AU740861B2 - Internet closed user group - Google Patents

Description

WO 98/45995 PCT/AU98/00235 INTERNET CLOSED USER GROUP Technical Field This specification relates to a method and arrangement for providing closed user group (CUG) communication over a network via which Internet services are provided, while enabling high speed communication to the subscribers.

Our co-pending patent application, entitled "High Speed Internet Access", discloses a method and arrangement of providing high speed data services to a large number of users via a communications network.

Background Art The provision of mass mar<et Internet services presents traffic management problems for both telecommunications carriers and Internet Service Providers.

At present most private users are connected to the Internet over modems which operate in the voice band. These services have low transmission rates and also tie up the exchange connection preventing simultaneous use of the phone on the same line and loading the exchange.

High speed access technologies have emerged which make the delivery of fast data and data-like services practical for the mass/broad market. Such technologies include transmission technologies such as ADSL and Hybrid Fibre Coax and multiplexing/Switching technologies such as ATM.

A number of IP based access services have been implemented and often use ATM as an underlying transport technique. However in these implementations, the carriage of services is usually based on the Internet Protocol and so the underlying ATM flows must be terminated at each location at which IP traffic is processed.

Therefore in IP based access networks which use ATM as the underlying transport, the advantages of using end to end ATM are foregone.

For an ATM end to end based access service, it has been necessary to establish individual ATM connections carrying AAL5 encapsulated data across the entire network between each subscriber and their ISP(s) of choice. In a mass market deployment of ATM technology, for example high speed Internet access based on ATM to residential subscribers, the number of ATM flows can amount to 10 Os to 1 00s of thousands simultaneously.

A service provider, such as an Internet Service Provider, may need to terminate SUBSTITUTE SHEET (RULE 26) WO 98/45995 PCT/AU98/00235 2 an individual information flow from each of its subscribers. This requirement could be imposed for service reasons, billing reasons, security reasons or routing reasons.

Typical large ISPs must design their systems to cope with up to hundreds of thousands of subscribers.

In an ATM connected scenario, at least two issues arise for service providers: ATM interface cards for switches and workstations terminate a finite maximum number of simultaneous ATM connections. Typically, this maximum number of connections is of the order of 1000 or fewer. This maximum number of ATM connections is significantly fewer than the number of subscribers to a large ISP.

The cost of leased public ATM links is generally relatively expensive compared with the competitive price charged for a single residential inernet service. In order to remain viable, ISPs offering high speed internet access over ATM would therefore require each leased ATM link into the public network to be shared by a relatively large number of residential subscribers. Certainly, the number of subscribers sharing a 155 Mbit/s link (for example) would need to be much greater than 1000.

Routing in the access network via traditional means is unsuitable in a multi service provider environment or in an environment where the service provider is different from the access network provider. This is because "routing" is a service which provides a value added connectivity and could potentially lead to traffic nominally served by a particular service provider to bypass or be routed around that service provider altogether.

Other issues arise in an ATM connected scenario which affect both access carriers and service providers: The logistical and technical difficulty of managing individual end to end ATM connections for each subscriber is significant.

Disclosure of the Invention Our above-mentioned co-pending application discloses an architecture capable of delivering Internet and other data services to hundreds of thousands of subscribers.

In our preferred solution the Internet subscribers are linked into a packet network using ATM in an end to end manner. In a specific embodiment the ATM link between gateways and subscriber is implemented using ADSL over the copper pair. ADSL operates outside the voice band so it is possible to operate a conventional phone and SUBSTITUTE SHEET (RULE 26) WO 98/45995 PCT/AU98/00235 3 the Internet connection simultaneously. An access gateway is used to terminate the ATM connections and map the traffic using multiplexing into a single connection for the ISP.

At the service provider interface, traffic arrives from a large number of subscribers multiplexed into a much smaller number of ATM flows. Traffic directed from the service provider towards the subscribers is multiplexed by the service provider into the appropriate flow towards that subscriber's access gateway.

A description of the ATM cell and header structure is given in the book "Asynchronous Transfer Mode-Solution for Broadband ISDN", Martin de Prycker; Prentice Hall, 3rd ed, 1995, at page 63 et seq. Essentially the information field is relatively small to minimize buffer size, and the header is used only to identify the link through the next stage, which contains a translation table to replace the header on the outgoing cells for use in the subsequent stage. ATM operates in a connection-oriented mode where the virtual connection is identified by the ATM header and the translation tables.

