Method and devices for specifying the quality of service in a transmission of data packets

Technical Field of the Invention

The present invention relates to a method for specifying the quality of service in a transmission of data packets between a service entity and a

5 user equipment on a mobile telephone network comprising a core network node and an access node that are adapted to control the handling of data packets sent between the service entity and the user equipment. Software devices and programs that perform the invention are also described.

Background of the invention

10 In many cases, data packets need to be sent between a mobile telephone user equipment and a service entity. Transmissions can be carried out both in the downlink and uplink direction. For example, a server can send different packet streams for sound and video to the user equipment in a real-time streaming session. He

fifteen User equipment can send data to the service entity or can initiate a real-time transmission session through control signaling. The service entity may also be another mobile telephone user equipment. The transmission is carried out over a mobile telephone network and the service entity may be part of the mobile telephone network or is capable of exchanging data packets with the

twenty net.

Typical mobile telephone networks comprise a core network with core network nodes, for example a Serving General packet radio service Support Node (SGSN - Radio Service Support Node in General Server packages) or a General Gateway packet radio service Support Node (GGSN - Support Node

25 General Gateway radio packet service). The core network nodes allow the exchange of data with external networks such as the Internet or mobile or landline networks of other operators. In addition, the usual mobile telephone networks comprise one or more access networks with access network nodes to control radio transmission to the user equipment, commonly designated,

30 for example, as base station controllers, Radio Network Controllers (RNC - Radio Network Controllers), Node B or base transceiver stations.

Depending on the type of packet traffic, the requirements for transmission differ significantly. For example, voice transmission requires low delay and low fluctuation while a limited amount of errors may be acceptable. Real-time transmission sessions that use temporary packet memories typically allow for greater delays and fluctuation and the receiver can generally also correct or hide errors, while file transfer can often be carried out as best-effort traffic but usually Requires error free data. In addition, operators can choose to offer different Qualities of Service (QoS) depending on the user's contract, that is, they can choose to carry out user differentiation. Accordingly, the provision of a defined quality of service is an important concept in the control of data traffic, as described for example in the technical specification 3GPP 23.107 V 6.3.0 of the 3rd Generation Partnership Project. 3rd Generation) “Quality of Service (QoS) concept and architecture”.

The quality of service related to a data transmission that affects nodes of the mobile telephone network and the user equipment is defined in different contexts. The user equipment and a mobile telephone network node negotiate a PDP (Packet Data Protocol) context that specifies parameters for the transmission of data packets to and from the user equipment. In addition, other contexts are established for different links between the service entity and the user equipment, for example a radio bearer between an access node and the user equipment, which specifies the transmission parameters in the radio link. The parameters of the other contexts are normally determined according to the PDP context. The packet flows between the service entity and the user equipment are then mapped to these contexts and transmitted accordingly.

All different contexts affect attributes to specify individual traffic parameters. In the current mobile telephone networks a plurality of such attributes are defined and may indicate either binary or numerical values. Accordingly, there is a high number of possible combinations for the values ​​of such attributes. In the negotiation or specification of the attributes for a context, the user equipment may also be involved. Depending on the manufacturer, model and software, the user equipment may have different requirements and behavior in the context specification procedure, that is, the adjustment of the attributes may also depend on the user equipment. This problem can be partially solved by carrying out a specific configuration for the user equipment according to the network operator, the manufacturer of the user equipment and the type. This is, however, inconvenient for the user and solves this problem only partially.

The concept of differentiated services allows you to mark in the header of a data packet what quality of service should be used when handling the package. A corresponding header field is defined for example in different versions of the Internet Protocol (IP - Internet Protocol). However, the marking of data packets does not allow information on the quality of service to be transmitted to all the nodes that control the quality of a data transmission in a mobile telephone network because not all nodes are adapted to evaluate the headers. For example, if packet transmission involves packet fragmentation and / or encryption with subsequent reassembly or decryption, the headers are not normally accessible to nodes that handle encrypted packets or packet fragments.

Accordingly, it is a problem to specify the quality of service that the nodes in a mobile telephone network use to handle data packets. In addition, it is also difficult for operators to specify how the resources controlled by the nodes are shared in a mobile telephone network between different levels of quality of service.

