GB2255876A - Telecommunications system - Google Patents

Abstract

A modified version of the CCITT ISDN Recommendation I.420 is proposed which provides a transition phase between PSTN and ISDN systems which is referred to as AI.420. This provides a telecommunication system for use with ISDN which is compatible therewith and comprises at the 'S' or 'T" reference point to a central interpreter to translate between ISDN protocols and AI.420 protocols. There is an associated signalling system for use with ISDN and compatible therewith employing AI.420 protocols. <IMAGE>

Description

TELECOMMUNICATIONS SYSTEM The advent of the Integrated Services Digital Network (ISDN) on a worldwide scale will be one of the most significant innovations in the field of telecommunications in the reasonable future. In this instance the term "integrated services" stands for the integration of voice, text, data and image communication in a single network, while "digital" refers to the digitization of the transmission paths and switching equipment. This will afford the network operators considerable savings with regard to cost and space, at the same time as improved transmission quality, even over long distances. For the subscriber there will be greater communication flexibility alongside a wider range of basic and enchanced services.

CCITT described the ISDN in the I-Series recommendations.

Other representative bodies have also concerned themselves with the subject of the ISDN. Some of its characteristics are given here by way of introduction, with special reference to their relevance in integrated services digital private branch exchanges (ISPBXs).

Just like the ISDN for the public networks, the ISPBX for inhouse networks is based on the principle of open communications.

"Open communications" means that within certain technical and operational limits any user can communicate with any other user on a communications terminal of his choice. An ISPBX is therefore essentially a neutral switching system as far as the service is concerned, connecting point A with point B on the basis of a directory number. It is the responsibility of the terminals to ascertain automatically (i.e unnoticed by the user) whether they are compatible and can therefore communicate with each other.

One of the implicit by-products of open communications provided by the ISPBX is portability, i.e. terminals can be unplugged, plugged in again and operated at different locations.

Shifting the intelligence required for the compatibility check from the switching system into the terminals creates the possibility of connecting several terminals in parallel (operating in bus mode) to the basic access point of a subscriber; these terminals make sure that a telephone is indeed only connected with a telephone and not with a telefax machine, for example.

This multi-service capability of the PBX station in turn allows all communications to be handled on a single ISPBX call number. In practical terms this means that over a period of time the telephone numbering scheme will be carried over into the ISDN numbering scheme, while at the same time the separate numbering schemes existing today for telex, teletex, datex, etc., will be dropped.

ISDN/ISPBX provides for both analog connectivity and for fully digital connectivity of digital terminals directly to digital switching equipment. This can be done either by way of basic access interfaces (SO) or by primary rate access interfaces (S2). The interface has the structure of a passive bus with two information channels (B-channels) and a separate signalling channel (D-channel) (2B + D). Up to 8 different terminals can be connected to it.

Primary rate access provides for 30 such information channels and a separate signalling channel (30B + D).

The useable bit rate for basic access is 144 kbit/s, divided between the two B-channels each with 64 kbit/s and the D-channel with 16 kbit/s. This bit stream can be transmitted with special ISDN procedures over the copper wire pairs of the conventional telephone lines.

The bit rate for primary rate access is 2.048 Mbit/s, shared between 30 B-channels each with 64 kbit/s, a D-channel with 64 kbit/s and an internal system channel also with 64 kbit/s.

The high bit rate of the signalling channel with 16 or 64 kbit/s enables message-oriented signalling for call control. This is handled in layer 3. Message-oriented signalling allows the interchange of a variety of parameters. Such signalling is advantageous for the interconnection of ISPBXs and radio paging equipment (RPE).

The D-channel is available throughout the duration of the connection. So signalling information can be interchanged continually between terminals, between terminals and network entities, and between network entities.

Users changing from an analogue PSTN service to ISDN will probably have pre-conceived ideas of the kind of service available.

There are also comments made about the 'S' interface which raise expectations of the functionality which is on offer.

