WO2020150872A1

WO2020150872A1 - Ethernet and controller area network protocol interconversion for in-vehicle networks

Info

Publication number: WO2020150872A1
Application number: PCT/CN2019/072577
Authority: WO
Inventors: Hsiao-Ying Lin; Mingming Zhang; Yadong Wei
Original assignee: Huawei Technologies Co., Ltd.
Priority date: 2019-01-21
Filing date: 2019-01-21
Publication date: 2020-07-30
Also published as: CN113302885A

Abstract

For in-vehicle networks, a method of performing protocol conversion at the data link layer between Controller Area Networks, CAN, frames and Ethernet frames, based on the value of the CAN identifier such that the transmission priority is expressed in the egress Ethernet frame, and vice versa. Additional methods include switching between output ports for the converted egress frame and /or performing access control, both based on values in data link layer fields of the corresponding ingress frame.

Description

ETHERNET and CoNTROLLER AREA NETWORK PROTOCOL INTERCONVERSION FOR IN-VEHICLE NETWORKS

TECHNICAL FIELD

The present disclosure relates to communication protocols, and in particular to the conversion between different communication protocols, such as those used in networks inside automobiles. In more detail, the present disclosure relates to OSI layer 2 conversion to/from Controller Area Network (abbreviated to CAN) frames and Ethernet frames, wherein priority of transmission information from an ingress (source) frame is expressed in the egress (output) frame after conversion. The present disclosure further relates to switching and access control of ingress frames based on information in the ingress frame.

BACKGROUND

CAN network is a major in-vehicle network currently because of its simplicity, for communication with controllers in vehicles.

Due to high bandwidth requirements triggered by new applications, such as Advanced Driving Assistant Systems (abbreviated to ADAS) or high resolution navigation systems, Automotive Ethernet network is becoming more popular for in-vehicle networks. Automotive Ethernet may be defined with reference to sets of IEEE standards as discussed below.

However, it takes time to make such an in-vehicle networking evolution. Co-existence of CAN and automotive Ethernet is a reality and will last for decades.

Figure 1 shows an example of the co-existence of CAN and Automotive Ethernet in a deployment scenario, such as an in-vehicle network, where CAN networks handles core vehicle functions, and automotive Ethernet supports external connectivity and advanced applications such as On-Board Diagnostics II (OBDI II) , Advanced Driving Assistant Systems (ADAS) , and Human Machine Interface (HMI) . The networks are interconnected by a conversion device. As exemplified, there are two CAN buses; on a first CAN bus a Vehicle Controller Unit, Body Controller and optionally other controllers are connected, whilst a Battery Management Unit, Battery Controller Unit and optionally other controllers are connected on a second CAN bus.

CAN is the main fieldbus for in-vehicle communications. CAN protocol (ISO 11898) defines the physical layer and the data link layer.

CAN frame format, at the data link layer according to the OSI model, is shown in Figure 2. As depicted, the CAN frame format at data link layer has a start of frame (SOF) , an Identifier, a remote transmission request (RTR) code, control area, data area (i.e. payload) , cyclic redundancy check (CRC) , response (ACK) and the end of frame (EOF) .

The CAN identifier (CAN ID) has two formats, 11 bits and 29 bits, and the data area is up to 8 bytes in the CAN standard.

Carmakers, such as Audi or Toyota, define CAN identifiers. For example, they can define CAN identifier 0x010 for CAN frames of engine system. They also define how they interpret the data area. For example, the first byte of the data area means the current speed of CAN frames with identifier 0x010. The switch gateway needs to know at least those CAN IDs it handles. The manufacturers also define how they interpret the data area. For example, the first byte of the data area means the current speed of CAN frames with identifier 0x010.

In the CAN bus communication process, the sender sends a CAN message, and the receiver on the bus selects the CAN message to be received according to the CAN identifier. The message corresponding to one CAN identifier may have more than two receivers. When two CAN frames from two nodes are going to be sent on the bus at the same time, one with a smaller value of CAN identifier gets higher transmission priority. The CAN ID determines the priority on CAN network. For instance, when the identifier value is 0, the CAN frame has the highest priority. CAN ID 0x000 is smaller than CAN ID 0x001. The transmission priority is guaranteed by hardware arbitrator. Hence, CAN identifiers can be seen as priority tags in CAN frames.

Automotive Ethernet is different from legacy Ethernet. It aims at providing more accurate time synchronization and better quality of network service, such as guaranteed maximal latency and jitter. For a specified set of data streams, network resources can also be reserved in advance to ensure transmission quality. IEEE provides an additional set of standards to form Automotive Ethernet, the so-called Audio/Video Bridging (AVB) . A newer version is called AVB generation 2 or Time-Sensitive Networks (TSN) . Table 1 below shows a comparison of legacy Ethernet and Automotive Ethernet mapped to layers of the OSI model.

Table 1

In AVB, to achieve the goals of providing better quality of network services, new protocols and additional headers are defined. IEEE 802.1Qat defines stream reservation protocol (SRP) . It classifies traffic into SR-Class A and SR-Class B. For up to 7 hops, traffic of SR-Class A is guaranteed to have at most 2ms latency while traffic of SR-Class B is guaranteed to have at most 50ms latency. IEEE 802.1Qat also identifies traffic by the stream ID. Traffic of a stream ID can be assigned with reserved network resources, such as reserved bandwidth of a whole transmission path, to guarantee the quality of network services, including low latency and jitter. IEEE 802.1Qav defines scheduling algorithms for legacy Ethernet and automotive Ethernet traffic, respectively.

Figure 3 shows the format of automotive Ethernet frame, where b stands for bit and B stands for byte. The 3 bit User Priority in the Virtual Local Area Network (VLAN) tag gives 8 different values for priority in Ethernet networks. With a higher value in user priority, the frame has higher priority being sent in the network. Automotive Ethernet can use two out of eight priority values, thus giving two possible priority levels. Others are reserved for legacy Ethernet, thus providing for at least six priority levels and possibly eight i.e. 0, 1, 2, 3, …, 7. User priority 0 is smaller than user priority 1 and so on.