In a preferred arrangement, the connection between the user and the exchange utilizes an ATM protocol which may, for example, be implemented over an ADSL or other suitable high speed link.

The data may be managed at the AAL5 layer level (ATM Adaptation Layer 5) at the user and exchange interface.

DISCLOSURE OF THE INVENTION The present invention provides for CUG capabilities while, at the same time providing the high speed advantages of our co-pending patent application.

The invention will be described with reference to the following drawings. The term "service provider" will be used to refer to both ISPs and quasi-ISPs such as corporations providing telecommuting facilities for their employees.

Brief Description of the Drawings.

Figure 1 shows an end to end architecture for a mass market Internet access system as described in the above co-pending application; Figure 2 illustrates the process of transferring data from the subscriber to the ISP.

Figure 3 illustrates the implementation of closed user groups (CUG) over SUBSTITUTE SHEET (RULE 26) WO 98/45995 PCT/AU98/00235 4 the network in accordance with an embodiment of the invention.

Figure 4 illustrates an arrangement for associating a subscriber with a single service provider in accordance with an embodiment of the invention.

Figure 1 is a representation of end to end Internet access described in the above co-pending application. A plurality of subscribers 11 1 S or RIM (Remote Integrated Multiplexers) (also known as Metropolitan Area Networks) are connected to the host access exchange 5 via a data access network 4. A plurality of access exchanges, 51 5E are interconnected through data network 6. A plurality of Internet Service Providers (ISP) or other types of service provider, 81 8P, are each connected between the Internet, or other data infrastructure 9, and the data network 6.

Each subscriber may have one or more computers, 10, and modems, 1 1.

The data access network may include ADSL links over copper pairs which originate at a local exchange or Remote Integrated Multiplexer (RIM) connected to the rest of the telecommunications network via high capacity links such as optical fibres.

RIM equipment serves as a point of termination of the ADSL connection which can be located sufficiently close to the subscribers to enable ADSL to be used over the installed copper pair base at information speeds compatible with high speed data services.

Other high speed links to subscribers may also be used, SDH, Hybrid Fibre Coax etc.

Information to be sent to the ISP by the subscriber is initially assembled in packet form and then converted to an efficient form for transmission.

Preferably, communication from the subscriber is carried using ATM, packets for transmission being divided into ATM cells.

However, in the telecommunications network, a large number of end to end ATM paths between individual subscribers and their ISP's may exhaust resources in the ATM switches and be unwieldy to manage and maintain.

The ATM flows from each subscriber are terminated and remultiplexed into simpler flows to the service providers close to the edge of the network. This may be at the local exchange which is connected to a number of RIMs so as to gather sufficient SUBSTITUTE SHEET (RULE 26) WO 98/45995 PCT/AU98/00235 Internet traffic to provide a worthwhile load for packet transmission. Alternatively, where there is sufficient traffic at the RIM, the subscriber ATM flows may be terminated there. The remultiplexed packets are forwarded through the network at the packet layer. The originating VP/VC identifier may be used as the basis for multiplexing.

As shown in Figure 2, packets, 34, to be sent from the user are converted into ATM format using AAL5 encapsulation and ATM VP/VC addresses, using an ATM card in the user's PC or data modem, which breaks up the packet into ATM cells, 351 35 V, and adds cell headers, 36 1 36 V, for transmission to the exchange network in ATM format. The packet level source and destination addresses are transmitted with other packet header information.

At the user exchange interface, the ATM user's flows are terminated and the data assembled into packets 37 including the original packet source and destination addresses, for onward transmission through the network. Specifically for each packet which is processed 37,38, reference to the incoming VP/VC is maintained 38 and this reference can be used to simplify onward forwarding. In this way, the ATM flow from the subscriber is terminated but the reference to VP/VC is not lost.

The VP/VC reference associates each ATM flow on the subscriber's line with a previously assigned ISP via a look up table.

At the exchange/ISP interface, the VP/VC packet flows from many individual subscribers are multiplexed into shared service provider VP/VC packet flows.

According to AAL5 protocol, each packet is sent as a contiguous flow of uninterrupted ATM cells such that the cells of one packet are not intermixed with the cells of another on the same ISP VP/VC. At the ISP, ATM cells are terminated and the data reassembled into packets, 40. At the Exchange therefore, subscriber's incoming packet is directed towards an ISP on the basis of its VP/VC identifier rather than on the basis of its packet source or destination address.