To address the developing ISDN basic access market, it is insufficient merely to provide telephones having full I.420 functionality. The Digital Plan System should interface to the I.420 service, support I.420 non-telephony terminals such as FAX, PC's, etc. but provide the lowest cost solution for telephony access in the first instance. Future plans should also include the ability to provide the lowest cost solution for PC terminal access.

It is vital that the transitional phase is not perceived by users as locking them into a costly product range with only short-term viability. A strategy is needed which carries ISDN users from terminal configurations that are economical before the wide introduction of ISDN basic access to a true I-series terminal environment step by low-cost step. Users should be able to stay at the forefront of ISDN access from a first purchase to a widening spectrum of products.

The concept described derives from a strategy of equipping terminals with the maximum degree of I-series operation compatible with acceptable cost for functionality. I-series operation is introduced from the physical medium up through the layers as far as economically feasible, so that the maximum degree of viable future-proofing is achieved. This is equivalent to implementing selectively those layers with the most stable standards. The concept insulates customers from changes in layer 2 and 3 ISDN standards by incorporating a single protocol termination for a number of terminals. It also minimises the equipment that must differ in different markets in the (far from single-standard) ISDN world, giving economies of scale.

The call-handling capabilities of the various phases of product in the intercept strategy are described.

Accordingly a modified version of 1.420 has been developed that provides the above advantages and for ease of reference is designated AI.420 subsequently. A glossary of terms and abbreviations is included at the end of the description.

According to the present invention there is provided a telecommunication system for use with an Integrated Services Digital Network (ISDN) and compatible therewith, comprising at the 'S' or 'T' reference point a central interpreter to translate between ISDN protocols and AI.420 terminal protocols as herein defined.

There is further provided a telecommunications system as above wherein the basic AI.420 terminal has no in-built intelligence.

There is further provided according to the present invention a signalling system for use with ISDN and compatible therewith employing AI.420 terminal protocols as herein defined.

The signalling system has a Layer 2 frame wherein the information field consists of seven elements, each comprising nine information bits and the necessary forward error correction bits.

The following description will refer to AI.420 terminal protocols and establish the requirements for them. Reference will be made to these protocols as 'Al .420 terminal protocols as herein defined'.

The present invention will now be described by way of example, with reference to the accompanying drawings, in which : Figure 1 shows a diagram of the grouping of ISDN user access functions; Figure 2 shows a Layer 1 Frame format for use with AI.420 and I.420 at Reference Points S and T; Figure 3 shows an I.420 Layer 2 Frame format; Figures 4 and 5 show I.420 Layer 3 message formats; Figure 6 shows a Layer 2 AI.420 Frame format; Figure 7 shows diagrammatically a typical AI.420 telephone; Figure 8 shows the typical hardware structure of an ISDN linecard; Figure 9 shows a first phase central interpreter connection; Figure 10 shows a second phase central interpreter connection; Figure 11 shows a third phase central interpreter connection; Figure 12 shows a fourth phase central interpreter connection.

There follows a description of a proposed signalling system for use in the AI.420 concept. The requirements for such a signalling system are outlined along with the signalling protocols considered.

Initially the signalling protocols used in I.420 are outlined. Various signalling protocols for use in AI.420 are then investigated and compared before a detailed description, at Layers 1 to 3, of the proposed protocol is given.

Layers 1 to 3 of the ISO OSI seven layer model are used to describe the signalling system with the differences between AI.420 and I.420 being highlighted.

The AI.420 Signalling System must be efficient enough to: (i) Communicate display information to the terminal and data about key depressions from the terminal so that the system seems to be responding immediately to the user's input, hence allowing menu driven operation on terminals with a display.

(ii) Permit upgrade to true I.420 terminals. So there must be a high degree of compatibility between the signalling methods used for AI.420 terminals and for I.420 terminals. Compatibility at the physical layer is obviously essential.

(iii) Allow LTU hardware designed for I-series signalling also to be used for AI.420 communications (with AI.420 software).