Conventional solutions

There is no international or industrial standard for protocol conversion between automotive Ethernet and CAN.

Some existing conventional methods focus on conversion between legacy Ethernet and CAN. They use CAN over Ethernet for protocol conversion as shown in Figure 5. When converting CAN frames to Ethernet, one or more CAN frames are encapsulated into the payload part of Ethernet frame. Aggregating multiple CAN frames into one Ethernet frame can further decrease the transmission bandwidth overhead. When converting Ethernet frames to CAN frames, the corresponding de-capsulation is performed.

Another existing conventional method provides for conversion between automotive Ethernet and CAN which is performed in application layer and one CAN frame maps to one Ethernet frame. Hence the latency introduced at the gateway switch is high and command options concerning transmission of the converted frame are limited.

Yet another existing conventional method provides for CAN-to-AVB protocol conversion. A fixed number of CAN frames are aggregated into one Ethernet frame. However, in this conventional solution, only the scheduling issue to lower the latency introduced by the conversion is addressed.

SUMMARY

Amongst conventional protocol conversion methods, none of them consider preserving the property of the prioritized transmission from the source network frame in the target network frame. Without taking transmission priority into consideration, this causes a network unavailability issue. For instance, when the traffic in Ethernet network is heavy, and there is some urgent information needed to be sent from CAN network to another node in Ethernet network, the converted frame would be delayed by internal scheduling at the gateway switch and bandwidth competition when accessing Ethernet.

Another security concern comes from the asymmetric bandwidths. Ethernet has much higher bandwidth (e.g. at least 100M bps) than CAN buses (500K bps commonly in use) . When there is no rigorous access control at the gateway switch, it is easy for attackers to launch denial-of-service attacks from Ethernet to paralyze CAN buses.

To address one or more of the foregoing issues, embodiments of the present disclosure thus provide methods, apparatus, and computer programs for protocol conversion in one or both directions between CAN frames and Ethernet frames. Embodiments of the present disclosure further provide methods of switching of an egress frame, which is converted from an ingress frame, to defined ports, and also provide methods of access control on ingress frames. The methods are performed at the data link layer.

In one aspect, the present disclosure provides a method of protocol conversion, for vehicles, comprising

- obtaining an ingress Controller Area Networks, CAN, frame comprising a CAN identifier in a frame header of the ingress CAN frame, where a value of the CAN identifier represents a transmission priority,

- Converting, at the data link layer, the ingress CAN frame into an egress Ethernet frame, based on the value of the CAN identifier such that the transmission priority is expressed in the egress Ethernet frame.

As discussed in the Background, transmission priority is performed in a CAN network according to a value of the CAN identifier. The frame bearing the lowest value of CAN identifier results in that frame being preferentially transmitted on that CAN bus when there are two or more frames on that CAN bus at the same time. There are many possible values of CAN identifier, as each may correspond to a target node in the CAN network.

Transmission priority in Ethernet frames may be expressed by a value of the VLAN tag field. Comparatively higher values of the VLAN tag field result in the frame with the higher value of VLAN tag, compared to other Ethernet frames, being preferentially transmitted in the Ethernet network.

The protocol conversion may be performed by a gateway entity or function, which operates at the data link layer (OSI layer 2) and preferably no higher layers. The gateway may have sub-modules which perform the functions such as conversion e.g. a secure mapping controller. The gateway may be provided in hardware or software, and may comprise a logical entity. The gateway may be disposed in a network, such as in a vehicle network, where there is co-existence of an Ethernet network and a CAN network which are interconnected. The gateway may also be a switch, which performs access control and/or interface or port switching functions between the networks based on preconfigured information and parsing at layer 2 of the ingress frame. The term gateway and switch may be used interchangeably or refer to their combination herein when referring to general functions.

The Ethernet network may comprise legacy Ethernet, and/or Automotive Ethernet. Legacy Ethernet is Ethernet which is not Automotive Ethernet.

When comprising legacy Ethernet, the VLAN tag, which is optional in legacy Ethernet, is present. Thus providing transmission priority information.

The CAN network may handle core vehicle functions, such as Vehicle Controller Unt, Body Controller, Battery Management Unit, Battery Controller Unit. The Automotive Ethernet network may provide external connectivity and advanced applications such as Human Machine Interface (HMI) and Advanced Driving Assistant Systems (ADAS) . The skilled person will be aware of other possibilities.

The gateway or switch may have one ingress/egress Ethernet port, and one ingress/egress CAN port. Preferably, there are two ingress/egress CAN ports on the switch. Each CAN port may connect to a different CAN bus for different controllers. The gateway or switch may have input and output buffers connected to each port.

The gateway or switch may obtain an ingress CAN frame by receiving the CAN frame at an ingress CAN port. By converting at the data link layer, the gateway or switch may only parse the frame at the data link layer or below layers.

CAN transceivers /controllers may support CAN2.0A and B standards, and Ethernet transceivers may support both legacy and automotive Ethernet.

The gateway or switch may be pre-configured with information about all possible values of CAN identifier or ways of identifying, classifying or grouping CAN identifiers.

After converting the ingress CAN frame into an Ethernet frame, the switch may output the egress Ethernet frame on a specified egress Ethernet port.

The system assumptions are that CAN transceivers support CAN2.0A and B standards and Ethernet transceivers support both legacy and automotive Ethernet.

The switch or gateway may perform functions of transmitting, parsing, assembling, as known to the skilled person on the frames.

In a possible design, the converting comprises

mapping the value of the CAN identifier into one or more data link layer fields of the egress Ethernet frame, said fields comprising VLAN tag field, Ethernet type field, stream ID field, source address field and destination address field.

