Method for detecting a relay attack by comparing tbr and pbr at different measuring point quantities

Abstract

The invention relates to a method for detecting and / or preventing relay attacks. The object is achieved by a method for detecting a relay attack, wherein radio signals with different frequencies are transmitted between a first and a second object and phase measurements and propagation time measurements are made on these radio signals and the change in the phase measurements when the frequency is changed is compared with the signal propagation measurements or their change, and wherein a relay attack is assumed if a predetermined deviation or a deviation determined from measurements on the radio signals is exceeded, wherein a measurement series is formed based on the phase measurements and / or a reference measurement series is formed based on the signal propagation time measurements, and the filtered and / or smoothed measurement series is compared with the at least one signal propagation time measurement or its change and / or the reference measurement series as a reference, and / or wherein the measurement series or its change is compared with the filtered and / or averaged reference measurement series.

Description

[0001] Method for detecting a relay attack by comparing TBR and PBR for different sets of measuring points

[0002] The invention relates to a method for detecting and / or preventing relay attacks.

[0003] Numerous methods are known for detecting or preventing relay attacks, in particular man-in-the-middle or wrap-around attacks or phase-manipulation attacks or early-detect-late-commit attacks.

[0004] Furthermore, EP 3 564 703 A1 discloses determining the phase position of a received signal and comparing it with a phase position determined by another method. If a predetermined deviation is exceeded, a relay attack is assumed. It thus discloses the comparison of a temporal position or phase position with a predetermined or other calculated temporal position and / or phase position, and not the comparison with a propagation time. Furthermore, the temporal position is not a propagation time but an equivalent to the phase position.

[0005] It is also known from WO 2022 / 096 514 A1 that the distances obtained from distance measurements when using commercially available transceivers such as the somewhat older cc2500 or the current cc26xx from Texas Instruments or the Kw35 / 36 / 37 / 38 from NXP or the Dialog DA1469x from Dialog depend on the frequency used to determine the distance. Furthermore, the document discloses a method for detecting a relay attack, wherein radio signals with different frequencies are transmitted between a first and a second object, phase measurements and propagation time measurements are performed on these radio signals, and the change in the phase measurements upon changing the frequency is compared with the signal propagation measurements or their change, and wherein a relay attack is assumed if a predetermined deviation or a deviation determined from measurements on the radio signals is exceeded.

[0006] It is also known that the measured values ​​obtained by phase and propagation time measurements change concurrently and almost proportionally with changes in the frequency used for the measurement. From this, it was recognized that a deviation in the changes has an unnatural origin and can be used to detect a relay attack. It was recognized that it is not necessary to calculate the distances; instead, measured values ​​can be compared with little or no processing, thus allowing a relay attack to be easily and reliably detected, which is the preferred approach.

[0007] The latter method works well when sufficient measurement values ​​are available. However, especially with few phase or time-of-flight measurements, the method cannot optimally utilize the large number of measurements compared to the other method (phase measurement or time-of-flight measurement). This is where the invention comes in.

[0008] For example, the typical Bluetooth implementation provides far fewer propagation time measurements than phase measurements, particularly since signal propagation time measurements are usually provided with only one antenna pair and only at a few frequencies. For example, with only 19 propagation time measurements, there are usually more than 250 phase measurements. These measurements are necessarily subject to some measurement error. Comparing a signal propagation time measurement with a phase shift change between two frequencies according to the state of the art cannot profitably utilize the large number of phase measurements.

[0009] The object of the invention is to provide an improved method that makes the best possible use of available measured values ​​in order to be more precise and / or more reliable. In the above-mentioned example, the method according to the invention can increase the accuracy by more than a factor of five. For example, the phase measurements are processed through an IIR filter with a first frequency sequence forwards and then a second time backwards, i.e. with the frequency sequence reversed. The gradients (of the phase curve with the change in frequency) taken from the phase curve obtained in this way (and in particular continuously continued between the measurement points) at the frequencies of the signal propagation time measurements are compared with the signal propagation times. Instead of selecting phase measurements available for two frequencies adjacent to / identical to the frequency of the respective signal propagation time measurement and comparing their change with the signal propagation time.Something similar can also be achieved using FIR filters.

