J.T.O. Phase II (Switching Specialisation) : AXE-10
Introduction to AXE 1
What is AXE? 1
AXE as viewed by the subscriber 3
AXE as viewed by the telecom administration 5
Flexibility - the be-all and end-all 8
2.0 AXE system structure 8
Processors in the AXE system-basic principles 8
System structure 14
Internal interworking and hardware in APT 21
The digital group switch 34
The digital subscriber stage 44
APZ 211 and APZ 212-control parts of the AXE system 55
The I/O (Input/Output) system in AXE 69
Addressing principles and the operating system 74
Traffic handling 83
Introduction to AXE
What is AXE?
This question may be answered in many different ways. Some would say, “A telephone exchange”, while others might be more specific and say, “A telephone system capable of serving all types of telecom networks- national as well as international”. And many of the answers given would be right.
But if the question reads, “What do the three letters ‘AXE’ stand for?”, there will usually be no answer.
What, then, does “AXE” mean? - The answer is that it is just a three-letter code denoting an Ericsson product.
All products, instruments, tools, etc. made or used by Ericsson are identified by a three-letter code.
The three letters are usually also followed by a number to indicate product variants.
We will discuss this matter in more detail later on in this book, Section 4.2.
Let us now revert to the first question, “What is AXE?”
To be able to give a comprehensive answer we are going to use a comparative example: we will compare an AXE exchange installed today with one of the first AXE exchanges ever installed, that is, the Sodertalje Exchange just south of Stockholm, which was cut over in 1976.
If we could place these two exchanges side by side, we would find that they look quite different. And if we take a closer look, the differences will become even more manifest. The older version uses relay-based technique for some of its functions, whereas relays are very rare in the newer one. The modern exchange features a wide range of facilities for clients to choose among, whereas the old one can offer only a limited number.
Yet both are called AXE. Where is the logic in this?
The answer is as follows: Even though the two versions differ as far as external characteristics are concerned, they are very similar in terms of internal structure because the same system structure has been used. Furthermore, the same type of design aids have been used in designing the two exchanges.
Since this internal structure is in no way dependent on the technology used, the AXE system is sometimes referred to as “future-proof”.
Another ten years from now new technology will be available, resulting perhaps in new AXE versions.
AXE as viewed by the Subscriber
A subscriber will make certain demands on his telephone as well as on the telecom network as a whole. These demands are usually more or less unreasonable:
“My telephone should function at all times, and it is a must that I should always be connected to the number I have dialled”.
Of course, such a demand is excessive, but on the other hand reality is not many steps behind. In most countries, the portion of unsuccessful calls due to technical faults and congestion, can be far below 1 per cent.
Another demand is that a telephone that is out of service should be quickly repaired. In these situations, subscribers will receive better service if the exchange itself can decide whether the telephone or the line is faulty.
These types of demands- together with the demand for quick set-up of connections- have always been made by subscribers.
The introduction of computer-controlled telephone exchanges also meant the introduction of a new concept- SUBSCRIBER FACILITIES. An AXE exchange can be provided with a variety of subscriber facilities, which means that subscribers can be offered better service.
We are now going to take a look at some of the facilities offered and see how they can be used.
Subscriber Facilities in AXE
Wake-up and Reminder Service
The subscriber can dial the hour for automatic wake-up on his telephone.
Call Transfer (“Follow me” or Temporary Call Transfer)
The subscriber can divert calls intended for his number to any other number within a specified area.
A short code replaces a long number or a number used frequently by the subscriber. The capacity is up to 100 numbers per subscriber.
Non-dialled Connection (“Hot Line”)
The subscriber need only lift the handset (receiver) to be connected to a given number, either directly or after, say, 5 seconds. If the subscriber dials a digit during these 5 seconds, he can use his telephone in the usual manner.
The subscriber presses a button to alternate between two calls.
Add-on Conference (Three-party Conference)
Three subscribers can converse with each other simultaneously.