Hence a degree of compatibility at Layer 2 is obviously desirable.

(The minimum degree of compatibility to achieve this demands the use of the same flags, bit stuffing and inter-frame fill rules to define Layer 2 frame boundaries and the same Frame Check Sequence (FCS) generation rules). The LTU would also be required to use AI.420 software to allow this requirement to be met.

(iv) Support both communications to individual terminals connected in parallel on a line and broadcast communications to all terminals on a line to learn their identities.

(v) Be distinguishable from I.420 signalling. The differences can be put in Layer 2 or Layer 3.

(vi) Keep the observed error rate low enough to avoid confusing or worrying users. At a raw BER of 10-6 (the target for the ISDN), this is taken here to mean noticeable (temporary) errors should occur less often than once per week. Thus for continuous channel'usage (an impractically high level), the BER must be reduced by a factor of at least 104. This must be achieved without introducing large amounts of storage in the terminals or transport delay or software overhead.

Because ISDN is a projected world wide system and is certain to be implemented piecemeal in most, if not all, countries and at different rates in various countries, it is essential that it is defined adequately in standards. This involves defining protocols and interfaces at various levels in the system to provide interworking between units from various manufacturers and also compatibility with existing systems for some considerable time to come.

Additionally, the system will evolve over many years and provision must be made for accommodating this, although its form may be currently unknown.

The body primarily concerned with producing these standards is CCITT, though there is input from other bodies and adoption of national standards where they are appropriate.

An important related concept is the use of the Open Systems Interconnection (OSI) reference model, which provides a framework within which communication protocol standards may be developed. The ISDN protocols fit within this framework.

As a result communications functions are "layered" and in the case of each layer it performs a related subset of the functions to provide services to the next higher layer. As a result in the ideal situation changes in one layer should not require changes in another layer. The resulting OSI reference model has seven layers numbered 1 to 7.

To define the requirements for ISDN user access functions are grouped to provide finite arrangements of items of equipment and reference points are defined between these functional groupings where interfaces are specified.

Functional groupings are : (a) Terminal Equipment Type 1 = TE1 (b) Terminal Equipment Type 2 = TE2 (c) Terminal Adaptor = TA (d) Network Termination 1 = NT1 (e) Network Termination 2 = NT2 and reference points R, S, T and U as shown in Figure 1. The User-Network Interfaces are described in the 1988 CCITT ISDN Recommendations I.411 ISDN User-Network Interfaces - Reference Configurations.

In Figure 2 is shown the Layer 1 frame structure which will be used in both AI.420 and I.420 for communication between NT and TE and between TE and NT which occurs at reference points S and T.

The bits making up the frame can be one of three levels of a pseudo-ternary code. The four bits in each frame used by the signalling system are the 'D' bits. In the direction NT to TE, each 'D' bit received is echoed to the TE's in the next 'E' bit.

Speech or data are carried in two 64 kbps 'B' channels and either I.420 or AI.420 signalling or packetised data traffic can be carried in the bits making up the 16 kbps 'D' channel. In Figure 3 is shown the structure of an I.420 Layer 2 frame.

The Address Field consists of two octets and contains the address of the intended receiver of the frame.

The Control Field consists of one or two octets. It identifies the type of frame and also ensures that frames are kept in the correct sequence at the receiver.

The Frame Check Sequence, FCS, is a two-octet word that is calculated by HDLC/LAPD devices from the other information between the opening and closing flags in a Layer 2 frame. It is calculated before transmission of frames and again on receipt of the frame at the other end. The calculated value is compared to the received value. If they are the same, then the frame is passed to the next stage. If not, error notification is passed to the next stage.

The Information Field consists of an integer number of bytes up to a maximum of 260 octets. It contains the Layer 3 message.