[0010] However, it has been shown to be disadvantageous to filter and / or smooth both signal propagation time measurements and phase measurements across frequencies, for example, and to compare the two results with each other. However, it is not necessarily harmful to average several measurements at nearly identical frequencies. Smoothing / filtering across frequencies for both measurement types, however, leads to a situation where relay attacks can no longer be detected as reliably as with the solution of the invention. Thus, the basic idea of ​​the invention is to smooth / filter one of the measurement types (phase or propagation time) for which more measured values ​​are available and to compare it with the measured values ​​of the other measurement type, which can also be achieved through a transformation.

[0011] The problem is solved by a method for detecting a relay attack, wherein radio signals with different frequencies (f1, f2, f3) are transmitted between a first and a second object (01, 02), and phase measurements are carried out on a first subset of the radio signals with different frequencies, and wherein these phase measurements are compared with a reference, and wherein, if a deviation that is predetermined or determined from measurements on the radio signals is exceeded, a relay attack is assumed, access or release is denied, a requested action or operation is not carried out, and / or an alarm or blocking is carried out, and / or wherein, if the deviation falls below the predetermined or determined from measurements, in particular signal flow measurements, the absence of a relay attack is assumed,access or release is granted and / or the requested action or operation is carried out and / or the alarm or blocking is not carried out, wherein a signal propagation time measurement is carried out on a second subset of the radio signals with different frequencies, which is in particular at least partially identical to the first subset, characterized in that a series of measurements is formed based on the phase measurements and / or a reference series of measurements is formed based on the signal propagation time measurements, in each case depending on the frequency of the radio signal frequencies used for the measurement, wherein the series of measurements and / or the signal propagation time measurement series is / are formed either in a strictly monotonically increasing or in particular in a strictly monotonically increasingmonotonically decreasing sequence of frequencies and that based on this series of measurements a filtered and / or smoothed series of measurements and / or based on the reference series of measurements a filtered and / or smoothed reference series of measurements is determined and that the filtered and / or smoothed series of measurements, in particular its change over the frequency, is compared with the at least one signal propagation time measurement or its change and / or the reference series of measurements as a reference and / or wherein the series of measurements or its change is compared with the filtered and / or averaged reference series of measurements.

[0012] Subsets have, in particular, at least partially similar frequencies to one another, for example identical and / or adjacent frequencies in the radio protocol used. In particular, one frequency is similar to another if the frequency spacing for frequencies above 1 GHz is a maximum of 5 MHz, in particular a maximum of 3 MHz. In this case, at least 50%, in particular at least 90%, preferably 100%, of the frequencies of the radio signals contained in one subset and in the other of the first and second subsets contain similar and / or identical frequencies, in particular to one, in particular the smaller, of the first and second subsets, wherein each frequency of the other is only ever considered similar to a maximum of one frequency of the one, in particular when it is determined for frequencies of one whether similar or identical frequencies are present in the other.

[0013] It is particularly advantageous if the number of measurements and / or the frequencies of the measurements of the subsets differ, in particular by at least a factor of two, preferably by a factor of four. In such cases, the method can demonstrate its advantages particularly clearly and ensure significant quality improvements.

[0014] It is particularly advantageous if the method is carried out in such a way that for each measurement in one of the first or second subsets there is a measurement in the other subset that is similar or identical in terms of frequency, in particular a similar measurement (in particular with a deviation of no more than 5 MHz) with a higher frequency and a similar measurement with a lower frequency, which was carried out in close temporal proximity and / or with a largely unchanged radio channel. A largely unchanged radio channel can be ensured in particular by a close temporal proximity, whereby this depends on the movement of the objects and changes in the actual environment and changes in radio interference, in particular in relation to the wavelength. The effects should result in a phase change of less than 90° between the measurement times.As a rule, a sufficiently close temporal relationship is present if the temporal separation is less than 10 ms, especially at frequencies in the GHz range. For example, at 2.4 GHz, the wavelength is approximately 12 cm; assuming a maximum movement of 4 m / s (brisk jogging) = 3 cm per 7.5 ms, the temporal separation could be chosen to be less than 7.5 ms.

