METHOD FOR MONITORING COMMUNICATION ON A COMMUNICATION BUS, ELECTRONIC DEVICE FOR CONNECTION TO A COMMUNICATION BUS, AND CENTRAL MONITORING DEVICE FOR CONNECTION TO A COMMUNICATION BUS
Patent Information
- Application Number
- DE502020009724
- Authority / Receiving Office
- DE · DE
- Patent Type
- Patents
- Current Assignee / Owner
- Priority Date
- 2019-11-22
- Filing Date
- 2020-11-10
- Publication Date
- 2024-11-28
- Estimated Expiration
- 2040-11-10
AI Technical Summary
Current communication bus monitoring systems in vehicles fail to effectively detect unauthorized access and identify manipulated stations that send infiltrated messages, posing a security risk as they rely on pre-installed lists and are vulnerable to manipulation.
A method where each station maintains a list of sent message identifiers, with a protocol unit keeping track of these identifiers to enforce a uniqueness rule, allowing for self-learning and secure monitoring without pre-installation, and a central monitoring station compares these lists to identify and isolate compromised stations.
This approach enhances security by enabling the detection of unauthorized access and manipulated stations, allowing for targeted security measures and reducing dependency on station hosts, thus improving the overall security of the communication bus.
Description
[0001] The invention relates to the technical field of monitoring communication on a communication bus for unauthorized bus access. This method can be used particularly with communication buses used in vehicles. Networked control units are also found in other fields of technology, e.g., in automation technology, process engineering, etc. The invention further relates to an electronic station for connection to a communication bus.
[0002] A multitude of control units are installed in modern vehicles. Several control units are used for the drivetrain alone, such as the engine control unit, transmission control unit, selector lever control unit, airbag control unit, and others. There are also other control units installed in the vehicle body that provide specific comfort functions. Examples include door or window lift control units, air conditioning control units, seat adjustment control units, and others. There are also control units that belong to the infotainment area, such as camera control units for monitoring the surroundings, navigation systems, communications modules, and entertainment devices with TV, radio, video, and music functions.
[0003] Typically, the control units of the different categories are each networked via a separate bus designed specifically for that category. Therefore, several different bus systems can be used in the vehicle. These various bus systems can be interconnected via gateways to enable data exchange. The CAN bus (Controller Area Network) is typically used for powertrain control units, as well as for comfort control units. Other bus systems are also used in the infotainment sector, such as those based on Ethernet technology, for example, AVB (Audio Video Bridging), which is based on the IEEE 802.1 standard family. Bus systems that transmit data via fiber optics are also possible. Examples include the MOST bus (Media Oriented System Transport) and the D2B bus (Domestic Digital Bus).
[0004] Bus systems in the automotive sector are increasingly becoming the target of hacking attacks and attempts to deliberately manipulate message content. Such attacks on the bus system typically occur via access to the physical transmission medium, i.e., the bus line, or via an OBD port (On-Board Diagnostics interface). Cybersecurity is also gaining increasing importance because of the growing number of complex driver assistance systems in vehicles, culminating in fully automated driving. Manipulation of these systems must be prevented.
[0005] Therefore, there is a need to continuously improve the security of communication on the communication buses in the vehicle as well as elsewhere.
[0006] The CAN bus is particularly widespread in the automotive sector and is frequently used for networking safety-relevant electronics within the vehicle. Therefore, there is a particular need for securing the communication.
[0007] DE 10 2015 205 670 A1 discloses an attack detection method for a motor vehicle bus system and a corresponding device. The aim is to detect and ward off external attacks on the bus system. For this purpose, a module is installed in a gateway of the bus system which uses a general approach to check whether the messages transmitted on the bus were transmitted according to the communication rules. These include various communication rules: According to one aspect, the module checks certain properties of a specified message cycle. These include, for example, the time between two consecutive messages and if it detects that the time interval between two consecutive messages does not match the specified cycle duration, it issues a warning message. According to another aspect, the module checks whether only identical messages are transmitted in succession.The review aspects may relate to a specific message type.
[0008] DE 10 2016 220 895 A1 discloses measures for detecting tampering in a CAN network. This involves extending a basic CAN controller or a full CAN controller with an RX filter device that compares the CAN identifiers intended for transmission with those of the received CAN messages.
[0009] A monitoring method for the CAN bus is disclosed in German patent application DE 10 2017 216 808 A1. This method exploits the uniqueness rule inherent in the CAN bus, which prohibits another station from sending a payload message with an identifier already reserved for that station. If a violation of this uniqueness rule is detected, unauthorized bus access has been identified.
[0010] From DE 10 2017 218 134 B3, a method for transmitting a message sequence via a data bus is known. An information message containing an information signal is transmitted during an active phase, and a safety message to initiate a sleep phase, as well as sleep messages containing a sleep signal, are transmitted at intervals of one sleep cycle time during the sleep phase. The information signal and the sleep signal are distinct from each other, and the safety message and the sleep messages are also distinct from each other.