Description of the Invention It is desirable to enable closed user groups (CUGs) to operate over the network, but it is important to ensure the integrity of the closed group by ensuring that CUG traffic remains within the CUG. This specification discloses a method and arrangement of providing CUG access in a public data service environment in a more efficient manner.

SUBSTITUTE SHEET (RULE 26) WO 98/45995 PCT/AU98/00235 6 A problem which arises when providing closed user group (CUG) access in a public data service arrangement where additional protocol overhead is added, in the form of VP/VC identifiers to segregate the information is that, in large networks with many users, the efficiency of utilization of the network is reduced because the additional protocol information must be exchanged for each packet.

This specification therefore describes an arrangement in which the source and/or destination addresses resident in each frame are used to identify closed user group membership in such a way as to enable the integrity of closed user groups to be maintained over the network.

A first embodiment of the invention comprises a data communication arrangement including a network, the network including a plurality of gateways, a plurality of stations being coupled to each gateway, and a plurality of servers, each server being coupled to a gateway, there being at least one gateway to which at least two servers are coupled; each station being associated with one server, whereby header information of messages transmitted by each station identifies the server with which the transmitting station is associated; the network including first multiplexing means at a close to the edge of the network to multiplex messages from two or more stations directed to servers coupled to the same gateway, the mnultiplexing being performed on the basis of the header information identifying the server.

In a second embodiment of the invention the multiplexing means are located at the gateways.

In a third embodiment of the invention each gateway to which stations are coupled contains a look-up table associating each server with the gateway to which the server is coupled.

In a fourth embodiment of the invention each gateway to which two or more servers are connected includes first demultiplexing means to separate messages addressed to specific servers coupled to the gateway from received multiplexed flows of messages.

In a fifth embodiment of the invention each gateway to which a server is coupled contains a look-up table associating each station with the gateway to which SUBSTITUTE SHEET (RULE 26) WO 98/45995 PCT/AU98/00235 7 the station is coupled.

In a sixth embodiment of the invention the network includes second multiplexing means to multiplex messages addressed to stations coupled to the same gateway.

In a seventh embodiment of the invention wherein the second multiplexing means are located at gateways to which servers are coupled.

In an eighth embodiment of the invention each gateway to which stations are coupled includes second de-multiplexing means to separate messages addressed to specific stations coupled to the gateway from received multiplexed flows of messages.

The invention also provides a method a method of communicating between a plurality of stations and a plurality of servers via a network, wherein two or more servers are coupled to a specific node of the network, the method including: multiplexing into a unified data flow traffic from two or more stations addressed to one or more servers coupled to said specific node for transmission through the network to the specific node; and at the specific node, de-multiplexing the traffic from the two or more stations and distributing the traffic to the intended servers.

The invention enables implementation of the transmission of different CUGs in the same logical pipes without additional protocol or connection identifiers to specifically identify the CUG membership of each frame. The source address, destination address or a combination thereof which are naturally resident in each frame can be used to identify the CUG membership of the frame at the remote end of the connection.

The frame protocol(s) which are allowed to be communicated through the shared CUG connection must be restricted to known protocols. For example, all traffic from all CUGs communicated through the shared link should be "native IP".

Each originator or destination must have a unique address within the frame address space. If the chosen protocol is IP for example, each originator and destination must have a unique IP address within the domain of the shared CUG connection.

Prior to multiplexing the traffic from different CUGs together into the one logical connection, security verification of the frame addresses must be performed to SUBSTITUTE SHEET (RULE 26) WO 98/45995 PCT/AU98/00235 8 prevent users of one CUG spoofing users of another.

The invention will be described with particular reference to Figure 3.

Figure 3 shows a plurality of subscribers (Subscriber 1 Subscriber n) (51,52) transmitting corresponding messages MSG1 to MSGn, (53,54) to a gateway 55. Each message includes source (S1,Sn) and destination (D1,Dn) information, and an associated payload.

Message MSG1 and message MSGn have destinations D1 and Dn respectively which are associated with Server Q and Server W respectively. Both the Servers Q and W are connected to the same network gateway, e.g. switch 57. The invention makes use of the fact that, because both messages must be sent to the same gateway 57, the messages can be multiplexed together, while maintaining the integrity of the individual messages, and routed as a single packet via logical link 56 through the network to gateway 57, and, at gateway 57, they are de-multiplexed and routed to their intended servers Q and W.