[0015] When a series of measurements changes over frequency, it is particularly important to consider its change over a frequency range or against the frequency, for example the gradient in a display against the frequency.

[0016] The comparison does not have to be made by converting to distances, but the deviation can also be predetermined in other units that allow for a comparison. Conversion is particularly possible using the relationship

[0017] Phase shift = 2 Pi * (distance) * frequency / c, where it should be noted that beyond a certain distance an ambiguity must be taken into account and c is equal to the speed of light

[0018] RTT = 2 * distance / c and consequently, neglecting ambiguity:

[0019] Phase shift = Pi * (RTT*c) * Frequency / c

[0020] And / or dPhaseshift(f 1 ,f2) = Pi * (RTT*c) * dFrequency(f 1 ,f2) / c thus (again illustrating the ambiguity) dPhaseshift(f 1 ,f2)RT / dFrequency(f 1 ,f2) = Pi * RTT it should be taken into account that ambiguity arises at distances greater than c / dFrequency. In most applications, however, the frequency spacing can be chosen so that no ambiguities arise, at least at distances under 150 meters. In particular, the possible distance and / or the frequency difference dFrequency is / are chosen so that no ambiguity arises or that it can be neglected. In particular, the distance is smaller than the speed of light divided by the frequency difference of the phase shift measurement, in particular at least half smaller, in particular less than 300m, in particular less than 150m.

[0021] Here, phase shift(f 1 ,f 2)RT is a phase shift between the transmissions at frequencies f 1 and f 2 from one object to another and back, which occurs due to the distance. It can be approximately equated to twice the phase shift that occurs during transmission from one object to another due to the distance. Furthermore, dPhaseshift(f 1 ,f 2) is the distance-related phase shift difference, if corrected, of the radio signals received at frequencies f 1 and f 2, dFrequency is their difference, and c is the speed of light. RTT is the signal round-trip time from one object to the other and back. Instead of accepting the ambiguity problem, one can also resolve the ambiguity using other methods and formulate it by adding the determined correction term to resolve the ambiguity.

[0022] Conversions are also conceivable, for example, to a free or dimensionless quantity and / or by means of transformation, e.g., FFT. If a quantity with a new dimension is calculated for comparison using FFT, the larger subset is preferably reduced so that it has the same or a similar size, in particular one that differs by a maximum of 10%, to the smaller subset. In particular, the reduction is carried out in such a way that at least 90% of the frequency of the measurements, in particular each frequency, of the smaller subset is retained as similar as possible, whereby, in particular, only 1:1 assignments are made.

[0023] For example, a filtered and / or averaged measurement series and / or reference measurement series can be generated using a transformation, particularly FFT. For example, the phase measurements can be subjected to an FFT and compared with the time-of-flight measurements. The FFT can be selected so that distances are ultimately compared, but also other dimensions can be used for comparison quantities such as distance divided by the speed of light.

[0024] In particular, if two phase measurements with adjacent frequencies f 1 , f 2 and a close temporal relationship do not have a sufficiently undisturbed signal propagation time measurement at one or more frequencies in the range of 90% f 1 to 1 10% f 2 , where f 1 is less than or equal to f 2 , known methods cannot provide suitable solutions. In particular, f 1 , f 2 , and f 3 lie in the range of 1 GHz to 10 GHz, especially in the range of 2 MHz to 6 GHz, and in particular, they are frequencies of a Bluetooth channel.

[0025] However, even if a propagation time measurement at frequency fl lacks sufficiently undisturbed phase measurements with adjacent frequencies f 1 , f 2 and close temporal correlation, and fl in the range of 90% f 1 to 1 10% f 2 , where f 1 is less than or equal to f 2 , known methods cannot provide good solutions. In particular, f 1 , f 2 , and fl lie in the range of 1 GHz to 10 GHz, especially in the range of 2 MHz to 6 GHz, and in particular, they are frequencies of a Bluetooth channel.

