CN111835611A

CN111835611A - Vehicle CAN bus data analysis method and device

Info

Publication number: CN111835611A
Application number: CN202010651002.2A
Authority: CN
Inventors: 笋大伟; 肖觊威; 李明春; 赵永胜
Original assignee: Beijing Softsec Technology Co ltd
Current assignee: Beijing Zhilian Anhang Technology Co ltd
Priority date: 2020-07-08
Filing date: 2020-07-08
Publication date: 2020-10-27
Anticipated expiration: 2040-07-08
Also published as: CN111835611B

Abstract

The invention discloses a vehicle CAN bus data analysis method and a device, and for a multi-value data field, the method comprises the following steps: preprocessing vehicle CAN bus data, and determining each multi-value data field in the vehicle CAN bus data; the method further comprises the following steps: monitoring each multi-value data field in the CAN bus data of the vehicle; selecting operation corresponding to the vehicle information with discrete states to change the operation state, and if the frequency of the operation state change is consistent with the monitored numerical value change frequency of a multi-valued data field, determining that the operation corresponds to the multi-valued data field; and changing the operation state of the operation again, and determining the field value corresponding to each operation state according to the value change at the position of the multi-valued data field. The method and the device CAN analyze the specific functional meaning of the corresponding field of the vehicle CAN bus data.

Description

Vehicle CAN bus data analysis method and device

Technical Field

The invention relates to the technical field of vehicle networking, in particular to a vehicle CAN bus data analysis method and device.

Background

A Controller Area Network (CAN) bus protocol is the most widely used protocol for vehicle networking and vehicle-mounted networks at present. The operation of a user for operating the vehicle, such as ignition starting, opening and closing of a vehicle door and the like, is essentially realized by sending a corresponding control command to a CAN bus by a corresponding Electronic Control Unit (ECU). The details of the CAN bus protocol are mastered, so that the method is not only beneficial to a novice to learn the vehicle control logic, but also beneficial to a third party to carry out safety detection on the vehicle.

However, the CAN bus protocol standard is public, but the details of the CAN bus protocol, different manufacturers have respective implementation modes and are kept secret from the outside. Therefore, the CAN bus protocol reverse technology is developed.

At present, two main CAN bus protocol reverse schemes are adopted, firstly, data flow changes in a CAN bus are observed through pulling and inserting an ECU, and the lost flow is associated with the pulled ECU; secondly, a gateway is arranged for each ECU, data entering and exiting the ECU are monitored, and the data are directly corresponding to the ECU.

In the two CAN bus protocol reverse schemes, the former scheme may be misjudged because different ECUs may have data association, one ECU is pulled out, and the other ECU possibly associated with the ECU also stops sending data; the latter is relatively accurate, but a gateway needs to be additionally arranged on each ECU to monitor the flow, so that the cost is high and the efficiency is low. Meanwhile, the two CAN bus protocol reverse schemes have the following common disadvantages:

(1) complicated operation

The former needs to plug and unplug the ECU, and the latter needs to add a gateway to the ECU, which means that the two protocols in the reverse direction need to disassemble the whole vehicle to expose the bus. In addition, the former reverses each ECU, and the ECUs are required to be pulled out, data of a period of time are counted, then the ECUs are inserted back, data of a period of time are counted, and finally the difference of the two pieces of data is compared. And the latter reverses each ECU, all needs to adjust the installing position of the gateway, counts data for a period of time, and then analyzes according to the monitoring data of the gateway. Both operations are very troublesome.

(2) Coarse grain size in the reverse direction

The protocol inversion schemes of the two protocols correspond data with the ECU, and do not correspond specific functions. According to the final output results of the two, only some data can be known and correspond to a certain ECU, but the specific functional meaning of the data is not known.

Disclosure of Invention

In view of this, the invention aims to: the specific function meaning of the corresponding field of the vehicle CAN bus data CAN be analyzed.

In order to achieve the purpose, the technical scheme of the invention is realized as follows:

the embodiment of the invention provides a vehicle CAN bus data analysis method, which comprises the following steps: preprocessing vehicle CAN bus data, and determining each multi-value data field in the vehicle CAN bus data; the method further comprises the following steps:

monitoring each multi-value data field in the CAN bus data of the vehicle;

selecting operation corresponding to the vehicle information with discrete states to change the operation state, and if the frequency of the operation state change is consistent with the monitored numerical value change frequency of a multi-valued data field, determining that the operation corresponds to the multi-valued data field;

and changing the operation state of the operation again, and determining the field value corresponding to each operation state according to the value change at the position of the multi-valued data field.

