WO2007077322A1 - Encoding of functional entities on the basis of control data exchanged on a telecommunications network - Google Patents

Abstract

The invention relates to a method for encoding functional entities on the basis of network control data of a telecommunications network. According to said method, (A) a modelling of the network control information is established by associating all network control information (CI<SUB>i</SUB>) with a single network function (F<SUB>v</SUB>) and a support signal (S<SUB>j</SUB>) received by said network function, (B) a modelling of the constraints induced by the architecture of the network functions on any network control information is established by conditionally associating two network functions with the existence of a reference point (PR<SUB>r</SUB>) connecting the two network functions in said architecture, and (C) a modelling of each network control datum is established on the basis of modellings (A) and (B) by applying network urban planning regulations ({U<SUB>i</SUB>}<SUP>i=I max</SUP>

Description

ENCODING ¹ FUNCTIONAL ENTITIES FROM CONTROL DATA EXCHANGED OVER A TELECOMMUNICATIONS NETWORK

The present invention relates to a method for encoding functional entities based on data exchanged over a telecommunications network making it possible to optimize, in particular, the volume of the data flows exchanged over this network necessary to ensure optimization of the execution of the data. set of network functions and their maintenance. The network functions may for example represent a distribution service for audiovisual content on this network.

More specifically, the present invention relates to a method of encoding functional entities from the network control data exchanged over a telecommunications network, under constraint of the architecture of the functions of this network, generating or using these data, and reference points linking the functions of this network.

The architecture of the network functions of a telecommunications network can be modeled according to an oriented graph consisting of two types of nodes: the functions; - the reference points.

In the context of modeling from a commercially available modeling language, such as the UML (Unified Modeling Language), for a presentation of the UML language, reference may be made to the G. Booch, J. Rumbaugh and I. Jacobson entitled "The Unified Modeling Language User Guide", Addison-Wesley, 1999) in its UML 2.0 version, the aforementioned functions and reference points are respectively components and functions. connectors.

In general, the functions constitute logical and / or arithmetic processing nodes capable of acting on the data, that is to say, producing, recording or using them. They thus have interfaces enabling activation of intrinsic processing procedures or to activate the procedures for processing other functions, thanks to the operations offered respectively to the required operations, in the sense of the aforementioned UML 2.0 language.

Each instance of a function defines a link between the function and the reference points associated with it. The link between two or more instances of these functions is thus modeled by the abovementioned reference points.

A reference point is an association of complementary interfaces of instances of different functions. More precisely, a reference point constitutes a set of required interfaces and their corresponding offered interfaces.

There are several types of reference points:

- the unidirectional bilateral reference point, denoted RP (bu), which constitutes a directional link between two function instances, in which the information originates from the user function having the interface required to be processed by the provider function, having the corresponding offered interface;

- bidirectional bilateral reference point, denoted RP (bb), which constitutes a bidirectional link, between two instances of functions, in which the information starts from the function having the interface required to be processed by the function having the interface corresponding offer. Both functions, in this case, have both required and offered interfaces;

- the RP unidirectional multilateral reference point (mu) which constitutes a unidirectional link between two groups of function instances, in which the information originates from the user functions having the interfaces required to be processed by the provider functions, having the interfaces corresponding offers. In this case, however, it is imperative to specify the membership of each instance to one of these groups.

Figure 1a shows an example of a network function architecture consisting of 4 functions A, B ₁ C, D. The RPbu reference point is an example of the unidirectional bilateral type. It defines the link between the functions A and B where the information starts from the function A user, continuous fine line, to be treated by the function B provider, dashed line.

Similarly, the reference point RP _bb is an example of the bilateral bidirectional type between the functions A and C, in solid continuous line. The reference point RP _mu is an example of the unidirectional multilateral type between the functions B, C and D, where the information originates from the user function B, continuous fine line, for a treatment by the functions C and / or D provider, in dashed line. The aforementioned oriented graph of functions and reference points illustrated in FIG. 1a makes it possible to model the architecture of the network functions. The orientation of the branches of the graph makes it possible to locate the flow of information during an information processing process.

Thus, information produced or transformed by a function can be sent to another function under the control of a reference point in two ways:

- by the user side, to initiate or transfer a processing by the function on the provider side;

- by the provider side to respond to a request for processing initiated by the function of the user side.

Thus, in FIG. 1b, the information I ₁ produced by the function A is sent for processing to the function B, ie from the user side to the provider side. The function B returns after processing the information b in response to the information I ₁ , ie from the provider side to the user side. An information model is a set of concepts and associations that can be used to map information related to a given domain or profession in a consistent manner. The set of concepts can be represented by UML classes when the UML modeling language is used. The information model is obtained, in general, following a business analysis phase.

The associations between information define a dependency framework that helps to understand business logic. So, a association that starts from information h to information I ₂ means that information Ii needs information I ₂ . From a point of view of an object-oriented modeling language, the information h depends on the information I ₂ , as illustrated in FIG. 1c. Consequently, for a business logic where the ultimate goal is to provide a product or a service to a user, an oriented graph makes it possible, most often, to distinguish the three groups of information below:

- initial information from which no information depends;

- intermediate information; - the final information, which does not depend on any information.

To represent an oriented graph, such as that illustrated in FIG. 1a, it is possible to use, in particular, as well as for the representation of the architecture of the network functions, the class diagrams defined by the formalism of the UML modeling language, as well as illustrated in Figure 1c. in the example of FIG. 1c, U is initial information and I ₂ final information.

However, network function architectures and information models can also be represented by a mathematical representation, using the formalism of set theory. In the context of the present application, the following sets of examples are considered:

the set of information El = {l / l represents a concept of the trade}. For example, in any business where you need to identify the user, authenticate the user to be able to perform a transaction with that authenticated user, El = {User ID, User Authenticator, S transaction on product or service} ;

- the information model:

Ml = {(x, y) / (x, ye EI) Λ (x ≠ y) _Λ (xRdy)}.

In the preceding definition: • e designates the relation of inclusion or belonging;

• A designates the intersection or conjunction relation; • Rd denotes the dependency relationship "depends on" as defined previously;

• x and y represent functions.

Thus, for an information model Ml = Dom (Rd) in the set El ² , where Dom (Rd), domain of the dependency relation Rd, is the algebraic definition of the pairs that verify this relation, the model of information above is written:

Ml = ((S ₁ UID) ₁ (S ₁ AID) ₁ (UID ₁ AID)) where: * S denotes a transaction on a product or service; * UID is an identifier of the user;

* AID means an authenticator of the user.

- The EF set of functions as defined in the previous paragraphs. An architecture of the network functions AF is then defined as follows: AF = {(x, y) / (x, ye EF) Λ ( _X Rfy)} with the relation Rf = "are linked by a reference point".

The architecture of the network functions AF defines the field of application of the dependency relation Rf on the set EF ² .

For example, in the case of the example of FIG. 1a, AF = (CA ₁ B) ₁ (A ₁ C) ₁ (B ₁ C) ₁ (B ₁ D)), x and y representing all the pairs of functions among the functions A, B, C and D of Figure 1a.

A reference point RP is then defined as:

RP = (AFP, b) / (AFP = P (AF)) Λ (be {bu, bb, mu}) where P (AF) denotes a partition of AF, defined by elements having the same reference point. For example, in the case of Figure 1a,

RPbu = [{A, B), bu] and RPmu = [((B ₁ C) ₁ (B ₁ D)) ^ u]. An ERP set of reference points is seen as a specification of the definition of an architecture, that is:

ERP = {(AFP.b)}.

The data modeling proposed in a Conceptual Data Model, a Logical Data Model or a Physical Data Model, derived from a classical method, such as the Merise method described by Hubert TARDIEU, Arnold ROCHFIELD and René COLLETTI in the book « The Merise method: principles and tools "published and published in 2000 in France by the Organization Editions, ISBN 2708124730, does not make it possible to translate, through this modeling, the notion of constraints, ie to say cause-and-effect links, generated by an architecture of network functions and reference points translated in their effects on the definition of network data or on the definition of the oriented relations between two network data, of a telecommunications network.

The implementation of a modeling language such as UML makes it possible, with respect to the above-mentioned drawback, to schematize the notion of a network function, a network functional entity or network devices, the only producers of network data and dependency relations oriented between these network data (the "oriented dependency relationship" of a network control data originated to a target network control data means that the network function, the network functional entity or the reference network device the original data is also the reference of the link with the target data). Indeed, the UML language (for example) allows, for each network datum and for each dependency relation between these network data, to specify that there exists a network function, and only one, or a functional entity, and only one, or a network organ, and only one, which are the hosts when said data or said dependency relationship is produced outside the network, and the producers when said data or said dependency relationship is produced within the network.

More recent work has been done in the field of urban planning modeling of the control architecture of telecommunications networks. The corresponding developments gave rise to the filing on February 28, 2005 of the French patent application FR 05 02022, entitled "Modeling tool for the design and optimization of the control architecture of a telecommunications network", on behalf of France Telecom. The aforementioned developments make it possible to establish a model, or more precisely a metamodel, of the telecommunication network business, responding to the constraints and specifications of the new generation networks, called NGN networks, for New Generation Network in English.

However, the above-mentioned developments and patent application do not describe or suggest any approach allowing the passage between an information model and a network data model specific to an architecture of the network functions of a given telecommunications network.