[0026] Particularly preferred embodiments are those in which changes in the phase measurements when the frequency changes, or in which changes in the phase shift resulting from the distance, are compared with the signal propagation time measurement when the frequency changes, in particular relative to the change in frequency. Or in other words, particularly preferably changes in the phase measurements or, in particular distance-related, phase shifts when the frequency changes are compared with the signal propagation time measurements, and this comparison is carried out for different frequencies, and it is checked whether these comparisons lead to comparison values ​​that lie within a predetermined range or a range determined from measurements on the radio signals, or exceed a predetermined deviation or a deviation determined from measurements on the radio signals, in particular from a predetermined value or a value determined from measurements on the radio signals.In particular, if the comparison values ​​are outside the range or interval or if the deviation is exceeded, a relay attack is assumed to have been detected and / or otherwise it is concluded that no relay attack has occurred.

[0027] The change in phase shift caused by or resulting from the change in frequency is due to the fact that, especially when the distance is approximately the same for both measurements, a different number of wave trains fit the distance and therefore the phase shift caused by the distance is different between the frequencies. This change in phase shift due to the frequency is the phase change caused by the change in frequency. This causes problems when measuring because the phase measurement is always dependent on a reference and an often undefined phase jump can occur when switching to transmit between different frequencies. Thus, switching for transmitting, and especially for receiving, is preferably phase-coherent, i.e. with a phase jump of zero. However, it is also sufficient to determine or know the phase jump.Then you can determine the phase change caused by the frequency change by correcting the measured phase change by the phase jump when switching the transmitter and the phase jump when switching at the receiver to measure the measured phase change.

[0028] The comparison can be performed as follows: determining a comparison coefficient k and checking whether it lies within a specified range as a predetermined deviation. In all subsequent descriptions, dPhase shift(f 1 ,f 2) / dFrequency(f 1 ,f 2) is advantageously obtained as the gradient described above, after filtering, in particular IFFT filtering, from the filtered, and in particular continuously completed, phase shift curve versus frequency. dPhase shift(f 1 ,f 2) / dFrequency(f 1 ,f 2) / RTT = k

[0029] However, this is not preferred, it is rather preferred to work without forming a ratio, but to determine a difference and check whether this lies within a predetermined range as a predetermined deviation, for example

[0030] Calculate the difference between the measured signal propagation time (ToF) (in one direction only, i.e., approximately 0.5 RTT) and the term dPhaseshift(f 1 ,f 2) / dFrequency(f 1 ,f 2) / 2 / Pi and check whether the result lies within a given range. Of course, differences can also be calculated between similar expressions, such as between

[0031] RTT x Pi and dPhase shift(f 1 ,f2) / dFrequency(f 1 ,f2)

[0032] Correcting the phase measurement at the receiver is particularly advantageous when phase-coherent frequency switching without phase shifting is not performed at the transmitter and / or receiver. This is because it is not the phase measured during reception itself that is important, but rather the distance-dependent phase change. Delta_PhaseChange(f1 ->f2) therefore preferentially incorporates the two phase shifts at f1 and f2 caused by the transmission.

[0033] However, other comparisons are also possible, provided the change in the radio signal's phase shift is compared with the signal propagation time over the frequency change, and this is done at multiple frequencies or frequency triples. This also allows distances to be calculated and compared.

[0034] It is also possible to neutralise motion effects, for example by taking several measurements at different times at the same frequency, in particular by averaging the phase measurements.

[0035] Preferably, the method is carried out such that the radio signals comprise a first plurality of radio signals and a second plurality of phase measurements and a third plurality of signal propagation time measurements are carried out, wherein the second plurality of signal propagation measurements are carried out on a fourth plurality of signals with a fifth plurality of frequencies and the third plurality of signal propagation measurements are carried out on a sixth plurality of signals with a seventh plurality of frequencies, and the range of the fifth plurality of frequencies and the range of the seventh plurality of frequencies overlap in the frequency interval, and in this frequency interval the change in the second plurality of signal propagation measurements is compared with the change in the third plurality of signal propagation measurements or the change. Particularly reliable detection is also possible in such an embodiment.The third majority is in particular in the range from 3 to 150 and the second majority is in particular in the range from 10 to 500, whereby their number differs in particular by a factor of at least two, in particular 4.