Patent application EP 1 250 022 already describes a system for providing quality of service in a telecommunications system such as a UMTS or other third generation system. For session control with policy-activated bearer establishment, the system comprises an information base that includes QoS service attributes and QoS service classes and is provided with interfaces for questions from RNC, SGSN and GGSN. However, the introduction of an information base requires significant adaptations in existing mobile phone networks.

Summary

It is an object of the present invention to provide a simple solution to specify the quality of service for handling data packets in a mobile telephone network.

According to the invention, the method described in claim 1 is carried out. In addition, the invention is carried out in a mobile telephone network as described in claim 9, a control device as described in claim 10 and a computer program as described in claim 13. Advantageous embodiments are described in the dependent claims.

In the proposed method, data packets are transmitted between a service entity and a user equipment over a mobile telephone network. The mobile telephone network comprises a core network node and an access node that are adapted to control the handling of data packets sent between the service entity and the user equipment. The data packets can be transmitted by the access node and the core network node. In addition or alternatively, other entities under the control of one of the nodes can handle the data packets, for example a scheduler in a Node B that is controlled by an RNC as an access node.

A plurality of service classes related to quality of service is preconfigured. For example, the operator may choose to use a specified number of pre-configured service classes for transmission in the mobile telephone network and to carry out all packet transmissions according to one of the pre-configured service classes. A selected service class is selected from said plurality of service classes for transmission. For example, the selection can be carried out by the service entity in view of the quality of service required for a specific flow or group of packet flows.

A first transmission context is established between the core network node and the user equipment. The first transmission context is associated with a first set of attributes to define a first quality of service for the exchange of data packets with the user equipment in the transmission. The selected service class determines the first set of attributes for a first single mapping function performed on the core network node. Preferably, a plurality of first sets is preconfigured in the core network node and the mapping function performs a selection from the preconfigured sets according to the selected service class.

In addition, a second transmission context is established between the access node and the user equipment. The second transmission context is associated with a second set of attributes and defines a second quality of service for the exchange of data packets with the user equipment in the transmission. The selected service class determines the second set of attributes by means of a second single mapping function performed on the access node. Preferably, a plurality of second sets are preconfigured in the access node and the mapping function performs a selection from the preconfigured sets according to the selected service class.

In this way the selected service class defines the handling of the data packets on the links controlled by the access node and the core network node and thus the quality of service for the data packets sent between the service entity and the user equipment in the links to which the respective contexts refer.

Preferably, the first and second service qualities are identical and correspond to each other. It is possible that said first and second attribute sets are subsets of larger attribute sets that may also comprise attributes that are otherwise specified, for example that are preconfigured to fixed values. It should also be noted that different service classes may, by means of the mapping function, refer to the same first and second attribute sets, that is, the number of service classes may be greater than the number of possible attribute sets.

The proposed method allows a simple specification of the quality of service that the nodes of a mobile telephone network use for handling data packets. Because the attributes are specified by the service classes, the specifications can easily be transmitted between the nodes of the mobile telephone network without major adaptations of existing systems. The method also simplifies it for operators to specify how resources are controlled by the nodes of a mobile telephone network shared between different service levels because operators can specify resources based on service classes without the need to define resources based in attributes. Another advantage of the proposed method is that it can be used in existing mobile phone communication systems after only minor adaptations in the devices.

In today's mobile phone systems, there are already messages that transmit sets of attributes for contexts between nodes. In a preferred embodiment of the invention, the selected service class is specified by a third set of attributes that is transmitted to at least one of the terminal network node and the access node, preferably both. The first and / or second mapping functions then determine the first set of attributes and / or the second set of attributes of the third set of attributes, that is according to the class of service. This allows the use of existing messages in mobile telephone communication systems for the transmission of information of the kind of service to and between nodes. In other words, the selected service class is encoded as a combination of attribute values ​​in the third set. It is not necessary that all possible combinations of attribute values ​​refer to a service class or that all attributes of the third set be used to determine the encoded service class.

Preferably, the number of service classes is small compared to the number of possible combinations of values ​​of the third set of attributes. A small number of service classes allows easy configuration of the quality of service by the operator. On the other hand, the number of service classes must correspond to the granularity of service differentiation required.