The embodiment of the invention also provides a vehicle CAN bus data analysis method, which comprises the following steps: preprocessing vehicle CAN bus data, and determining each state value field in the vehicle CAN bus data; the method further comprises the following steps:

monitoring each state value field in the CAN bus data of the vehicle;

collecting field values of each state value in the CAN bus data of the vehicle at preset time intervals, and simultaneously selecting vehicle information with continuous states to collect real physical values at the preset time intervals;

and fitting each piece of collected state value field information with the collected vehicle information, and determining that the vehicle information corresponds to the state value field with the highest fitting degree.

The embodiment of the invention also provides a vehicle CAN bus data analysis device, which comprises:

the first preprocessing module is used for preprocessing the vehicle CAN bus data and determining each multi-value data field in the vehicle CAN bus data;

the first monitoring module monitors each multi-value data field in the CAN bus data of the vehicle;

the first processing module selects the operation corresponding to the vehicle information with discrete states to change the operation state, and if the frequency of the change of the operation state is consistent with the monitored numerical value change frequency of a multi-valued data field, the operation is determined to correspond to the multi-valued data field;

the second preprocessing module is used for preprocessing the vehicle CAN bus data and determining each state value field in the vehicle CAN bus data;

the second monitoring module monitors each state value field in the CAN bus data of the vehicle;

the second processing module collects each state value field value in the vehicle CAN bus data at preset time intervals, and simultaneously

Selecting vehicle information with continuous states to acquire real physical values at the preset time interval;

As can be seen from the above technical solution, for a multi-valued data field, the method of the embodiment of the present invention includes: preprocessing vehicle CAN bus data, and determining each multi-value data field in the vehicle CAN bus data; the method further comprises the following steps: monitoring each multi-value data field in the CAN bus data of the vehicle; selecting operation corresponding to the vehicle information with discrete states to change the operation state, and if the frequency of the operation state change is consistent with the monitored numerical value change frequency of a multi-valued data field, determining that the operation corresponds to the multi-valued data field; and changing the operation state of the operation again, and determining the field value corresponding to each operation state according to the value change at the position of the multi-valued data field. Therefore, the purpose of analyzing the specific functional meaning of the corresponding field of the vehicle CAN bus data is achieved.

Drawings

Fig. 1 is a schematic flow chart of a method for analyzing vehicle CAN bus data according to an embodiment of the present invention.

Fig. 2 is a schematic structural diagram of a vehicle CAN bus data analysis device according to a third embodiment of the present invention.

Fig. 3 is a schematic flow chart of a vehicle CAN bus data analysis method according to a fourth embodiment of the present invention.

Fig. 4 is a schematic structural diagram of a vehicle CAN bus data analysis device according to a sixth embodiment of the present invention.

Detailed Description

In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is further described in detail below with reference to the accompanying drawings and examples.

Due to the specificity of the CAN bus protocol, the realization modes of the CAN bus protocols of various automobile factories are different and are kept secret from the outside. One CAN bus data may be a control command, may reflect vehicle status information, or may simply feed back heartbeat data periodically (or may be a multi-situation mashup).

In a CAN bus data, the data portion has a maximum of 8 bytes (64 bits). A protocol designer may use some of the bytes or even only some of the bits to design a function.

The CAN bus protocol reverse scheme CAN analyze the specific functional meaning of the corresponding field of the vehicle CAN bus data. For the multi-valued data field, the corresponding operation can be analyzed, and the field value corresponding to each operation state can be analyzed. For the state value field of the full-value data field, the corresponding vehicle information can be analyzed.

According to the CAN bus data characteristics, the invention divides the CAN bus data into three types: single value data field, multi-value data field, full value data field. The full-value data field can be further subdivided into a state value field, a counter value field and a check value field.

The single-value data field is data which is the same value from beginning to end, and the data is often used as filling without the need of an inference function.

The multi-valued data field is data that can only change in some values, and such data is often used as control, status display, etc. to represent vehicle information with some discrete states. Such as opening and closing of a vehicle door; such as the state of the wiper: static, slow brushing, fast brushing, and the like.

The full-value data field is data capable of traversing a range of values, and such data often represents vehicle information with continuous states, such as vehicle speed, fuel quantity and the like; or a counter, such as used for heartbeat data; or may simply be used as a check value.

In the CAN bus data field category, single-valued data fields and multi-valued data fields are well judged. The difficulty is mainly to distinguish three types of state value fields, counter value fields and check value fields in the full-value data fields.

In the full-value data field, the characteristics of the counter value field and the check value field are relatively obvious, so the state value field is distinguished by using an elimination method, namely, if the data does not belong to the counter value field or the check value field, the data is regarded as the state value field. The features of the counter value field and the check value field are described in detail below.