[0011] From EP 3 148 154 A1, a Controller Area Network (CAN) device with a comparator module is known. The comparator module is configured to interface with a CAN transceiver and includes a receive data (RXD) interface, a CAN decoder, an identifier memory, comparator logic, and a signal generator. The receive data (RXD) interface is configured to receive data from the CAN transceiver. The CAN decoder is configured to decode an identifier of a CAN message received by the RXD interface. The identifier memory is configured to store an entry that corresponds to at least one identifier.The comparison logic is configured to compare a received identifier from a CAN message with the entry stored in the identifier memory and outputs a match signal if the comparison indicates that the received identifier from the CAN message matches a stored entry. The signal generator is configured to output a signal to invalidate the CAN message in response to the match signal.
[0012] A mechanism for detecting intrusion attempts is known from US Patent 2016 / 0308891 A1. The mechanism uses a method implemented on a node connected to a network bus. In this method, message identifiers are stored, with at least one of these identifiers identifying the node. This at least one message identifier is contained in a message at the time the message is sent from the node to the network bus. The method also monitors network bus traffic, which includes messages sent by the node and by other nodes connected to the network bus. If a message sent on the network bus by at least one of the other nodes is identified as having a message identifier that matches the at least one message identifier, a processor on the node is alerted.
[0013] From DE 10 2015 205 670 A1, an attack detection method for a bus system of a motor vehicle is known, wherein communication rules for sending messages are defined for the bus system. In the method, messages sent via the bus system are received, and it is analyzed whether the received messages are received in accordance with the communication rules.
[0014] A server for detecting vehicle anomalies is known from WO 2019 / 142741 A1. The server comprises a communicator that communicates with a vehicle to receive a protocol from an in-vehicle network. The server also includes a vehicle selector that, upon receiving information indicating that an anomaly is present in a first vehicle among a multitude of vehicles, selects a vehicle associated with the anomaly from among the multitude based on the anomaly, with the first vehicle being the one communicating with the communicator. The server also includes a protocol collector that transmits an initial request to the vehicle associated with the anomaly, via the communicator, to transmit a protocol from an in-vehicle network.In addition, the server includes a protocol analyzer that determines whether an anomaly occurs in the vehicle associated with the anomaly based on information indicated by the protocol transmitted by the vehicle associated with the anomaly in response to the first request and received by the communicator.
[0015] From WO 2019 / 074000 A1, a vehicle control device for a vehicle-mounted system is known, which contains a plurality of information processing devices interconnected via a communication bus. The vehicle control device is configured to process information containing first and second management information. The first management information relates to security anomalies, which are communication data anomalies attributable to a security attack from outside the vehicle-mounted system. The second management information relates to security anomalies, which are communication data anomalies attributable to anomalies within the vehicle-mounted system.
[0016] The invention aims to provide an effective monitoring method for a communication bus that can not only detect unauthorized bus access but also offers a chance to identify the manipulated stations that send out the injected messages.
[0017] This problem is solved by a method for monitoring communication on a communication bus according to claim 1, an electronic station according to claim 6, and a central monitoring station for a communication bus according to claim 8.
[0018] The dependent claims include advantageous further developments and improvements of the invention in accordance with the following description of these measures.
[0019] The solution consists of a method for monitoring communication on a communication bus through which a number of electronic stations are networked. The message format is designed such that the messages are identified by an identifier, with a defined set of rules for each station regarding which messages with which identifier it is permitted to send. A uniqueness rule prevents another station from sending a payload message with an identifier already reserved for that station. A protocol unit in a control unit maintains a list of the identifiers of the messages sent by each station. This list allows for the detection of violations of the uniqueness rule, enabling the initiation of security measures against the identified station.Maintaining the list at each station has the advantage that the monitoring system is self-learning, eliminating the need for pre-installation of the lists. The functionality of the stations can be modified by subsequently enabling features or a subsequent software update, without having to reprogram the lists in the workshop. Another advantage is that there is no dependency on the station's host, allowing the monitoring process to be designed very securely. Another advantage is the possibility of subsequently implementing the monitoring process at the station with the help of a software update.
[0020] According to the invention, the list is sent to a central monitoring station connected to the communication bus with a message containing an entry for an identifier of the sending station. A higher-level monitoring instance installed in the central monitoring station compares the identifiers recorded in the list with a table containing the identifiers for the stations on the communication bus and the reference lists with the identifiers assigned to each station that may be sent in messages from the station sending the list. This measure allows the identification of the station that has injected a forged message (attacker control unit). This makes the monitoring system significantly more secure because targeted security measures can then be initiated against the identified, manipulated control unit.
[0021] It is advantageous if the protocol unit builds the list successively by generating a new list entry whenever the station sends a message with an identifier that has not previously been entered into the list. This avoids duplicate entries in the memory and saves storage space.
[0022] In one variant, the message containing the logged list is sent to the central monitoring station connected to the communication bus, which has a higher-level monitoring authority, when the end of a duty cycle is detected. In the simplest case, a duty cycle can consist of the phase from switching on to switching off the power supply to the communication system.
