Configuration Guide

Group 1:	Parameters defining the choice and the decision to use a protocol "p" on a certain line (to configure on C1R1P0 to 35).
	For each user line the additional parameters allow to configure the functional options offered.
Group 2:	Parameters defining the X.25 Interface of the user line facing the X.25 node, represented by the internal switch of the FastPad. This switch manages the virtual circuit between the network line and the user line.
	These are the network parameters of level (OSI) 3 of the X.25 recommendation.

With this organization, all network services can be offered to each line, independent of their protocol.

Figure 3-1: Configuration Parameter Groups

General Definition of Parameters

Configuration Parameters for the Protocol

These parameters define the operation mode of the following levels:

On the physical level; parameters such as throughput, interface signalling, CRC, LRC, parity...etc.
On data link level: frame delimiter, flow control, frame length...etc.

Configuration Parameters for the Virtual Circuit

These parameters define the level 3 interface facing the internal switch. The following parameters are distinguished:

Parameters used for the establishment of the virtual circuit; they define the additional network facilities as well as the processing of the X.121 addresses.

Example: CUG, throughput class, reversed charging...

Parameters used during the transfer of the data enclosed in the X.25 packets.

Example: Packet length, window size...

Localizing the Parameters in the Configuration File

Configuration Parameters for the Protocol (first ground)

They are put together for all protocols in Cl2Rx and the following classes, as a function of the protocol.

X.25	C12	X.25 parameters
Asynchronous	C7	Abbreviated calling table
	C8	Automatic calling table
	C13	Parameter X.3
X.25 PVC	C17	PVC table
	C8	Automatic calling table
X.25 PSTN	C13	Extension parameters
	C20	Modem initialization table
	C21	XID frame table
	C22	X.25/PSTN conversion table
SDLC	C8	Automatic calling table
	C19	Clusters
VIP	C8	Automatic calling table
	C13	Station address
	C19	Controllers
BSC 2780/3780	C8	Automatic calling table
	C13	Station address
	C19	Terminal description
BSC 3270	C8	Automatic calling table
	C13	Station Address
	C19	Controller description
HDLC and Frame-Relay	C8	Automatic calling table
Frame-Relay (subscriber) (FRA)/FRCE	C17	DLCI table
Frame-Relay/FRTE,FRSE	C32	Frame relay PLL couple
Transit type	C30	Dynamic X.25 parameters on network FR or B or D channels.
Network (FRTE)	C30	Dynamic X.25 parameters on network FR or B or D channels.
D channel (ISDN)	C22	X.121-PSTN/ISDN conversion table
B channels (ISDN)	C30	Dynamic X.25 parameters on network FR or B or D channels.
PLL in D channel (ISDN)	C13	PLL/TEI/profile number relations (defined In C30).
	C30	Dynamic X.25 parameters on network FR or B or D channels.
	C22	X.121-PSTN/ISDN conversion table.
MLP	C25	MLP bundle
Virtual IP router	C24	Parameters of IP router connection
Connection	C17	LLC table
	C8	Automatic calling table
SNMP, IP Routing	C31	SNMP and IP router

Configuration Parameters for the Virtual Circuit (second group)

They are put together in C12Rx for each protocol and in C18 when a CUG has been allocated to the X.25 call. See Connection Profiles and Parameters chapter of this manual.

Profiles

To facilitate the operations, the operator has a number of profiles available. A profile is a predefined set of parameters of the groups 1 and 2.

General Principle of the Configuration of Network Services

Definition and Format of the Addresses

The address fields, present in the call packet. are used to identify the two communicating parties. They are connected to each other when the X.25 virtual circuit is established.

The "calling" address field identifies the originator of the call.
The "called" address field contains the identity and the location of the addressee.

The presence of the fields depends on the type of line. It is configured in C12RxP45:

Lines between two switches: A line between two switches is called a network line (C12RxP44, Chapter 4 of the configuration manual).

: The calling and called addresses are always present in the X.25 call.

User lines: generally called a subscriber line, it can be configured by parameter Cl2RxP44 and the format of the addresses in the call packet can be chosen (1 or 2 addresses).

When one address has been selected the type of address changes with the origin of the call:

Call from subscriber received by the FastPad: only the called address is supplied by the subscriber.
Call from the FastPad transmitted to a subscriber: only the calling address is supplied to the subscriber.

Processing the Addresses

Address Conversion

The FastPad address format is derived from the network address and it is structured in three hierarchical levels (see "General routing principle" of this chapter).

To adapt the external subscriber format to the internal network format, there is an internal address conversion mechanism; it is available for each line (subscriber and network).

Two tables are available. These tables are configurable for each line and they can be accessed via C9R11 and C9R12.

This conversion takes place according to the value of parameter C12RxP89.