Characteristics of the counter value field: the bit reversal rate of the least significant bit (the probability of the bit changing from 0 to 1, or from 1 to 0) is 1 and the bit reversal rate of the less significant bit is twice as high as the more significant bit as it is incremented or decremented over time. For example, two bytes are used to represent the counter value, i.e. a total of 16 bits are used. This value will accumulate from 0x0000 to 0xFFFF and then go to the next cycle. Assuming we have taken one cycle of data, 65536 total, the 16 th bit value (least significant bit) is totally inverted 65535 times, and the bit inversion rate is 1; the 15 th bit value is inverted for 32767 times totally, and the bit inversion rate is 0.5; the inversion rate of the 14 th bit is 0.25, and so on. Of course, we do not necessarily have to be able to obtain data for a whole cycle; however, regardless, as long as the data has the characteristics of the counter, the boundaries of the counter can be segmented. And if the inversion rate of the high bit is half of the inversion rate of the low bit, the high bit is still in the boundary of the counter data, and the left boundary can be continuously inferred to the left. The iteration is performed, inferring to the left bit by bit until the inversion rate of the high bit is no longer half of the low bit, which now serves as the left boundary (most significant bit) of the counter field.

Characteristics of the check value field: the check value field has a distinct random number characteristic compared to the data reflecting the vehicle state information, and also the counter field. Theoretically, each bit of the check value field has a bit reversal rate of 0.5. In the case of sufficient data, the bit reversal rate distribution of all bits of the check value field should exhibit a normal distribution with 0.5 as expected.

Example one

Based on the above analysis, a schematic flow diagram of a method for analyzing vehicle CAN bus data according to an embodiment of the present invention is shown in fig. 1, and the method includes:

step 11, preprocessing vehicle CAN bus data, and determining each multi-value data field in the vehicle CAN bus data;

the step is used for preprocessing the vehicle CAN bus data, and CAN be used for processing the vehicle CAN bus data in an off-line mode or in an on-line mode. Performing off-line pretreatment, namely collecting sample CAN bus data to determine; and (3) online preprocessing, namely directly preprocessing the current vehicle CAN bus data.

In the step, each multi-value data field in the vehicle CAN bus data is determined, and the position of the multi-value data field under each frame ID CAN be determined. Similarly, for any frame ID, when monitoring in step 12, it is only necessary to monitor the position of the multi-valued data field corresponding to the frame ID.

Step 12, monitoring each multi-value data field in the CAN bus data of the vehicle;

step 13, selecting operation corresponding to the vehicle information with discrete states to change the operation state, and if the frequency of the change of the operation state is consistent with the numerical value change frequency of a monitored multi-valued data field, determining that the operation corresponds to the multi-valued data field;

this step can determine that the selected operation corresponds to a certain multi-valued data field. For example, the wiper operation corresponds to the 5 th byte having a frame ID of 666, wherein the 5 th byte is a multivalued data field.

And step 14, changing the operation state of the operation again, and determining the field value corresponding to each operation state according to the value change at the position of the multi-valued data field.

For example, wiper operation includes 4 states, rest, normal, fast, slow. This step can determine that the value of the hexadecimal field corresponding to the static state is 00, the value of the hexadecimal field corresponding to the normal state is 20, the value of the hexadecimal field corresponding to the fast-brushing state is 60, and the value of the hexadecimal field corresponding to the slow-brushing state is 80.

In an alternative embodiment, if the frequency of the operation state change is consistent with the monitored multiple-valued data field value change frequency, the operation is changed according to a preset frequency; and if the frequency of the operation state change is consistent with the monitored numerical value change frequency of a multi-valued data field, determining that the operation corresponds to the multi-valued data field.

In an optional embodiment, determining each multi-value data field in the vehicle CAN bus data specifically includes:

dividing CAN bus data with the same frame ID in the vehicle CAN bus data into the same data set;

traversing CAN bus data of the same data set by taking bytes as units, and counting the number of different values of each byte to determine a multi-value data field of each byte.

Wherein, the counting the number of different values of each byte to determine the multi-valued data field to which each byte belongs specifically includes: and when the number of different numerical values of the bytes is not more than the preset number, determining the bytes as a multi-value data field.

Example two

The two-vehicle CAN bus data analysis method provided by the embodiment of the invention comprises the following steps:

step 21, dividing CAN bus data with the same frame ID in the vehicle CAN bus data into the same data set;

in an alternative embodiment, a vehicle has multiple control elements, and different frame IDs represent data characteristics of different control elements. It is assumed that there are 10000 pieces of vehicle CAN bus data, and 20 pieces of the CAN bus data are exemplified as shown in table 1.