[0036] In particular, the frequencies at which the delay and / or phase measurements are performed are in a range of 25 to 100 MHz, in particular, they completely span such a range. In particular, the frequencies are in the range of 2 to 6 GHz. In particular, the spacing between adjacent frequencies used for the delay and / or phase measurements is in the range of 0.1 to 10 MHz, in particular in the range of 0.5 to 10 MHz.

[0037] Multiple comparisons can also be performed and an aggregated comparison number determined, which is then compared with a predetermined deviation. For example, the mean of the deviation can be calculated and compared with a predetermined deviation. It is also possible to sum the extent of the excess of a predetermined deviation and the extent of the under-utilization or non-utilization with opposite signs and compare them with a predetermined total deviation. Unequal weighting during aggregation is also conceivable.

[0038] Particularly advantageously, a plurality of individual comparisons, in particular at least three, are performed. For an individual comparison, in particular, a measured value of the reference measurement range is compared with the measurement series or the filtered and / or averaged measurement series, or a measured value of the measurement series is compared with the reference measurement range or the filtered and / or averaged reference measurement ranges and / or, particularly preferably, the change in these.

[0039] It is particularly advantageous to carry out the individual comparisons in such a way that a gradient is determined from the averaged, filtered, smoothed and / or transposed series of measurements and compared with a measured value of the reference series of measurements.

[0040] The results of the individual comparisons can then be aggregated and compared with a predetermined deviation or a deviation determined from measurements of the radio signals. In particular, at least three individual comparisons are aggregated.

[0041] It is particularly advantageous to use a filtered measurement series or a filtered reference measurement series, which is obtained, for example, using an FIR filter or an FFT, particularly by removing higher frequency components from the FFT result, such as the top 25% of the frequency spectrum, and applying an FFT to the remaining portion. This effectively reduces interference.

[0042] However, filtering using an IIR filter is particularly preferred, especially forward and backward (with respect to the order of the measurements' frequencies in a measurement or reference measurement series). This allows for even better correction of external influences and measurement errors. By passing the signal through the filter twice in reverse order, phase artifacts generated by the filter can also be largely avoided, which is particularly advantageous. A further advantage of using IIR is the lower computing power required and lower latency, which is particularly advantageous in quasi-real-time applications and / or where energy and / or computing power are limited. In particular, the method is carried out in such a way that the filtering is carried out in a mobile (end) device, in particular an automobile, mobile phone, or radio key, where the advantages are particularly evident.

[0043] Such a procedure can be further improved by extending the measurement or reference measurement series to be filtered at the beginning or end (with respect to the frequencies of the measurement or reference measurement series), in particular by enrichment by repeating the first measured value together with its frequency of the measurement or reference measurement series before its beginning and / or the last measured value with its frequency after the end of the measurement or reference measurement series.

[0044] It is particularly advantageous to use phase-based measurements and / or time-of-flight measurements for distance measurement, which also allows other attacks to be detected, other functions to be implemented, and the radio signals to be used efficiently.

[0045] It is preferred to select at least five phase measurements and / or transit-time measurements in the frequency interval and / or the number of frequencies at which phase measurements and / or transit-time measurements are performed in the frequency interval. The method is advantageously carried out with multiple first objects and a common second object. In particular, the common second object can represent an authentication means, such as an electronic key, for example, a key fob. This increases the reliability of detection and makes it more difficult to defraud the system.

[0046] One radio signal is distinguished from another, in particular, by its frequency. In particular, radio signals exhibit frequency differences that exceed the frequency stability of the hardware involved.

