a Global Navigation Satellite System, GNSS, the method comprising: providing a spreading code; providing a carrier wave having a carrier frequency; applying a frequency function to the carrier wave, the frequency function defining a time-dependent course in frequency space and changing periodically; and modulating the spreading code onto the carrier wave. procedure after claim 1 , periodically changing a modulation method according to which the spreading code is modulated onto the carrier wave. procedure after claim 1 or 2 , further comprising: introducing a chip phase offset when modulating the spreading code onto the carrier wave; and periodically changing the chip phase offset. A method according to any one of the preceding claims, wherein modulating the spreading code onto the carrier wave comprises: overlaying the spreading code with an encryption chip sequence to obtain an encrypted spreading code; and modulating the encrypted spreading code onto the carrier wave. A method as claimed in any preceding claim, wherein applying the frequency function to the carrier wave causes the carrier frequency to vary within a frequency band. Method according to one of the preceding claims, providing multiple discrete selection frequencies; and designate one of the plurality of discrete selection frequencies as the carrier frequency of the carrier wave for a time interval [t(i), t(i+1)], respectively. Method according to one of the preceding claims, further comprising: providing and transmitting a GNSS signal s _i in a time interval te [t(i), t(i+1)] according to the following form: $$s_i=A{w}_i\left(t\right)c\left(t\right)\text{cos}\left(2\pi \left(f_T+f_i\left(t\right)\right)i\right),$$

where: A is the amplitude of the signal, w _i (t) is an encryption function for the spreading code, c(t) is the spreading code, f _{T is} the carrier frequency, and f _i (t) is the frequency function. procedure after claim 7 , wherein the encryption function w _i (t) further comprises at least one of: - a chip phase offset when modulating the spreading code onto the carrier wave; - a respective phase modulation method for the time interval [t(i), t(i+1)]. A method according to any one of the preceding claims, further comprising: transmitting the carrier wave modulated with the spreading code; and after sending out the carrier wave modulated with the spreading code, sending out a parameter character string for extraction of the frequency function. procedure after claim 9 , further comprising: providing a navigation message; modulating the navigation message along with the spreading code onto the carrier wave. procedure after claim 10 , further comprising: applying an authentication key to the navigation message to obtain an authentication code; sending the authentication code before, with or after the navigation message; Sending out the authentication key after the complete sending out of the authentication code. Method according to any of the preceding claims, wherein a parameter string for extraction of the frequency function is broadcast over a communication channel outside of GNSS and/or over a data channel of a public GNSS service, wherein the parameter string is overlaid with a broadcast string. Procedure according to one of claims 9 until 12 , further comprising: receiving a GNSS signal transmitted with the carrier wave to which the spreading code is modulated and the frequency function is applied; storing the GNSS signal for a period of time equivalent to a lifetime of the frequency function; mapping the frequency function using the parameter string; and mapping the spreading code using the mapped frequency function. Device for use in a GNSS for carrying out the method according to one of Claims 1 until 12 is set up. Receiver device of a GNSS system set up to carry out the following method steps: receiving a GNSS signal; storing the GNSS signal for a period of time equivalent to a lifetime of the frequency function; mapping the frequency function from the parameter string; and Map the spreading code using the mapped frequency function.

Images