TABLE 1

The data in table 1 above is represented by 16. In order to analyze the vehicle CAN bus data, the vehicle CAN bus data are classified according to the frame ID, and the data with the same frame ID are divided into the same data set so as to be convenient for subsequent analysis. For example, in 10000 pieces of data in table 1, there are 1000 pieces of data having a frame ID of 666, and therefore, 1000 pieces of CAN bus data having a frame ID of 666 indicating data characteristics of the wiper are divided into the same data set. As shown in table 2;

TABLE 2

Step 22, traversing CAN bus data of the same data set by taking bytes as units, and counting the number of different values of each byte to determine a multi-value data field of each byte;

in an alternative embodiment, the data portion has a maximum of 8 bytes (64 bits) in a CAN bus data. In the step, the CAN bus data of the same data set is cut by taking bytes as units, so that a multi-value data field to which each byte belongs CAN be determined.

Traversing j CAN bus data in the same data set, counting the number of different values of each byte, wherein each CAN bus data comprises n bytes,

when the array arr [ n ] is 1, the nth byte is the same numerical value from beginning to end, and the nth byte is divided into single-value data fields;

when the array arr [ n ] is less than or equal to 25, the number of different values of the nth byte is less than 10% of the byte range, and the array is divided into multi-value data fields;

when the array arr [ n ] >25, it is divided into full-value data fields.

Wherein, one byte is 8 bits, the range represented by 8 bits is 0-255, and total 256 values. Because of the multi-valued data field, that is, data that can only change in some values, such data is often used as control or vehicle information indicating some state dispersion. Such as opening and closing of a vehicle door; such as the state of the wiper: static, slow brushing, fast brushing, and the like. Therefore, in the present embodiment, a byte whose number of change of the byte number is less than 10% of 256 values is empirically defined as a multi-value data field. Specifically, the predetermined number of limits may be adjusted according to the actual application.

For example, each piece of CAN bus data contains 8 bytes, the number of different values of the first byte is 1, and the first byte belongs to a single-value data field; the number of different numerical values of the second byte is 1, and the second byte belongs to a single-value data field; the number of different values existing in the 5 th byte is 20, and belongs to the multi-value data field. The number of different numerical values of the third byte and the fourth byte is respectively 100 and 120, and the third byte and the fourth byte belong to a full-value data field; the number of different values of the sixth byte, the seventh byte and the eight byte is 120, 130 and 140 respectively, and then the sixth byte, the seventh byte and the eight byte belong to the full-value data field.

Step 23, monitoring each multi-value data field in the CAN bus data of the vehicle;

for example, a multi-valued data field of the 5 th byte with a listening frame ID 666;

monitoring a multi-value data field of the 4 th byte with 777 frame ID;

and a multi-valued data field of 3 rd byte with the monitoring frame ID being 888.

And 24, selecting the operation corresponding to the vehicle information with discrete states to change the operation state, and if the frequency of the change of the operation state is consistent with the monitored numerical change frequency of a multi-valued data field, determining that the operation corresponds to the multi-valued data field.

In an alternative embodiment, selecting the wiper operation changes the wiper from the stationary state to the normal state, and if the frequency of the state change coincides with the monitored frequency of the change of the value of the 5 th byte multivalued data field of the frame ID 666, it is determined that the wiper operation corresponds to the 5 th byte multivalued data field of the frame ID 666.

If the frequency of the state change coincides with the monitored frequency of the change of the value of the multivalued data field of the 5 th byte with the frame ID 666 and the frequency of the change of the value of the multivalued data field of the 4 th byte with the frame ID 777, and at this time, it cannot be determined which multivalued data field the wiper operation corresponds to, the frequency of the change of the wiper operation state is adjusted to select the change from the stationary state to the normal state every 1s, the change from the stationary state to the normal state every 2s, and the change from the stationary state to the normal state every 3s, respectively. At this time, if the frequency of changing from standstill to the normal state every 2s coincides with the monitored multivalued data field value change frequency of the 5 th byte of which the frame ID is 666, it is determined that the wiper operation corresponds to the multivalued data field of the 5 th byte of which the frame ID is 666.

And 25, changing the operation state again, and determining the field value corresponding to each operation state according to the value change at the position of the multi-valued data field.

The operating state of the wiper is changed, and if the value at the position of the 5 th byte multivalued data field of the frame ID 666 is changed from 00 to 20 from the stationary state to the normal state, it is determined that the value of the hexadecimal field corresponding to the stationary state is 00 and the value of the hexadecimal field corresponding to the normal state is 20. For another example, if the value at the position of the 5 th byte multivalued data field of the frame ID 666 changes from 00 to 60 from the still state to the fast-brushing state, it is determined that the hexadecimal field value corresponding to the fast-brushing state is 60.