[0047] With further advantage, particularly to simplify comparisons and make them even more robust, the change between at least two, in particular all, of the fifth plurality of frequencies and / or between at least two, in particular all, of the seventh plurality of frequencies is carried out in a phase-coherent manner. Preferably, all frequency changes of at least one of the objects are carried out in a phase-coherent manner. With a somewhat lesser advantage but still advantageous, the phase shifts occurring during the frequency change, in particular at the transmitter and / or receiver, are measured and used to correct the phase measurements.

[0048] Advantageously, the method is carried out in such a way that runtime measurements and / or phase measurements on radio signals and / or radio signals with a power below a predetermined and / or, in particular, from or taking into account received radio signals, determined lower power limit are disregarded or are only taken into account with a lower weight, in particular those radio signals are disregarded or given a lower weight that are more than 50% below the average power of the received radio signals and / or wherein runtime measurements and / or phase measurements on radio signals and / or radio signals with a power above a predetermined and / or, in particular, from or taking into account received radio signals, determined upper power limit are disregarded or are given a lower weight, in particular those radio signals are disregarded or given a lower weight,which are more than 50% above the average power of the received radio signals.

[0049] In other words, it is preferred if measurements with low received power, in particular a received power below a predetermined value or proportion of the average or maximum received power, are not taken into account or are only taken into account with a lower weighting, and / or if measurements with very high received power, in particular a received power above a predetermined value or proportion of the average or maximum received power, are not taken into account or are only taken into account with a lower weighting. Such configurations allow the method to be designed to be particularly robust.

[0050] Preferably, the width of the frequency interval is at least 0.1 MHz and / or a maximum of 100 MHz and / or the frequency spacing between two consecutive different frequencies is at least 0.1 MHz and / or a maximum of 10 MHz and / or it is preferred if the different frequencies represent at least five frequencies and / or a maximum of 200 frequencies and / or wherein the radio signals are emitted on the different frequencies one after the other and / or consecutively, in particular directly consecutively, and / or wherein the bandwidth of the radio signals never exceeds 50 MHz, in particular 25 MHz.

[0051] In particular, it is preferred if the sequence of the radio signals emitted and / or used for comparison has a, in particular monotonically, preferably strictly monotonically, increasing or decreasing frequency sequence. This has proven particularly advantageous when using filters and / or transformations, in particular FIR and / or IIR filters.

[0052] One embodiment of the invention is characterized in that only the signals sent by the first object or (exclusively or) the signals sent by the second object are used to determine the relay attack. This includes embodiments in which only the first object transmits, as well as those in which only the second object transmits, and those in which both transmit but only a portion of the signals—namely, those sent by the first object or (exclusively or) those sent by the second object—are used to determine the distance.

[0053] The method preferably includes deciding whether to use the signals from the first or second object, in particular based on at least one estimate or determination of the effects of interference on reception at both objects. This decision can be made before or after the signals are transmitted, or after a portion of them is transmitted.

[0054] If speed is to be increased, it is preferable to make the decision as early as possible and to minimize the transmission of unused signals, especially after the decision to stop transmitting them. If the process is to be designed to be as robust as possible, the decision is made only after the signals from the first and second object have been transmitted. Transmitted and received signals can be used to make the decision. However, other data or measurements, such as noise or signals external to the process at the receiver, can also be used alternatively or in addition.

[0055] In particular, the radio signals of the first or (exclusively or) the second object are selected, the reception of which at the other of the two objects was, is or is expected to be less disturbed.

[0056] It is particularly advantageous if the first and / or second object switches between at least two of the multiple frequencies in a phase-coherent manner, or if a phase jump occurring at the switching object during switching is measured and taken into account in the calculation. This enables even more robust and simpler process control and distance measurement, and further advantages can be realized when using the signals by simplifying evaluations based on them. For example, if the time of the phase-coherent change or the change with a measured phase jump at the transmitting object is known and the change in the received signal is determined at the receiving object, the time between transmission and reception of the change can be determined, which represents the signal propagation time (ToF), and the phase shift can also be determined. The signal propagation time can be used to directly determine the distance using the speed of light.This is also possible via phase shift, but modulo the wavelength. Using multiple frequencies reduces the ambiguity in phase-based measurements. Combining signal-propagation time and phase-based measurements allows for particularly accurate and robust distance measurements.