The present invention aims to remedy the shortcomings of the techniques and developments of the prior art, to allow, through a sui generis modeling process, to model all the network data of a telecommunications network, in the purpose of controlling and optimizing, both qualitatively and quantitatively, the nature and / or density of data produced, recorded or used by the network functions of any telecommunications network, taking into account the architecture of these network functions and the points reference linking them. More specifically, an object of the present invention is the implementation of a method of encoding functional entities from data exchanged over a telecommunications network, the modeled exchanged data concerning only the data needed for the implementation. network functions and designated for this reason network control data, taking into account the architecture of these network functions and the reference points linking them, in order to allow control, optimization and possible adaptation of the implementation. of these network functions and their execution.

The method of encoding functional entities from network control data exchanged and manipulated over a telecommunications network from network control information, taking into account the constraints induced by the architecture of the network functions and the control points. reference linking these network functions available on this telecommunications network, object of the invention, is remarkable in that it consists at least to establish a network control information modeling by association of any network control information to a network function and a single and a support signal of this network control information received by this network function, establish a modeling of the constraints induced on any network control information by the architecture of said network functions and said reference points, by creating an association between two network functions conditionally to the existence of a reference point linking the two network functions in this architecture of the network functions, establish a modeling of each network control data from the modeling of network control information and modeling induced constraints on any information network control by application of network planning rules, including a referencing of each network control data either by a network function, or by a functional entity, subset or grouping of network functions, or by a network member setting at least one network function.

According to particular features, the step of establishing a network control information modeling comprises discriminating and encoding a dependency relationship between two network control information by establishing a causal link between the transmission of the bearer signal each of these two pieces of information.

According to other particular characteristics, the step consisting in establishing a modeling of the constraints includes, for each network function, the attribution of a role of referencing an information or a link between two modeled information.

According to still other particular features, the step of establishing a modeling of each network control data consists in applying the constraints of a network function architecture model to a model of the network control information, for each imposed constraint. by the urban planning rules.

According to still other particular characteristics, each signal received by a network function is either a signal of the request signal type or a signal of the resource signal type.

The method of encoding functional entities based on an object of the invention firstly provides the modeling of the network control data of any type of telecommunications network.

It can be implemented from advanced modeling languages such as UML, for example.

It allows the implementation of telecommunication network modeling tools, each modeling tool, for a given telecommunications network, for obtaining a model of the network control information, a model of the network control data referenced each by a network function, respectively by functional entities, of the architecture of the network functions of this telecommunications network. The present invention also provides: - data storage means comprising computer program code instructions for performing the steps of any of the methods described briefly above, and

a computer program containing instructions such that, when said program controls a programmable data processing device, said instructions cause said data processing device to implement any of the methods described briefly above.

A more detailed description of the method of encoding functional entities from network control data exchanged and manipulated over a telecommunications network, object of the invention, will be given below in connection with the figures in which, in addition to the figures 1a to 1c relating to the prior art:

FIG. 2 represents, by way of illustration, a flowchart of the essential steps of implementation of the method of encoding functional entities from network control data exchanged and manipulated over a telecommunications network, object of the invention;

FIG. 3 represents, in an illustrative manner, an illustration of a network planning rule governing the grouping of the control data. network according to the method of encoding functional entities from network control data exchanged and manipulated on a telecommunications network object of the invention;

FIG. 4a represents, in an illustrative manner, the architecture of the network functions for a network functionality designated "Network Attachment", implementing the link to a network of a user terminal connecting to a telecommunications network, and used by the service network "My TV line";

FIG. 4b represents, in an illustrative manner, a sequential diagram of the execution of the "Network Attachment" network functionality, the architecture of which is represented in FIG. 4a;

FIG. 4c represents a model of the network control information of the Network Attachment network functionality, whose architecture is represented in FIG. 4a; FIG. 4d represents a model of the architecture of the network functions of the Network Attachment network functionality, whose architecture is represented in FIG. 4a;

FIG. 4e represents, in an illustrative manner, a model of the network control data referenced by the network functions of the "Network Attachment" network functionality, the architecture of which is represented in FIG. 4a;

FIG. 4f represents, in an illustrative manner, a model of the network control data referenced by network functional entities obtained after transformation of the model of the network control data referenced by the network functions of the network functionality "Network"

Attachment ", whose architecture is shown in Figure 4a;

FIG. 4g represents, in an illustrative manner, a model of the network control data referenced by network functional entities of the "Network Attachment" network functionality, the architecture of which is represented in FIG. 4a;

FIG. 5a represents, for purely illustrative purposes, a flowchart of the steps for implementing the method that is the subject of the invention, allowing to execute the coding of network functional entities, grouping of network functions, from the network control data modeled according to a network control data model obtained in accordance with the method forming the subject of the invention, the execution of the entity coding network functionalities advantageously making it possible to optimize the finally obtained network control data model;

FIG. 5b represents, for purely illustrative purposes, a flowchart for implementing network planning rules that are conditional on the grouping, allowing the implementation of the method that is the subject of the invention and to perform the coding of network functional entities. ;

FIG. 6a is a view of the architecture of the network functions of the Network Attachment network functionality, whose architecture is represented in FIG. 4a "My TV line";

FIG. 6b represents, for purely illustrative purposes, a modeling sequence diagram representing the successive signal exchanges between network functions of a telecommunications network during the execution of a first scenario of execution of these network functions for the execution of a determined network functionality;

FIG. 6c represents, for purely illustrative purposes, a modeling sequence diagram representing the successive exchanges of signals between network functions during the execution of a second scenario of execution of these network functions for the execution of the same network functionality than that described in connection with Figure 6b;

FIGS. 6d and 6e show the network control data modeled according to the method that is the subject of the invention for the functions or "Identification-Authentication" functional entities respectively "Authentication-Identification" of the "My TV line" network service in the case of execution scenarios of these network functions of FIGS. 6a and 6c respectively; FIG. 7a represents, by way of illustration, the grouping of network control data referenced by the same network function according to the network urban planning rule U ₄ for the network service "My TV line"; FIG. 7b represents, for illustrative purposes, the grouping of the "Access Management" and "Line Identification" functions in the same functional entity;

FIG. 7c represents, as an illustration, the grouping of the "Access Management" and "Line Authentication" functions in the same functional entity;

FIGS. 7d and 7e represent, for purely illustrative purposes, the grouping of the "Line Identification" and "User Profile Resource Control Database" network functions from the network control data model and specific conditional network planning rules of the grouping. ;

FIG. 7f represents, for purely illustrative purposes, the grouping of the "Line Identification", "Line Authentication" and "User Profile Resource Control Database" network functions, in the same common network functional entity; and

FIG. 7g represents, for purely illustrative purposes, a model of the grouping of the network functions "Access Management", "Line Identification", "Line Authentication" and "User Profile Resource Control Database" in the same final common network functional entity, obtained thanks to the implementation of the method which is the subject of the invention.

Prior to the detailed description of the method of encoding functional entities from network control data exchanged and manipulated over a telecommunications network, object of the present invention, various general definitions involved in the implementation of the object process of the invention will be given below. We design :

- network control information: any information produced, transmitted or memorized within a telecommunications network allowing the implementation of all the network features of the latter; - network functionality: set of network functions; a service provided by a telecommunications network, such as missed call management, consists of a set of network features; - network control data: any digital data produced, transmitted or stored within a telecommunications network, and serving as information carrier for the network control;

- network planning rule: rule of conditional establishment of the relative arrangement and interaction of network functions, network functional entities, grouping of network functions, and network devices allowing the legitimate implementation of network functions or network functional entities for performing network functionality; Its purpose is to model network control data. More specifically, the modeling of network control data exchanged and manipulated on a telecommunications network from network control information is implemented taking into account the constraints induced by the architecture of the network functions and the network functions. reference points linking these network functions available or to be implemented on the considered telecommunications network.

For the implementation of the method which is the subject of the invention, with reference to FIG. 2, it is thus considered, for any telecommunications network:

a set of network control information written {Cli} |: [; a set of network functions denoted {F _v } ^V _V ⁼ J _{ , the functions of this set of network functions being organized according to an architecture of network functions denoted by AF;

a set of reference points noted (RP ₁ ) JIf;

a set of signals denoted {Sj} ^J r [, the signals Sj of this set of signals making it possible, of course, to convey the network control information via the network control data for the considered telecommunications network. Network control information is defined by the reception of a signal by a network function, as represented in the functional view (sequence diagram) of the implementation.

In network service examples, network functionality or network function given in the following description, for the implementation of the method of the invention, the designation of services, functionality or network function used is the standard generic designation usually used. In view of the aforementioned elements, the method which is the subject of the present invention is remarkable in that it consists at least, as represented in FIG. 2, in establishing in a step A a modeling of the network control information by association of any network control information. CIj to one and only one network function F _v and to a support signal of this network control information S _j received by this network function.

In step A of FIG. 2, the modeling operation is represented by the symbolic relation:

M "ci,};:; ) »MI {cii, (F _v , Sj)};:;

The aforementioned modeling operation performed in step A is justified by the fact that all the elements circulating in a network can be split into network control information CIj such that each extracted information is modified only by one and only one network function. According to a remarkable aspect of the method that is the subject of the invention, a network control information CI i manipulated in any telecommunications network is thus defined by a pair formed by a network function F _v and a signal S _j received by this network function.