The Flag that delimits the AI.420 frame must be the same as used in I.420 protocols in order to be recognised by HDLC/LAPD controllers. The Flag, '01111110', must be uniquely identifiable on reception, so any data between two flags must not resemble it. The bit-stuffing/destuffing technique described in the CCITT Recommendation Q.921 section 2.6 entitled Transparency, is used to ensure that data between two flags does not contain any group of more than 5 consecutive is.

Everything between the opening and closing flags of the Layer 2 frame will be bit stuffed before transmission and destuffed on reception. This means that a zero will be inserted after any occurrence of five consecutive '1's, before transmission of each frame. After five consecutive ones received, the following zero will be removed. This allows for the recognition of an abort sequence and of the opening and closing flags by the receiver.

In I.420 signalling a Layer 2 frame contains a single Layer 3 message and messages received in error are retransmitted.

The Control Element is used to ensure that all consecutive Layer 3 messages are received in the correct order. Addressing is done at Layer 2, limiting the information content of a frame to a single Layer 3 frame. HDLC/LAPD devices accept the frame, calculate and compare the Frame Check Sequence (FCS), and pass the Address, Control, and Information fields to the controller.

In Figures 3 and 4 is shown the Layer 3 protocol described in CCITT Recommendations Q.931 and Q.932 (I.450-452) and designed to effect the establishment and control of circuit switched and packet switched connections.

Figure 4 shows the message format and Figure 5 the information element structure.

The terminals on the passive bus are assigned a priority.

When a terminal recognises a consecutive number of ones equal to its priority in the 'D' echo channel from the LTU, it may transmit in the 'D' channel to the LTU. Once it has finished transmitting a Layer 2 frame, it drops to the lowest priority, allowing other terminals to transmit. Where two or more terminals have the same priority only one will transmit successfully. The remainder will back off, increasing their priority, and try again once the successful terminal has finished. This method is described in section 6.1.4. of Recommendation I.430.

The AI.420 concept is aimed to make telephones providing ISDN-or-better telephony which meet the physical layer ISDN specifications for basic access, and just enough of the link layer specifications to permit ISDN linecard hardware to carry communications to and from them.

Realising the AI.420 concept means (i) Keeping telephones as simple as possible for their function.

(ii) Ensuring multiple telephones can be connected to one extension cable run.

(iii) Requiring change to no more than software in an ISDN linecard.

To achieve these requirements (i) Storage must be minimised.

(ii) Basic telephones must have no intelligence.

(iii) AI.420 telephones must be able to route speech via either B channel for a call.

(iv) AI.420 telephones must meet I.430 (and derived specs.) and just enough of I.440 (Q.920) (and derived specs.) for their (D-channel only) signalling to pass through an HDLC/LAPD controller.

(v) The signalling must be adequate to communicate, without apparent delay, soft-keys' menus to two telephones in simultaneous use on one passive bus.

Consideration will now be given to the differences between I.420 and AI.420 operation.

At the link layer, the AI.420 signalling framing is identical to that for the ISDN, with the exception that the number of bytes within a frame may optionally be non-integral. This non-integral number of bytes mode of operation permits linecards automatically to distinguish lines with AI.420 telephones from those with I.420 terminals.

An AI.420 Layer 2 Frame Format is shown in Figure 6.

In order to provide an apparently immediate response to key depressions, thereby meeting the first requirement in Section 2, the AI.420 protocol must allow transmission of display information in response to the fatest possible keying rate. A suitable rate is specified by British Telecommunications in their Specification S1626 Issue 1 - Core Purchase Requirements for a Two Piece Telephone. No similar requirement for the American or European markets has yet been found, so the signalling system is designed to meet the above Specification.

This specification states that : 1) A key must be pressed in excess of 30ms for it to be valid.

2) There must then be a further time, in excess of 15ms, before a second key press may be detected.

3) Key releases less than 5ms should be ignored. Therefore each AI.420 terminal on a passive bus must be given the opportunity to transmit a Layer 2 frame containing at least two Layer 3 elements every 45ms. These Layer 3 elements contain terminal identity and key depression information.

The number of Layer 3 elements that can be carried by a Layer 2 frame is limited by the number of terminals on a passive bus.

