METHOD AND DEVICE FOR PROTECTING A SIGNAL FOR TRANSMITTING A MEASUREMENT VALUE TO A SIGNAL PROCESSING UNIT

Abstract

A method for protecting a signal for transmitting a sequence of measurement values to a signal processing unit. The method includes a step of reading in at least one sequence of measurement values from a sensor and a step of processing the measurement values using a cyclical processing specification in order to determine a varied measurement value for each of at least two measurement values. Finally, the method comprises a step of sending the varied measurement values to the signal processing unit as measurement values from the sensor.

Claims

1-12. (canceled)

13. A method for protecting a signal for transmitting a sequence of measurement values to a signal processing unit, the method comprising the following steps: reading in at least one sequence of measurement values from a sensor; processing the measurement values using a cyclical processing specification to determine a varied measurement value for each measurement value of at least two of the measurement values; and sending the varied measurement values to the signal processing unit as measurement values from the sensor' wherein in the processing step, the processing specification is configured to alternately link measurement values to a first predefined link value and a second predefined link value that differs from the first link value to determine the varied measurement values.

14. The method as recited in claim 13, wherein in the processing step, the processing specification used is configured to bring about an algebraic and/or logical link between the measurement value and/or a digital representation of the measurement value and a predefined link value to determine the varied measurement values.

15. The method as recited in claim 14, wherein in the processing step, the processing specification used is configured to add a measurement value to the predefined link value to determine a corresponding varied measurement value.

16. The method as recited in claim 14, wherein in the processing step, the processing specification used which is configured to modify a lowest-value digit of the measurement value in order to obtain a corresponding varied measurement value.

17. The method as recited in claim 14, wherein the measurement values are processed as binary datum and the processing specification modifies a digit of a digit having a lowest-valued value.

18. The method as recited in claim 17, wherein in the processing step, the processing specification used is configured to use a value that forms a value complementary to the first predefined link value as the second predefined link value.

19. The method as recited in claim 13, wherein in the processing step, the processing specification used which is configured to apply the cyclical processing specification when measurement values of the sequence of measurement values within a predefined time interval are identical.

20. The method as recited in claim 13, wherein in the processing step, the processing specification used which is configured to perform the processing cyclically using the processing specification for measurement values that are read in at a predefined time interval from one another in order to obtain one corresponding varied measurement value in each case.

21. A method for detecting an error in a transmission of a signal containing measurement values to a signal processing unit, the method comprising the following steps: reading in a sequence of received values from the signal; analyzing whether the received values have a cyclical pattern as obtained by applying a read in processing specification to measurement values from a sensor; and ascertaining the error in the transmission of the signal containing received values when it is not detected in the analyzing step that the signal containing received values has the cyclical pattern.

22. A device configured to protect a signal for transmitting a sequence of measurement values to a signal processing unit, the device configured to: read in at least one sequence of measurement values from a sensor; process the measurement values using a cyclical processing specification to determine a varied measurement value for each measurement value of at least two of the measurement values; and send the varied measurement values to the signal processing unit as measurement values from the sensor; wherein in the processing, the processing specification is configured to alternately link measurement values to a first predefined link value and a second predefined link value that differs from the first link value to determine the varied measurement values.

23. A non-transitory machine-readable storage medium on which is stored a computer program for protecting a signal for transmitting a sequence of measurement values to a signal processing unit, the computer program, when executed by a computer, causing the computer to perform the following steps: reading in at least one sequence of measurement values from a sensor; processing the measurement values using a cyclical processing specification to determine a varied measurement value for each measurement value of at least two of the measurement values; and sending the varied measurement values to the signal processing unit as measurement values from the sensor' wherein in the processing step, the processing specification is configured to alternately link measurement values to a first predefined link value and a second predefined link value that differs from the first link value to determine the varied measurement values.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0029] FIG. 1 is a block diagram of a device for protecting a signal for transmitting a sequence of measurement values to a signal processing unit for use in a vehicle.

[0030] FIGS. 2A and 2B show two diagrams, each representing the content of the signal as output by the sending unit, for example.

[0031] FIG. 3 is a flow diagram of an exemplary embodiment of a method 300 for protecting a signal for transmitting a sequence of measurement values to a signal processing unit.

[0032] FIG. 4 is a flow diagram of an exemplary embodiment of a method 400 for detecting an error in the transmission of a signal containing measurement values to a signal processing unit.

DETAILED DESCRIPTION OF EXAMPLE EMBODIMENT

[0033] In the following description of advantageous exemplary embodiments of the present invention, elements shown in the various figures that have a similar action are given the same or similar reference numerals and these elements are not described again.