[0023] This monitoring method can be particularly advantageous for use with a communication bus based on a variant of the CAN bus family, corresponding to a Controller Area Network bus. As a safeguard, if the monitoring instance identifies a station whose reference list does not contain an entry for an identifier listed in the received list, the bus-off state is triggered for the identified station. In the bus-off state, the station is disconnected from the communication bus. In this state, it can no longer transmit forged messages. Other safeguards could also be initiated in this way. Examples of other safeguards include initiating a secure mode so that no threat emanates from the station, e.g., from an attacker's control unit.Many control units are equipped with a limp-home mode in which they only perform a basic function. With this basic function, for example, the vehicle's full power can no longer be accessed in a vehicle network, but the vehicle can still reach a repair shop. The CAN bus has been used in vehicles since the early 1990s and is very widespread. The data transmissions that take place via this communication bus are safety-relevant, and therefore there is a high demand for securing CAN bus communication against manipulation.
[0024] Examples of other safeguards that can be used in the process include: Issuing a warning message to the driver, for example with a visual, acoustic or haptic signal, activating the alarm system, sending a tampering message via radio to a central database computer of the vehicle manufacturer or an authority or to a smartphone of the vehicle owner, saving the location, date, and time of the tampering event in a memory, blocking the key with which the vehicle was switched on.
[0025] This list is not exhaustive, and several of the listed security measures can be used in parallel.
[0026] One adaptation of the procedure for use with a CAN bus is that the identifier being logged corresponds to a CAN bus message identifier.
[0027] The monitored user data messages can correspond to a so-called CAN bus standard data frame.
[0028] In another embodiment, the invention relates to an electronic station for connection to a communication bus. A number of electronic stations are networked via the communication bus, the messages transmitted via the communication bus being marked with an identifier. For each station, it is defined which messages with which identifier may be sent by the station. A uniqueness rule applies, prohibiting another station from sending a payload message with an identifier that is already reserved for that station. The station includes a protocol unit designed to maintain a list of the identifiers of the messages sent by the station.
[0029] The station also features a logic unit designed to send the list maintained by the protocol unit to a central monitoring station connected to the communication bus. This central monitoring station has a higher-level monitoring authority. The message contains an entry for an identifier of the sending station. The higher-level monitoring authority compares the identifiers listed in the list with a table containing the identifiers for identifying stations and reference lists, along with the identifiers assigned to each station that the sending station is permitted to send in messages. If the corresponding reference list does not contain an entry for an identifier listed in the received list, a transmission unit of the central monitoring station sends a backup message containing a backup measure to the station.
[0030] The logic unit of the electronic device is designed to receive the security message from the central monitoring station and to execute a security measure communicated in the security message. This results in the same advantages as the method according to the invention.
[0031] Another solution consists of a central monitoring station for connection to a communication bus, wherein a number of electronic stations are networked via the communication bus, wherein the messages transmitted via the communication bus are marked with an identifier, wherein it is defined for each station which messages with which identifier may be sent by the station, wherein a uniqueness rule applies which prohibits another station from sending a payload message with an identifier that is already reserved for that station, and wherein a protocol unit in a station maintains a list of the identifiers of the messages actually sent by the station, characterized in that the central monitoring station is designed to receive the list from a station with a message that contains an entry for an identifier of the sending station.and wherein the central monitoring station includes a higher-level monitoring instance which is designed to compare the identifiers recorded in the list with a table present in the central monitoring station, in which the identifiers relating to the identification of the stations of the communication bus and reference lists with the identifiers assigned to each control unit are recorded, which may be sent in messages from the station sending the list.
[0032] In order to carry out security measures against the unmasked station, it is advantageous if the central monitoring station has a transmitting unit which is designed to send a security message with a security measure to the station whose identifier matches the identifier of a received list for which the associated reference list does not have an entry for an identifier listed in the list.
[0033] An embodiment of the invention is illustrated in the drawings and is explained in more detail below with reference to the figures.
[0034] They show: Fig. 1 shows the principle of networking electronic components via CAN bus; Fig. 2 shows a block diagram for a vehicle communication network with control units of different categories; Fig. 3 shows the format of the standard frame transmission frame for the CAN bus; Fig. 4 shows the format of the remote frame transmission frame for the CAN bus; Fig. 5 shows a block diagram of a CAN bus interface equipped with a monitoring module according to the invention; Fig. 6 shows a flowchart for a program installed as a monitoring module on the CAN bus interface; and Fig. 7 shows a flowchart for a variant of attack detection.
[0035] The present description illustrates the principles of the disclosure according to the invention. It is therefore understood that those skilled in the art will be able to design various arrangements which, although not explicitly described here, embody principles of the disclosure according to the invention.
[0036] The CAN bus was first standardized in 1994. The corresponding ISO standard is ISO 11898. There is a standard for the high-speed range up to 1 Mbit / s, which is ISO 11898-2. Then there is a standard for the low-speed range up to 125 kbit / s, which is ISO 11898-3. Due to the increasing volume of data, the bus loads on the CAN buses are constantly rising. This led to a further development of the CAN bus. The extended CAN bus is known as the CAN FD bus. FD stands for Flexible Data Rate. In this CAN bus variant, the data rate is switched. For the arbitration phase, the rate remains low, as with the classic CAN bus. For the transmission of the user data, a higher data rate is used. If the user data of a CAN FD message is transmitted faster, the duration of bus occupancy is reduced, and the bus load decreases.If the transmission time remains within the same range as for traditional CAN messages, larger amounts of data could be transported with a CAN FD message. This is also the case with CAN FD. Instead of the 8-byte payload field, CAN FD uses a payload field of up to 64 bytes. The data rate for transmitting the payload field increases from 500 kbit / s to 2 Mbps, for example.