It is advantageous to define the first and / or second mapping function by means of a mapping table. For example, the service class can indicate a row in the table that contains a set of attributes. The mapping table can be specified during node configuration. It allows a quick and simple processing of the service classes by the respective nodes and an easy configuration.

Preferably, a specification of the selected service class is transmitted from the service entity to the core network node because the service entity generally knows the requirements for quality of service in a data transmission. It is, however, possible that the quality of service specifications are modified, for example in a core network edge node, if the operator determines that the selected quality of service must be changed according to the user's contract. . The service entity can specify the requirements for example by marking packages or using a defined flow for the packages. In this case, an edge node of the core network can determine the service class selected from the marking or flow and transmit the specification of the corresponding service class to the core network node. For example, an operator can define two classes of service, each corresponding to a specific package marking. Based on those two specific markings, the edge node can then select a first of these service classes for users with Premium contracts and a second service class for all other users.

Apart from the core network node and the access node, the data packets can also be transmitted by other entities, which may affect the quality of service for the transmission. For example, planning entities on different links may delay data packets. A planning entity controlled by the access node, typically at node B, plans the transmission of data packets on a radio link. In this case, the access node preferably controls the planning entity according to the selected service class.

In an advantageous embodiment, the first mapping function performed in the core network node is identical to the second mapping function performed in the access node, that is the first and second sets of attributes are identical and the Same quality of service is specified for all links. However, the properties of different links may be significantly different from each other and different sets and different mapping functions may be more advantageous in this case.

Preferably, a configuration procedure defines at least one element of a group comprising the plurality of service classes, the first mapping function and the second mapping function, for example the contents of a mapping table. The procedure can be initiated by an Operation Support System (OSS) of the mobile telephone network that allows the operator of a mobile telephone network to both specify and change the corresponding parameters. Configuration procedures may also specify and change resources, which are attributed to all or some selected from the service classes in said plurality of service classes. In this way, resources can be allocated for mandatory services such as emergency calls or system signaling as well as bandwidths reserved for users with Premium contracts.

The invention is also carried out in a control device for a mobile telephone network that is adapted to carry out the transmission of data packets between a service entity and a user equipment and comprising a core network node and a access node The nodes are adapted to control the handling of data packets sent between the service entity and the user equipment. The control device may be one of these nodes or it may be another device in the mobile telephone network. A plurality of service classes related to a quality of service is preconfigured in the mobile telephone network.

The control device comprises a memory in which a single mapping function is stored. The function refers to the service classes for attributes that define the quality of service for the transmission of data packets. A processing unit is adapted to determine a service class selected for transmission from said plurality, for example according to parameters in an establishment message for a transmission context. The processing unit is also adapted to specify a set of attributes of the selected service class using the single mapping function, preferably using a set of values ​​for the attributes stored in memory for the respective service class, that is, the mapping function can for example be defined in a mapping table. The treatment unit is also adapted to establish a transmission context with the user equipment. The transmission context is associated with the set of attributes that define the quality of service for the exchange of data packets with the user equipment in the transmission. A control unit controls the sending of data packets according to the aforementioned set of attributes.

The control device can. For example, being a Serving General packet radio service Support Node (SGSN - Radio Service Support Node in General Service packages), a Gateway General packet radio service Support Node - Radio Service Support Node in General Door packages Link), an improved general packet radio service support node, an RNC radio network controller, a base station controller, a base transceiver station or a B node. The control device may be adapted for use in any embodiment of the method as described above.

The invention can also be carried out in a program unit comprising a code for carrying out those steps of a method described above that relate to a single device. The program unit according to the invention is for example stored in a data carrier or loadable in a processing unit of a control device, for example as a signal sequence.

The foregoing and other objects, features and advantages of the present invention will become more apparent in the following detailed description of preferred embodiments as illustrated in the accompanying drawings.

Brief description of the drawings

Fig. 1

shows an architecture to provide a defined quality of service in a mobile telephone system.

Fig. 2

shows the cooperation of nodes in a method of mobile telephony in which the invention is realized.

Fig. 3

shows an example of a service class mapping.

Fig. 4

shows a flow chart of a method according to the invention.

Fig. 5 shows a device according to the invention.