Indeed, when receiving a signal Sj, a network function receiving the aforementioned signal modifies the values of an information conveyed by this signal. This is particularly the case of the information of the update when received by a network function such as the network function "Access Management", ie "access management", resulting from a address allocation transmitted by a signal S _j designated "Advice Allocation Response".

Thus, for a function F _v and for a signal Sj transmitted towards this function F _v , a network control information CIj is such that CIj = (F _v , Sj) with the assumption that the network control information CIj is set by the network function F _v when receiving the signal Sj related to a scenario associated with the Execution of the execution and use of a network functionality or a telecommunications service.

The aforementioned step A is followed by a step B of establishing a modeling of the induced constraints on all network control information by the AF architecture of the network functions and reference points.

According to a remarkable aspect of the implementation of the method that is the subject of the invention, the modeling of the induced stresses in step B of FIG. 2 is performed by creating an association between two network functions conditionally on the basis of existence of a reference point RP _r linking the two network functions in the architecture of the network functions AF.

In FIG. 2, in step B thereof, the modeling operation of the induced constraints on any network control information is noted by the symbolic relation:

AF {F _v , F _v ', (RP _r )}. Thus, an association in the sense of the modeling operation carried out in step B of FIG. 2, between two network functions F _v and F _v ', exists if and only if a reference point RP _r links them to the latter. functional architecture of the AF network. In addition, each network function F _v or F _v 'must be the reference of an information or a link between the modeled network control information. Each network function F _v is the reference with respect to this network control information or the link between network control information modeled in the sense that the above network function guarantees the values of the network control information CIj = (F _v , S _j ) whatever Sj signal sent to the network function F _v . Under these conditions, the network function F _{v is said to} reference the network control information CIj associated with it via the signal S _j .

The method which is the subject of the invention then consists, in a step C, in establishing a modeling of each network control data from the modeling of the network control information executed in step A and the modeling of the induced constraints on any network control information executed in step B, the modeling operation of step C being executed by application of specific network planning rules. In FIG. 2, in step C thereof, the network planning rules are noted:

Uj designating a specific rule of network urbanism according to the general definition given previously in the description.

In step C of FIG. 2, the modeling operation of the network control data is represented by the symbolic relation:

MD = U (Ml {.} X AF {.}).

In the aforementioned symbolic relation: - Ml {.} Denotes the modeling of the network control information CIi obtained in step A;

AF {.} Denotes the modeling of the constraints obtained following the execution of step B;

U denotes the corresponding extraction of the constraints imposed by the urban planning rules Uj.

It is understood, in particular, that the aforementioned network urban planning rules make it possible to model the network control data which are referenced either by the network functions F _v or by the network functional entities, a network functional entity consisting of a function group network, as will be described later in the description.

The modeled network control data MD can then be represented in accordance with the conventional modeling process, either by one or more functional views of the aforementioned network control data, or by an organic view of the data relative to the constituent network members of the considered telecommunications network.

More specifically, it is indicated that step A of establishing a modeling of the network control information by creating an association between two network functions conditionally to the existence of a reference point linking these two network functions in the architecture of the network functions, is advantageously performed by establishing a link of causality between the transmission of the support signal of each of these two network control information.

The aforementioned causal link can be highlighted as follows: If a signal S _j is emitted by a network function F _v following the reception by this same network function F _v of another signal Sj ', it then exists, in the sense of the modeling used by the invention, a dependency between the corresponding network control information. This is the case for a signal S ₂ transmitted from a function Fi to a function F ₂ if the signal S ₂ temporally follows the signal Si transmitted to the function Fi. Under these conditions, Cl ₁ = (F ₁₁ S ₁ ) depends on Cl ₂ = (F ₂₁ S ₂ ).

A more detailed description of a set of urban planning rules U = {Uj} ;: ', ^max will now be given in a general way. This description will also be illustrated by means of network functions in a telecommunications network given by way of example, namely those of the service.

"My TV Line", which is an IP television broadcast service or telephone line marketed by France Telecom. A list of features and messages used to implement this "My TV Line" service is provided in the Appendix. In this sample application, the network function "Access

Management "receives a" Network Address Request "signal. The resulting network control information C1 then corresponds to the pair Cl = ("Access

Management "," Network Address Request ").

In the same example, following the receipt of the "Network Address Request" signal by the "Access Management" network function, this function sends the "Network Access Authorization Request" signal to the "Network Access Admission Control" network function.

The network control information Cl ₁ = ("Access Management",

"Network Address Request") then depends on the network control information Cl ₂ = ("Network Access Admission Control", "Network Access

Authorization Request ").

In addition, for a first network control information Cl ₁ referenced by a first network function F ₁ and dependent on a second network control information Cl ₂ referenced by a second network function F ₂ , the step of establishing a constraint modeling B includes an allocation to the first network function Fi d a role of responsible for the association of the first network control information Ch to the second network control information Cl ₂ .

Consequently, a network control datum D _k referenced by a network function F _v is, in the context of the modeling used by the present invention, represented by a pair D _k = (F _v , S) where S denotes a set of signals Sj received by the function F _v .

In addition, a signal Sj received by a network function F _v is either a signal of the request signal type or a signal of the resource signal type.

More specifically, it is indicated that a request signal is a signal making it possible to send a request to a network function F _v , for example that of an address allocation, while a resource signal is the response or the acquisition of a request signal, for example the result of the aforementioned address allocation.

In general, it is indicated that the aforementioned network urban planning rules comprise at least one conditional rule for transforming network control information into network control data. With reference to the aforementioned conditional rule Ui.

Urban planning rule Ui: the network control data comes from a transformation of network control information and defined by a network function if this network control data is: - referenced by one and only one network function F _v ;

- characterized according to the type of the signal received by the network function either as request network control data or resource network control data.

In the example of the service "My TV line", the network control information Ch = ("Network Access Admission Control", "Network

Access Authorization Request ") can be transformed into a query network control data since the signal" Network Access Authorization Request "is a request signal. The Network Access Authorization Response (CI ₂ = Network Access Admission Control) may be transformed into resource network control data since the Network Access Authorization Response signal is a resource signal.

Network planning rules may furthermore include a conditional rule for transforming a link between network control information into a link between network control data defined by the aforementioned network functions. Urban planning rule U ₂ : the link between network control data is derived from a link between network control information on the existence criterion of a coherence relation with a reference point linking these network functions.

Thus the association of network control information Ch = ("Access Management", "Network Address Request") to the network control information Cl ₂ = ("Network Access Admission Control", "Network Access Authorization Request") is convertible into an association of network control data Di = ("Access Management", Network Address Request ") to the network control data D ₂ = (" Network Access Admission Control "," Network Access Authorization Request ") when the "Access Management" and "Network Access Admission Control" functions are linked by a reference point as will be written later in the description.

The network planning rules furthermore comprise at least one conditional rule for adding a link between a first network control datum defined by a first set of signals associated with a determined network function and a second network control datum defined by a second set of signals associated with the aforementioned determined network function. Urban planning rule U ₃ :

The link is added if: - the first set of signals S is a set of request signals and the second set of signals S 'is a set of resource signals and if,

there exists at least one signal S ' _j of the second set of signals S' which is a response to a determined signal S _j of the first set of signals S. The link added between the first network control data item to the second control data element network is further oriented.

The planning rule U ₃ above means that, following the reception by the determined function F _v of the request signal of the first set S of network control signals, one or more signals are emitted until the reception by the network function F _v of a resource signal belonging to the set S 'of network control signals, the aforementioned signal constituting a response to the request signal of the first set of signals S. Under these conditions, the association in the sense of modeling used by the invention is oriented and written: D = (F _v1 S) to D '= (F _V , S').

The network control data model obtained allows the addition of an association of D = ("Network Access Admission Control", "Network Access Authorization Request") to D "(" Network Access Admission Control "," Network Access Authorization " Response ") since the" Network Access Information Response "resource signal is the acknowledgment of the" Network Access Authorization Request "request signal.

Network planning rules also include at least one conditional rule for grouping network control data.

Urban planning rule U ₄ : The network control data are grouped together if:

- These network control data are of the same type of data request or resource;

these network control data are referenced by the same network function; - the network control data model, following the grouping performed, remains acyclic with respect to the dependency relationships between the network control data. The dependency character refers to the dependencies derived from data modeling and the acyclic character refers to the absence of a loop capable of generating an infinite closed circuit of the modeled network control data. An illustration of the town planning rule U ₄ is now given in connection with FIG. 3. The aforementioned figure represents a model of the network control data of the service "My TV line".

Indeed, the network control data Di = ("Access

Management "," Network Access Authorization Response ") and D ₂ = (" Access Management "," Line Identification Response ") as shown in the drawing of Figure 3 can be grouped into one and the same network control data D = ( "Access Management", {"Network

Access Authorization Response "," Line Identification Response "}) since the network control data Di and D ₂ are of the resource type, because these data are referenced by the same network function" Access

Management "and the grouping does not create a cycle, that is, a closed loop, in the new network control data model.

In FIG. 3, the other network control data contained in the non-referenced blocks simply designate network control data modeled in accordance with the method that is the subject of the present invention, that is to say from a pair formed by a network function and a signal.

A network control data D may, in accordance with the method that is the subject of the present invention, be referenced by a network functional entity EF, that is to say, in fact, by a grouping of network functions and then represented by a pair D = (EF ₁ S) where S denotes a set of signals received by all the network functions of EF.