There is an additional feature introduced when the address to convert is not configured in the tables; the call will be refused or the call will be returned in its original format. The choice of the behavior depends on the parameter C12RxP89.

1 - Conversion on Incoming Call

Table C9R11P0 to P199 adapts the external address to the internal address.

The different possibilities are:

Conversion of the calling address.
Conversion of the called address.
Conversion of both addresses.

2 - Conversion on Outgoing Call

Table C9R12P0 to P199 adapts the internal addressing to the external addressing plan.

The different possibilities are here also:

Conversion of the calling address.
Conversion of the called address.
Conversion of both addresses.

Management of Aimed Points

1. Description

: The management of "aimed points" is a function enabling the use of flexible addressing in a FastPad network.
: This function allows the organization of routing and the adaptation of any addressing plans to the FastPad plan (in DNIC ZO AB form).

2. "Aimed Point" Address

The "aimed point" address is an address, in internal FastPad format, of a region of a FastPad network. The "aimed point" is thus either a DNIC, a DNIC ZO or a DNIC ZO AB. This address is used internally to determine the output line of the local switch.

The "aimed point" may be purely logical (i.e. not correspond to the address of the distant equipment.

Routing then takes place in two phases:

The switch examines the called address of the call request packet and determines, by means of a correspondence table, an "aimed point" address (Address conversion table C9R11),
The "aimed point" address is converted into an output line by means of the known DNIC, ZO, AB tables and the call routed to the calculated port: (DNIC, ZO, AB routing tables).

3. Configuration of "Aimed Point"

Called Address

· C12R3P89		=	64
			Calculation of aimed point.
· C9R11	PY	=	567AAAAAA	Initial address + absorption
	Py + 1		9000	Calculated address of aimed point
· C9RO	Px	=	9000	Known DNIC
· C9R1	Px	=	1, 1, 0, 2	Routing on 1 line, 1 line, fixed 0 to line 2.

Thus, in this example:

Figure 3-2: Example

4. Constraints and Limits

The number of aimed points is limited to 100 per equipment.
It is recommended that, on the same mode, address conversion should not be used on incoming calls on some lines and translation to aimed point on other lines.

To route the alarms to a Network Management System, the operating functions must be given a subscription to the "aimed point" translation service (C24R0).
Conversion on incoming calls is not possible.

Intelligent Network Access

The purpose of this service is to enable a private network composed of several subassemblies, and knowing only the addresses belonging to its own numbering plan, to access a remote subassembly via the "Transpac" network. What is involved is the provision of an end-to-end service between users of a private network via "Transpac." In a transparent manner for the users (in particular as concerns their numbering plan).

A configuration parameter, by line, allows the backup at the input of the switch (or the reproduction at the output of the switch) of the following parameters in an address extension facility:

Private network calling party address
Private network called party address
CUG
Indication of reverse charging (RC)

A second configuration parameter, by line, allows the determination of the behavior to be adopted on the X.25 fields at the input of the switch.

The processing resulting there from may include:

The caller's address
The called party's address
The facilities.

and the possible actions may be:

Deletion
Modification
Addition at the output of the FastPad

In practice, to allow access to the intelligent network of Transpac, the functional steps are:

Backup in an address extension facility of the addresses, CUG and RC.
Replacement of the called party's address by the intelligent network address of the intelligent network server of Transpac or insertion at the output of the switch of an unknown facility so that Transpac can route onto the intelligent network while detecting an unknown facility.
Delete the calling party's address.

The addresses of the calling party, the called party (and possibly the CUG and RC) are reproduced on any subscriber output line to the private network management.

Applications:

Figure 3-3: Diagram

Addresses a and b are in the private network addressing plan.

Addresses A, B, X, Y are the addresses of the ports.

(0) -	The call-request packet sent on the "private network management" subscriber line has the following form:
	@ calling: a	@ called: b

During the processing of the incoming call, the addresses are modified after backup in the address extension facility

	@ calling: A	@called:	@IR	(Intelligent Network Server address)
	CUG and RC according to the configuration.
(1) -	The call transmitted to Transpac is the following:
	@ calling: A	@called:	@IR

Address extension:

@ calling: a

@ called: b

CUG and RC according to the configuration

(2) -	The call transmittal by Transpac is the following:
	@ calling: XA	@ called: B

Address extension:

	@ calling: XA	@ called: B
	CUG and RC according to the configuration
(3) -	On the "private network management' subscriber line there is a reproduction of the addresses, of the CUG and the RC (according to configuration)
	@ calling: a	@ called: b

Private network management configuration:

By line, two connection parameters are defined:

parameter 111 gives the type of private network management. The coding of this parameter is as follows:
- bit 0 called address: yes/no (1/0)
- bit 1 calling address: yes/no (1/0)
- bit 2 CUG: yes/no (1/0)
- bit 3 RC: yes/no (1/0)
- bits 4 to 7 reserved

The values used are:

called party/calling party address: 3
called party/calling party address and internal CUG: 7
called party/calling party address, internal CUG and RC: 15

By default, this parameter is equal to 0 (no processing).