[0057] Phase-coherent switching or alternation between two frequencies is understood in particular to mean that the time of switching is precisely determined or measured, and the phase after switching is known relative to the phase position before switching. This is the case if the phase change during switching is zero or a pre-known value. Alternatively, the phase jump occurring during switching can also be measured, particularly locally—i.e., particularly before transmission or at the receiver—and calculated and / or corrected before and / or during the comparison.

[0058] The phase difference or jump when changing between frequencies can be known, for example, because it is predetermined or can be derived from other known quantities, for example the duration of a radiation, in particular an immediately preceding one, at a frequency.

[0059] The phase difference usually occurs when switching between two frequencies for technical reasons, but can also be avoided. The switching between two frequencies can be carried out with a short interruption or without interruption. At the time of the interruption, the phase jumps, or during the switchover with an interruption, the phase of the signals that are thought to continue into the interruption jumps before and after the switching. At the time of the switchover without an interruption or at an imaginary changeover time during the interruption, in particular in the middle of the interruption and / or at the end of the signal before the interruption or at the beginning of the signal after the interruption, a defined phase jump occurs. This is the phase difference.

[0060] Particularly advantageously, the receiving object and / or the receiving object also switch phase-coherently, particularly between the different frequencies. In particular, the first and second objects switch between frequencies in a phase-coherent manner. This is achieved, in particular, by phase-coherent switching of at least one PLL on the first and / or second object. In particular, the objects are configured accordingly.

[0061] Advantageously, the second or (exclusive or) first object does not send signals for distance determination and / or the second or first object sends (exclusive or) signals only for time and / or clock synchronization. This saves energy and processing time.

[0062] Preferably, the first and / or second, or each of the two, objects transmits the signals on multiple frequencies consecutively and / or one after the other, particularly immediately after the other. In particular, when transmitting by the first and second objects, all signals from the first or second object are transmitted first, followed by those from the other. This allows, among other things, the influences of environmental or distance changes and of movements of one or both objects to be reduced.

[0063] Advantageously, the signal bandwidth never exceeds 50 MHz, especially 25 MHz. This saves energy, avoids interference with other processes, and allows for the use of simpler components compared to broadband methods.

[0064] Advantageously, the signals are transmitted via multiple antenna paths, in particular with multiple antennas, especially one after the other, at the transmitting object and / or received with multiple antennas at the receiving object. Particularly advantageously, if a relay attack is detected, access or authorization is denied, a requested action or operation is not performed, and / or an alarm or blocking is triggered. If a relay attack is not detected, access or authorization is granted and / or the requested action or operation is performed and / or the alarm or blocking is not triggered.

[0065] The problem is also solved by one or two objects, each equipped with transmitting and receiving means and a controller, configured to carry out the method according to the invention.

[0066] The objects are advantageously parts of a data transmission system, in particular a Bluetooth, WLAN, or mobile radio data transmission system. The radio signals are preferably signals of the data transmission system, in particular of a data transmission standard, for example, mobile radio standard, WLAN, or Bluetooth, which are used for data transmission in accordance with the data transmission standard.

[0067] The task is also solved by using the change in phase measurements compared to signal propagation measurements or their change between two objects, whereby the phase measurements and signal propagation time measurements are carried out on signals with overlapping frequency bandwidths to detect a relay attack.

[0068] The object is also achieved by an access system for granting and / or denying access, set up to carry out the method according to the invention and for granting and / or denying access based on the detection according to the method.