The network planning rules allowing the implementation of the method that is the subject of the invention further comprise at least one conditional rule for referencing a network control data item by a network functional entity.

Planning Rule U ₅ :

D network control data referenced by a function network F _v such that D = (F _V , S) is defined as referenced by a functional entity EF such that D '= (EF ₁ S) if the network function F _v belongs to the functional entity EF.

Thus, with reference to the above-mentioned FIG. 3, for a network functional entity EF = "Access Relay and Management" then the functional entity EF references the network control data D '= ("Access Relay and Management", "Network Address Request Since there is a network control data D defined by a network function such as

D = ("Access Management", "Network Address Request") and such that the function F _v = "Access Management" belongs to the functional entity EF.

The town planning rules for implementing the method that is the subject of the invention also include a conditional rule for grouping network control data defined by a network functional entity. Urban planning rule Uβ:

Network control data is grouped if:

the grouped network control data are of the same type, query or resource;

the grouped network control data are referenced by the same network functional entity;

- the network control data model remains acyclic with respect to the dependencies between network control data following clustering.

With reference to the aforementioned FIG. 3, from the model of the data referenced by the network functions, it is possible to define two network control data referenced by the network functional entity.

EF = {"Identification", "Authentication and Access Authorization"} the data:

- Of i = ({"Identification", "Authentication and Access Authorization"}, "Network Access Authorization Request"), and, - D ' ₂ = ({"Identification", "Authentication and Access Authorization"}, "Network Access Information Request ").

Since Network Access Admission Network Features Control "and" Network Access Contrai Database "belong to the same network functional entity EF, the network control data D'i and D ' ₂ are derived from network control data Di and D ₂ such that Di = (" Network Access Admission Control "," Network Access Authorization Request ") and D ₂ = (" Network Access Control Database "," Network Access Information Request ").

The aforementioned network control data D'i and D ' ₂ can be grouped in a data item:

D '= ({"Identification", "Authentication and Access Authorization"}, {"Network Access Authorization Request", "Network

Access Information Request "}) since these are two request data referenced by the aforementioned functional entity EF and this grouping does not imply a cycle in the model of the network control data.

In addition, in accordance with the implementation of the method that is the subject of the invention, a network control data item D may be referenced by a network element O covering the realization of network functions and the deployment of the latter on a given network.

Under these conditions, the network planning rules allowing the implementation of the method that is the subject of the invention comprise at least one conditional rule for referencing a network control data item by a network member.

Planning Rule U ₇ :

A network control data D referenced by a network function F _v is such that D = (F _V , S) is defined as referenced by a network element O such that D '= (O ₁ S) if the function F _v is implemented that is, made and deployed on the network member O.

As an example for a network element O = "Video Service Node" then the network device O references the network control data D '("Video Service Node", "Network Address Request") since there is a data D network control defined by a network function such as D = ("Access Relay", "Network Address Request") is such that the "Access Relay" function is performed and deployed on the network member O. A nonlimiting preferred embodiment of the modeling of network control data referenced by network functions and by network functional entities will now be described in a nonlimiting manner in the case where the modeling language is UML, any language modeling of the essential features of the aforementioned UML language can however be implemented without departing from the scope of the subject of the present invention.

In the example given hereinafter with reference to FIGS. 4a to 4g, the network functions and network functional entities taken as an example correspond to those of the functional subsector "Network Attachaient" of the functional entity or telecommunications service " My TV line.

FIG. 4a shows the architecture of the network functions of the "Network Attachment" feature.

The architecture of the network functions of the "Network Attachment" feature mentioned above and represented in FIG. 4a, corresponds to a functional view obtained for example by the implementation of the modeling tool described by the French patent application number 05 02022. previously mentioned in the prescription.

In FIG. 4a, and by convention: the network functions, excluding support network functions, are represented by ovals including the name of the function or functional entity under consideration;

the reference points or reference point grouping points are represented by diamonds connecting two functions or functional entities;

the support network functions are represented by a three-dimensional object symbolizing, for example, a hard disk;

the signals or set of signals are represented by oriented arrows including their designation and their reference Si to S ₂ 2- FIG. 4b represents a time diagram of the signals exchanged between the different functions represented in FIG. 4a with their reference point. FIG. 4c represents a UML model of the network control information of the model Network Attachment network functionality obtained in accordance with the implementation of the method that is the subject of the present invention. This model is obtained following the implementation of step A of the method that is the subject of the invention as represented in FIG. 2, from an application of the method that is the subject of the invention to the architecture of the network functions of FIG. the Network Attachment network functionality as shown in FIG. 4a.

FIG. 4d represents a UML model of the network functions of the "Network Attachment" network functionality, a model obtained by implementing the method that is the subject of the invention from the architecture of the network functions represented in FIG. 4a. With reference to Figure 4d above, it is understood that the functions "Access Management" and Network Access

Admission Control "are connected by reference point RP ₄ g. The link between network control information is therefore convertible into a link between network control data, because the network planning rule U ₂ is satisfied.

The first embodiment describes UML modeling of network control data referenced by network functions.

The UML modeling of the network control information is described from the functional architecture of the network functions, their reference points, and the signals exchanged during a scenario of the telecommunications service "My TV Line" between network functions of the network. "Network Attachment" network functionality.

The network functions, excluding support network functions, are as follows:

- "Access Relay";

- "Access Management";

- "Line Identification";

- "Line Authentication"; - "Network Access Admission Control";

- "Address Allocation";

- Localization Access Admission Control. Network Information Function support network functions are as follows:

- "Network Address Database";

- "Localization Database". Support network functions of type "User Information

Function "are the following:

- "User Profile Resource Control Database";

- Network Access Control Database.

The signals exchanged between network functions are the following: - Si, S ₂ "Network Address Request";

- S ₃ "Network Access Authorization Request";

- S ₄ "Network Access Information Request";

- S ₅ "Network Access Information Response";

- S ₆ "Network Access Authorization Response"; - S ₇ "Line Identification Request";

- S ₈ "Line Identification Information Request";

- S ₉ "Line Identification Information Response";

- S ₁₀ "Line Identification Response";

- Su "Line Authentication Request"; - S ₁₂ "Line Authentication Information Request";

- S ₁₃ "Line Authentication Information Response";

- S ₁₄ "Line Authentication Response";

- S ₁₅ "Address Allocation Request";

- S ₁₆ "Address Allocation Localization Admission Request"; - If ₇ "Address Allocation Localization Admission Information Request";

- S ™ Address Allocation Localization Admission Information Response;

- Sig "Address Localization Admission Response";

- S ₂₀ "Address Info Request";

- S ₂₁ "Address Info Response"; - S ₂₂ "Address Allocation Response";

- S ₂₃ "Network Address Response".

The network control information is represented in a UML model as follows with reference to Figure 4c:

if the network control information I = (F, S), then I is represented by the UML class "F-S";

if the network control information h and b are such that h = (Fi, Si) depends on I ₂ = (F ₂ , S ₂ ), then the dependency is represented by the association between UML classes oriented of "F ₁ -S ₁ "to" F ₂ -S2 ".

For example, I = ("Access Management", "Network Address Request") is represented by the class "AccessManagement" - "NetworkAddressRequest" and the fact that h = ("Access Relay", "Network Address Request") depends on I ₂ = ("Access Management", "Network Address Request") is represented by the association oriented from "AccessRelay" - "NetworkAddressRequest" to "AccessManagement" - "NetworkAddressRequest".

The UML modeling of the network functions and reference points is described from the functional architecture of the network functions, their reference points, and the signals exchanged between network functions of the "Network Attachment" functionality with reference to FIG. 4a. .

The network functions and the reference points that connect them are represented in a UML model as follows with reference to FIG. 4d:

the network function F _v is represented by a UML class "F";

the reference point RP _r is represented by the association "RP" between the network functions that it connects.

The implementation of the modeling of the network control data referenced by network functions is described in this paragraph for each of the town planning rules to be implemented. The scenario of the service "My TV Line" described from the signals exchanged between network functions of "Network Attachment" with reference to FIG. 4a is represented by a UML sequence diagram with reference to FIG. 4b, in order to better visualize the concatenation of signals received and transmitted by each network function.

For the Urban Planning Rule Ui, only the receiving network function signal S ₂₃ is not known. In this case, this signal can not be the source of a network control data as part of the set of network functions of the "Network Attachment" feature. With this exception, the network control information shown in Figure 4c respects the Urban Planning Rule U ₁ and are therefore all transformable into network control data. Indeed, they are all referenced by one and only one network function since I = (F _V , S) is referenced by F _v . In addition each signal is identifiable as a request signal or a resource signal.

The request type network control data are the following: - D ₁ = ("Access Relay", "Network Address Request");

- D ₂ = ("Access Management", "Network Address Request");

- D ₃ = ("Network Access Admission Control", "Network Access Authorization Request");

- D ₄ = ("Network Access Control Database", "Network Access Information Request");

- D ₅ = ("Line Identification", "Line Identification Request");

- D ₆ = ("User Profile Resource Control Database", "Line Identification Information Request");

- D ₇ = ("Line Authentication", "Line Authentication Request"); - D ₈ = ("User Profile Resource Control Database", "Line Authentication Information Request");

- D ₉ = ("Address Allocation", "Address Allocation Request");

- D ₁₀ = ("Localization Access Admission Control", "Address Allocation Localization Admission Request"); Di ₁ = (Localization Database, Address Allocation Localization Admission Information Request);

- D ₁₂ = ("Network Address Database", "Address Info Request").

The network control data of the resource type are the following: - D ₁₃ = ("Network Access Admission Control", "Network Access Information Response");

- D ₁₄ = ("Access Management", "Network Access Authorization" Response ");

- D ₁₅ = ("Line Identification", "Line Identification Information Response");

- D ₁₆ = ("Access Management", "Line Identification Response");

- Di ₇ = ("Line Authentication", "Line Authentication Information Response");

- Die = ("Access Management", "Line Authentication Response");

Di ₉ = (Localization Access Admission Control, Address Allocation Localization Admission Information Response);

- D _2O = ("Address Allocation", "Address Localization Admission Response");

- D ₂₁ = ("Address Allocation", "Address Info Response");

- D ₂₂ = ("Access Management", "Address Allocation Response").

The network control data referenced by network functions are represented in the UML model as follows: - if the network control data D = (F _V , S), then D is represented by the UML class "F _v -S "

if the network control data Di and D ₂ are such that Di = (Fi ₁ Si) depends on D ₂ = (F ₂ , S ₂ ), then the dependency is represented by the association between UML classes of "F ₁ -S ₁ "to" F ₂ -S ₂ ". For example, D = ("Access Management", "Network Address

Request ") is represented by the class" AccessManagement- NetworkAddressRequest "and the fact that Di = (" Access Relay "," Network Address Request ") depends on D ₂ = (" Access Management "," Network Address Request ") is represented by the association oriented from "AccessRelay" - "NetworkAddressRequest" to "AccessManagement" - "NetworkAddressRequest".

According to the Urban Planning Rule U ₂ , the links between network control information with reference to Figure 4c are all transformable into links between network control data since they are all consistent with the associations between network functions represented by the points. reference numbers shown in FIG. 4d.

Additions of dependency links between control data network in accordance with the Urban Planning Rule U ₃ are as follows:

- from D ₂ = ("Access Management", "Network Address Request") to Di ₄ = ("Access Management", "Network Access Authorization Response"), Die = ("Access Management, Line Identification Response"), Dis = ("Access Management", "Line Authentication"

Response ") and D ₂₂ = (" Access Management "," Address Allocation Response ") since the resource signals" Network Access Authorization Response "," Line Identification Response "," Line Authentication Response "and" Address Allocation Response "respond to same "Network Adress Request" request signal;

- from D ₅ = ("Line Identification", "Line Identification Request") to D ₁₅ = ("Line Identification", "Line Identification Information Response") since the "Line Identification Information Response" resource signal responds to the request signal "Line Identification Request"; - from D ₇ = ("Line Authentication", "Line Authentication Request") to D- ₁₇ = ("LineAuthentication", "Line Authentication Information Response") since the "Line Authentication Information Response" resource signal responds to the request signal "Line Authentication Request"; - from D ₃ = ("Network Access Admission Control", "Network Access Authorization Request") to Di ₃ = ("Network Access Admission Control", "Network Access Information Response") since the resource signal "Network Access Information Response" responds to the "Network Access Authorization Request" request signal; - from D ₉ = ("Address Allocation", "Address Allocation Request") to D ₂₁ = ("Address Allocation", "Address Info Response") and D _2O = ("Address Allocation", "Address Localization Admission Response") since the "Address Info Response" and "Address Localization Admission Response" resource signals respond to the same "Address Allocation Request" request signal;

- from D10 = ("Localization Access Admission Control", "Address Allocation Localization Admission Request") to Di ₉ = ("Localization Access Admission Contrai "," Address Allocation Localization Admission Information Response ") since the resource signal" Address Allocation Localization Admission Information Response "responds to the request signal" Address Allocation Localization Admission Request ". These links represented by UML associations are in bold line in the network control data model of "Network Attachment" shown in Figure 4e.

The application of the U Planning Rule ₄ targeting network control data consolidation is impossible from the network control data model and initialized, because all possible network control data groupings on this model would involve cycles.

The second embodiment describes UML modeling of network control data referenced by network functional entities.

Functional entities are defined as follows: - "Access Relay & Management" = {"Access Relay", "Access Management"};

- "Identification Authentication & Access Authorization" = {"Line Identification", "Line Authentication", "Network Access Admission Control", "User Profile Resource Control Database", "Network Access Control Database"};

- "Address Localization & Allocation" = {"Address Allocation", "Localization Access Admission Control", "Network Address Database", "Localization Database"}.

The network control data, when derived from a set of signals received by a network functional entity, is represented in a UML model as follows:

if the network control data D = (EF, {Si, S ₂ J) where Si and S ₂ are signals received by the network functional entity EF, then D is represented by the UML class "FS-i + S" ₂ ". This representation can be transposed for network control data referenced by network functions.

Following the application of the Urban Planning Rule U ₅ , the UML model network control data referenced by network functions (see Figure 4e) is transformed by replacing each network function by the network functional entity to which it belongs (see Figure 4e). The network control data, referenced by a functional entity, transformed by D _n , network control data referenced by a network function, is designated in the remainder of the paragraph by D ' _n .

The groupings conforming to the Urban Planning Rule O _& of network control data referenced by network functional entities are then the following: - D ₃ = ("Identification Authentication & Access Authorization", "Network Access Authorization Request") and D ₄ = ("Identification Authentication & Access Authorization", "Network Access Information Request") may be grouped into one and the same data D = ₃₄ ("Identification Authentication & Access Authorization", {"Network Access Authorization Request", "Network Access Access Information Request "});

- ₅ = ('Identification Authentication & Access Authorization', 'Line Identification Request') and D ' ₆ = (' Identification Authentication & Access Authorization ',' Line Identification Information Request ') may be grouped into one and the same data From ₅₆ = ("Identification Authentication & Access Authorization", {"Line Identification Request",

"Line Identification Information Request"});

- D ' ₇ = ("Identification Authentication & Access Authorization", "Line Authentication Request") and D'β = ("Identification Authentication & Access Authorization", "Line Authentication Information Request") can be grouped into one and the same data Of ₇ β = ("Identification

Authentication & Access Authorization ", {" Line Authentication Request "," Line Authentication Information Request "};

- Of g = ("Address Localization & Allocation", "Address Allocation Request"), D'-io = ("Address Localization & Allocation", "Address Allocation Localization Admission Request") and D'u = ("Address

Localization & Allocation "," Address Allocation Localization Admission Information Request ") can be grouped into one and the same g-ion data = ("Address Localization &Allocation",{"Address Allocation Request", "Address Allocation Localization Admission Request", "Address Allocation Localization Admission Information Request"}; - On ig = ( "Address Allocation &Localization","Address Allocation Localization Admission Information Response") and D ₂ = o ( "Address Allocation &Localization","Localization Address Admission Response") can be grouped in a single same data D'i ₉₂ o = ('Address Localization &Allocation','Address Localization Admission Information Response', 'Address Localization

Admission Response "}).

The network control data from these groupings are in bold line in the model of the network control data referenced by the "Network Attachment" feature as shown in FIG. 4g. A more detailed description of a preferred embodiment of the method according to the invention for executing the coding of network functional entities from modeled network control data as described previously in the description, will now be given in connection with FIG. with Figure 5a, Figure 5b and the following figures. The method that is the subject of the invention then makes it possible to define and encode the network functional entities of any telecommunications network service from a grouping of network control data, as will be explained hereinafter.

It is understood in particular that for the preferred mode of implementation of the method which is the subject of the invention, the above-mentioned network control data are modeled as mentioned previously in the description as a function of the constraint imposed by the architecture of the network functions and reference points of the considered telecommunications network model. The evaluation of the different choices of grouping of network functions into network functional entities to ensure the implementation of a telecommunications network service is then performed as it will be described below.

With reference to FIG. 5a, network control data modeled according to a network control data model are noted: MDR = {D _x = (F _v , S _j )} ^.

In the above relationship, MDR designates the network control data model considered: - D _x indicates any network control data contained and defining the aforementioned MDR network control data model; - F _v and Sj designate as mentioned above in the description, any network function or any corresponding network control signal.

As represented in FIG. 5a, the method which is the subject of the invention includes, iteratively, the verification in step A of a first and a second urban planning rule, the urban planning rules Ue and Ug conditional of the possible grouping of a first and a second network control data considered.

Consequently, in verification step A, the verification of the Us and Ug network urban planning rules is noted: U ₈ AND U ₉ (D ₁₁ D ₂ ).

In the above relation, it is understood that the urban planning rules U ₈ and U ₉ are applied to two network control data Di and D ₂ , these data being defined as previously mentioned in the description by the relations: D ₁ = (Fi ₁ S ₁ ) and D ₂ = (F ₂ , S ₂ ).

The above verification step A allows from the above-mentioned MDR network control data model and the grouping from the network control data D ₁ and D ₂ to generate a possible network control data model designated MDR _P , which understood, is subjected to a treatment for verification of the relevance of the regrouping thus carried out.

The aforementioned verification step A is then followed by a conditional execution step B of the aforementioned possible grouping on the criterion of measuring the simplicity of the network control data model obtained, that is to say the possible model MDR _P following the grouping provided for in step A.

It is more specifically understood that step B is intended to verify the merits of the aforementioned possible grouping before any actual execution operation.

The iterative nature of the method that is the subject of the invention as represented in FIG. 5a results, from a first iteration of rank k = 0, in repeating steps A and B by returning to step A of FIG. 5a. noted k following, the aforementioned iterative character being maintained and the successive iterations performed as a network function grouping from network control data such as data Di and D ₂ , can be executed.

With reference to FIG. 5a and more specifically, it is indicated that the first conditional urbanization rule Us of the grouping allows this grouping if the first Di and the second D ₂ network control data are of the same type, that is, that is, the request or resource type, as previously mentioned in the description.

In addition, the first planning rule U ₈ allows this grouping if the first network control data Di depends on the second network control data D ₂ , as mentioned previously in the description relative to the definition of the dependence of the data of network control in accordance with the implementation of the method which is the subject of the invention.

Finally, the first conditional urban planning rule of the Us grouping allows this grouping if the model of the network control data that is to say the possible MDR _P model remains acyclic following the aforementioned possible grouping.

With regard to the second conditional urban planning rule of the grouping, the rule Ug ₁ indicates that for a first and a second network control data Di and D ₂ each referenced by a function that is to say control data network such as Di and D ₂ defined in step A of Figure 5a referenced by the functions F ₁ respectively F ₂ , the second urban planning rule of aforementioned grouping Ug involves the grouping of functions Fi and F ₂ in the same functional entity.

A detailed description of a preferred non-limiting mode of implementation of the conditional Us and Ug urban planning rules of this grouping will now be given in connection with FIG. 5b. In step A of FIG. 5a, the successive application of the urban planning rules U ₈ and Ug to the network control data Di and D _{2 is considered} so as to execute an application according to the logical AND relation of the rules U ₂ and U ₉ above.

Thus, with reference to FIG. 5b, the application of the planning rule U ₈ to the network control data Di and D ₂ to the substep D ₁ of FIG. 5b is considered.

By way of nonlimiting example, the sub-step A 1 may consist for example of the execution of the town planning rule U ₈ (D ₁ , D ₂ ) in a test sub-step A ₁₀ on the network control data D ₁ and D ₂ noted according to the relationships:

- D ₁ AND D ₂ e T ₁ . ? or,

- D ₁ AND D ₂ e T _s ?

The test A ₁₀ thus verifies either the membership of the network control data D ₁ and D ₂ to the same request type T _r or of these same data D ₁ and D ₂ to the same resource type T ₃ .

On the negative response of the test A ₁₀ , a return to the execution of the sub-step A ₁₀ was carried out for the choice of another pair of network control data denoted D 1, D ₂ in FIG. 5 b.

Instead of a positive response to the substep _10, a sub-step test Ai ₁ is executed, which consists in verifying the dependence relation of the first network control data Di vis-à-vis the second network control data D ₂ .

On negative response to the test A ₁₁ , a return to the sub-step A ₁₀ is performed for another pair of data D ^ and D ' ₂ network control. On the contrary, on a positive response to the sub-step A ₁₁ , a sub-step A ₁₂ is executed which consists in verifying that the possible network control data model MDR = MDR (D _X {] D ₂ ) is acyclic. This operation is noted:

MDR _p = MDR (D ₁ [JD ₂ ).

It will be recalled that the verification of the acyclic character of the possible network control data model MDR _P is an operation performed by means of conventional means, in particular in the UML modeling language.

On a negative answer to the test Ai ₂ a return is made to the sub-step A-10 for another pair D'i, D ' ₂ of network control data.

On the contrary, on a positive response to the sub-step Ai ₂ , we have network control data Di = (Fi, Sj) and D ₂ = (F ₂ , S _j ') for which the conditional urbanization rule Us of the grouping has has been verified and is therefore denoted U ₈ (D ₁ , D ₂ ) = TRUE.

Under these conditions, urban planning rule U ₉ can be applied to the functions Fi and F ₂ referencing the network control data Di and D ₂ mentioned above. The second conditional urban planning rule of the U ₉ grouping then consists in declaring the grouping of the aforementioned network functions F ₁ and F ₂ , which of course made it possible to check the network planning rule Us at the true value.

The grouping operation in the sub-step A ₂ i in step A ₂ of step A of FIG. 5a is represented by the relation in FIG. 5b:

F _ι [j F ₂ → EF _c .

In the previous relationship, EF _C denotes a common functional entity resulting from the group considered.

The execution of the urban planning rules U ₈ and Ug mentioned above, as described in connection with FIG. 5b, thus makes it possible for two network control data Di = (Fi, Si) dependent on the network control data D ₂ = ( F ₂ , S ₂ ) with Fi ≠ F ₂ and if Di and D ₂ are of the same type request or resource then the grouping makes it possible to define a functional entity EF ₀ such that EF _C = {Fi, F ₂ ) and a new datum D = (EF _C , S) grouping of the network control data Di and D ₂ with the set of signals S received by the common functional entity EF ₀ , the set of aforementioned signals S, S being defined by S = { If, S ₂ }. The definition of functional entities from groupings of network control data following the application of the Us and Ug urban planning rules is made based on a measure of the simplicity of the network data model obtained. According to a particularly remarkable characteristic of implementation of the method that is the subject of the invention, for a UML class network control data model, the criterion for measuring the simplicity of the network control data model, that is to say of any MDR model as defined above and in particular of any possible MDR _P model, includes the computation of a metric quantifying the simplicity of the aforementioned network control data model. This metric is taken equal to the ratio of the total number of associations between UML class and the total number of classes of the data model of the MDR network control _P considered.

The aforementioned metric checks the relationship:

/ \ total number of association [MDR _n ) MS [MDR] = _j - ± r ^. total number of class \ MDR _P )

It is thus clear that, thanks to the implementation of the method that is the subject of the invention, for each choice of possible network control data grouping MDR _P , the corollary is the grouping of the network functions F _v which each reference the data in a Network functional entity such as EF ₀ common network functional entity.

Thus, at each integration of one or more network functions in a network functional entity, the model of the network control data is updated by replacing the network functions grouped by the corresponding current network functional entity. Of course, the process is iterative as shown in FIG.

5a. For each rank iteration ^k _k z% then a choice of network control data may advantageously be performed from among a set of possible groupings of network control data, by discriminating the minimum value of the simplicity measure of the control data model. Network obtained for each possible grouping of network control data. So the possible grouping retained for execution conditional and for next iteration k = k + 1 corresponds to the aforementioned minimum value.

This particularly advantageous procedure is illustrated in detail in FIG. 5a in the following manner.

Step B previously described may advantageously comprise a substep Bi of calculation of the above-mentioned metric MS according to the relation:

In the previous relationship:

- ACNp is the total number of associations for the possible MDR _P network control data model considered;

- CTNp is the total number of classes for the above MDR _P network control data model.

It is understood, in particular, that following step Bi the process can then be repeated to perform the calculation of a plurality of metrics quantifying the simplicity of a network control data model for a plurality of groupings and finally of models successive potential network control data MDR _P.

This operation is represented in FIG. 5a by a return after the execution of the sub-step Bi in step A by the step BiA for all the possible groupings f: of groups of network control data Di and

D ₂ constituting the starting model MDR.

Thus, the calculation of the aforementioned metrics can be performed iteratively as long as there are pairs of network control data that can be grouped together and of course satisfying the network planning rules U ₈ and U ₉ .

At the end of the sub-iteration by the return step BiA, at the end of the last step Bi a set of metrics quantifying the simplicity of an MDR network data model is available, this set of metrics being noted:

[MS [MDR _P ]} ^P ; S. The substep B ₁ is then followed by a substep B ₂ which allows to choose the possible network control data model noted MDR _p min such that the value of the metric of the latter is the minimum value among the metric values constituting the set of metric values mentioned above. In step B ₂ allowing this choice, the operation is represented by the symbolic relation:

MDR _pnύn /

.

The choice of the value of the metric of the minimum value and correlatively the choice of the corresponding network control data model MDR _{P m} j _n , can then be executed by a conventional sorting program, which will not be described in detail for this reason.

The substep B ₂ can then be followed by a substep B ₃ which is executed at the end of iteration of given rank k. The substep B ₃ makes it possible to declare the union of the functions Fi and F ₂ referencing the network control data Di and D ₂ which made it possible to group and obtain the possible network control data model MDR _{P m} j _π , this grouping operation corresponding to the grouping of the functions Fi and F ₂ according to the relation EF _C = [F ₁ U -F ₂ ). This operation is accompanied by the deletion of the functions Fi and F ₂ from the selected network control data model MDR _P min and the introduction of the common network functional entity EF ₀ into the same selected network control data model MDR _{P m} j _n .

The substep B ₃ is then followed by the return step denoted B ₃ A at the next iteration denoted k following in step A of FIG. 5a.

It is thus understood that at the end of the iteration of rank k the best grouping of network control data is chosen for the next iteration.

The iterative process is continued as long as there is a pair of network control data Di and D ₂ satisfying step A of Figure 5a.

A description of the concrete implementation of the method that is the subject of the present invention will be given in the framework of the architecture of the network functions and the representation of the signal exchanges of a modeling sequence diagram, such as a UML modeling. . An extract of the network function architecture of the "Network Attachment" feature is shown in Figure 6a.

As part of the network function architecture of the "Network Attachment" feature shown in FIG. 6a, the network functions, excluding the support network function, are as follows:

- "Access Management",

- "Line Identification",

- "Line Authentication".

The support network function of the type "User Information Function" is the following:

- "User Profile Resource Control Database".

The reference points RP _x, or RP3, PR _4, PR ₄ Ws, RPβ, RPβbis as shown in Figure 6a, are reference points that connect the corresponding network functions. Network control signals exchanged between network functions are as follows:

- S ₂ "Network Address Request",

- S ₇ "Line Identification Request",

- Ss "Line Identification Information Request", - Sg "Line Identification Information Response",

- S ₁₀ "Line Identification Response",

- Su "Line Authentication Request",

- S ₁₂ "Line Authentication Information Request",

- S ₁₃ "Line Authentication Information Response", - S ₁₄ "Line Authentication Response".

It is understood, in particular, that the aforementioned functions represented by ovals in FIG. 6a and the aforementioned signals S ₂ to S ₁₄ and the reference points connecting these functions correspond to elements known from the state of the art, because implemented in the "Network Attachment" functionality of the "My TV line" telecommunications network service available on the market. Consequently, the designation of the network functions, the aforementioned network control signals is the designation origin of these.

It is understood, in particular, that the implementation of the method which is the subject of the invention is not limited to the example given with reference to FIG. 6a but on the contrary perhaps applied to any set of network functions, these functions being connected by reference points forming a set of reference points {RP _r } ^r _r Z *, the set of signals exchanged by the aforementioned functions being denoted by [S _j y ^.

An example of the implementation of the method of encoding network functional entities, object of the present invention, from network control data will now be given in a concrete case in connection with Figures 6b and 6c.

Recall that FIG. 6b corresponds to a first scenario in which the "Line Identification" function is called after the "Access Management" function and where the "Line Authentication" function is called concomitantly with the call of the "Line Identification" function. Identification "whereas in Figure 6c the" Line Authentication "and" Line Identification "functions are interchanged as well as the signals exchanged by these functions with the" Access Management "function.

In general, it is indicated that the modeling sequence is constituted either by the successive signal sequences exchanged by the network functions, or by the network function sequences corresponding to these successive signal exchanges.

It is understood, in particular, that the signals exchanged between network functions being oriented, there is a one-to-one correspondence between an oriented signal exchanged by two network functions and the corresponding network functions, by means of the reference point which makes it possible to connect the two network functions.

With reference to FIGS. 6b and 6c, for a functional entity EF = {"Access Management", "Line Identification"} then: - the network function sequences are:

• "Access Management" - "" Line Identification "corresponding to successive signals exchanged S ₇ or SV;

• "Line Identification"->"Accessmanagement" successive signals exchanged If ₀ or S'i ₀ ;

- "Access management" is the only element of the sequence when in the second scenario it receives the signal S ' ₂ and sends the signal S'u.

It is thus understood that by the notation convention of the signals exchanged in the second scenario as represented in FIG. 6c, the signal S _n exchanged from the network function F _m to the network function F _p in the first scenario represented in FIG. 6b becomes S ' _n in the second scenario. S ' _n is always exchanged between the functions F _m and F _p but because of the content of the two scenarios this signal is different from the signal S _n .

The concrete example of implementation is described for an initialization step k = 0 followed by three successive iterations k = 1 to k = 3. Initialization of the network data model k = 0 The step of modeling the network control data required upstream of the process is described below. The modeling of the network control data deduced from the first sequence diagram represents the scenario 1 as in FIG. 6d and the modeling of the network control data deduced from the second sequence diagram represents the scenario 2 as represented in FIG. 6e. The network planning rules implemented in this step are the rules of network planning designated rules Ui, U ₂ and U ₃ previously in the description.

The grouping of network control data referenced by the same network function according to the town planning rule U ₄ is as follows, with reference to FIG. 7a:

- ("Access Management", S ₂ ) and ("Access Management", S ' ₂ ") can be grouped in the same data (" Access Management ", (S ₂ , S' ₂ ));

- ("Access Management", S-io) and ("Access Management", S'i ₄ ) can be grouped in the same data ("Access Management",

{Sio, S'14});

- ('Access Management', Si ₄ ) and ('Access Management', S'io ') may be grouped in the same data (Access Management, {S14, S'10});

- ("User Profile Resource Control Database", S ₈ ) and ("User Profile Resource Control Database", S'1 ₂ ") can be grouped in the same data (" User Profile Resource Control Database ", (S ₈ , S '12});

- ("User Profile Resource Control Database", S ₁ 2) and ("User Profile Resource Control Database", S'a ") can be grouped in the same data (" User Profile Resource Control Database ", {S12, S ' ₈ });

1 ° Iteration of grouping of network data k = 1. The application of the principle of grouping network data from the proposed data model and the rules, urban planning rule U ₈ and urban planning rule Ug, induces the following grouping choices:

- ("Access Management", (S ₂ , S ' ₂ }) and ("One Identification", S ₇ ") and other network data aggregations involving the grouping of the" Access Management "and" Line "network functions

Identification "in the same network functional entity with reference to Figure 7b:

• a new MDRi network control data model with 18 associations and 10 UML classes; • MS (MDR ₁ J = I eZI O;

- Due to the symmetry of the data model, the data groupings involving the grouping of the network functions "Access management" and "Line Authentication" in the same network functional entity with reference to Figure 7c has the same consequences as the grouping of data. network functions "Access Management" and

"Line Identification".

• a new MDR2 network data model such as

MS (MDR ₂ ) = MS (MDRi) = 18/10;

- ('Line Identification', S ₇ ) and ('User Profile Resource Control Database, (S ₈ , S' ₁₂ )) as well as other data groupings involving the bundling of 'Line Identification' and 'User Profile Resource' network functions Control Database "in a single network functional entity according to Figure 7d:

• a new MDR ₃ network data model with 18 associations and 11 UML classes;

MS (MDR ₃ ) = 18/1 1; - because of the symmetry of the data model, the data groupings involving the bundling of the Line Authentication and User Profile Resource Control Database network functions in the same network functional entity according to Figure 7e, has the same consequences as the grouping of the "One Identification" and "User Profile Resource Control Database" network functions:

• a new MDR ₄ network data model such as

MS (MDR ₄ ) = MS (MDR ₃ ) = 18/1 1.

The minimum of the measure of simplicity of the different network data models obtained is achieved with the models MDR ₃ and MDR ₄ . The first proposed grouping is thus validated in the first iteration, ie the MDR ₃ model, that is to say that of the grouping of the network functions "Line Identification" and "User Profile Resource Control Database". 2 ° Network data grouping iteration, k = 2. The second iteration must therefore evaluate the following choices of network data groupings:

- ("Line Authentication", Su) and ({"Line Authentication", "User Profile Resource Control Database"}, (S ₇ , S ₈ , S'1 ₂ } ") and other groupings of data involving grouping network functions "Line Identification", "Line Authentication", "User Profile Resource Control Database", in the same network functional entity according to FIG. 7f:

• a new MDR ₅ network data model with 8 associations and 7 UML classes;

• MS (MDR ₅ ) = 8/7. It can be noticed that following the grouping of the network functions "Line Identification", "Line Authentication" and "User Profile Resource Control Database" in the same network functional entity, the network identification (Line Identification, Sg) and Line Authentication (S'13) as well as network data (Line Authentication, S ₁₃ ) and Line Identification (S'g) can be grouped together title of the Urban Planning Rule Uβ since these data are referenced by the same functional entity {"Line Identification", "Line Authentication", and "User Profile Resource Control Database"};

- ("Access Management", {S ₂ , S ' ₂ }) and ({"Line Identification", "User Profile Resource Control Database" (S ₇ , S ₈ , S' ₁ 2) can not be grouped as this would create a cycle around ("Line Authentication", S'u).

3 ° Network data grouping iteration, k = 3. The third iteration is immediate with only one group offered:

- ("Access Management", {S ₂ , S ' ₂ }) and ({"Line Identification", "Line Authentication", "User Profile Resource Control Database"}, (S ₇ , S'u, S ₈ ,

S _'12}) and other data combinations involving the consolidation of network functions "Access Management", "Line Identification", "Line Authentication" and "User Profile Resource Control Database" in one functional network entity referenced in Figure 7g .

The method that is the subject of the invention for encoding network functional entities is not limited to the embodiment described above.

In particular, in addition to the aforementioned coding from previously described modeled network control data, another method is to code the network functional entities directly from the existing network functions and / or network functional entities, as described and claimed in the application. French patent entitled "Method of encoding network functional entities from network functions of a telecommunications network and corresponding modeling tool" filed the same day on behalf of France Telecom.

The aforementioned method applied to the same concrete case shows the coherence of the iterations of the two methods of definition and coding of network functional entities implemented on the "Network Attachment" functional entity part of the "My TV line" service.

Although the symmetrical aspect of the various diagrams or models relating to the "Line Identification" and "Line Authentication" functions must be taken into account, this aspect does not modify the aforementioned statement, because of the adherence and the consistency of the modeling the network control data to the network functions that reference the latter.

ANNEX

Definition of functions

- "Access Relay": This function introduces local configuration information received from the user. It converts the query into another understandable by the functions of the functional entity

- "Access Management":

This function allows you to organize and organize the actions and procedures necessary for attachment.

- Network Access Admission Control:

It makes it possible to proceed to admission control to a well-defined network (or part of the network). In this case it is the part dedicated to the MLTV service - "Line Identification":

It identifies the access line of the user.

- "Line Authentication":

It authenticates the access line of the user.

- "Localization Access Admission Control": It allows to proceed to the geographical control of admission to a network

(or part of the network) well defined. In this case it is the part of the network dedicated to the MLTV service.

- "Address Allocation":

It is used to allocate the IP address of the user. The information bearer functions provide a functional representation of the location of the information needed for the other functions (above) for handling the network attachment:

- Network Access Control Database:

It contains the information to identify the terminals of the users who have subscribed to the MLTV service.

- User Profile Control Database: It contains credentials and authentication credentials for users' connectivity lines.

- "Localization Database":

It contains the geographic location information of the connectivity lines. This information may affect the network configuration process of the user's terminal.

- "Network Address Database":

It contains the necessary configuration information as a result of the implementation of the services including IP addresses, DNS servers and service access node addresses.

Definition of messages

- If, S ₂ "Network Address Request":

The message represents the request from the user terminal to attach to the network. It contains its MAC address (Layer 2) - S ₃ "Network Access Authorization Request":

The message represents the request from the Access Management function to verify the existence of the MAC address of the user's terminal.

- S ₄ "Network Access Information Request": The message represents the request resulting from the "Access Access" function

Management "to scan the database for the MAC address of the user's terminal.

- S ₅ "Network Access Information Response":

The response containing or not information on the MAC address of the user's terminal.

- Se "Network Access Authorization Response":

The response containing verification information of the MAC address of the user's terminal

- S ₇ "Line Identification Request": The message represents the request resulting from the "Access Identification" function

Management "to verify the existence of the access line used by the user terminal to access the MLTV service. - S ₈ "Line Identification Information Request":

The message represents the request from the "Line Identification" function to scan the database for the access line used by the user terminal to access the MLTV service. - Sg "Line Identification Information Response":

The response containing or not information on the access line used by the user terminal to access the MLTV service.

- S-io "Line Identification Response":

The response containing verification information of the access line used by the user terminal to access the MLTV service.

- Su "Line Authentication Request":

The message represents the request from the "Access Management" function to authenticate the existence of the access line used by the user terminal to access the MLTV service. - S- _I2 "Line Authentication Information Request":

The message represents the request from the "Line Authentication" function to poll the database for the access line used by the user terminal to access the MLTV service. - S ₁₃ "Line Authentication Information Response":

The response containing or not information on the access line used by the user terminal to access the MLTV service.

- SI ₄ "Line Authentication Response":

The response containing authentication information of the access line used by the user terminal to access the MLTV service.

- Sis "Address Allocation Request":

The message represents the request from the "Access Management" function to request an IP address for the user terminal

- S- _I 6 "Address Allocation Localization Admission Request": The message represents the request resulting from the function "Address

Allocation "to request to verify the non nomadism of the terminal of the user wishing to access the service of" My TV Line ". - If ₇ "Address Allocation Localization Admission Information Request": The message is the request after the "Location Access Admission Contrat" to scan the database regarding the geographical location of access line used by the user terminal to access the MLTV service.

- S- _I8 "Address Allocation Localization Admission Information Response":

The response containing information on the geographic location of the access line used by the user terminal to access the MLTV service.

- S- ₁₉ "Address Allocation Localization Admission Response":

The response containing information on the nomadism or non-nomadism of the access line used by the user terminal to access the MLTV service. - S _2O "Address Info Request":

The message represents the request from the "Address Allocation" function to request the assignment of an IP address to the user terminal.

- S ₂ i "Address Info Response": The response containing information on the assignment of an IP address to the user terminal.

- S ₂₂ "Address Allocation Response":

The response containing the IP address of the user terminal

- S ₂₃ "Network Address Response": The response containing the IP address of the user terminal as well as other information regarding the service access portal address (Service Access).

Claims

A method of encoding functional entities from network control data exchanged and manipulated over a telecommunications network from network control information, taking into account the constraints induced by the architecture of network functions and reference points linking said network functions available on said telecommunications network, characterized in that it consists of at least:

- modeling network control information by association of any network control information to one and only one network function and a signal support of this network control information received by this network function; establishing a modeling of the induced constraints on all network control information by the architecture of said network functions and said reference points, by creating an association between two network functions conditionally on the existence of a reference point linking said two network functions; network functions in said network function architecture;

establishing a modeling of each network control data from said modeling of the network control information and of said modeling of the induced constraints on any network control information, by application of network planning rules including a referencing of each control data; network either by a network function, or by a functional entity, subset of network functions, or by a network device implementing at least one network function.

Method according to claim 1, characterized in that the step of establishing a modeling of the network control information comprises discriminating and encoding a dependency relationship between two network control information by establishing a network control information. causality link between the transmission of the support signal of each of these two network control information.

3. Method according to one of claims 1 or 2, characterized in that the step of establishing a constraint modeling includes, for each network function, the assignment of a referencing role of a network control information or a link between two modeled network control information.

4. Method according to claims 2 and 3, characterized in that for a first network control information referenced by a first network function and dependent on a second network control information referenced by a second network function, the step of establishing constraint modeling includes assigning to said first network function a role of responsible for associating the first network control information with the second network control information.

5. Method according to one of claims 1 to 4, characterized in that the step of establishing a modeling of each network control data is to apply the constraints of an architecture model network functions to a model of network control information, for each constraint imposed by said planning rules.

6. Method according to one of claims 1 to 5, characterized in that each signal received by a network function is either a signal of the request signal type, or a signal of the resource signal type.

7. Method according to one of claims 1 to 6, characterized in that said network planning rules comprise at least one conditional rule for transforming a network control information into a network control data, said network control data. originating from a network control information transformation and defined by a network function, if said network control data is: - referenced by one and only one network function; characterized, according to the type of the signal received by said network function either as a query network control data item or as a resource network control data item.

8. Method according to one of claims 1 to 7, characterized in that said network planning rules comprise at least one conditional rule for transforming a link between network control information into a link between network control data defined by said network functions, said link between network control data being derived from a link between network control information on the criterion of existence of a coherence relation with a reference point linking these network functions.

9. Method according to one of claims 1 to 8, characterized in that said network planning rules comprise at least one conditional rule of adding a link between a first network control data defined by a first set of signals associated with a determined network function and a second network control data defined by a second set of signals associated with said determined network function, said link being added if: the first set of signals is a set of request signals and the second set of signals is signal is a set of resource signals, and, if

- There is at least one signal of the second set of signals that is a response to a signal of the first set of signals, the link added between the first network control data to the second network control data being further oriented.

10. Method according to one of claims 1 to 9, characterized in that said network planning rules comprise at least one conditional rule grouping network control data, network control data being grouped, if:

said network control data are of the same type of request or resource data;

said network control data are referenced by the same network function; - the network control data model following clustering remains acyclic, with respect to dependency relationships between network control data.

11. Method according to one of claims 1 to 10, characterized in that said network planning rules comprise at least one conditional rule referencing a network control data by a network functional entity if this network control data being referenced by a network function, said network function belongs to said network functional entity.

12. Method according to one of claims 1 to 11, characterized in that said network planning rules comprise at least one conditional rule grouping network control data, data being grouped, if:

said network control data are of the same type of request or resource data;

said network control data are referenced by the same network functional entity; - the network control data model following clustering remains acyclic with respect to dependency relationships between network control data.

13. Method according to one of claims 1 to 12, characterized in that said network planning rules comprise at least one conditional rule referencing a network control data by a network member if the network control data is referenced by a network function, said network function is implemented on said network member.

14. Method according to one of claims 1 to 13, characterized in that it includes at least, iteratively:

- the verification of a first and a second network planning rule conditional on the possible combination of a first and a second network control data;

the conditional execution of said possible grouping, based on criteria for measuring the simplicity of the network control data model obtained, following said possible grouping.

15. The method of claim 14, characterized in that said The first conditional urban planning rule of the grouping authorizes this grouping if:

the first and second network control data are of the same type, query or resource; - the first network control data depends on the second network control data;

- the network control data model remains acyclic, following the possible grouping.

16. The method of claim 14 or 15, characterized in that for a first and a second network control data each referenced by a function, said second grouping conditional urban planning rule involves the grouping of said functions in the same functional entity.

17. Method according to one of claims 14 to 16, characterized in that, for a UML class network control data model, said criterion for measuring the simplicity of the network control data model includes the calculation of a a metric quantizing simplicity, said metric state taken equal to the ratio of the total number of associations between UML classes to the total number of classes of the network control data model.

18. Method according to one of claims 14 to 17, characterized in that, for each execution of a grouping of network control data, a grouping of the network functions, each of these network functions referencing a network control data, in a current network functional entity is executed, accompanied by an update of the network control data model by replacing network functions grouped by said current network functional entity.

19. Method according to one of claims 14 to 18, characterized in that, for each iteration, a choice of the best grouping of network control data is performed among a set of possible groupings of network control data, by discrimination of the network control data. minimum value of the simplicity measure of the network control data model obtained for each possible grouping of network control data, the possible grouping retained for conditional execution and for the next iteration corresponding to the said minimum value.

Data storage means comprising computer program code instructions for performing the steps of any of the methods according to claims 1 to 19.

21. A computer program containing instructions such that, when said program controls a programmable data processing device, said instructions cause said data processing device to implement a method according to any one of claims 1 to 19.