Detailed description of preferred embodiments

Fig. 1 illustrates a context of quality of service in 3rd generation mobile telephony systems as specified in the 3GPP technical specification

5 23.107 V 6.3.0. of the 3rd Generation Collaboration Project. Traffic comprising data packets is sent between a service entity (AF) and a user equipment comprising a terminal equipment (TE - Terminal Equipment) and a mobile terminal (MT - Mobile Telephone Terminal). The service entity may be a server that could be located in the operator's network or in an external network but

10 It can also be another user equipment. The purpose of the context is to provide a quality of service (QoS) defined at the application level that uses the bearer services of the underlying levels. These bearer services are specified by contexts that comprise attributes to define the QoS of the respective bearer service. As the quality of end-to-end service in the

fifteen application layer depends on the specifications of the underlying levels, the contexts of the bearer services need to be specified with respect to the required end-to-end quality of service.

The local TE / MT bearer service sends the data packets within the user equipment. The packets are received or sent over a radio link 20 with the radio access network (RAN1 - Radio Access Network 1) of the mobile telephone network. The External Carrier Service is provided by another network that may also be a UMTS (Universal Mobile Telephony System) network, that is, a network in accordance with 3GPP specifications, another mobile telephone network or a network fixed as a fixed communication system

25 just like the internet The external carrier sends data packets between the service entity (AF) and a gateway node (CN-GW) of the core network of the mobile telephone network. The present invention is primarily concerned with specifying the quality of service for the UMTS bearer service.

The core network also comprises a core network node (CN1 - Node of

30 Core Network 1) that controls the sending of packets between the core network and the radio access network (RAN1). The gateway node (CN-GW) and the core network node (CN1 can be the same node. Data packet traffic through the mobile telephone network is sent over a Radio Access Bearer Service between a mobile telephone terminal (MT) and a core network node (CN1) and over a Network Bearer Service Core between a Gateway node (CN-GW) and a core network node (CN1). These Services are in turn provided by a Radio Bearer Service on the radio link between the user equipment and a radio access network (RAN1), a RAN Access Bearer Service between a radio access network (RAN1 ) and a core network node (CN1) and a Trunk Network Bearer Service within the core network. Ultimately, all services depend on different physical bearer services on the respective links, that is, typically a plurality of contexts and services refer to individual links in a transmission. Sufficient quality of service is especially important in transmission bottleneck links that limit total quality of service. For mobile phone networks, the bottleneck link is typically a wireless link between the radio access network and the mobile phone terminal.

Fig. 2 shows an example of a data packet transmission using the proposed method with contexts and nodes involved. For the transmission of the data packets, a PDP context is negotiated between user equipment UE1 - User equipment 1) and a core network node, here an SGSN (SGSN1). The transmission is carried out later between a core network node and an access node or at least controlled by them. Dotted line 11 indicates a possible way in which packets are sent in the uplink and downlink direction between the user equipment (UE) and the service entity (AF).

The establishment of the PDP context can for example be initiated by a corresponding request (RQ1) from the user equipment to the SGSN. It is also possible that the network (for example the GGSN) requests the establishment of the PDP context, for example by means of a message to the user equipment that then initiates the sending of a request (RQ1) to activate a PDP context.

The PDP context comprises attributes that define the quality of service for packet transmission. The establishment of a radio bearer is typically included in the establishment of a PDP context. For that purpose, the SGSN (SGSN1) sends a request (12) for the establishment of a radio bearer to an access node, in the example an RNC (RNC1). In the state of the art, the request includes those attributes of the PDP context that are required to establish the radio bearer in line with the quality of service negotiated in the PDP context. The transmission of the data packets in the radio link to the user equipment is for example carried out by a node B (NB) that is controlled by the RNC using radio resource control signaling (13). It is also possible to integrate the functionality of node B and the RNC into a single node. Node B comprises a scheduling entity (SE1 - Planning Entity 1) that distributes the packets that arrive in the available resources, for example in shared channels

or dedicated Since the handling of the packets by the planner is important for the quality of service of the transmission, the planner must also be controlled according to the quality of service of the required transmission. The SGSN also sends a request (14) to an edge node of the core network, here a GGSN (GGSN1), for the establishment of a core network bearer.