Parameter 111 consequently makes it possible, at the input of the switch, to back up in the "address extension" DTE facility the addresses and even the CUG and RC. This backup takes place before any actions that may be requested by the parameter 112. At the output, this same parameter allows the reproduction of these data from the address extension facility.

Parameter 112 defines the behavior to be adopted. This behavior corresponds to a row in class 26. Class 26 "Table of X.25 field modifications" has only one recurrence and 20 parameters (each parameter has 24 bytes). This parameter is valid only if parameter 111 is different from 0 and is analyzed only on an incoming call.

By default, parameter 112 is equal to 0, which means; no X.25 field to be modified. Value 1 points to parameter 0 in C26 (value 2 to parameter 1 in C26, etc.).

The actions in C26 have the following format:

Type_action, X.25 field, <value> <value>:optional
Type_action:	A	Reserved (e.g. Addition at input of FastPad)
	B	Addition at output of FastPad
	C	Change
	D	Destroy
	F	Forbid
X.25 field:	E	called party address
	A	calling party address
	F	Facility
	B, C, D	reserved

DTE address check

This service allows the checking of the address furnished by the DTE (caller's address at the request of the called party; its called party address upon call confirmation or clearing request) and enables the DTE to use signalling packets having both caller and called party addresses.

Call request (as per STUR X.25)

For a two-address DTE, the service allows the checking of the caller's address (respectively the called party's address furnished by the DTE during the accepted call or clearing request). If the addresses are not identical, the call is cleared. Otherwise, the call continues normally. The check is carried out in relation to the secondary address if it is configured; otherwise in relation to the main address.

For a single-address DTE, the secondary address (or the main address if the secondary address is not configured) of the DTE is inserted. If the length of the sub-address plus the address to be inserted is more than 15 digits, the call is cleared. This processing replaces the usual address completion processing.

For a non X.25 DTE, the address to DNICZOAB format generated by the FastPad is replaced by the secondary (or main) address.

Incoming call (as per STUR X.25)

For a two-address DTE, the main address received is replaced by the secondary address of the DTE (if the latter is configured).

If the length of the secondary address plus the SA is more than 15 digits, the final address is truncated.

For a single-address DTE, the address furnished is eliminated to leave only the SA.

Call accepted or clearing request (as per STUR X.25)

Case of the use of the CLAMN facility

Called party address

In two-address mode:

If the called party address is absent, the call is cleared.
Otherwise, the called address is checked.

In single address mode:

If the SA is absent, the call is cleared.
Otherwise, the secondary address (or main address if the secondary address is not configured) is inserted.

Calling address

In two-address mode:

: The caller's address if any is not known. The calling address length field is then set to 0; the calling address fields is eliminated.

In single-address mode.

If the calling address if furnished, the call is freed.

Only the following combinations are managed:

BF: Addition of a facility at output of FastPad
DA: Destruction of caller's address at input
CE: Change of called party address

Example:

Line 1 of equipment 900240 is a subscriber line with private network addressing. It is configured as 1 address. and subscribes to called party and calling party address transport. The addresses of the outgoing call-request packet from the FastPad must be modified to furnish:

@ calling: address of port 90024001
@ called party: address of IR 90402180

Moreover, the addition of the facility F70100 is desired at the output of FastPad.

The configuration is then as follows:

in C12 rec 1:

	Pl11 =	3 private network management (transport @ calling party and @ called party)
	Pl12 =	1 behavior in C26

PO:DACE90402180BFF701000

in C26 rec 0:

Other cases

In two-address mode:

The furnished addresses are not known. The address length fields are then set to 0, and the address fields are eliminated.

In single-address mode:

If an address is furnished. the call is cleared.

Configuration

A - Subscriber line

Connection parameter for C12Ri:

This service is implemented from a connection parameter allowing its activation. This parameter is defined in C12P89. Parameter 89 is already used for address conversions and the Aimed Point. Bit 7 remains available. By default, this bit is equal to 0 and the service is thus not activated. This bit, called "DTE" address presence authorized" is coded as follows: bit 7: DTE address check/insertion (1/0: yes/no).

C9R11:

To check or be able to insert the address furnished by the DTE, C9R11 is used (conversion of address on incoming call). This table can contain 100 addresses. For the same DNICZOAB, a main address and secondary address may exist. In order to identify a secondary address, it is marked with an "E" at the end of the DNICZOAB referring to the secondary address.

The order in which the main and secondary addresses are filled in on the same line in C9R11 is not subject to any constraint. However, it is recommended that the main address be configured first and then the secondary address in order to minimize the travel time of the table during a call from the remote network (in fact, the main address which allows the routing through the use of Aimed Point).