In frames transmitted by the terminal, the maximum number of elements is seven. This is because each element will be coded into 17 or 13 bit words, depending on the level of forward error correction required, and any number of these must not make an integer number of bytes. Any number between 1 and 7 multiplied by 17 or 13 will give a non-integer number of bytes.

The telephone inserts the frame check sequence (FCS) for transmission, and ignores it on reception. Layer 2 frames are not error protected as units, and no addressing is performed at layer 2. No provision is made for simultaneous connection of AI.420 and I.420 terminals to the same passive bus, so there is no need to code distinguisably the frames for transmission to the two kinds of terminal. The function of layer 2 frames in AI.420 signalling is to get layer 3 information through HDLC/LAPD controllers.

Error control and addressing are carried out at layer 2 on network layer (layer 3) elements individually. From one to seven of these can be carried in any layer 2 frame. There is no other correspondence between layer 3 frames and layer 2 frames. Typically each element consists of seventeen bits. Nine of these contain the information. The remaining eight are used for forward error correction. This provides an adequately secure signalling mechanism requiring very little local storage of information and processing to achieve an adequate throughput for practical purposes. The error correction process is standard and fairly trival for both ASIC's and microprocessors. Of the nine information bits per layer 3 element, one is reserved for frame start, and is accompanied by addressing information.

The layer 3 elements would be used to route and communicate information to and from a variety of peripheral devices. Since I.420 terminals will be used for data provision, there is no requirement for contention resolution within AI.420 terminals, and a simple polling mechanism is used to avoid potential contention between terminals on one line in normal use. The I.420 mechanisms are exploited initially to set up communications with each telephone, which will have a unique serial number accessible to the signalling subsystem. Rather than building the complexity into the hardware, good management of the signalling assignments is needed.

The elements of a typical AI.420 telephone are shown diagrammatically in Figure 7.

The AI.420 telephones will be operable from I.420 linecards with special software. The typical hardware structure of an ISDN line card is illustrated in Figure 8. In many cases, HDLC/LAPD controllers would be provided per line interface circuit, and no packet multiplexer would be present.

Rather than using a single passive bus at the 'S' or 'T' reference point, a central interpreter (CI) would distribute calls to AI.420 terminals (mainly) star-connected to the CI. The terminals would include low-cost, simple telephones which would communicate with the CI in a manner based on the lower layers of the I.420 specification so that only one kind of wiring and CI extension card would serve both these low-cost terminals and I.420 terminals. Each line interface circuit would in fact be capable of connecting any one of 1) An I.420 basic access to an ISDN.

2) Up to 2 of the above simple, cheap terminals.

3) Up to eight I.420 terminals on a passive bus.

The CI would contain elements with variants for territories, development phase, and standards developments. It would be connected to the terminals via I.420-like (AI.420) busses suitable eventually for both AI.420 and I.420 terminals. The AI.420 terminals would not vary with territory, standards variations, or any of the equipment developments described later. The CI would translate between the complex ISDN protocols and the simple AI.420 terminal protocols, and switch traffic between the various terminals and the network. It would provide the terminal users with supplementary services similar to the facilities of a PABX (intercom etc.).

An AI.420 terminal :1) Would be plugged into the same kind of port as an I.420 terminal.

2) Would be simpler than an I.420 terminal.

3) Would give access to 64 kbit/s B channel(s) + fast signalling.

4) Would use both B channels, or two terminals would share two B channels on one ISDN line.

5) Could not be connected to the same bus at the same time as an I.420 terminal.

In all phases it is envisaged that only connection to an ISDN would be supported: No interfaces for PSTN connection would be provided in the CI. As in conventional ISDN scenarios, the range of call types that could be handled would be a function only of the kinds of terminals installed and the call-handling capabilities of the ISDN. Plugging in a terminal should be enough to install it.