The subscriber hears a weak tone if called by a third party during a conversation in progress. This facility also includes alternation on inquiry.
This facility is available in two variants: diversion on busy and diversion on no reply. A common characteristic of both variants is that diversion takes place to some other number programmed by the subscriber.
These are some of the subscriber facilities offered by the AXE system today.
Future AXE facilities are dealt within section 3.5, ISDN.
AXE as VIEWED by the TELECOM ADMINISTRATION
Who buys an AXE exchange? In most cases the buyers are national telecom administrations, but some countries have private telephone companies- Finland and the USA, for example.
Of course, the buyers also make demands on the telephone systems they are going to purchase.
The administration usually makes a so-called CHOICE OF SYSTEM, which means that it decides to buy a large number of exchanges from one and the same supplier.
In this way, maintenance, spare parts handling, training, etc. will be easier to organize as compared with a purchase comprising various types of exchanges from different suppliers.
Considering the fact that the service life of an exchange is very long, we realize that this kind of decision is a very important one.
It is essential that the administration should choose the “right” system from the beginning.
We will now mention some of the factors that an administration must take into account before adopting a new system.
As readers, you should have these factors in mind when studying the system structure later on in this book.
Does the system include basic functions (coin telephones, private exchange functions, etc.)?
Can the system handle operator-controlled traffic, for instance, to a local exchange?
What other facilities does the system offer? Note that subscriber facilities can be profitable to an administration.
EXAMPLE: The “Call Waiting” facility results in a larger portion of successful calls, thus increasing the number of charged calls as well as the administration’s business earnings.
Will future extensions be costly? (Is “spare capacity” for future extensions available?)
Does the system include concentrators? (Can the administration offer subscriber facilities to subscribers in rural areas?)
Can the system provide the administration with adequate statistical information? Such information constitutes a useful tool when dimensioning the network, which in turn results in a higher grade of service for the subscribers.
Is the system capable of handling digital transmission?
How many alternative routes (number of routes and number of lines per route) can the system handle?
Will the system be able to satisfy present and future demands as regards numbering? (A numbering plan often covers a period of 30-50 years into the future).
Will it be easy to change the numbering of subscriber lines? (A subscriber who moves to a new address within the same exchange area usually wants to keep his old number).
Is the system capable of handling present and future call metering methods?
Does the system incorporate facilities for time-differentiated call metering? (Lower rates in the evening than during office hours).
Can the system handle call metering for coin telephones and special facilities?
Is the system compatible with all existing and planned signalling systems? (For instance, CCITT’s Signalling System No. 7).
Will the system be easy to operate and maintain?
Operation and maintenance activities are performed by personnel who (1) cost money and (2) need training. Reduction in the number of personnel and/or training time will, of course, reduce costs.
Will centralized operation and maintenance be possible? (Unattended local exchanges are supervised from a central point. This means less personnel and lower total cost of operation and maintenance).
Is automatic testing of system equipment provided? (Such testing will facilitate fault tracing, thus reducing repair time).
Is the system easy to communicate with? (Shorter personnel training time).
As we can see, a great many factors influence the purchase of telephone exchanges. Since today’s systems are beginning to reach a very high degree of complexity - a fact which makes them difficult to evaluate - some administrations find it convenient to buy one exchange from each of a number of suppliers. This gives the administration time to evaluate the different systems and to compare them with one another before deciding on one or, perhaps, two systems.
Can we then say that AXE satisfies these requirements? YES, INDEED. Its designers took them into account even at the “drawing board stage”.
Since the development of the system was controlled at all times by the demands made on its performance, the solutions to the problems resulting from these demands form an integral part of the system. Or in other words: there are no temporary solutions in AXE.
FLEXIBILITY- The Be-all and End-all
Does a telephone system have to be flexible? Yes, a telephone system must be flexible from two different points of view.