EXAMPLE III

Based on the same inventive concept, a structural schematic diagram of a vehicle CAN bus data analysis device provided by the third embodiment of the present invention is shown in fig. 2, and the device includes:

the first preprocessing module 201 is used for preprocessing the vehicle CAN bus data and determining each multi-value data field in the vehicle CAN bus data;

the first monitoring module 202 monitors each multi-value data field in the vehicle CAN bus data;

the first processing module 203 selects the operation corresponding to the vehicle information with discrete states to change the operation state, and if the frequency of the change of the operation state is consistent with the monitored numerical value change frequency of a multi-valued data field, the operation is determined to correspond to the multi-valued data field;

The first processing module 203 is specifically configured to, if the frequency of the operation state change is consistent with the monitored frequency of the multiple-valued data field value change, change the operation state of the operation according to a predetermined frequency; and if the frequency of the operation state change is consistent with the monitored numerical value change frequency of a multi-valued data field, determining that the operation corresponds to the multi-valued data field.

The first preprocessing module 201 determines each multi-value data field in the vehicle CAN bus data, and is specifically configured to:

The first preprocessing module 201 counts the number of different values existing in each byte to determine the multi-valued data field to which each byte belongs, and is specifically configured to:

and when the number of different numerical values of the bytes is not more than the preset number, determining the bytes as a multi-value data field.

Example four

Based on the above analysis, a flow diagram of a vehicle CAN bus data analysis method provided by the fourth embodiment of the present invention is shown in fig. 3, and the method includes:

step 31, preprocessing vehicle CAN bus data, and determining each state value field in the vehicle CAN bus data;

In the step, each state value field in the vehicle CAN bus data is determined, and the position of the state value field under each frame ID CAN be determined. Similarly, for any frame ID, when monitoring in step 32, only the position of the status value field corresponding to the frame ID needs to be monitored.

Step 32, monitoring each state value field in the CAN bus data of the vehicle;

step 33, collecting each state value field value in the vehicle CAN bus data at a preset time interval, and simultaneously selecting vehicle information with continuous states to collect real physical values at the preset time interval;

the acquisition time interval of the monitoring data in the step is the same as the acquisition time interval of the real physical value, and the starting time point and the ending time point are also the same.

And step 34, fitting each piece of collected state value field information with the collected vehicle information, and determining that the vehicle information corresponds to the state value field with the highest fitting degree.

In an alternative embodiment, prior to the fitting, the method further comprises: and carrying out normalization processing on each collected state value field information and the collected vehicle information. The normalization is performed to mask the difference between the actual physical value and the vehicle CAN bus data, because the actual physical value may need to be converted into the vehicle CAN bus data through formula operation, and the conversion may interfere with the fitting. Through normalization processing, the numerical values of the physical value and the field value are mapped into the range of 0-1, and the physical significance corresponding to the full-value data field can be deduced according to the similarity degree of the data change trend.

In an optional embodiment, the fitting each piece of collected state value field information with the collected vehicle information to determine that the vehicle information corresponds to the state value field with the highest degree of fitting specifically includes: and calculating the sum of squares of errors of each state value field and the vehicle information, and determining that the vehicle information corresponds to the state value field with the minimum deviation.

In an optional embodiment, determining each status value field in the vehicle CAN bus data specifically includes:

traversing CAN bus data of the same data set by taking bytes as units, and counting the number of different values of each byte to determine a full-value data field of each byte;

traversing CAN bus data of the same data set by taking a bit as a unit, and acquiring the bit turnover rate of each bit corresponding to each full-value data field by taking a byte as a unit; combining the full-value data fields of adjacent bytes, and calculating the bit turnover rate weight of each bit corresponding to each combined full-value data field; performing subfield cutting on each merged full-value data field according to the weight to determine the boundary of each full-value data subfield; classifying all full-value data subfields according to the bit turnover rate characteristics of each bit corresponding to all the full-value data subfields after cutting;

performing subfield cutting on each merged full-value data field according to the weight to determine the boundary of each full-value data subfield, specifically comprising:

determining that the previous bit and the next bit belong to different full-value data subfields if the bit flipping rate weight of the previous bit is smaller than the bit flipping rate weight of the next bit;

if the bit flipping rate weight of the previous bit is not less than the bit flipping rate weight of the next bit, determining that the previous bit and the next bit belong to the same full-value data subfield;

the method includes the steps of classifying all full-value data subfields according to bit-flipping rate characteristics of each bit corresponding to all full-value data subfields after being cut, and specifically includes:

determining the full-value data subfield as a counter value field according to that the bit flip rate of the low significant bit is twice that of the high significant bit;

or, according to the normal distribution with 0.5 as the expectation of the bit turnover rate distribution of each bit of the full value data subfield, determining the full value data subfield as the check value field;

or, if the full value data subfield does not belong to either the counter value field or the check value field, determining that the full value data subfield is the status value field.