In the early years, it will be necessary to support (non PSTN controlled) PSTN call types (e.g. Group 3 Fax) as well as ISDN call types. These call types would be handled as ISDN audio calls.

In the first product phase, the CI would be able to connect calls only to AI.420 terminals.

The phase one CI as shown in Figure 9 would incorporate one I.420 interface and up to four AI.420 interfaces. The I.420 interface would connect to one ISDN basic access (at S or T) and the AI.420 interfaces would connect a maximum of eight AI.420 terminals, equivalent in telephony terms to an analogue 2 + 8 system. The CI would be able to connect calls from any of the interfaces to any of them, with the exception that it would not be able to connect a call from the I.420 interface back to the I.420 interface. No local conferencing would be provided.

The CI would handle only speech calls: Its I.420 port software would be identical with that in an I.420 telephone with the exception that it would be able to connect two calls, rather than one call, at a time.

The CI's AI.420 port software would control the hardware to provide all supervisory indications needed by the AI.420 terminals and would offer them local PABX-like facilities (non-blocking secret intercom, enquiry, return, transfer, shuttle, ring when available, mute, busy indications, calling party identification, diverts, call distribution, group pickup, call barring, night service). Some of these facilities would be mapped to the equivalent ISDN supplementary services to provide wide-area PABX-like operation.

Any I.420 terminals would be parallel-connected at the I.420 interface, and all their call handling would be carried out by the network. Thus, any intercommunication between I.420 and AI.420 terminals would be routed via the ISON.

The phase two CI as shown in Figure 10 would incorporate the same interfaces, with the same interconnection capabilities, as phase one. Each AI.420 interface would be equippable with either up to eight I.420 terminals or up to two AI.420 terminals, but not a mix of the two kinds.

The CI would be able to connect calls to any terminal connected to an AI.420 interface, whether an I.420 or AI.420 terminal. The normal blocking rules would apply to I.420 terminals connected to AI.420 interfaces.

The CI would connect both speech and data calls: Its I.420 port software would be the same as in phase one with the addition that it would relay incoming and outgoing call requests and appropriate priorities between the I.420 interface and all AI.420 interfaces equipped with I.420 terminals, and monitor the messages relayed for call connection and disconnection occurrences.

The CI's AI.420 port hardware would support contention resolution and prioritisation at layer 1. Its AI.420 port software would control the hardware to provide all supervisory indications needed by the AI.420 terminals and would offer all telephones on its AI.420 ports local PABX-like facilities (as detailed for phase 1).

Where the addressed market's ISDN supplementary services allowed, those facilities would be mapped to them to give wide-area PABX-like operation. Facilities to offer data terminals on AI.420 interfaces are for further study.

Any I.420 terminals parallel-connected at the I.420 interface would have their call handling carried out by the ISDN.

In addition to the interfaces of phase two, as shown in Figure 11 the phase three CI would have analogue port interfaces.

It would have the same interconnection capabilities as phase two plus the ability to handle simple calls to and from analogue ports in the same manner as digital audio calls. AI.420 interfaces would be equippable as in phase two.

Supervisory indications and PABX-like facilities would be those in phase two plus the use of end-to-end signalling to effect virtual private network operation (e.g. FLASKS2).

In addition to the interfaces of phase three, the phase four CI as shown in Figure 12 would have additional I.420 and AI.420 interface options up to a maximum of 24 ports.

The CI would have the same interconnection capabilities as phase three. AI.420 interfaces would be equippable as in phase three.

Supervisory indications and PABX-like facilities would be the same as in phase three.

The AI.420 system as proposed has the following implications for the user:1) He is buying an ISDN switch, which is in agreement with emerging market fashion.

2) He does not suffer the high expense of I.420 terminals during the early days of their introduction nor their possible poor availability.

3) He gains access to most ISDN features, and full access to the ISDN network via his ISDN exchange port.

4) Only the CI needs upgrading as the ISDN message set expands, so the great majority of the customer's investment is protected: Instead of several complete system replacements, only a small upgrade cost per change is incurred. The greater the number of AI.420 terminals per basic access, the greater the benefit this gives. It may be possible (depending on software complexity and equipping practice) to support obsolete I.420 terminals by translating protocols in the CI.

5) It will be safer for the user to upgrade to a full I.420 system. He will not have to depend upon the future availability of upgrade linecards from his supplier and the manufacturer. He will not need new linecards, and he can buy I.420 terminals from any source.

6) It is easier to upgrade to full I.420 ISDN than it is with a completely proprietary extension interface. The changing of the hardware and the work in procuring new hardware is avoided.

7) The AI.420 port permits versatile configuring of the user's system as his office environment evolves. - 8) Intercom and PABX type facilities are available.

To summarise the main differences between the ISDN and AI.420 signalling systems are : ISDN 1. Erroneous frames are retransmitted.

AI.420 Forward error correction prevents erroneous frames.

2. Only one Layer 3 message in any Layer 2 frame.

Several Layer 3 messages in a Layer 2 frame. Also a fraction of a Layer 3 message in a Layer 2 frame is possible.

3. Addressing done at Layer 2 limiting the relationship between Layer 2 and Layer 3.

Addressing done at Layer 3 allowing flexibility in the relationship between Layer 2 and Layer 3 frames.

4. To allow retransmission of frames terminals require memory and intelligence and hence the cost is increased.

No retransmission is required and hence terminals can be relatively uncomplicated and inexpensive.

5. Can support only ISDN compatible terminals at all layers.

Can support terminals compatible with Layer 1 of I.420; i.e. ISDN and AI.420 terminals.

6. Terminals will be required in different versions for different countries.

One version of terminal will suit all environments as intelligence in the central interpreter will control communications.

ISDN (cont) 7. Terminals can support either or both speech and data communications.

AI.420 (cont) Terminals can support only speech communication.

Glossary of Terms ASIC Application Specific Integrated Circuit.

ISO International Organisation for Standardisation.

OSI Open Systems Interconnection.

I.420 ISDN Recommendations from the CCITT.

ISDN Integrated Services Digital Network.

ISPBX Integrated Services Private Branch Exchange.

CCITT International Consultative Committee for Telegraphy and Telephony.

AI.420 A modified version of I.420 that meets the CCITT recommendation at Layer 1 and part of Layer 2 only.

LTU Line termination Unit. It may be an ISDN linecard or an ISP.

ISP Integrated Services Plansystem.

FCS Frame Check Sequence. Part of the Layer 2 frame recommended by CCITT.

BER Bit Error Rate.

'D' channel 16 kbps channel used for signalling and packetised data traffic.

'E' bit This 'D' echo bit is a copy of the 'D' bit last received by the LTU. The 'E' bits make up the 'D'-Echo channel.

Glossary of Terms IFF Interframe Fill of at least eight '1's expected between Layer 2 frames by most HDLC devices.

HDLC High level Data Link Controller.

LAPB Link Access Protocol used to handle information in the 'D' channel. Based on LAP Layer 1 The Physical Layer of the ISO OSI seven layer model.

Layer 2 The Data Link Layer of the ISO OSI seven layer model.

Layer 3 The Network Layer of the ISO OSI seven layer model.

Claims (5)

1. A telecommunication system for use with an Integrated Services Digital Network (ISDN) and compatible therewith, comprising at the 'S' or 'T' reference point a central interpreter to translate between ISDN protocols and AI.420 terminal protocols as herein defined.

2. A telecommunications systems as claimed in Claim 1 wherein the basic AI.420 terminal has no in-built intelligence.

3. A signalling system for use with ISDN and compatible therewith employing AI.420 terminal protocols as herein defined.

4. A signalling system as claimed in Claim 3 having a Layer 2 frame wherein the information field consists of seventeen bits, comprising nine information bits and eight forward error correction bits.

5. A telecommunication system substantially as hereinbefore described with reference to and as illustrated in Figures 6, 7 and 9 to 12.