The mapping may be according to preconfiguration at the gateway or switch, such as provided by a table, rules or logic.

Given the difference in number of bits for each of the possible target fields, and their possible combinations thereof compared to the fixed size of 11 or 29 bits of CAN ID, the mapping may be performed in various ways. This is possible provided that entities on the target network are aware of the manner of mapping and so can parse the resulting expression of transmission priority.

In a possible design, the converting comprises mapping the value of the CAN identifier into a value of the User Priority field of the VLAN tag field.

Preferably, the value of the CAN identifier is mapped to the VLAN tag field only, taking account of the priority levels available in the type of Ethernet frame.

Similarly, a value of the VLAN tag may be mapped to the CAN ID in the reverse conversion direction.

In a possible design, the converting comprises mapping a smaller value of the CAN identifier to a larger value of the User Priority field.

Due to the larger possible range of values of CAN ID, the mapping of the value of the CAN ID to expression in Ethernet frame header may be many to few or one. For mapping from VLAN ID to CAN ID, the mapping relationship may be few or one to many. When mapping to the User Priority field, at most 8 possible values priority can be expressed.

When user priority is 0 and the VLAN identifier is 0, this may be mapped to a certain value of CAN ID that broadcasts with priority the CAN frame (s) on a particular CAN bus.

In a possible design the mapping all possible values of CAN identifier are divided into two sub groups, and each sub-group is mapped respectively to a different value of the User Priority field.

This may be performed for AVB.

In a possible design the mapping all possible values of CAN identifier are divided into eight sub groups, and each sub-group is mapped respectively to a different value of the User Priority field.

This may be performed for legacy Ethernet.

In a possible design the method further comprises

transferring Data in a Data field of the ingress CAN frame into a payload field of the egress Ethernet frame;

or

encapsulating the ingress CAN frame in the egress Ethernet frame;

or

in the case that the obtaining an ingress CAN frame comprises obtaining first and second ingress CAN frames, each comprising a first and second CAN frame header, wherein the first and second frame headers are the same, the method comprises:

transferring information about the ingress CAN frame into metadata in the egress Ethernet frame, and

transferring Data in a Data field of each of the ingress CAN frames into a payload field of the egress Ethernet frame;

or

in the case that the obtaining an ingress CAN frame comprises obtaining first and second ingress CAN frames, each comprising a first and second CAN identifier respectively having corresponding respective first and second values of the first and second CAN identifiers, wherein the first and second values both map to a same value of a User Priority field, the method comprises:

mapping the CAN identifier to the same value of a User Priority field,

transferring information about the ingress CAN frame into metadata in the egress Ethernet frame; and

encapsulating each ingress CAN frame in the egress Ethernet frame.

The conversion from CAN to Ethernet may be performed in various ways: 1) One CAN frame is encapsulated in one Ethernet frame. 2) Several CAN frames are encapsulated in one Ethernet frames. 3) Several CAN frames with a common header are decomposed and aggregated in one Ethernet frames. A metadata is embedded in the egress Ethernet frame in order to let the gateway switch differentiate different encapsulations.

When transferring Data in a Data field of the ingress CAN frame into a payload field of the egress Ethernet frame the other fields may be dropped or ignored, thus saving payload space. The value of the CAN identifier is mapped as disclosed herein.

Multiple ingress CAN frames can be aggregated in one egress Ethernet frame. This is more efficient. Further efficiencies arise if only the Data is kept. Further, ingress CAN frames which have similar levels of transmission priority importance can be transmitted together thus improving priority receipt in the Ethernet network.

When encapsulating CAN frames in the egress Ethernet frame, each CAN ID is extracted and preserved. Further, the priority information expressed by the CAN IDs is preserved by mapping the CAN IDs to the User priority (and optionally other fields) . Thus all information is kept.

In a possible design information about the ingress CAN frame into metadata in the egress Ethernet frame comprises information about one or each of the ingress CAN frames,

and the metadata comprises one or more of:

the header of an or each ingress CAN frame,

the number of CAN frames partially or wholly represented in the egress Ethernet frame,

a manner of how the CAN frames are comprised in the egress Ethernet frame, and

a timestamp of generation of the egress Ethernet frame.

The manner refers to whether the CAN frame is encapsulated or the relevant field thereof e.g. payload Data is encapsulated only.

The timestamp is for marking the time of the generation of this Ethernet frame. It will give extra information, other than that in the embedded CAN frames. The meta data may be embedded in payload of the Ethernet frame.

In a possible design the ingress or egress Ethernet frame comprises a legacy Ethernet frame having a VLAN tag or an Automotive Ethernet frame.

As used in the present disclosure, Automotive Ethernet may be defined to mean the set of standards for AVB and legacy Ethernet above network layer 2 (data link layer) , plus optionally the special physical layer (100BASE-T1, IEEE 802.3bw) .

In a possible design the method further comprises

- obtaining an ingress Ethernet frame,

- Converting, at the data link layer, the ingress Ethernet frame into an egress CAN frame, based on reference information provided from one or more data link layer fields of the ingress Ethernet frame.

In the Ethernet to CAN conversion, the user priority value may determine the priority, and/or may determine the manner of broadcast on which egress CAN ports.

Conversion may be performed in both ways. For example, a diagnostic routine may result in ingress Ethernet frame comprising CAN frames which are converted and switched to the target buses. Returning CAN frame (s) containing report information from the controllers on the CAN network may be converted into Ethernet frame (s) .

In a possible design the said more data link layer fields of the ingress Ethernet frame comprise Destination Address field, Source Address field, VLAN tag field, Ether-type field and fields of the IEEE 1722 data stream.

The Destination Address field and Source Address field contain MAC addresses.

In a possible design, the fields of the VLAN tag field comprise Tag Protocol ID, User Priority, Canonical Format Indicator, VLAN ID, and /or fields of the IEEE 1722 data stream comprise Header, Stream ID, Timestamp, Gateway Information and Packet Information fields. Some or all or these fields may also be used for switching to egress ports.

In a possible design the converting, at the data link layer, the ingress Ethernet frame into an egress CAN frame comprises

de-encapsulating a or each CAN frame from the ingress Ethernet frame into the or each corresponding egress CAN frame,

or comprises

transferring Data in the ingress Ethernet frame into a Data field of a created egress CAN frame, e.g. the Data is Data which was previously extracted from a CAN frame and stored in the ingress Ethernet frame.

In a possible design the converting at the data link layer comprises converting at the data link layer only. This may refer to without processing at a higher layer in a protocol stack.

In a possible design the method further comprises

- Determining, based on the reference information provided by the one or more data link layer fields of an or each ingress Ethernet frame, a first transmission manner for each egress CAN frame; and/or

- Determining, based on the reference information provided the by one or more data link layer fields of an or each ingress Ethernet frame, and based on a value of a CAN identifier for the or each CAN frame represented in each ingress Ethernet frame, a second transmission manner of each egress CAN frame.

- Determining, based on the reference information provided by the one or more data link layer fields of an or each ingress Ethernet frame and based on value of a stream ID, a third transmission manner for each egress CAN frame

Based on the determination, the switch may perform switching i.e. a transmission manner. Preferably the User Priority is the reference field used together other fields such as with one or more MAC fields, Stream ID. Preconfigured rules table or logic and so on may be used to determine the transmission manner.

Switching may be performed separately to either of access control and protocol conversion.

CAN frames may be represented by only having Data and optionally respective CAN ID in the ingress Ethernet frame.

A stream ID may refer to that in AVB.

In a possible design the first or second transmission manners comprise

unicast transmission of the egress CAN frame from a corresponding egress CAN port,

broadcast transmission from corresponding egress CAN ports, and

multicast transmission from corresponding egress CAN ports;

wherein each CAN port is respectively for a different CAN bus.

Broadcast or multicast utilizes frame duplication by the switch. Multicast uses three ports. Broadcast or unicast uses at least two ports. Two CAN ports are a common configuration in automotive deployments.

In a possible design the method further comprises

- Determining, based on the value of the CAN identifier of the or each ingress CAN frame, a third transmission manner for each corresponding egress ethernet frame;

- Wherein the third transmission manner is one of:

o unicast transmission of the egress ethernet frame from a corresponding egress ethernet port,

o broadcast transmission from corresponding egress ethernet ports, and

o multicast transmission from corresponding egress ethernet ports.

Ports may be connected to different buses or entities.

In a possible design the method further comprises

- Performing access control for an ingress CAN frame or an ingress Ethernet frame based on information at the data link layer of the respective ingress CAN frame or ingress Ethernet frame.

Access control may employ white and /or black list filtering. If an ingress frame is blocked no further action is taken. Access control may be performed separately to either of switching and protocol conversion.

If access control is passed, protocol conversion and optionally switching may then be performed.

Access control improves the security of the vehicle network to attack. Given that vehicles may be cars, bikes, planes, trains, buses and so on moving at speed in a congested environment, reducing success of malicious attacks is beneficial.

In a possible design the information at the data link layer comprises the CAN identifier of the ingress CAN frame, or frame header information of the ingress Ethernet frame.

Frame header information may in particular comprise MAC address information.

In another aspect, the present disclosure provides a network gateway, for connection to an ingress/egress Ethernet port and ingress/egress CAN ports in a vehicle, wherein the gateway is configured to perform any method as disclosed herein.

In another aspect, the present disclosure provides a processor comprising a computation unit and memory, wherein the memory stores instructions for instructing the computation unit to perform the method as disclosed herein.

In another aspect the present disclosure provides a method for switching to a target CAN bus, one or more CAN frames expressed in an ingress Ethernet frame, for in-vehicle networks, comprising

- Determining, based on a reference information provided by the one or more data link layer fields of the ingress Ethernet frame, a first transmission manner for each egress CAN frame;

and/or

- Determining, based on the reference information provided the by one or more data link layer fields of the ingress Ethernet frame, and based on a value of a CAN identifier for the or each CAN frame represented in the ingress Ethernet frame, a second transmission manner of each egress CAN frame; and/or

- Determining, based on the reference information provided by the one or more data link layer fields of the ingress Ethernet frame and based on value of a stream ID, a third transmission manner for each egress CAN frame

- Transmitting the or each egress CAN frame according to the determined transmission manner.

A CAN frame is expressed by being represented either in whole by encapsulation or in part, as disclosed herein. This method may be adapted according to the other methods as disclosed herein, for example to provide conversion whilst maintaining transmission priority, or access control on ingress frames.

Compared with the conventional solutions, in the solutions provided in the present disclosure, protocol conversion can be efficiently performed without traversing the whole protocol stack and maintain transmission priority information. Network security of the vehicle is improved, and frames after conversion can be delivered to targeted buses in the target network thus reducing message congestion and resource wastage.

BRIEF DESCRIPTION OF DRAWINGS

FIG. 1a is an illustration of the co-existence of CAN and Automotive Ethernet in a deployment scenario such as an in-vehicle network according to the conventional art.

FIG. 1b is an schematic illustration of interconnected CAN and Automotive Ethernet networks in a vehicle according to the conventional art.

FIG. 2 is a schematic of CAN frame format at the data link layer.

FIG. 3 is a schematic of automotive Ethernet frame at the data link layer.

FIG. 4 is a comparison of layers of the OSI model, TCP/IP model and automotive Ethernet.

FIG. 5 is a flowchart of an embodiment of the method according to the present disclosure.

FIG. 6 is a flowchart of another embodiment of the method according to the present disclosure.

FIG. 7a is an illustration of a conversion module for converting between Ethernet and CAN frame format according to an embodiment of the present disclosure.

FIG. 7b is an illustration of a gateway switch for converting and switching between Ethernet and CAN frame format according to an embodiment of the present disclosure.

FIG. 8 is an exploded schematic of an automotive Ethernet frame at the data link layer illustrating reference fields.

FIG. s 9a-c are illustrations of different ways of converting CAN frames to Ethernet frames according to embodiments of the present disclosure.

FIG. s 10a-b are illustrations of different ways of performing access control between Ethernet and CAN networks according to embodiments of the present disclosure.

DESCRIPTION OF EMBODIMENTS

To make objectives, technical solutions, and advantages of embodiments of the present invention clearer, the following clearly and completely describes the technical solutions in the embodiments of the present invention with reference to accompanying drawings in the embodiments of the present invention. Apparently, the described embodiments are some but not all of the embodiments of the present invention. All other embodiments obtained by a person of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

Below is a brief explanation of FIG. s1-3, as these are discussed in more detail in the Background section.

FIG. 1 shows a conversion device connecting an Automotive Ethernet network 30 and a

CAN network

23, 27. The conversion device 10 has functionality according to the conventional art as disclosed herein.

FIG 1b illustrates Automotive Ethernet 30 and CAN

networks

23, 27 in a vehicle 40 interconnected by a conversion device 10 according to the conventional art.

FIG. 2 illustrates a typical CAN frame 50 at the data link layer, comprising various fields including the CAN identifier field 60 and Data field 70.

FIG. 3 illustrates the format of automotive Ethernet frame 80. The 3 bit User Priority field 100 in the Virtual Local Area Network (VLAN) tag 90 gives 8 different values for priority in Ethernet networks. In legacy Ethernet frames the VLAN tag 90 is optional.

From the comparison shown in FIG. 4 it can be seen that CAN standards ISO 11898-1 and ISO 11898-2, 3 map to the physical and data link layers. In Ethernet, 100 BASE-T1 in IEEE 802.3bw-2015 maps to the physical layer. AVB is shown as mapping to several layers. Of particular relevance in the present disclosure are the layers of AVB which map to the data link layer, including IEEE 1722 Layer-2 AVB Transport Protocol, IEEE 802.1AS Precision Time Protocol, IEEE 802.1Qav and IEEE 802.1 Qat.

FIG. 5 illustrates actions which may be performed in a first embodiment of the method according to the present disclosure. In this first embodiment, optional actions are illustrated by dotted lines.

In general, the method acts upon an ingress CAN frame, and produces an egress Ethernet frame by converting the ingress CAN frame into the egress Ethernet frame. As disclosed herein, the method of this first embodiment may be adapted to be performed on multiple ingress CAN frames. In the case of multiple ingress CAN frames, the ingress CAN frames are represented in the egress Ethernet frame by extracting relevant information such as the CAN ID (which is converted and optionally also kept) and data fields, or extracting and converting the CAN ID and encapsulating each ingress CAN frame..

In the step of obtaining 500 an ingress CAN frame comprising a CAN identifier in a frame header of the ingress CAN frame, where a value of the CAN identifier represents a transmission priority, a corresponding CAN identifier in the ingress CAN frame is obtained.

Optionally, step 510 of performing access control for the ingress CAN frame occurs next in the method. The ingress CAN frame may be parsed according to the fields in the CAN header, such as CAN ID, and according to a white list or blacklist. If on the white list, the CAN frame may be converted and /or switched. If on the black list, the CAN frame is not for the target network and is not converted nor switched.

Then an action of converting 520, at the data link layer, is performed. The ingress CAN frame is converted into an egress Ethernet frame, based on the value of the CAN identifier such that a transmission priority is expressed in the egress Ethernet frame. The value of the obtained CAN identifier is converted into one or more header fields at the data link layer in the corresponding egress Ethernet frame. The conversion is performed according to a mapping. The CAN identifier value may be mapped to the VLAN tag, and in particular to the User priority field therein. A principle of mapping may be that a lower value of the CAN ID maps to a higher value in the User Priority tag. Many CAN IDs may be mapped to one value of the VLAN tag or User Priority field. For example, one range of CAN IDs maps to one VLAN value, one CAN ID maps to another VLAN value, and another range of CAN IDs maps to another VLAN value. Other permutations and combinations of the mapping are possible.

In the converting 520, the payload is also converted. Figures 9a-9c depict three different ways of converting ingress CAN frames.

Then, the converted Ethernet frame is output.

Optionally, the transmission manner of the converted Ethernet frame is determined 530. This may be according to a mapping, such that certain CAN ID 60 are mapped to certain corresponding egress Ethernet ports.

Further implementation details of the method steps are provided in the Summary.

The method may alternatively, or in conjunction for bidirectional conversion, be performed on an ingress Ethernet frame and output one or more egress CAN frames. The ingress Ethernet frame may include representations or encapsulations of one or more CAN frames. In this case, the or each CAN frame is generated or de-encapsulated to become egress CAN frame (s) .

FIG. 6 illustrates actions which may be performed in a second embodiment of the method according to the present disclosure. In this second embodiment, optional actions are illustrated by dotted lines.

The method acts upon an ingress Ethernet frame, and produces an egress CAN frame by converting the ingress Ethernet frame into the egress CAN frame.

Multiple CAN frames can be represented in the ingress Ethernet frame. For example, there may be two or more CAN frames wholly encapsulated in the ingress Ethernet frame. In another example, only the data fields of several CAN frames may be stored in the ingress Ethernet frame. Optionally, their CAN IDs may also be stored in the ingress Ethernet frame. Otherwise, their CAN ID is expressed in the Ethernet field header as disclosed herein. Optionally, metadata as disclosed herein may be included in the Ethernet frame.

In the step of obtaining 600 an ingress Ethernet frame comprising a VLAN identifier in a frame header of the ingress Ethernet frame, where a value of the VLAN tag represents a transmission priority, the value of the VLAN tag is obtained. The value of the VLAN tag, such as the User Priority, may also indicate via a mapping or rules which CAN ports to switch the represented CAN frames to.

Optionally, step of performing access control 610 for the ingress Ethernet frame occurs next in the method of the second embodiment. The ingress Ethernet frame may be parsed according to the fields in the Ethernet header such as VLAN tag and/or source address and/or destination address, according to a white list or blacklist. If on the white list, the Ethernet frame may be converted and /or switched. If on the blacklist, the Ethernet frame is not for the target network and is not converted nor switched. The target network is thus more secure.

Then an action of converting 620, at the data link layer, is performed. The ingress Ethernet frame is converted into one or more CAN frames corresponding respectively to the CAN frames represented in the ingress Ethernet frame. This may comprise recreating the CAN frames according to their data and CAN IDs, or decapsulating the CAN frames. Optionally, the converting may be based on the value of the VLAN tag such that a transmission priority is expressed in the egress CAN frames via the CAN ID of each CAN frame. The conversion is performed according to a mapping. The conversion mapping operations may be the same as those in the first embodiment.

Next, the transmission manner of the converted CAN frame (s) is determined 630. The transmission manner of this embodiment may be applied to transmission manners in other embodiments as disclosed herein. The transmission manner may be according to a mapping, such that the represented CAN frames are mapped to certain corresponding egress CAN ports. Mapping rules may be preconfigured. The mapping be according to reference information in the layer 2 header of the ingress Ethernet frame, and also optionally according to information in the represented CAN frames. The reference information may one or more of the fields identified in Figure 8. In an example, two fields of the VLAN tag field of the ingress Ethernet frame are used, comprising the VLAN ID and the User Priority tag. For example, when the value of VLAN ID is zero and the User Priority value is zero, then the converted CAN frames are broadcast to all CAN buses.

The order of the determining 630 and the converting 620 can be interchanged or occur in parallel.

Then, the converted Ethernet frame is output.

Further implementation details of the method steps are provided in the Summary.

FIG. s 7a and 7b illustrate both ingress and egress frames 190, 210 and the corresponding respective egress frames 190, 210 converted therefrom. The conversion may be in one or both directions. In one embodiment, the conversion is performed by a conversion unit such as a CAN –Ethernet gateway switch 110 or a secure mapping controller 120 inside the gateway switch 110. The gateway switch is configured to perform the method of any embodiment as disclosed herein.

In FIG. 7a, one Ethernet bus is in communication with the conversion unit via data buffer 130. The conversion unit is in communication via

respective buffers

140, 140 to two CAN ingress/

egress ports

150, 160 which are for respective different CAN buses.

Incoming Ethernet frames can be stored by a buffer 130 in communication with an ingress/

egress Ethernet port

170, 180. In this embodiment, ingress/egress refers to a port which is bidirectional. In the case of one directional conversion, the port may be unidirectional and is thus ingress or egress only.

The Ethernet frames 190 include representations 200 of CAN frames. Two representation manners are shown: in a first manner, one CAN frame is comprised in an Ethernet frame, whereas in a second manner, two or more CAN frames are comprised in the Ethernet frame 190.

When the Ethernet frames are ingress Ethernet frames, the conversion unit converts the ingress Ethernet frames into egress CAN frames as disclosed in embodiments herein. Switching between the two buses may also be performed as disclosed herein.

When the CAN frames are ingress CAN frames, the conversion unit converts the ingress CAN frames into an egress Ethernet frame or frames, as disclosed in embodiments herein.

In FIG. 7b, two Ethernet buses are in communication with the gateway switch 110 via

ports

170, 180 and

data buffers

130, 130. The gateway switch is in communication via

respective buffers

140, 140 to two CAN ingress/

egress ports

150, 160 which are for respective different CAN buses.

The secure mapping controller 120 performs conversion and /or switching between the ingress packets and egress packets according to preconfigured conversion rules and preconfigured switching rules.

FIG. 7b is a simplified representation compared to Figure 7a, in that multiple frames on one bus are not shown, nor is encapsulation of CAN frames. However, both can be used with this embodiment.

FIG. 8 illustrates the layer 2 fields of Ethernet frame header which can be used as reference information, as disclosed herein. FIG. 8 illustrates both automotive Ethernet and legacy Ethernet fields. The fields include: Destination Adress, Source Address, VLAN tag, Ether-type, IEEE 1722 data stream. The VLAN tag field includes Tag Protocol, User Priority, Canonical Format Indicator, VLAN ID fields. The IEEE 1722 data stream field includes Header, Stream ID, Timestamp, Gateway Information, Packet Information. The Payload of IEEE 1722 data stream field may not be included as one of the reference fields.

FIG. s 9a-c illustrate three different manners of converting payload parts or all of ingress CAN frames into egress Ethernet frame.

The first way is one to one mapping, as shown in FIG. 9a, where one CAN frame 50 having a data field 70 is mapped to one Ethernet frame 220. In this way, the metadata 250 contains a timestamp inserted by the gateway switch and a description including how many CAN frames are embedded in this Ethernet frame and how they are embedded. In this case the metadata would indicate only one. The timestamp is for marking the time of the generation of this Ethernet frame. It will give extra information, other than about the embedded CAN frames. The number of CAN frames is an indicator for the gateway switch to differentiate the encapsulation type.

The second way, as shown in FIG. 9b, is to embed multiple CAN frames 50, 50 in one Ethernet frame 230. When more than one CAN frame is encapsulated, any overhead from including the metadata is correspondingly lower on average.

The third way, as shown in FIG. 9c, is for multiple CAN frames which have the same frame header. Multiple CAN payloads 70 are embedded in one Ethernet frame 240 while the identical or similar CAN frame headers of the ingress CAN frames are contained in the metadata. Each ingress CAN frame may have the same CAN ID. The third way saves overhead compared to the first and second ways.

FIG. 10a illustrates a simple one to one transmission manner performed by the inter network gateway switch 110, as disclosed herein. Switch rules decide which information from which interface flows to which interface (s) . The gateway switch may also handle access control. For each

interface

150, 160 170, 180 at the gateway switch, there is a filter based on layer 2 information. A filter can be a whitelist, a blacklist, a mixture of two lists, or a more flexible setting, such as passing the traffic to processes in higher layers.

The secure mapping controller 120 of the gateway switch 110 can use at least two different switch rules in determining the switching manner. The first one is one to one. When the ingress frame is allowed to flow to another interface i.e. passes access control, the secure mapping controller makes the conversion and then puts the converted frame to a transmission buffer of the target interface. The second switch rule is one to many as shown Figure 10b, in either direction. It can be multicast or broadcast. After the secure mapping controller makes the protocol conversion, the resulted frames are then broadcasted or multicasted.

To further illustrate applications of the method and gateway switch as disclosed herein, four further embodiments are provided.

A third embodiment relates to a diagnostic service. A diagnostic request is applied on a vehicle for the status of the whole vehicle. A command activating diagnostic process is launched into the vehicle. This command should be sent across Ethernet and CAN networks to all diagnosed controllers in the vehicle. Status information from those controllers should be transmitted back to the source sending the command.

In this third embodiment, the transmission priority is not critically high. Rather, the embodiment’s purpose is that the command should be broadcasted or multi-casted to the controllers being queried.

Table 2 below is an example about how to determine how to transmit this command from Ethernet to CAN networks.

The command, in the form of an Ethernet frame bearing a representative CAN frame, enters the gateway switch from a first Ethernet interface, and then is converted into CAN frames.

Those CAN frames are then sent to a first CAN interface and a second CAN interface, such as shown in the left part of FIG. 10b.

Table 2

In more detail, when Ethernet frames arrive at Ethernet interface 1, the gateway switch analyzes the header part and if the user priority value is 3 and the CAN ID of the embedded CAN frame is 0x201, it de-encapsulates the CAN frame out and broadcast to all CAN buses. Here the priority 3 shows a commonly used priority level for AVB traffic. The CAN ID value is merely an example, and can be omitted.

To parse the CAN ID from the embedded CAN frame in the Ethernet frame, the gateway switch refers to the payload part of Ethernet frame.

Status information are collected back from the multiple controllers with different CAN IDs. To further save the bandwidth in Ethernet, those CAN frames with different CAN IDs are aggregated in fewer Ethernet frames. Moreover, to provide a better response quality in terms of stability, resulting Ethernet frames are assigned the same stream ID, such that stream reservation protocol can be applied on this stream ID. Examples of rules for aggregating CAN frames and giving a stream ID are shown in Table 3.

Table 3

In the switch rules, CAN frames with certain CAN IDs are switched to Ethernet interface 1. The Ethernet interface 2 and so on might not be used. The CAN frames are aggregated into Ethernet frames with a stream ID A.

A fourth embodiment relates to a diagnostic request applied on a vehicle for the status of a specific controller for a certain time period. A command activating diagnostic process is launched into a vehicle. This command is sent across Ethernet and CAN networks to a diagnosed controller. Hence, it only involves one-to-one mapping and unicasting, as shown in Table 4.

Table 4

In the return direction (CAN to Ethernet) , status information from the diagnosed controller is transmitted back to the source. The CAN frames bearing the status information have the same CAN IDs. Hence, to further save the bandwidth in Ethernet, those CAN frames are aggregated by using their payloads, i.e. the returning CAN frames are not wholly encapsulated. Their common (i.e. identical) header will be embedded in the metadata part in resulting egress Ethernet frames. Examples of rules of aggregation are shown in Table 5.

Table 5

A fifth embodiment relates to software over the air update. That is bulk data entering a vehicle such as for a controller. The update package is downloading from the service provider into in-vehicle Ethernet network. The controller being updated is in the CAN network. To ensure a stable transmission from Ethernet to CAN, the update package transmission uses stream reservation protocol where each Ethernet frame contains multiple CAN frames and a stream ID. Table 6 illustrates the applicable actions.

Table 6

A sixth embodiment relates to access control. For each interface of the gateway switch, there is a filter or a set of filters. Table 7 provides an example of the filtering actions taken.

Table 7

The foregoing disclosure merely discloses exemplary embodiments, and is not intended to limit the protection scope of the present invention. It will be appreciated by those skilled in the art that the foregoing embodiments and all or some of other embodiments and modifications which may be derived based on the scope of claims of the present invention will of course fall within the scope of the present disclosure.

Claims

A method of protocol conversion, for vehicles, comprising

- obtaining an ingress Controller Area Networks, CAN, frame comprising a CAN identifier in a frame header of the ingress CAN frame, where a value of the CAN identifier represents a transmission priority,

- Converting, at the data link layer, the ingress CAN frame into an egress Ethernet frame, based on the value of the CAN identifier such that a transmission priority is expressed in the egress Ethernet frame.
The method according to claim 1, wherein the converting comprises

mapping the value of the CAN identifier into one or more data link layer fields of the egress Ethernet frame, said fields comprising VLAN tag field, Ethernet type field, stream ID field, source address field and destination address field.
The method according to claim 2, wherein the converting comprises mapping the value of the CAN identifier into a value of the User Priority field of the VLAN tag field.
The method according to any of claims 1-2, wherein the converting comprises mapping a smaller value of the CAN identifier to a larger value of the User Priority field.
The method according to any of claims 2-4, wherein in the mapping all possible values of CAN identifier are divided into two sub groups, and each sub-group is mapped respectively to a different value of the User Priority field.
The method according to any of claims 2-5, wherein in the mapping all possible values of CAN identifier are divided into eight sub groups, and each sub-group is mapped respectively to a different value of the User Priority field.
The method according to any preceding claim, further comprising

transferring Data in a Data field of the ingress CAN frame into a payload field of the egress Ethernet frame;

or

encapsulating the ingress CAN frame in the egress Ethernet frame;

or

in the case that the obtaining an ingress CAN frame comprises obtaining first and second ingress CAN frames, each comprising a first and second CAN frame header, wherein the first and second frame headers are the same, the method comprises:

transferring information about the ingress CAN frame into metadata in the egress Ethernet frame, and

transferring Data in a Data field of each of the ingress CAN frames into a payload field of the egress Ethernet frame;

or

in the case that the obtaining an ingress CAN frame comprises obtaining first and second ingress CAN frames, each comprising a first and second CAN identifier respectively having corresponding respective first and second values of the first and second CAN identifiers, wherein the first and second values both map to a same value of a User Priority field, the method comprises:

mapping the CAN identifier to the same value of a User Priority field,

transferring information about the ingress CAN frame into metadata in the egress Ethernet frame; and

encapsulating each ingress CAN frame in the egress Ethernet frame.
The method according to claim 7, wherein the information about the ingress CAN frame into metadata in the egress Ethernet frame comprises information about one or each of the ingress CAN frames,

and the metadata comprises one or more of:

the header of ingress CAN frame,

the number of CAN frames represented in the egress Ethernet frame,

a manner of how the CAN frames are comprised in the egress Ethernet frame, and

a timestamp of generation of the egress Ethernet frame.
The method according to any preceding claim, wherein an ingress or egress Ethernet frame comprises a legacy Ethernet frame having a VLAN tag or an Automotive Ethernet frame.
The method according to any preceding claim, further comprising

- obtaining an ingress Ethernet frame,

- Converting, at the data link layer, the ingress Ethernet frame into an egress CAN frame, based on reference information provided from one or more data link layer fields of the ingress Ethernet frame.
The method according to claim 10, wherein the said more data link layer fields of the ingress Ethernet frame comprise Destination Address field, Source Address field, VLAN tag field, Ether-type field and fields of the IEEE 1722 data stream.
The method according to claim 11, wherein the fields of the VLAN tag field comprise Tag Protocol ID, User Priority, Canonical Format Indicator, VLAN ID, and /or fields of the IEEE 1722 data stream comprise Header, Stream ID, Timestamp, Gateway Information and Packet Information fields.
The method according to any of claims 10-12, wherein the converting, at the data link layer, the ingress Ethernet frame into an egress CAN frame comprises

de-encapsulating a or each CAN frame from the ingress Ethernet frame into the corresponding egress CAN frame,

or comprises

transferring Data in the ingress Ethernet frame into a Data field of the egress CAN frame, e.g. the Data is Data which was previously extracted from a CAN frame and stored in the ingress Ethernet frame.
The method according to any preceding claim, wherein the converting at the data link layer comprises converting at the data link layer only, such as without processing at a higher layer in a protocol stack.
The method according to any preceding claim, further comprising

- Determining, based on the reference information provided by the one or more data link layer fields of an or each ingress Ethernet frame, a first transmission manner for each egress CAN frame; and/or

- Determining, based on the reference information provided by the one or more data link layer fields of an or each ingress Ethernet frame, and based on a value of a CAN identifier for the or each CAN frame represented in each ingress Ethernet frame, a second transmission manner of each egress CAN frame.

- Determining, based on the reference information provided by the one or more data link layer fields of an or each ingress Ethernet frame and based on value of a stream ID, a third transmission manner for each egress CAN frame.
The method according to claim 15, wherein the first or second transmission manners comprise

unicast transmission of the egress CAN frame from a corresponding egress CAN port,

broadcast transmission from corresponding egress CAN ports, and

multicast transmission from corresponding egress CAN ports;

wherein each CAN port is respectively for a different CAN bus.
The method according to any preceding claim, further comprising

- Determining, based on the value of the CAN identifier of the or each ingress CAN frame, a third transmission manner for each corresponding egress ethernet frame;

- Wherein the third transmission manner is one of:

○ unicast transmission of the egress ethernet frame from a corresponding egress ethernet port,

○ broadcast transmission from corresponding egress ethernet ports, and

○ multicast transmission from corresponding egress ethernet ports.
The method according to any preceding claim, further comprising

- Performing access control for an ingress CAN frame or an ingress Ethernet frame based on information at the data link layer of the respective ingress CAN frame or ingress Ethernet frame.
The method according to claim 18, wherein the information at the data link layer comprises the CAN identifier of the ingress CAN frame, or frame header information of the ingress Ethernet frame.
A network gateway, for connection to an ingress/egress Ethernet port and ingress/egress CAN ports in a vehicle, wherein the gateway is configured to perform the method of any one of the preceding claims.
A processor comprising a computation unit and memory, wherein the memory stores instructions for instructing the computation unit to perform the method of any one of the preceding claims.
A method for switching to a target CAN bus, one or more CAN frames expressed in an ingress Ethernet frame, for in-vehicle networks, comprising

- Determining, based on a reference information provided by the one or more data link layer fields of the ingress Ethernet frame, a first transmission manner for each egress CAN frame;

and/or

- Determining, based on the reference information provided the by one or more data link layer fields of the ingress Ethernet frame, and based on a value of a CAN identifier for the or each CAN frame represented in the ingress Ethernet frame, a second transmission manner of each egress CAN frame; and/or

- Determining, based on the reference information provided by the one or more data link layer fields of the ingress Ethernet frame and based on value of a stream ID, a third transmission manner for each egress CAN frame

- Transmitting the or each egress CAN frame according to the determined transmission manner.