EXAMPLE five

The five-vehicle CAN bus data analysis method provided by the embodiment of the invention comprises the following steps:

step 51, dividing CAN bus data with the same frame ID in the vehicle CAN bus data into the same data set;

step 52, traversing CAN bus data of the same data set by taking bytes as units, and counting the number of different values of each byte to determine a full-value data field of each byte;

in an alternative embodiment, the data portion has a maximum of 8 bytes (64 bits) in a CAN bus data. In the step, the CAN bus data of the same data set is cut by taking bytes as units, so that the full-value data field of each byte CAN be determined. It should be noted that it is possible that the full-valued data field may have its function divided among multiple data fields, so that further merging and cutting of the full-valued data field is required.

In the same data set, traversing j pieces of CAN bus data, counting the number of different values of each byte, wherein each piece of CAN bus data comprises n bytes, and when the array arr [ n ] >25, dividing the array into full-value data fields.

Step 53, traversing the CAN bus data of the same data set by taking a bit as a unit, and acquiring the bit turnover rate of each bit corresponding to each full-value data field by taking a byte as a unit; combining the full-value data fields of adjacent bytes, and calculating the bit turnover rate weight of each bit corresponding to each combined full-value data field; performing subfield cutting on each merged full-value data field according to the weight to determine the boundary of each full-value data subfield; classifying all full-value data subfields according to the bit turnover rate characteristics of each bit corresponding to all the full-value data subfields after cutting;

1) traversing CAN bus data of the same data set by taking a bit as a unit, and acquiring the bit turnover rate of each bit corresponding to each full-value data field by taking a byte as a unit.

In this embodiment, the same data set is traversed for j pieces of CAN bus data, the full-value data field determined in step 22 is used, and the bit inversion rate of each bit of the full-value data field is calculated. Here, the full-value data field is in bytes, but it is possible that the full-value data field has a function that is divided into a plurality of bytes, so that the full-value data field needs to be further merged and cut to accurately define the number of bytes of the full-value data field.

Traversing j pieces of CAN bus data for any ith bit in a full-value data field with byte as a unit, if the value of the ith bit of the jth CAN bus data is different from that of the jth-1 CAN bus data, turning the bit by +1, and after traversing the jth CAN bus data is finished, dividing the obtained bit reversal times by the total number j to obtain the bit reversal rate of the ith bit.

For example, the third byte corresponds to bits 17 to 24, and each bit corresponds to a bit flip rate of 0, 0.002, 0.013, 0.086, 0.155, 0.296, and 0.612, respectively. In the example of the bit flipping rate 0.002 of the 19 th bit, after traversing 1000 pieces of CAN bus data, the number of bit flipping times is 2, and the bit flipping rate is 2/1000-0.002.

The fourth byte corresponds to 25 th to 32 th bits, and each bit corresponds to a bit flip rate of 0, 0.004, 0.025, 0.082, 0.178, 0.332 and 0.69.

The sixth byte corresponds to bits 41 to 48, and each bit corresponds to a bit flip rate of 0.007, 0.015, 0.031, 0.062, 0.125,0.25,0.5,1, respectively.

The seventh byte corresponds to bits 49 to 56, and each bit corresponds to a bit flip rate of 0.494, 0.513, 0.493, 0.513, 0.501, 0.495, 0.49, 0.484, respectively.

The eighth byte corresponds to bits 57 to 64, and each bit corresponds to a bit flip rate of 0.498, 0.523, 0.53, 0.505, 0.503, 0.494, 0.503, 0.504, respectively.

2) And combining the full-value data fields of the adjacent bytes to obtain a bit turnover rate array with the length of m.

For example, adjacent three or four bytes are combined to obtain bit flip rate arrays brr [0, 0.002, 0.013, 0.086, 0.155, 0.296, 0.612, 0, 0.004, 0.025, 0.082, 0.178, 0.332, 0.69], with an array length m of 16;

adjacent six-seven-eight bytes are combined to obtain bit flip rate arrays brr [0.007, 0.015, 0.031, 0.062, 0.125,0.25,0.5,1, 0.494, 0.513, 0.493, 0.513, 0.501, 0.495, 0.49, 0.484, 0.498, 0.523, 0.53, 0.505, 0.503, 0.494, 0.503, 0.504], and the array length m is 24.