First, flexibility is a prerequisite when producing and selling the system. It must be possible to use one and the same system in different parts of the world and to satisfy different requirements with regard to system operation.
Second, the system must be flexible for telecom administrations to operate. In this context, it is of particular importance to remember that an exchange cannot just be shut down for extension or repair.
All modifications, repairs or changes must be made while the exchange is in service, and without disturbing the traffic handling. These factors, too, have been taken into consideration when designing the AXE system. Only very extensive changes in the exchange will interfere with the traffic, though still to a very small degree.
AXE SYSTEM STRUCTURE
PROCESSORS in the AXE SYSTEM- BASIC PRINCIPLES
The AXE system is referred to as an SPC system. Here, SPC stands for Stored Program Control, which means that programs stored in a computer control the operation of the exchange. (Note that exchange is used generally to denote either the plant as a whole - i.e. including the means of control employed - or that part of the plant which performs the telephony or switching functions).
All operations to be performed by the exchange are stored in the computer memory. To modify a function we must consequently modify the computer memory.
n SPC Exchange
The memory contains a large number of instructions which tell the computer what to do in different situations. To illustrate this, we may compare an AXE exchange with an old manual exchange.
A manual exchange is controlled by an operator. During the decades immediately before and after the turn of the century this was the most common type of exchange, but even today manual exchanges are used (small company PBXs, hotel PBXs, etc.; PBX = Private Branch Exchange).
igure 2.1.2 shows a manual exchange used in Vasa (Finland) in 1890.
Manual Exchange in 1890
Putting it somewhat simply, we might say that in AXE the operators have been replaced by a powerful computer. The computer memory contains all the information and skills previously possessed by operators.
In those days, “reprogramming” the operator meant telling her how to change her procedures. Thus, to change something in AXE we must reprogram the computer, i.e. modify the list of instructions. There are many other similarities between manual exchanges and AXE.
For instance, what would happen in the manual exchange if the operator was taken ill? It would, of course, “stop”.
To improve the reliability of a manual exchange we may have two operators, one of whom is standby. And this is also a principle used in AXE: the switching equipment is controlled by two computers, one of which is standby. We will revert to this duplication concept later on.
APT and APZ
As has been said, AXE consists of two main parts: switching equipment for switching telephone calls, and a computer for controlling the switching equipment. These two parts have been given designations resembling the AXE letter code. The switching equipment is called APT, and the computer is called APZ.
But not just what we can see and touch in the exchange is called APT. APT also has programs, which are stored in the computer (APZ) but which belong to the exchange (switching) part (APT).
To illustrate this correlation we are going to design a simple system for traffic signals to be used at an intersection, and these signals will be controlled by a computer. Let us assume that we buy a computer consisting of a central processing unit containing the processor and the memory, and that we supplement this computer with a DISPLAY UNIT, a KEYBOARD and a FLOPPY-DISK UNIT. These last three units are known under the collective term of INPUT/OUTPUT DEVICES.
We assemble our computer equipment, connect it, and switch on the power. What will happen?
A beep is heard, and something is printed out on the display. Obviously, the computer already contains some kind of program. And this is called the operating program because it handles the work performed in the computer. What we now have in front of us on the desk corresponds to the APZ part of AXE. Thus, APZ consists of hardware (the computer, the memory, the input/output devices, etc.) and software for handling memories and input/output devices, and for administering the work done by the computer.
We are now going to take a look at the functions to be controlled by our computer.
he traffic signal system will be of modern type, with dug-in sensors for detecting motor-cars. In addition, the traffic signal posts will have buttons to be pressed by pedestrians before crossing the street.
Traffic Signals Controlled by a Computer
To control these traffic signals we must write a program which tells the computer how to act in different situations.
And the program that we write must have certain data to work with.
The data in our program will be, for instance, what the signals indicate at any given moment. The computer must “remember” what the signals indicate to enable the program to work satisfactorily. We provide the computer with two kinds of material: a program and data. The program will not change when the system is started up, but the data will.
We will now compare our traffic signal system with the AXE system and define some common concepts.
The program we have written is intended for a specific application. Hence, as opposed to general programs, this type of program is called Application Program.
Our application program consists of program and data, or Software.
The traffic signals, the sensors, the lines and the program that we have written to control these correspond to APT in AXE.
onsequently, APT in AXE consists of the exchange (printed board assemblies, lines, etc.) and of software stored in the computer (APZ).
APT = Telephony part of AXE
APZ = Control part of AXE
The Two Parts of an AXE Exchange
Let us now take a closer look at the computer that controls the exchange.
TWO TYPES of PROCESSORS
As you will understand, we cannot use a personal computer like the one used in our traffic signal system.
The work to be performed in a telephone exchange can be said to fall into two main groups:
Routine scanning of equipment to detect changes. An example is the checking performed to see if a subscriber has lifted his handset. This is done several times every second.
Complex analyses and diagnostics requiring high computing capacity and large volumes of data. Examples are the selection of outgoing routes or traffic measurements.
These two chief tasks have one thing in common: the importance of the TIME factor.
Here, TIME refers to the moment at which something is done or happens. (When a subscriber lifts his handset he expects to receive dial tone directly - not after, say, 10 seconds).
A computer designed to cope with such time requirements is usually called a real time processor or just processor. The solution is to have two different types of processor to control the system: one Central Processor (CP) and a number of Regional Processors (RP). The RPs assist the CP in performing routine tasks and report important events occurring in the exchange to the CP.
All decisions are made by the central processor.
he Architecture of the Control-system
RP Handles Simple but Frequent Tasks, whereas CP Handles Complex Tasks
This type of configuration permits simple modification of the system capacity by just increasing or decreasing the number of regional processors. This rule applies up to the capacity limit of the central processor.
2.2 SYSTEM STRUCTURE
As we have already seen, the AXE system consists of two main parts: APT, which is the telephony part, and APZ, which is the control part. Both APT and APZ use hardware (printer board assemblies) and software (programs and data). We will now take a closer look at the telephony part, APT, and see what it includes. Later on in this book we will also discuss the control part, APZ.
To facilitate the handling of a system the size of AXE, APT has also been divided into a number of Subsystems.
The division into subsystems is function-related, and below we will briefly discuss some of the many reasons why such a division is necessary.
DESIGN: The responsibility for the design of a subsystem rests with a department or section at Ericsson.
DOCUMENTATION: The fact that the division into sub-systems is function-related facilitates the locating of the documents involved.
SYSTEM DESCRIPTION: Some subsystems are needed only in certain applications. The names of the subsystems included in a particular exchange give a condensed description of the tasks to be performed by the exchange concerned.
The name of a given subsystem reflects the function of that subsystem. Some subsystems contain only software whereas others contain both software and hardware. We will now briefly discuss all the subsystems presently used in APT (the telephony part). Some of them will be studied in more detail later on.
SUBSYSTEMS in APT
TCS, TRAFFIC CONTROL SUBSYSTEM: Only software. TCS is a central part of APT and can be said to replace the operator of a manual system. Examples of the subsystem’s functions are:
Set-up, supervision and clearing of calls.
Selection of outgoing routes.
Analysis of incoming digits.
Storage of subscriber categories.
TSS TRUNK and SIGNALLING SUBSYSTEM: Software and hardware. The subsystem handles the signalling over and the supervision of connections to other exchanges.
GSS GROUP SWITCHING SUBSYSTEM: Software and hardware. GSS sets up, supervises and clears connections through the group switch. Selection of a path through the switch takes place in the software.
OMS OPERATION and MAINTENANCE SUBSYTEMS: Software and hardware. The subsystem contains various functions related to statistics and supervision. OMS is one of the largest subsystems in APT.