[0037] The classic CAN bus also has a special CAN Remote Frame format. The CAN Remote Frame is sent by one station to request specific data from another station.
[0038] In the following description of the exemplary implementation, the bus stations are referred to as control units, as is common in the automotive sector. However, it is also possible that a bus station is not designed as a control unit. Examples given include certain sensors or actuators (e.g., control elements) that are connected to the bus.
[0039] Fig. 1 This illustrates the principle of networking electronic components using a CAN bus. A CAN network is a system consisting of CAN nodes with CAN interfaces (electronic components such as control units, sensors, and actuators) that exchange data with each other via their respective CAN interfaces and a transmission medium (CAN bus) connecting all CAN interfaces. Three CAN nodes 10 are shown. The bus structure of the CAN bus is linear. Therefore, there is one bus line 15 to which all three CAN nodes 10 are connected. In most applications, a twisted, unshielded two-wire cable (unshielded twisted pair - UTP) is used as the bus line 15, over which symmetrical signal transmission takes place. In symmetrical signal transmission, the signals are transmitted as voltage differences across two lines. The wire pair consists of a non-inverted CANH and an inverted signal line CANL.The receivers reconstruct the original data signal from the difference between the signals present on these two conductors. This has the advantage that common-mode interference occurring on both conductors of bus line 15 is canceled out by the difference calculation and thus does not affect the transmission.
[0040] To avoid signal reflections, the bus line 15 is terminated at both ends with a terminating resistor 13 of the same value as the characteristic impedance of the bus line (120 ohms).
[0041] A CAN interface consists of two parts: the communication software and the communication hardware. While the communication software encompasses higher-level communication services, the basic communication functions are typically implemented in hardware. Two hardware components are distinguished here: The CAN controller 14 ensures the standardized execution of the CAN communication protocol, thereby relieving the host 16, on which the aforementioned communication software runs, of this task. The CAN transceiver 12 connects the CAN controller 14 to the CAN bus 15. It shapes the signals for data transmission during the sending process and performs signal conditioning during reception.
[0042] Fig. 2 This diagram illustrates the typical structure of a communication network in a modern motor vehicle. Reference number 151 designates an engine control unit (ECU). Reference number 152 corresponds to a gear selector control unit, and reference number 153 designates a transmission control unit. Additional control units, such as a vehicle dynamics control unit (for vehicles with electrically adjustable dampers), an airbag control unit, etc., may be present in the vehicle. The networking of such control units, all belonging to the powertrain category, is typically achieved using the CAN bus system (Controller Area Network) 104, which is standardized as an ISO standard, most commonly ISO 11898-1. Various sensors in the vehicle, which are no longer intended to be connected solely to individual control units, are also designed to be connected to the CAN bus system 104, with their sensor data being transmitted via the bus to the individual control units.Examples of sensors in a motor vehicle include wheel speed sensors, steering angle sensors, acceleration sensors, yaw rate sensors, tire pressure sensors, distance sensors, knock sensors, air quality sensors, etc. The gear selector control unit, connected to the gear selector control unit, allows the driver to select driving modes. These include gear selection and engine settings such as sport mode, normal mode, all-wheel drive, etc.
[0043] However, the modern motor vehicle can also have other components, such as video cameras, e.g. as a rear-view camera or as a driver monitoring camera. The motor vehicle also contains other electronic devices. These are located closer to the passenger compartment and are often operated by the driver. Examples include a user interface device with which the driver can make settings, but also operate classic components. These include turn signal controls, windshield wiper controls, light controls, audio settings for the radio, other settings for the car phone, navigation system, etc. This user interface arrangement is provided with the reference number 130. The user interface arrangement 130 is often also equipped with a rotary / push button with which the driver can select the various menus shown on a display in the cockpit.On the other hand, a touch-sensitive display also falls into this category. Even voice input for user assistance falls into this area.
[0044] The navigation system has the reference number 120, which is also installed in the cockpit area. The route, which is shown on a map, can of course also be shown on the cockpit display. Other components, such as a hands-free system, may be present but are not shown in detail. The reference number 110 also designates an on-board unit. This on-board unit 110 corresponds to a communications module through which the vehicle can receive and send mobile data. Typically, this is a cellular communications module, e.g., according to the LTE standard. All of these devices belong to the infotainment area. They are therefore networked via a bus system 102 designed for the specific needs of this device category. In the example shown, it is assumed that the bus system 102 was also implemented as a variant of the CAN bus.The CAN FD bus mentioned earlier would be suitable, as it can transport data at a higher data rate, which is advantageous for the networked control units in the infotainment area.
[0045] A different bus system is often used for the infotainment area. Examples include the AVB (Audio Video Bridging), MOST (Media Oriented System Transport), or D2B (Domestic Digital Bus) bus systems. Gateway 140 is provided for transmitting vehicle-relevant sensor data via communication interface 110 to another vehicle or to an external central computer of a database. Gateway 140 is connected to both different bus systems 102 and 104. Gateway 140 is designed to convert the data it receives via CAN bus 104 into the transmission format of infotainment bus 102 so that it can be distributed in the packets specified there.To forward this data externally, i.e., to another vehicle or to the central computer, the on-board unit 110 is equipped with a communication interface to receive these data packets and convert them into the transmission format of the corresponding mobile communication standard. Conversion is also necessary if bus 102 is implemented as a CAN FD bus.