3) And calculating the bit turnover rate weight of each bit corresponding to each combined full-value data field.

Bit weight array W [ [ alpha ] ]]，

The bit inversion rate array brr [ ] [ [0.125,0.25,0.5,1] of the counter value field of example 4bit, so its bit weight array W [ ] [ [3,2,1,0 ].

The bit inversion rate array of the checksum field of example 4bit may be brr [ ] [ [0.54,0.39,0.44,0.51], so its bit weight array W [ ] [ [1,1,1,1 ].

Similarly, the calculation principle is the same as that described above for the bit weight array with length 16 corresponding to three four bytes and the bit weight array with length 24 corresponding to six seven eight bytes.

4) And performing subfield cutting on each combined full-value data field according to the weight to determine the boundary of each full-value data subfield.

When W [ i ] < W [ i +1], it means that the i +1 th bit belongs to another full-value data field and is the most significant bit of another full-value data field;

when W [ i ] is not less than W [ i +1], the ith bit and the (i + 1) th bit belong to the same full-value data field.

For example, if the bit transition rate corresponding to 24 bits of the 3 rd byte is 0.612 and the bit transition rate corresponding to 25 bits of the 4 th byte is 0, W [24] is 1 and W [25] is ∞, then 24 bits of the 3 rd byte and 25 bits of the 4 th byte belong to different full-value data subfields. The 3 rd byte and the 4 th byte belong to different full value data subfields.

For another example, if the bit flip rate corresponding to 48 bits of the 6 th byte is 1, and the bit flip rate corresponding to 49 bits of the 7 th byte is 0.494, then W [48] is 0 and W [49] is 1, so that 48 bits of the 6 th byte and 49 bits of the 7 th byte belong to different full value data subfields. The 6 th byte and the 7 th byte belong to different full value data subfields.

However, the bit flip rate distribution of each bit of the 7 th byte and the 8 th byte exhibits a normal distribution with 0.5 as expected, and the weight corresponding to each bit is 1, so that the 7 th byte and the 8 th byte belong to the same full-value data subfield.

5) And classifying all full-value data subfields according to the bit turnover rate characteristic of each bit corresponding to all the full-value data subfields after cutting.

The full value data subfields are further divided into three classes:

determining the full-value data subfield as a counter value field if the bit flip rate of the less significant bit is twice the bit flip rate of the more significant bit;

determining the full-value data subfield as the check value field if a bit flip rate distribution of each bit of the full-value data subfield exhibits a normal distribution expected with 0.5.

And if the data sub-field does not belong to the counter value field or the check value field, determining that the full-value data sub-field is the state value field.

For example, byte 3 is a status value field, byte 4 is a status value field, byte 6 is a counter value field, and bytes 7-8 are combined into a check value field.

This step may determine where the state value field is located.

Step 54, monitoring each state value field in the CAN bus data of the vehicle;

for example, 500 frame IDs each have a status value field.

Step 55, collecting field values of each state value in the CAN bus data of the vehicle at preset time intervals, and simultaneously selecting vehicle information with continuous states to collect real physical values at the preset time intervals;

for example, the vehicle information is the vehicle speed, and the vehicle speed is taken as an example, and the vehicle information is taken as 600 seconds of data at a preset time interval of 1 second, that is, each state value field in the monitoring data is taken as 600 state value field values, and the vehicle speed is taken as 600 real physical values at the same time.

And 56, normalizing the collected field information of each state value and the collected vehicle information.

Take 600 data instances of a certain state value field. Firstly, adding or subtracting a number to the 600 data at the same time, so that the minimum number in the 600 data is 0 (other data are all positive numbers); finding out the maximum number A in the 600 data, and dividing the 600 data by A; thus, the 600 data values are all in the range of 0-1. The subsequent steps are to calculate and analyze the data after the normalization processing.

And 57, calculating the sum of squares of errors of each state value field and the vehicle information, and determining that the vehicle information corresponds to the state value field with the minimum deviation.

For any one of the state value fields, the sum of the squares of the errors with the vehicle information is formulated as:

therefore, the specific state value field corresponding to the vehicle information, such as the vehicle speed, the oil quantity and the like, CAN be determined, and the purpose of analyzing the specific functional meaning of the field corresponding to the vehicle CAN bus data is achieved.