[0034] FIG. 1 is a block diagram of a device 100 for protecting a signal 105 for transmitting a sequence of measurement values 110 to a signal processing unit 115 for use in a vehicle 117. In this case, the sequence of measurement values 110 is provided by a sensor 120, for example, and represents a physical variable which the sensor 120 captures in predefined time steps and outputs as the measurement values 110. For example, the sensor 120 can be configured to capture an angular rate or acceleration at a point in a vehicle 117 and to output accordingly associated measurement values 110. The measurement values 110 are read in into the device 100 by a read-in interface 125 and are processed in a processing unit 130 using a cyclical processing specification 135 in order to determine a varied measurement value 110′ for each of at least two measurement values 110. These varied measurement values 110′ are then transmitted to the signal processing unit 115 in the signal 105 via a sending unit 136 (for example together with further measurement values 110 and/or in a particular data format). The signal processing unit 115 can, for example, prepare the measurement values for a driver assistance system or a trigger device for safety means in a vehicle 117, but this is not shown in greater detail in FIG. 1 for the sake of clarity.

[0035] To then obtain the varied measurement values 110′, as described above, the read in measurement values 110 or at least two of the read in measurement values 110 are processed in the processing unit 130 in accordance with the cyclical processing specification 135. This cyclical processing specification 135 is, for example, configured such that measurement values 110 that are at a predetermined time interval from one another are linked by a predefined link value 140. For example, a link value 140 of this kind may be a predefined value that is added to the measurement values 110 to obtain the varied measurement values 110′. In this way, the measurement values 110 in question are deliberately “distorted” in a conventional manner in a predetermined time interval and are output in the signal 105 as the varied measurement values 110′.

[0036] If it is now detected that the varied measurement values 110′ contained in the signal 105 have actually been varied in accordance with the processing specification 135, a device 150 for detecting an error in the transmission of a signal 105 containing measurement values 110 is then provided on the receiving side of the signal 105. In this device 150, the signal 105, which comprises a sequence of received values 160 representing the varied measurement values 110′, is first read in via a read-in interface 155. These received values 160 are then supplied to an analysis unit 165, which can in turn access the corresponding cyclical processing specification 135 and the link value 140 from a memory in order to check whether the received values 160 have been processed in accordance with the cyclical processing specification 135 and the link value 140. The same processing specification 135 and the same link value 140 are thus advantageously known both in the device 100 and the device 150. Furthermore, information regarding the analysis result in the analysis unit 165 is relayed to an ascertaining unit 170 in which an error is ascertained and is sent to the signal processing unit 115 via an error signal 175 when the analysis unit 162 does not detect that the signal 105 containing received values 160 has the cyclical pattern as obtained by applying the processing specification 135 to measurement values 110 from the sensor 120. In this way, the signal processing unit 115 can thus indicate that the received values 160 from the signal 105 are not valid and thus cannot be used as a basis for further functions.

[0037] Using the approach presented here, it is therefore possible to ascertain a loss of the function of the sensor 120 or of the validity of the measurement values 110. Here, these measurement values 110 may be in digital form, for example in a transmission format of a data frame having a header and an operand, which is also configured to transmit a plurality of data words, for example. In certain situations, an error can namely arise that means that a digital part 180 of the sensor 120 is no longer functioning correctly and is outputting static values as the measurement values 110, which are then interpreted in the signal processing unit 115 as valid measurement values and are used for actuating functions.

[0038] As already briefly mentioned above, the above-mentioned problem can be solved by increasing the resolution of the measurement values 110, for example from 16 bits to 32 bits, such that, when a digital part 180 of the sensor 120 is operating correctly, a noise, which would have to be detected in the measurement values 115 due to the increased resolution, can be expected when the sensor captures the physical variable. Although this procedure is possible in principle, it requires a dramatically increased capacity for the transmission of the measurement values at the accordingly increased resolution in order to be able to detect certain errors on its own. At this point approach proposed here by linking the measurement values in a cyclical manner, i.e., at certain intervals, to certain predefined link parameters, such that, on the basis of the presence of these modified measurement values, it can then also be detected whether the digital part 180 of the sensor 120 is functioning correctly. For this purpose, the device 100 can for example also be understood to be an extension of the digital part 180 of the sensor 120 such that the signal 105 can be transmitted for example via standard data transmission paths in a vehicle 117, such as the CAN bus. Upstream of the signal processing unit 115, the device 150 can then in turn ascertain whether the received values 160 contained in the signal 105 correspond to correct measurement values 110.