To achieve this, gateway 55 includes means to associate messages addressed to servers attached to the same gateway, so they can be combined into one packet or data frame. The gateway 55 includes, a look-up table which indicates to which gateway the servers are attached. The gateway 55 is thus able to determine from the destination addresses D1 and Dn or the identity of the transmitting stations, if each of the stations is associated with only one server, that messages MSG1 and MSGn are intended for servers connected to the same gateway 57. Thus gateway 55 multiplexes MSG1 and MSGn for transmission to gateway 57. At gateway 57, MSG1 and MSGn are de-multiplexed and forwarded to their respective servers Q and W on the basis of destination addresses DI and Dn.

For downstream messages from servers Q and W, gateway 57 is provided with means such as a table which associates the subscriber address with a unique gateway, thus enabling messages from the servers to different subscribers connected to the same gateway to be multiplexed at gateway 57 and transmitted to the subscribers' gateway, gateway 55, where they are de-multiplexed and forwarded to the appropriate subscribers.

Thus both upstream and downstream messages can be segregated without the need for additional protocol information.

SUBSTITUTE SHEET (RULE 26)

Claims (11)

1. A data communication arrangement to provide closed user group capability across a network, the network including a plurality of gateways, a plurality of stations being coupled individually to associated gateways, and a plurality of servers, each server being coupled to a gateway, there being at least one gateway to which one or more servers are coupled; each station being associated with one server, whereby header information of messages transmitted by each station identifies the server with which the transmitting station is associated, the header information including source and/or destination addresses which identify membership of a closed user group; the network including first multiplexing means associated with a first node close to the edge of the network to multiplex messages from one or more stations S directed to servers coupled to the same gateway associated with a second node .into frames for transmission across the network from the first node to the second 5* node without termination of the frames at intermediate nodes, the multiplexing being performed on the basis of the header information identifying the server to S. which the corresponding message is directed whereby the integrity of closed user groups is maintained across the network.

2. An arrangement as claimed in claim 1, wherein said first multiplexing e 20 means are located at the gateways.

3. An arrangement as claimed in claim 1 or claim 2, wherein each gateway to which stations are coupled contains a look-up table associating each server with the gateway to which the server is coupled.

4. An arrangement as claimed in any one of claims 1 to 3, wherein each gateway to which two or more servers are connected includes first demultiplexing means to separate messages addressed to specific servers coupled to the gateway from received multiplexed flows of messages.

An arrangement as claimed in any one of claims 1 to 4, wherein each gateway to which a server is coupled contains a look-up table associating each station with the gateway to which the station is coupled.

6- An arrangement as claimed in any one of claims 1 to 5, wherein the netr includes second multiplexing means to multiplex messages addressed to station*, oupled to the same gateway.

7. An arrangement as claimed in claim 6, wherein the second multiplexing means are located at gateways to which servers are coupled.

8. An arrangement as claimed in any one of claims 5 to 7, wherein each gateway to which stations are coupled includes second de-multiplexing means to separate messages addressed to specific stations coupled to the gateway from received multiplexed flows of messages.

9. A data communication arrangement substantially as herein described with reference to the accompanying drawings.

A method of communicating between a plurality of stations and a plurality of servers via a network, wherein two or more servers are coupled to a specific node, referred to as the destination node, of the network; the network including a plurality of gateways, a plurality of stations being coupled individually to associated gateways, and a plurality of servers, each •*..server being coupled to a gateway, there being at least one gateway to which one 15 or more servers are coupled; each station being associated with one server, whereby header information of messages transmitted by each station identifies the server with which the transmitting station is associated, the header information including source and/or destination addresses which identify membership of a closed user group; the network including first multiplexing means associated with a first node close to the edge of the network to multiplex messages from one or more stations directed to servers coupled to the same gateway associated with the destination node into frames for transmission across the network from the first node to the destination node, the multiplexing being performed on the basis of the header information identifying the server to which the corresponding message is directed whereby the integrity of closed user groups is maintained across the network; the method including: multiplexing into a unified data flow traffic from two or more stations addressed to one or more servers coupled to said destination node for transmission through the network to the destination node without termination of the data flow at intermediate nodes; and at the destination node, de-multiplexing the traffic from the two or more S stations and distributing the traffic to the intended servers.

11. A method of communication substantially as herein described with reference to the accompanying drawings. DATED THIS TWENTY NINTH DAY OF AUGUST 2001 ALCATEL by its patent attorneys Freehills Carter Smith Beadle oo *o o