EXAMPLE six

Based on the same inventive concept, a schematic structural diagram of a vehicle CAN bus data analysis device provided by a sixth embodiment of the present invention is shown in fig. 4, and the device includes:

the second preprocessing module 401 is used for preprocessing the vehicle CAN bus data and determining each state value field in the vehicle CAN bus data;

the second monitoring module 402 monitors each state value field in the vehicle CAN bus data;

the second processing module 403 is configured to collect field values of each state value in the vehicle CAN bus data at a predetermined time interval, and simultaneously select vehicle information with continuous states to collect real physical values at the predetermined time interval; and fitting each piece of collected state value field information with the collected vehicle information, and determining that the vehicle information corresponds to the state value field with the highest fitting degree.

The second processing module 403 fits the collected field information of each state value with the collected vehicle information, and determines that the vehicle information corresponds to the state value field with the highest fitting degree, and is specifically configured to:

and calculating the sum of squares of errors of each state value field and the vehicle information, and determining that the vehicle information corresponds to the state value field with the minimum deviation.

Prior to fitting, the second processing module 403 is further configured to: and carrying out normalization processing on each collected state value field information and the collected vehicle information.

The second preprocessing module 401 determines each status value field in the vehicle CAN bus data, and is specifically configured to:

In conclusion, the beneficial effects of the invention are as follows:

the vehicle CAN bus data analysis method adopts different processing modes to perform function reversal aiming at the state value fields in the multi-value data field and the full-value data field, and CAN analyze the corresponding operation of the multi-value data field and the field value corresponding to each operation state. For the state value field of the full-value data field, the corresponding vehicle information can be analyzed. Therefore, the purpose of analyzing the specific functional meaning of the corresponding field of the vehicle CAN bus data is achieved.

The above description is only for the preferred embodiment of the present invention, and is not intended to limit the scope of the present invention. Any modification, equivalent replacement, and improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A vehicle Controller Area Network (CAN) bus data analysis method is characterized by comprising the following steps: preprocessing vehicle CAN bus data, and determining each multi-value data field in the vehicle CAN bus data; the method further comprises the following steps:

monitoring each multi-value data field in the CAN bus data of the vehicle;

2. The method of claim 1, wherein if the frequency of the operation state change coincides with the listened multiple-valued data field value change frequency, the operation is subjected to the operation state change at a predetermined frequency; and if the frequency of the operation state change is consistent with the monitored numerical value change frequency of a multi-valued data field, determining that the operation corresponds to the multi-valued data field.

3. The method of claim 1, wherein determining each multi-valued data field in the vehicle CAN bus data specifically comprises:

4. The method according to claim 3, wherein the counting the number of different values existing in each byte to determine the multi-valued data field to which each byte belongs specifically comprises:

5. A vehicle Controller Area Network (CAN) bus data analysis method is characterized by comprising the following steps: preprocessing vehicle CAN bus data, and determining each state value field in the vehicle CAN bus data; the method further comprises the following steps:

monitoring each state value field in the CAN bus data of the vehicle;

6. The method according to claim 5, wherein the fitting each collected state value field information with the collected vehicle information to determine that the vehicle information corresponds to the state value field with the highest degree of fitting specifically comprises:

7. The method of claim 5, wherein prior to fitting, the method further comprises: and carrying out normalization processing on each collected state value field information and the collected vehicle information.

8. The method of claim 5, wherein determining each status value field in the vehicle CAN bus data specifically comprises:

9. A vehicle Controller Area Network (CAN) bus data analysis device is characterized by comprising:

10. The apparatus according to claim 9, wherein the first processing module is specifically configured to change the operation state according to a predetermined frequency if the frequency of the change of the operation state is consistent with the monitored frequency of the change of the multiple-valued data field values; and if the frequency of the operation state change is consistent with the monitored numerical value change frequency of a multi-valued data field, determining that the operation corresponds to the multi-valued data field.

11. The apparatus of claim 9, wherein the first preprocessing module determines each multi-valued data field in the vehicle CAN bus data, and is specifically configured to:

12. The apparatus of claim 11, wherein the first preprocessing module counts the number of different values existing in each byte to determine the multi-valued data field to which each byte belongs, and is specifically configured to:

13. A vehicle Controller Area Network (CAN) bus data analysis device is characterized by comprising:

14. The apparatus of claim 13, wherein the second processing module fits each collected state value field information with the collected vehicle information, determines that the vehicle information corresponds to a state value field with a highest degree of fit, and is specifically configured to:

15. The apparatus of claim 13, wherein prior to fitting, the second processing module is further to: and carrying out normalization processing on each collected state value field information and the collected vehicle information.

16. The apparatus of claim 13, wherein the second preprocessing module determines each status value field in the vehicle CAN bus data, and is specifically configured to: