Patent Grant
7352744
1. Field of the Invention
This invention relates to a switched full-duplex Ethernet type communication network, particularly in avionics.
2. Description of the Related Art
The Ethernet network, which is the reference in the world of communication networks, can be used to send data in digital form by packets or “frames”, where a packet is defined as being a set of data sent in a single step on the network.
In an Ethernet network, the data in each packet are not interpreted. The network carries the data without understanding their meaning. A packet is composed of two types of data, network data that are used to route the packet to its correct destination, and useful data which comprise the “useful load” in the packet.
An Ethernet network is composed of different equipment that is subscribed to the network, and connected to each other through a communication means formed of active equipment called switches, which perform three functions:
Operation of such a network is simple. Each network subscriber can send packets in digital form at any time towards one or several other subscribers. When a switch receives the packets, the “network information” data are analyzed to determine the destination equipment. The packets are then switched towards this equipment.
In the “switched full-duplex Ethernet type network” expression:
For example, this network may be a 100 Mbits/s switched full-duplex type network on twisted pair; the term “twisted pair” means that connections between the equipment and the switches are composed of two pairs of cables, each pair being twisted; the term 100 Mbits/s simply means the transmission or reception speed of packets on the network.
The Ethernet technology imposes:
At the present time, in the civil aeronautics field, data exchanges between the various onboard computers are based on the use of the ARINC 429 aeronautical standard.
However, the switched full-duplex Ethernet network is frequently used in industry. The emergence of new communication technologies shows that this type of network is an open and standard solution (IEEE standard 802.3) with a considerable potential for development as a local network. But this type of solution does not provide any means of guaranteeing segregation and transfer performances (in terms of network access, latency, etc.) necessary for avionics applications.
The purpose of this invention is to propose a switched full-duplex Ethernet type network, which guarantees data segregation and a limit to data transfer time, to enable application in avionics.
The invention relates to a switched full-duplex Ethernet type network comprising at least one source subscriber equipment and at least one destination subscriber equipment connected to each other through at least one physical link through at least one switch and through at least one virtual link which is the conceptual representation of a link from a source equipment to at least one destination equipment, characterised in that each source equipment transmitting Ethernet frames comprises:
Advantageously, each destination equipment comprises means of subscribing to at least one virtual link in reception and achieving segregation between virtual links as far as the application.
Advantageously, each switch comprises incoming passband control means for each virtual link. It uses a static configuration table to know which virtual links it needs to switch, and the allowable number of packets for a virtual link. It comprises:
In one example embodiment, each switch comprises the following in sequence:
Advantageously, a virtual link is characterized by:
In one advantageous embodiment, network redundancy is achieved by doubling up the network, each subscriber having a connection to each of the two networks.
The invention also relates to a process for implementation of a switched full-duplex Ethernet type communication network comprising source and destination subscriber equipment connected to each other through at least one physical link through at least one switch and through at least one virtual link that is the conceptual representation of a link from a source equipment to at least one destination equipment, characterized in that in a transmission service, an application is allowed to access virtual links in transmission, this service being used to multiplex virtual links to the physical link through an Ethernet interface, and to send packets for each virtual link as a function of the passband allocated to the virtual link. In a reception service, the packets are decoded, it is checked that their format is correct and useful data are made available to applications. In a passband protection service in the switch, the time characteristics of the packets are checked for each incoming virtual link, and if the allowable characteristics are exceeded, packets are destroyed to prevent a failure at a transmitter or a virtual link from compromising traffic in other virtual links output from this switch.
Advantageously, in one network redundancy service at subscriber level, a packet is sent and received in two virtual links in order to set up network redundancy, duplication of the network that is transparent for applications to guard against a failure in a switch or an interface.
Advantageously, in a “sampling” service, the destination terminal only presents the last received value to the user, and in this service the last value is systematically overwritten by the new received packet. In a “queuing” service, the destination equipment presents all data that it receives to the user, this service making it possible to:
In a “file transfer” service, a data file is transferred and the transmission service breaks this file down into packets which are then transmitted sequentially, and the reception service recreates this file.
Advantageously, a passband and an inter-packet time are assigned for each virtual link.
In one advantageous embodiment, a subscriber in reception refines the selection of packets on the same virtual link using the network addressing information contained in the packet.
Advantageously, data integrity is achieved on each packet by a CRC that carries out a calculation to validate data transmitted in the packet, each packet is verified at each equipment entry on the network, and every bad packet is destroyed so that it is not used in order to release the passband and thus avoid uselessly creating congestion at the switches.
A more complete appreciation of the present invention and many of the attendant advantages thereof will be readily obtained as the same becomes better understood by reference to the following detailed description when considered in connection with the accompanying drawings, wherein:
The switched full-duplex Ethernet network according to the invention uses the virtual link concept to limit the end to end transfer time, in other words from a source equipment to one (or several items of) destination equipment.
This virtual link (VL) concept provides means of isolating data transfers between a source equipment 13 and destination equipment 14. A virtual link VL is seen as a “pipe” on the network, as illustrated in
A virtual link VL is characterized by:
A network subscriber may comprise several virtual links VL1, VL2, VL3, as shown in
When equipment 21 wants to send a packet to equipment 23, 24 and 25, it sends a packet on the virtual link VL1. When it wants to send a packet to equipment 22 and 23, it sends a packet on the virtual link VL2.
The difference between virtual links VL1 and VL2 is identified by the destination identifier in the packet. On the network, the virtual link to which a packet belongs is determined by the identifier of the virtual link in the packet.
A switch uses a static configuration table to determine the virtual links that it is required to switch, and the allowable number of packets for a virtual link.
The virtual link concept is a means of fixing communications between equipment by configuring routes and passbands allocated to the virtual links. Thus, the flow formed by a virtual link is sure to be not disturbed by other flows sharing the same physical links all along its route in the network.
Furthermore, the virtual link concept enables central flow management, to make sure that the sum of the passbands allocated to virtual links on the same physical link does not exceed the capacities of the technology of this physical link. In the above example, the sum of the passbands of virtual links VL1 and VL2 must be less than the transmission capacity of the physical link from equipment 21.
In the network according to the invention, each equipment comprises:
In the network according to the invention, each switch comprises:
Therefore as illustrated in
Thus, as illustrated in
In these transmission and reception services, the virtual link may be represented as being a queue, as seen by the application.
Other protection services are used to guard against some network failures:
The communication interface may supply three additional services to subscriber level applications:
The transmission service breaks it down into packets which are then transmitted sequentially. The reception service reconstructs the file. It is also capable of restarting on error (for example in the case of a data download).
The invention does not only relate to the definition of virtual links that are used in equipment to send and receive data. It also relates to a particular use of the allocated passband, and sub-filtering functions on reception in a virtual link.
1) The Allocated Passband and Inter-packet Time
The allocated passband is defined as being the number of packets sent per second and the size of each packet. But this definition is incomplete because a passband does not give the distribution of these packets in time. It is also necessary to specify the minimum time to be respected between two packets. This minimum inter-packet time (IPT) gives the maximum passband of the virtual link for a given packet size.
The assignment of a passband (PB) and an inter-packet time (IPT) for a virtual link does not mean that the packets will systematically be transmitted on the virtual link every IPT and occupy the entire allocated passband. These packets will only be sent on the virtual link when a subscriber application makes them available to the transmission in this virtual link.
2) The Filter Function in a Virtual Link in Reception
On the same virtual link, a subscriber in reception can refine the selection of packets using network addressing information contained in the packet. This filter mode which uses a virtual sub-links concept gives greater flexibility for the definition and use of virtual links by avoiding the creation of specific virtual links.
The performances of the network according to the invention can be broken down in four ways:
1) Data Integrity
Data integrity is achieved on each packet by a CRC (Cyclic Redundancy Check) that enables a calculation to validate data transmitted in the packet. The CRC is located at the end of the packet and it corresponds to all bits in the packet (network information plus useful information).
In the network a packet is verified at each equipment entry to the network, so that every bad packet can be destroyed so that it cannot be used, in order to release the passband and to avoid unnecessarily overloading the switches.
2) Network Availability
The network is shared by several systems for communications. Therefore its availability has a non-negligible impact on the global availability.
This availability is increased by network redundancy, that consists of doubling up the network, with each subscriber having a connection to each of the two networks, and one of the two packets being selected on reception.
This network redundancy enables operation even if a switch or several links are defective.
3) Network Determinism
The network is a deterministic network. This means that any packet belonging to a virtual link for an allocated passband is sure of accessing the network and being transmitted to receiving equipment for a limited latency time (maximum packet transit time).
For a virtual link, the maximum latency calculation is given by the following formula:
Latency time of a virtual link=TaE+TTE+(NS×TTS)+TTR+TAR
Where
The times TAE and TAR are a result of the asynchronism between the applications and the network in transmission and reception.
A first estimate will give the following maximum values for a packet transfer time, independently of the virtual links:
In the example illustrated in
For the virtual link VL1 from equipment 13 to equipment 14 through two switches 11, if the time IPT=5 ms and the read interval=1 ms, we have:
4) End to End Routing of Data
The network according to the invention creates a single end to end route using virtual links, for each packet.
At each subscriber, each packet is assigned on transmission to a virtual link predefined by configuration. This packet has an addressing part that contains the identification of the virtual link. This identification is used for routing in the network.
In the routing of a packet, a service verifies if a packet is authorized to pass through each switch at which the packet arrives. Similarly, on reception, a service checks if the packet belongs to a virtual link authorized in reception. If not, the packet is destroyed.
In this example embodiment, several characteristics of the network according to the invention have been specified, in order to simplify and standardize the use of this network in avionics.
1) Number of Virtual Links
Three subscriber classes are defined, depending on the number of virtual links in transmission and reception:
This type of distribution is given in table 1 at the end of the description.
2) Passband in Virtual Links
Each virtual link has an allocated passband that is given by the minimum inter-packet time (IPT) and the packet size. The IPT is given by the following formula:
IPT=1 ms×2k where K is an integer from 0 to 7;
which gives either 1 ms, 2 ms, 4 ms, 8 ms, 16 ms, 32 ms, 64 ms, or 128 ms.
By definition, four maximum packet sizes have been defined for each virtual link, namely 16 bytes, 98 bytes, 226 bytes and 482 bytes. This size only indicates useful data for the packet, used by applications directly. Network information data are also taken into account for calculation of the passband.
In table 2 at the end of the description, the effective passbands on the network for a virtual link are given for each packet size (useful load) and for each IPT time.
The required transmission passband for equipment on aircraft is low (<400 kbits/s). Thus, an upper bound in transmission is given which is summarized by the rule that the sum of passbands of the virtual links VL in transmission must be less than 5 Mbits/s. Thus, in transmission we can have:
In order to allocate a passband to a virtual link, a real need has to be identified in terms of data transmission frequency and volume.
Margins on the passband are given in two ways, either by duplication of the virtual link, or by an effective margin on the passband of the virtual link. In the latter case, it is possible to:
3) Virtual Links Mode
Three transfer modes have been defined above, namely “sampling”, “queuing”, and “file transfer”. In order to facilitate the definitions in the virtual links, each virtual link is used either in “sampling” mode or in “queuing” mode or in “file transfer” mode.
4) Filtering
The possibility of filtering in a virtual link enables some equipment to avoid an excessive number of virtual links in transmission and in reception (if the same transmitter). This means can be used to group several virtual links into a single link and to define “virtual sub-links”, the packets being selected by network information contained in the packet. Filtering gives better flexibility on the use of communication “pipes” and enables optimisation of their use.
The term “structure” specifies the arrangement of the useful load (data transmitted and received by applications) in the packet.
The term “format” specifies the type of data (integer, binary, etc.).
The Ethernet standard does not impose the packet structure and the data format (unlike the ARINC 429 standard); however, the following rules are respected in the example considered.
As illustrated in
1) Functional Validity
The transmitter calculates the functional validity, and the receiver or receivers use it to determine the degree of confidence to be assigned to the transmitted packet. The functional validity is coded in the first four bytes of the packet.
The functional validity is the equivalent of “SSMs” used on the ARINC 429 aeronautical standard. However, there are some major differences:
2) Data Format
Data are coded in a predefined format, in order to standardize exchanges. This format is as similar as possible to the “computer” format manipulated by compilers and microprocessors, to minimize format transformations to be made by applications.
The selected formats are defined in the coding rule illustrated in table 3 at the end of the description.
3) Data Grouping
The Ethernet protocol is particularly efficient when transmitted information is relatively long. Therefore, it is very advantageous to group data to that they can be sent all at once to facilitate coherence and integrity.
The data grouping criteria may be as follows:
It is essential to define the subscriber hardware and software architecture, at least from the communication point of view, before carrying out a network design and sizing procedure.
This involves a system and network architecture design phase that must clearly identify:
The systems must take account of tests and if necessary define specific virtual links for transferring data to test instrumentation means.
For each hardware equipment, the following are necessary:
Equipment, even equipment external to the aircraft, or a test installation, etc., may support several subscribers each with one or several logical interfaces to one or several sub-networks.
The following are necessary for each virtual link:
The transmission system is responsible for defining virtual links. The transmission system must make sure that equipment in reception take account of virtual links.
A database supports this part of the process and defines:
This type of stack must enable every subscriber to interface with the communication network.
There are three possible approaches for integration of this function in the subscriber equipment:
Obviously, numerous modifications and variations of the present invention are possible in light of the above teachings. It is therefore to be understood that within the scope of the appended claims, the present invention may be practiced otherwise than as specifically described herein.
| Number | Date | Country | Kind |
|---|---|---|---|
| 01 14263 | Nov 2001 | FR | national |
The present patent document is a continuation of U.S. application Ser. No. 10/287,912 filed on Nov. 4, 2002, and claims priority to French patent application FR 01 14263 filed Nov. 5, 2001, the entire contents of each of which are hereby incorporated herein by reference.
| Number | Name | Date | Kind |
|---|---|---|---|
| 4780869 | Engdahl et al. | Oct 1988 | A |
| 5379278 | Safadi | Jan 1995 | A |
| 6188689 | Katsube et al. | Feb 2001 | B1 |
| 6282669 | Imanaka et al. | Aug 2001 | B1 |
| 6615271 | Lauck et al. | Sep 2003 | B1 |
| 6721273 | Lyon | Apr 2004 | B1 |
| 6766482 | Yip et al. | Jul 2004 | B1 |
| 20020090946 | Mielke et al. | Jul 2002 | A1 |
| 20030152077 | Saint Etienne et al. | Aug 2003 | A1 |
| 20060107108 | Geng et al. | May 2006 | A1 |
| Number | Date | Country | |
|---|---|---|---|
| 20040218554 A1 | Nov 2004 | US |
| Number | Date | Country | |
|---|---|---|---|
| Parent | 10287912 | Nov 2002 | US |
| Child | 10717549 | US |