[0039] FIGS. 2A and 2B shows two diagrams representing the content of a signal 105 as output by the sending unit 136, for example. Here, the signal 105 is output as a sequence of data words in which a header 200 first introduces a data word 205 of this kind, and a data field 210 then adjoins the header 200. The varied measurement values 110′ are then contained in the data fields 210, for example, and can be transmitted to the device 150 or the signal processing unit 115.

[0040] FIG. 2A shows a first exemplary embodiment of how a sequence of measurement values 110 is processed to form varied measurement values 110′ in the processing unit 130 by the corresponding processing specification 135 and prepared for transmission in the signal 105. By way of example, in this case it is assumed that the digital part 180 of the sensor 120 from FIG. 1 outputs a 4-bit-wide static measurement value 110 in the form 1111, which is caused by a “stuck-at-high” error, however, and thus needs to be marked as defective and cannot be utilized in the signal processing unit 115. Since, for example, it then cannot be detected in the signal processing unit 115 that this measurement value 110 or a corresponding sequence of measurement values 110 in the above-mentioned form is defective, then, in accordance with the view from FIG. 1, the device 100 processes at least two of the measurement values 110 to form varied measurement values 110′ by way of the processing unit 130 in accordance with the processing specification 135 and inserts them into a corresponding data field 210 of a data word 205. For example, this processing specification 135 can provide that the measurement value 110 is alternately linked to a first link parameter 215 as a (first) link value 140, thereby resulting, for example, in a varied measurement value 110′ in the form 1110, which is therefore entered in the data field 210 of the first data word 205. This means that the lowest-value digit of the measurement value 110 present here as a binary datum is modified from a value 1 to the value 0. For example, the first link value thus corresponds to a subtraction of the lowest value, which can then be represented by the 4-bit-wide measurement value 110. Furthermore, a subsequent measurement value 110 can be linked to a second link parameter 220 as a (second) link value 140 by way of the processing unit 130 in accordance with the processing specification 135, such that the content of the measurement value 110 is modified and this measurement value 110 is entered in the data field 210 of the second data word 205 as a “varied” measurement value 110′ in the form 1111. For another subsequent measurement value 110, a procedure for a varied measurement value 110′ can then be carried out again analogously to the procedure for the first data word 205 in the processing unit 130 in accordance with the processing specification 135, such that, in this case, a varied measurement value 110′ in the form 1110 is obtained in turn, which is then entered in the data field 210 of the third data word 205. It is then apparent that, despite the adjoining static measurement values 110 in the form 1111, data words 205 of which the content of the data fields varies cyclically are transmitted in the signal 105 such that, due to the knowledge of the corresponding processing specification 135 and the link parameter 140 used in each case, it is possible in a very simple manner to identify, in the device 150 in the analysis unit 165, if an error of the kind mentioned above has arisen and to prevent the transmitted measurement values 110 from being used by the signal processing unit 115 itself outputting a corresponding error signal 175. By way of the knowledge of the link value 140 used and the time interval 225 between two applications of the link value 140 in question, the sequence of measurement values 110 output by the digital part 180 of the sensor 120 can be analyzed and evaluated in a very simple manner.

[0041] In the exemplary embodiment shown in FIG. 2A, a second link parameter 220 is then used as the link value 140, which leaves the measurement value 110 substantially unmodified, such that, when considered in a purely technical sense, only the first link parameter 215 can be applied as the link value 140 to the measurement values 110 which are entered in the data field 210 of the first and third data words 205. The second measurement value 110 is not modified at all in this case. As a result, the processing in the device 100, specifically in the processing unit 130, can be simplified such that, for example, the numerical load for the specific conversion of the measurement values 110 into the varied measurement values 110′ can be reduced. However, the result is that the measurement values 110 are actually “distorted” and thus need to be, or should be, corrected back again in the device 150 or a sub-unit of this device 150 in order to avoid any further errors when these measurement values 110 are used. This is necessary in particular when such processing of the measurement values 110 into the processed or varied measurement values 110′ is carried out even if the read-in interface 125 receives not only statically adjoining measurement values 110 but also actual dynamically varying measurement values 110 which are converted in the processing unit 130 in accordance with the processing specification 135 to protect the transmission. Even if an evaluation is only carried out in a component of the device 100 such that the measurement values 110 read in via the read-in interface 125 remain the same and corresponding processing is carried out in accordance with the processing specification 135, this is not known on the receiving side of the signal processing unit 115, and therefore corresponding compensation of the conversion of the measurement values 110 into the varied measurement values 110′, carried out in the processing unit 130, should always be performed for accordingly protecting the usability of the measurement values.

[0042] FIG. 2B is a view of a composition of the signal 105 using a further exemplary embodiment for the processing specification 135. In this case, different measurement values 110 are adopted: in this case, for example, the sequence of digital measurement values in the form 1101, 1010, and 1110. If, for example, a link value 140 which corresponds to an addition of a lowest-value digit is then selected as the first link parameter 215 and a link value 140 which corresponds to a subtraction of a lowest-value digit is selected as the second link parameter 220, a value of 1110 is obtained as the first varied measurement value 110′, a value of 1001 is obtained as the second varied measurement value 110′, and a value of 1111 is obtained as the third varied measurement value 110′, which are then each entered in the relevant data field 110 of the data word 105 in question.

[0043] In this way, by alternately adding and subtracting a low value to and from the corresponding measurement value 110, the “distortion” caused by the processing specification 135 can be compensated for by averaging different link values 140, in particular when the measurement values have been observed for a longer period of time, for example for an integration of these measurement values, such that an evaluation of this kind in fact no longer requires correction of the varied measurement values 110′ in the device 150 or the signal processing unit 115, meaning that numerical or circuitry-related complexity can be further reduced.

[0044] At the same time, it can be noted that, in principle, not only does a low value of the corresponding measurement values 110 need to be modified by the processing specification 135 in order to detect an error in the sensor 120 or digital part 180, but other digits in the measurement value 110 can also be actively modified as long as it is known which digit is used by the processing specification 135 or which link value 140 is used. Information regarding the cycle of the processing specification or the time interval 225 between the use of the different link parameters 215 and 225 as link values 140 is also relevant.

[0045] It is of course also possible for measurement values 110 having any resolution to be able to used for the approach presented here, meaning that the approach presented here is not limited only to 4-bit-wide measurement values. It is also apparent that, due to the approach presented here, it is thus no longer necessary to significantly increase the resolution of the measurement values in order to detect, by way of a sensor 120 that captures a noise of the physical value, that the digital part 180 of the sensor 120 is still functioning correctly. Instead, by actively modifying individual digits or components of the measurement values 110 in the knowledge of the exact modification specification or processing specification 130, it is possible to identify a potential malfunction of the sensor 120 or digital part 180 of the sensor 120 very effectively and technically very efficiently.

[0046] In summary, in view of the approach presented here, the question is when an above-described procedure of increasing the resolution to 32 bits is required. That procedure is only useful if the noise is supposed to be used to check whether the digital part 180 of the sensor 120 is still running. Since, in statistics, a plurality of consecutive equal values cannot be excluded, it is for example proposed in other approaches that the noise width be increased in order to significantly decrease the probability of consecutive equal values. However, in VMPS using 32 bits, more than twice the SPI bus load would be generated on the SPI, and this alone poses difficulties for the microprocessor. In addition, the radiation would increase in electromagnetic compatibility. This situation becomes even worse on a 500 kHz CAN bus. If the entire width of 8 bytes is required, a value of 200 ps for a data word 205 is obtained per signal. For 6 signals, this is already 1200 ps, or 2400 ps with redundancy. If an interval of 5 ms is selected, just the transmission of the sensor values generates a bus load of 50%.

[0047] According to the approach presented here, the data width can be reversed to 16 bits again and, for example, one bit of “artificial noise” can be added: this means, for example, that the old value is stored, and a new value is calculated, with an LSB then being subtracted or added in a statistically distributed manner. Therefore, a check as to whether the digital is still “alive” can only be performed by way of a comparison.

[0048] FIG. 3 is a flow diagram of an exemplary embodiment of a method 300 for protecting a signal for transmitting a sequence of measurement values to a signal processing unit. The method 300 comprises a step 310 of reading in at least one sequence of measurement values from a sensor and a step 320 of processing the measurement values using a cyclical processing specification in order to determine a varied measurement value for each of at least two measurement values. Finally, the method 300 comprises a step 330 of sending the varied measurement values to the signal processing unit as measurement values from the sensor.

[0049] FIG. 4 is a flow diagram of an exemplary embodiment of a method 400 for detecting an error in the transmission of a signal containing measurement values to a signal processing unit. The method 400 comprises a step 410 of reading in a sequence of received values from the transmission signal and a step 420 of analyzing whether the received values have a cyclical pattern as obtained by applying the processing specification to measurement values from a sensor. Finally, the method 400 comprises a step 430 of ascertaining the error in the transmission of the signal containing received values if it is not detected in the analyzing step that the signal containing received values has the cyclical pattern.

[0050] If an exemplary embodiment contains an “and/or” conjunction between a first feature and a second feature, this should be interpreted such that, according to one specific embodiment, the exemplary embodiment includes both the first feature and the second feature and, according to a further specific embodiment, the exemplary embodiment includes either only the first feature or only the second feature.