[0046] As in Fig. 2 As shown, a monitoring module 18 is provided in each of the control units connected to the respective CAN bus 102, 104.
[0047] Fig. 3 This shows the message format of a CAN Standard Frame. More precisely, it illustrates Fig. 3 a CAN transmission frame format according to the CAN communication standard.
[0048] The ISO 11898-1 transmission frame contains many different individual bits that serve control purposes. The various fields and control bits of the transmission frame are listed with their English names in the following table. The length of each field is also given. The full names of these bits will not be repeated in subsequent references. Control Bit Ausführliche Bezeichnung Länge SOF Start of Frame 1 Bit Identifier Identifier 9 Bits RTR Remote Transmission Request 1 Bit IDE Identifier Extension 1 Bit r Reserved Bit 1 Bit DLC Data Length Code 4 Bit Data Field Data Field 0-8 Bytes CRC CRC Sequence DEL CRC Delimiter 1 Bit ACK Acknowledge 1 Bit DEL ACK Delimiter 1 Bit EOF End of Frame Code 4 Bit
[0049] A CAN frame contains a Start-of-Frame (SOF) field, an Arbitration field, a Control field, a Data field, a Cyclic Redundancy Check (CRC) field, an ACK field, an End-of-Frame (EOF) and an Intermission Sequence (ITM) field.
[0050] According to an exemplary embodiment of the invention, the SOF field is a field that indicates the beginning of a CAN frame, i.e., the beginning of a message. The arbitration field identifies a message and assigns it a priority. Based on the length of an identification field associated with the arbitration field, the CAN frame is split into a standard format and an extended format (the standard format is shown). In the standard format, the arbitration field has a length of 11 bits. For the extended format, the length of the identification field in the arbitration field is 29 bits.
[0051] The identifier determines the priority of the data frame and, together with the acceptance filter, ensures the sender-receiver relationships defined in the communication matrix in the CAN network. The communication matrix specifies which messages each ECU processes. Therefore, if a message arrives whose message identifier is not listed there, the message is filtered out by the acceptance filter and not forwarded to the application.
[0052] Using the RTR bit, the transmitting station informs the receivers of the frame type (data frame or remote frame). A dominant RTR bit indicates a data frame, while a recessive bit indicates a remote frame. Additionally, the arbitration field can contain a 1-bit Identification Extension (IDE) field to identify whether a frame is in standard or extended format. A value of 0 indicates standard format. A value of 1 indicates extended format.
[0053] The DLC field indicates to the receiver the number of payload bytes contained in the message. The payload bytes are transported in the data field. A maximum of eight payload bytes can be transmitted in one data frame, or up to 64 bytes for CAN-FD. The payload bytes are protected against transmission errors using a checksum transmitted in the CRC field, applying the cyclic redundancy check.
[0054] Based on the result of the CRC check, the receivers acknowledge receipt positively or negatively in the ACK slot. An ACK bit is transmitted at the end of the message by the CAN controllers that accurately received the message. The node that sent the message checks whether the ACK bit is present on the CAN bus. If ACK is not found, this indicates that a node was unable to receive the message correctly, and the transmitting station can attempt retransmission.
[0055] The transmission of a data frame is terminated with seven recessive bits, which corresponds to the End Of Frame Code EOF.
[0056] Fig. 4 The image also shows the message format of a CAN remote frame. A remote frame allows an electronic control unit (ECU) to request desired user data if it is not already being transmitted cyclically. This frame type is rarely used in automotive applications, as data transmission there is not on demand but essentially cyclical.
[0057] Except for the missing data field, the structure of a remote frame is identical to that of a data frame. The distinction between data and remote frames is made using the RTR bit. In the case of a data frame, the RTR bit is transmitted dominantly. A remote frame is characterized by a recessive RTR bit.
[0058] In principle, corresponding remote frames can be defined for all existing data frames in the CAN network. To do this, one simply has to ensure that the identifiers of the remote frames match the identifiers of the associated data frames. As soon as a CAN node receives a remote frame whose identifier is identical to an identifier in its own communication matrix, it responds with the corresponding standard frame.
[0059] The Fig. 5 This shows a possible implementation of the monitoring module 18 in the form of a software module in a control unit. The CAN interface of the control unit consists of the components CAN transceiver 12 and CAN controller 14. Reference number 16, in turn, denotes the host hardware and software of the control unit. The monitoring module 18 is implemented using software and consists of the three components protocol unit 18-1, detector unit 18-2, and logic unit 18-3.
[0060] The following explains two different methods for detecting attacks in the on-board electronics of the 100 series. Fig. 2 The example shown can be carried out. The variant according to Fig. 6 This corresponds to cyclic attack detection. In the case shown, the cycle time is linked to the vehicle's operating phase. In vehicles, the operating phase refers to a so-called terminal 15 cycle. This term traditionally dates back to the early days of automotive technology. Terminal 15 traditionally refers to the switched positive terminal of the battery. This switching process occurred when the ignition key was turned. Modern vehicles often have a start button that also activates the power supply. However, some devices remain connected to the power supply even after the normal mode has been terminated by pressing the start button again. Such devices then switch, for example, from normal mode to standby mode, in which they consume less power but still perform certain functions.As an example, a body control unit is mentioned that waits in standby mode for the receipt of a radio signal from a radio key in order to unlock the doors when the radio signal arrives.