Claims

Claims 1. A method for detecting a relay attack, wherein radio signals with different frequencies (f1, f2, f3) are transmitted between a first and a second object (01, 02), and phase measurements are made on a first subset of the radio signals with different frequencies, and wherein these phase measurements are compared with a reference, and wherein, if a deviation exceeds a predetermined value or a deviation determined from measurements on the radio signals, a relay attack is assumed, access or release is denied, a requested action or operation is not carried out, and / or an alarm or blocking is triggered, and / or wherein, if the deviation falls below the predetermined value or the deviation determined from measurements, in particular signal flow measurements, the absence of a relay attack is assumed,access or release is granted and / or the requested action or operation is carried out and / or the alarm or blocking is not carried out, wherein a signal propagation time measurement is carried out on a second subset of the radio signals with different frequencies, which is in particular at least partially identical to the first subset, characterized in that a series of measurements is formed based on the phase measurements and / or a reference series of measurements is formed based on the signal propagation time measurements, in each case depending on the frequency of the frequencies of the radio signals used for the measurement,wherein the measurement series and / or signal propagation time measurement series contain either a monotonically increasing or monotonically decreasing sequence of frequencies and that based on this measurement series a filtered and / or smoothed measurement series and / or based on the reference measurement series a filtered and / or smoothed reference measurement series is determined and that the filtered and / or smoothed measurement series, in particular its change over the frequency, is compared with the at least one signal propagation time measurement or its change and / or the reference measurement series as a reference and / or wherein the measurement series or, whose change is compared with the filtered and / or averaged reference measurement series.

2. The method according to claim 1, wherein the numbers of elements of the first and second subsets differ by at least a factor of two, in particular at least four.

3. Method according to one of the preceding claims, wherein a filtered and / or averaged series of measurements and / or reference series is generated by means of a filter, in particular an FIR and / or IIR filter, and used for the comparison.

4. Method according to the preceding claim 3, wherein the filtering is carried out by means of an IIR filter, which is in particular passed through once forwards and once backwards.

5. Method according to one of the preceding claims, wherein the measurement series and / or reference measurement series is transformed, in particular an FFT is generated, and the result of the transformation is used for the comparison.

6. Method according to one of the preceding claims, wherein a plurality of individual comparisons are carried out, in each of which a single signal propagation time measurement is used as a reference.

7. Method according to the preceding claim 6, wherein each of the individual comparisons is carried out such that a gradient is determined from the averaged, filtered, smoothed and / or transposed series of measurements and is compared with the reference.

8. Method according to one of the preceding claims, wherein the filtering is carried out with an FFT, subsequent removal of higher frequency components and an IFFT.

9. Method according to one of the preceding claims, wherein a transformed measurement range is generated by means of FFT and this or its values are compared with the reference and / or whereby a transformed reference measurement series is generated by means of FFT and these or their values ​​are compared with the measurement series or values ​​derived therefrom.

10. Method according to one of the preceding claims, wherein the phase-based measurements and / or propagation time measurements are used for distance measurement and / or wherein the signal propagation times and / or signal propagation time changes are compared with the change in phase positions and / or change in the phase position changes. 1 1. Method according to one of the preceding claims, wherein propagation time measurements and / or phase measurements on radio signals and / or radio signals received with a power below a predetermined and / or, in particular, from or taking into account received radio signals, determined lower power limit are disregarded, in particular those radio signals are disregarded which are more than 50% below the average power of the received radio signals and / or wherein propagation time measurements and / or phase measurements on radio signals and / or radio signals received with a power above a predetermined and / or, in particular, from or taking into account received radio signals, determined upper power limit are disregarded, in particular those radio signals are disregarded which are more than 50% above the average power of the received radio signals.

12. Method according to one of the preceding claims, wherein the measurements with low reception power, in particular reception power below a predetermined value or proportion of the average or maximum reception power, are not taken into account 13. Method according to one of the preceding claims, wherein the width of the frequency interval is at least 0.1 MHz and / or at most 100 MHz and / or wherein the frequency spacing between two consecutive different frequencies is at least 0.1 MHz and / or a maximum of 10 MHz and / or the different frequencies represent at least five frequencies and / or a maximum of 200 frequencies and / or wherein the radio signals are emitted on the different frequencies one after the other and / or consecutively, in particular directly one after the other, and / or wherein the bandwidth of the radio signals never exceeds 50 MHz, in particular 25 MHz.

14. Access system for granting and / or denying access, comprising a first object and a second object, in particular an access restriction means and / or alarm means, and a controller configured to carry out a method according to one of claims 1 to 13 for detecting a relay attack.