This conversion makes it possible:

For the routing function to find more rapidly the DNICZOAB on which to route (the called address being the main address),
During address conversion, to find the DNICZOAB of the main address first,
Thanks to the letter "E" to find the secondary address.

B - Network line

Reminder:

The criteria designating a Network line do not change and are: 52=PDN line or type of signal (P44) =network (these are the same criteria which are taken into account for "flow control parameter negotiation).
The present behavior of PDN lines (network line of the PDN single-address or PDN two-address type) cannot be used at the same time as the present upgrading.

Profile 564 is used at the remote network FastPad interface. This profile differs from profile 0 on four parameters:

Parameter 44: type of signalling: Network X.25
Parameter 52: Non-PDN
Parameter 45: 2 addresses
Parameter 89: Aimed point (01XXXXXX).

Secondary address

This functionality is used to present an address other than the main address to any party as a calling address. This address, called the Secondary address, is the one appearing on the invoice.

A - Configuration

This service does not use any configuration parameter through DTE access.

When a secondary address is present on a DTE access, this address is stored in C9R11, the letter E being marked at the end of the corresponding DNICZOAB. When the main and secondary addresses are configured. the main address is first filled in on C9R11 (to allow routing) and then the secondary address.

B - Processing

a - Call request

Mode 1: address with Ctrl/insertion parameter = yes

The secondary address S is inserted as calling address if it is present. Otherwise, it is the main address P which is inserted.

Mode 2: addresses with Ctrl/insertion parameter = yes

When the secondary address is filled in, the checking of the caller's address furnished by the DTE covers this secondary address. If the secondary address is not configured, the check is carried out with the main address.

b - Incoming address

Mode 1: address with Ctr/insertion parameter = yes

The main address is removed from the called address.

Mode 2: addresses with Ctr/insertion parameter = yes

The call transits in the network with its main address to allow routing. If the secondary address is provided for, the call is presented to the DTE with its secondary address (or else with its main address).

c - Call accepted or clearing request

It is the secondary address it is provided for (or else the main address) which is inserted by the network (or by the DTE when it is authorized to furnish its address) in the called party address field of a communication packet accepted or on a clearing request in association with the CLAMN service.

General routing principle

The routing available in the FastPad network is a decentralized service and thus managed in each switching node (subscribers node and transit node).

The routing service is activated systematically in each switch. It analyzes the called address of the incoming call packet.

The next step is, by means of decision tables, to deduce the configured routing and the outgoing line number enabling the called party to be reached.

The routing service uses a fixed internal addressing plan which allows the consultation of routing tables. To allow the adaptation of addressing to the DNIC ZO AB format used by the routing service, a certain number of mechanisms may be put in place:

address conversion
use of aimed points
modification of called party address by the intelligent network access service

Consecutively, the routing algorithm is executed as follows:

1. Conversion of the called address (if necessary),

2. or aimed point, or intelligent network access, hierarchical analysis of the called address field according to three levels as mentioned in Chapter 4 (configuration).

1st level. group identity of the node. Defining the X.25 network and identified by a region number: DNIC. Identification is possible on the x first digits of the DNIC (x between 1 to 4).
2nd level: number of the node in this network: ZO.
3rd level: Subscriber number connected to the user line of the node: AB.

Figure 3-4: Diagram

Example: Subscriber A, identified by its address 9000 10 01, establishes a virtual circuit with subscriber B with network address 9000 20 02. The following takes place:

Figure 3-5: Example

1. Routing activated in the node with ZO = 10:

: The DNIC field of the called address is the same as the DNIC of the calling party. The routing searches in the ZO table of the network and finds the line to the node with ZO = 20. The call is routed to line 03.

2. Routing activated in the node with ZO = 20:

: The routing of the arriving call is checked. It finds that the DNIC and the ZO agree with this node. The routing looks for the AB in the table and find that the call must be sent to line 02, because this is the line where subscriber AB = 02 has been connected.

Deterministic Routing

During implementation of a new network, the system can take into account additions and withdrawals of subscribers and nodes, without constantly changing the routing tables.

There is another feature available when the called address is not configured in the tables.

There are two configuration possibilities:

- the call is refused.
- the call is routed to a line allotted to unknown address fields (unknown DNIC, unknown ZO, unknown AB).

This deterministic routing is then reported in the adjacent network element (other switch, terminal).

Adapted Routing Facility

When it is not possible for the routing to deliver a call the following will happen:

The call packet is returned to the incoming line. Some information has been added to the packet. When the packet arrives back at the preceding node, this node xx, by re-activating the routing service, another path in the network. This mechanism is called "call return". It can be configured with C12RxP81 of the call input line.
When no call return has been configured, the call is refused and a clear packet is sent to the originator.

Table of Contents