[0061] The Fig. 6 shows, from top to bottom, two consecutive terminal 15 cycles. In the terminal 15 cycle x, various message transmissions between the control units in the drive train are shown. From left to right, the control units are designated as transmission control unit 153, engine control unit 151, manipulated selector lever control unit 152, and gateway 140. First, a message is transmitted via CAN bus 104 from transmission control unit 153 to engine control unit 151 in step 202. This is a regular message in the format of the standard data frame, as in Fig. 3 The message is shown and sent. This message is received by the engine control unit 151 because it was identified as relevant to the engine control unit 151 through the acceptance filtering process. The appropriate response to this message from the engine control unit 151 occurs in step 206. If the measurement data is important, it can be stored in memory, and the control program running in host 16 accesses it to execute a control function based on this measurement data. Simultaneously, the protocol unit 18-1 of the CAN interface adds an entry to a list of the station's own CAN identifiers. This occurs in step 204. This list is gradually updated as more messages are sent by the transmission control unit 153. The purpose of this list is to subsequently identify an attacking control unit.For this to work, the attacking control unit must also build a list. A more secure option against manipulation is to ensure that every control unit, including the attacking control unit, is equipped with a correspondingly designed hardware component containing this protocol unit. This could be implemented in a hardware block integrated into the CAN transceiver or the CAN controller. In the example shown, the attacking control unit is a manipulated selector lever control unit 152. In step 208, the attacking control unit receives the command to send a forged message to the CAN bus 104. This can also be triggered by a manipulated control program, although this would require detailed knowledge of the standard control program. In step 210, the attacking control unit 152 sends the forged message to the CAN bus 104.The engine control unit 151 accepts the falsified message because it passes its acceptance filter. In step 214, the engine control unit 151 reacts incorrectly. This can occur because the control program calculates with the falsified transmitted measured value or parameter value. Subsequently, further regular and falsified messages can be transmitted via the CAN bus 104, which will not be considered further. In step 215, a terminal 15 event occurs. This switches off the on-board electronics 100. However, the shutdown is not immediate. The control units are "powered down" in a controlled manner, performing some tasks such as saving or resetting set parameters, etc. The following steps 216 and 218 correspond to steps that occur in connection with the power-off of control units 153 and 152.In step 216, the transmission control unit 153 reports its logged list of CAN identifiers it has sent in messages to a higher-level monitoring instance. This monitoring instance is preferably located in a central network node. For the on-board electronics 100 of vehicles, the gateway 140 is suitable for this purpose. The message containing the list of used CAN identifiers is sent to the gateway 140 via the CAN bus 104. Since the payload field in the standard data frame is limited, several CAN messages must be transmitted. Each of these messages, which transmits a portion of the logged CAN identifiers, also contains an identifier of the sending control unit. In one variant, this identifier can specify the type of control unit. Thus, in the case of the transmission control unit 153, the identifier indicates that the message transmits the CAN identifiers of a transmission control unit.The same message is also sent from the attacker control unit 152 to the gateway 140. The attacker control unit 152 also sends all logged CAN identifiers to the gateway 140. Since the attacker control unit 152 also logs the CAN identifiers of the forged messages, these are also delivered to the higher-level monitoring instance in the gateway 140. This is the basis for the ability to detect the attacker control unit 140. However, in the example shown, the detection process for the attacker control unit 140 only takes place in the subsequent terminal 15 cycle, i.e., during the next startup process of the on-board electronics 100.
[0062] In Fig. 6 The subsequent terminal 15 cycle is designated with index x+1. The evaluation of the transmitted CAN identifier protocols takes place in step 220. In step 220, a received CAN identifier protocol is compared with a reference list archived in the monitoring instance. The reference list contains the regular CAN identifiers assigned to the corresponding control unit. This list is stored in each control unit and is used for acceptance filtering upon receipt of a CAN message, to determine whether the message is relevant to the control unit itself or is addressed to another control unit. The reference list is sometimes also referred to as the K-matrix, corresponding to the communication matrix. It must be ensured that all reference lists of the control unit network are present in the vehicle electronics 100 within the monitoring instance of gateway 140.These reference lists can be programmed into the gateway 140 in a protected memory area (e.g., EPROM) at the end of the vehicle manufacturing process. When the list, which is logged by the protocol unit 18-1 in the selector lever control unit 152, is tested, it is detected that the CAN identifier of the forged message sent in step 210 has also been logged. This CAN identifier is not included in the reference list. This reveals that the selector lever control unit 152 has been manipulated. Subsequently, the monitoring instance initiates one or more of the described security measures. This is done by the monitoring instance in the gateway 140 sending a message via the CAN bus 104 to the attacker's control unit 152. The logic unit 18-3 is provided for this purpose in each control unit.The logic unit 18-3 in the selector lever control unit 152 receives the message with the security measure from the gateway 140. The logic unit can either be connected to the host 16 via a direct signal line and send the host a command to shut down or initiate a secure mode. Alternatively, this command could be delivered to the CAN controller 14 via a CAN message. Another security measure involves the logic unit triggering a bus-off state. This then interrupts the connection to the CAN bus 104.
[0063] In Fig. 7 Another variant of attack detection is shown, which is not the subject of the claimed invention. The same reference numbers denote the same components, as explained previously. The difference is that this variant enables immediate attack detection. For this to work, the control units themselves must perform a comparison with the list generated by the protocol unit 18-1. This comparison is performed even if the list is incomplete. The comparison with the recorded list is carried out in the detector unit 18-2. For each incoming message, a comparison of the CAN identifier contained in the message with the recorded list is performed. This comparison first takes place in the transmission control unit 153 in step 222, after the forged message has been sent by the attacker control unit 152 in step 210.The comparison results in a match with an entry in the list logged by transmission control unit 153. Whenever a match is detected, the existing uniqueness rule indicates that a potential attack has been identified. This is then immediately reported to the monitoring unit in gateway 140 in step 216. This report includes the identifier for the reporting control unit and the CAN identifier of the received message where the match was detected. However, the monitoring unit cannot yet identify the attacking control unit based on this information. This is only resolved when the functioning control unit sends another regular message with the suspect CAN identifier. This occurs in step 224 via transmission control unit 153.In step 228, the selector lever control unit 152 detects that an identical CAN identifier has been received, which is also listed in the list maintained by the station's own protocol unit 18-1. Therefore, in step 218, a message is sent to the monitoring instance in the gateway 140, specifying the suspicious CAN identifier and the control unit ID of the selector lever control unit 152. The monitoring instance then receives two messages: one from the regular control unit and one from the rogue control unit 152. This information is sufficient to identify the rogue control unit 152. To this end, in step 220, the monitoring instance compares its existing reference lists with the suspicious CAN identifier. There will be a regular reference list in which the CAN identifier is listed.Once this reference list is found, the monitoring instance can determine which control unit identifier matches it. The message with this control unit identifier then originated from the functioning control unit. This simultaneously exposes the attacking control unit, the one from which the other message originated. Subsequently, one or more security measures are initiated against the identified attacking control unit. Alternatively, the monitoring instance could store a mapping list containing the regular control unit identifiers for each CAN identifier. This would eliminate the time-consuming search for the matching reference list.
[0064] It is also mentioned that an attack could already be inferred from the number of entries stored in the list maintained by the protocol unit.
[0065] It is also mentioned that the lists maintained by the protocol unit should be protected against memory overflow in order to be able to control an attack by flooding the CAN bus with many different CAN IDs.
[0066] The disclosure is not limited to the embodiments described herein. There is room for various adaptations and modifications that a person skilled in the art would consider based on their technical knowledge and the disclosure.
[0067] There are several options for implementing the described solution for attacker identification: 1. Software solution where the modules protocol unit, detector unit, and logic unit are installed as part of the interface software in the control unit. 2. Software solution installed in an additional hardware security module. 3. Hardware solution integrated into the CAN transceiver or CAN controller chip.
[0068] Of these implementation variants, the last two implementation types can be classified as relatively secure against manipulation.
[0069] The described monitoring method could also be used with bus systems other than the CAN bus. The LIN bus (Local Interconnect Bus) is specifically mentioned as an example. However, the LIN bus is a master / slave bus, which also includes an identifier in the message format, which can also identify a specific control command.
[0070] All examples mentioned herein, as well as conditional formulations, are to be understood without limitation to such specifically cited examples. For instance, it is recognized by experts that the block diagram shown here represents a conceptual view of an exemplary circuit arrangement. Similarly, it is understood that a flowchart, state transition diagram, pseudocode, and the like are different ways of representing processes that are essentially stored in computer-readable media and can thus be executed by a computer or processor.
[0071] It should be understood that the proposed method and associated apparatus can be implemented in various forms of hardware, software, firmware, specialized processors, or a combination thereof. Specialized processors can include application-specific integrated circuits (ASICs), reduced instruction set computers (RISCs), and / or field-programmable gate arrays (FPGAs). Preferably, the proposed method and apparatus are implemented as a combination of hardware and software. The software is preferably installed as an application program on a program storage device. Typically, this is a machine based on a computer platform that includes hardware such as one or more central processing units (CPUs), random access memory (RAM), and one or more input / output (I / O) interfaces. An operating system is also typically installed on the computer platform.The various processes and functions described here may be part of the application program or a part that is executed via the operating system. Bezugszeichenliste
[0072] 10 CAN node 12 CAN transceiver 13 Termination resistor 14 CAN controller 15 Bus line 16 Host 18 Monitoring module 18-1 Protocol unit 18-2 Detector unit 18-3 Logic unit 19 Monitoring instance 100 Vehicle electronics 102 Infotainment CAN bus 104 CAN bus 110 On-board unit 120 Navigation system 130 Control unit 140 Gateway 151 Engine control unit 152 Selector lever control unit 153 Transmission control unit 202 - 228 Various steps of an attack detection procedure
Claims
1. Method for monitoring the communication on a communication bus (104), wherein the communication bus (104) networks a number of electronic stations (151, 152, 153), wherein the messages transmitted via the communication bus (104) are identified with an identifier, it being stipulated for each station (151, 152, 153) which messages with which identifier are allowed to be sent by the station (151, 152, 153), wherein an unambiguity rule prohibiting another station (152, 153) from sending a payload-data message with an identifier that is already reserved for this station (151) applies, and wherein a protocol unit (18-1) in a station (151, 152, 153) keeps a list of the identifiers of the messages actually sent by the station (151, 152, 153), characterized in that the list is sent to a central monitoring station (140) connected to the communication bus (104) with a message which includes an entry for an identifier of the sending station (151, 152, 153), wherein a superordinate monitoring entity installed in the central monitoring station (140) compares the identifiers recorded in the list with a table in which are recorded the identifiers regarding the identification of the stations (151, 152, 153) and reference lists containing the identifiers assigned to each station (151, 152, 153) which are allowed to be sent in messages from the station (151, 152, 153) sending the list.
2. Method according to Claim 1, wherein the protocol unit (18-1) successively builds up the list by generating a new list entry when the station (151, 152, 153) sends a message containing an identifier which was previously not yet entered into the list.
3. Method according to Claim 1 or 2, wherein the message with the protocolled list is sent to the central monitoring station (140) connected to the communication bus (104) when an end of a working cycle is detected.
4. Method according to one of the preceding claims, wherein a transmitting unit of the central monitoring station (140) sends a security message containing a security measure to that station (151, 152, 153) whose identifier matches the identifier for a received list in which the associated reference list has no entry for an identifier recorded in the received list.
5. Method according to Claim 4, wherein the communication bus (104) is designed according to a variant of the family of the CAN bus, accordingly Controller Area Network bus, and the security measure carried out is the triggering of the bus off state for the station (151, 152, 153).
6. Electronic station (151, 152, 153) for connection to a communication bus (104), wherein the communication bus (104) networks a number of electronic stations (151, 152, 153), wherein the messages transmitted via the communication bus (104) are identified with an identifier, it being stipulated for each station (151, 152, 153) which messages with which identifier are allowed to be sent by the station (151, 152, 153), wherein an unambiguity rule prohibiting another station (152, 153) from sending a payload-data message with an identifier that is already reserved for this station (151) applies, and wherein the station (151, 152, 153) has a protocol unit (18-1) which is designed to keep a list of the identifiers of the messages sent by the station (151, 152, 153), characterized in that the station (151, 152, 153) has a logic unit (18-3) which is designed to send the list kept by the protocol unit (18-1) to a central monitoring station (140) having a superordinate monitoring entity and connected to the communication bus (104) with a message which includes an entry for an identifier of the sending station (151, 152, 153), wherein the superordinate monitoring entity compares the identifiers recorded in the list with a table in which are recorded the identifiers regarding the identification of the stations (151, 152, 153) and reference lists containing the identifiers assigned to each station (151, 152, 153) which are allowed to be sent in messages from the station (151, 152, 153) sending the list, and wherein a transmitting unit of the central monitoring station (140) sends a security message containing a security measure to the station (151, 152, 153) if the associated reference list has no entry for an identifier recorded in the received list, and wherein the logic unit (18-3) is designed to receive the security message from the central monitoring station (140) and to carry out a security measure communicated in the security message.
7. Electronic station (151, 152, 153) according to Claim 6, wherein the communication bus (104) is designed according to a variant of the family of the CAN bus, accordingly Controller Area Network bus, and the security measure carried out is the triggering of the bus off state for the station (151, 152, 153).
8. Central monitoring station (140) for connection to a communication bus (104), wherein the communication bus (104) networks a number of electronic stations (151, 152, 153), wherein the messages transmitted via the communication bus (104) are identified with an identifier, it being stipulated for each station (151, 152, 153) which messages with which identifier are allowed to be sent by the station (151, 152, 153), wherein an unambiguity rule prohibiting another station (152, 153) from sending a payload-data message with an identifier that is already reserved for this station (151) applies, and wherein a protocol unit (18-1) in a station (151, 152, 153) keeps a list of the identifiers of the messages actually sent by the station (151, 152, 153), characterized in that the central monitoring station (140) is designed to receive the list from a station (151, 152, 153) with a message which includes an entry for an identifier of the sending station (151, 152, 153), and in that the central monitoring station (140) includes a superordinate monitoring entity which is designed to compare the identifiers recorded in the list with a table which is present in the central monitoring station (140) and in which are recorded the identifiers regarding the identification of the stations (151, 152, 153) of the communication bus (104) and reference lists containing the identifiers assigned to each station (151, 152, 153) which are allowed to be sent in messages from the station (151, 152, 153) sending the list.
9. Central monitoring station (140) according to Claim 8, wherein the central monitoring station (140) has a transmitting unit which is designed to send a security message containing a security measure to that station (151, 152, 153) whose identifier matches the identifier for a received list in which the associated reference list has no entry for an identifier recorded in the received list.
10. Central monitoring station (140) according to Claim 9, wherein the communication bus (104) is designed according to a variant of the family of the CAN bus, accordingly Controller Area Network bus, and the security measure carried out is the triggering of the bus off state for the station (151, 152, 153).