The proposed method defines a number of fixed service classes and associates a pre-configured quality of service with each service class. A service class can also be associated with a type of service class. For example, a planner, for example planning entity (SE1), can distinguish three types of service classes: Signaling (SIG - Signaling), Guaranteed Bit Rate (GBR - Guaranteed Bit Rate) and Best-Effort (BE - Best Effort ). The configuration defines types of service classes and which service or service components are associated with what type of service class.

The attributes that define the quality of service relative to a particular service class are also preconfigured. Especially those attributes for the service class that apply to any user are previously configured from the operation support system (OSS) using configuration messages (GIS). Examples for such attributes are an Operator Defined Scheduling Priority (ODSP) or other attributes for the service class such as 'Committed Rate' and 'Peak Rate'. Also attributes for the service class in a specific link, for example for the radio bearer in the radio link, are preferably configured from the operating support system. The quality of service attributes that are specific to both the service class and the subscriber such as uplink GBR, downlink GBR, and minimum bit-rate MBR (uplink rate) can be uplink configured by the establishment request for a context, for example the 'RAB Assignment'. In this case, the establishment request (12) for a context comprises both an indication of the class of service and the attributes related to a subscriber. The uplink MBR is relevant to the BE service class since the radio access network, ie the uplink scheduler in Node B, carries out an uplink rate policy. The uplink MBR of quality of service attribute is optional for GBR service classes and could be applied for variable rate codecs (for example, AMR) or adaptive speed services.

The requirements for quality of service are determined by the service executed in the application layer, for example in an application client executed in the user equipment. For example, a conversation call on a telephone has strict delay requirements while a real-time transmission service with temporary packet memory can tolerate moderate delays and fluctuation. Accordingly, either the service entity (AF) or the user equipment (UE1) specifies the quality of service required, for example by marking the sent data packets or selecting a flow for the packets. The operator can then configure service classes that refer to selected package markings or flows with selected properties. Examples of services that an operator may want to offer could be Internet access with low data speed, Premium Internet access with high data speed, voice over IP telephony including Emergency Calls or “Share” that allows users to share media during an ongoing phone call.

In at least some of the nodes involved, in the previous example in the SGSN and in the RNC, the handling of data packets is carried out based on the flow of which the packets are part. A flow, for example an IP flow, is normally defined by 5 parameters, ie source and destination addresses, source and destination port numbers and protocol identification. It is not, however, mandatory to use all parameters in the definition of flow management. In IP networks, it is also possible to handle data packets according to the so-called Differentiated Services Code Point (DSCP - Differentiated Services Code Protocol, which is transported in the IP header. However, the DSCP is not part of the definition of a flow and many nodes in a mobile telephone communication system are not able to evaluate the respective header fields, for example because they handle packets after fragmentation or encryption carried out during the transmission.

Therefore, the service class preferably identifies flows or aggregates of flows that are associated with the same quality of service. The number of service classes defined by an operator corresponds to the granularity of the service differentiation that the operator wishes to achieve. Often, 4 to 8 classes of service are suitable to allow both a sufficient differentiation of service and to allow a simple system configuration. In other cases, a high number of service classes is more appropriate to allow for better service differentiation. The treatment of packets by different nodes can also be defined differently. For example, some nodes, for example the GGSN, may be adapted to handle packets for each kind of service in a different way. Other nodes can map several classes of service to the same attributes, ie packet handling, for example if the quality of service depends primarily on a single parameter such as the planning priority.

Different applications can be executed simultaneously on the user equipment (UE1) and these applications can exchange data packets with different service entities. Different service entities can send data packets with different user equipment as destinations to the same edge node of the mobile telephone network. Packet filters in the user equipment to distinguish data packets related to different applications and in the edge node to distinguish data packets corresponding to service entities ensure that the packets are sent to the correct destination. Also packages for the same destination can correspond to different kinds of service and need to be associated by the package filters accordingly.

For example, an application layer service, provided from the service entity (AF), for example, directly by the operator or by matching with a service provider, may include multiple service components each associated with a particular flow An operator policy can define that each of the flows must be associated with a different quality of service. For example, an IMS (IP Multimedia Subsystem - IP Multimedia Subsystem) service may comprise a signaling flow of the Session Initiation / Session Description Protocol (SIP / SDP - Session Initiation Protocol / Session Description Protocol), a flow for voice, a flow for video and a flow for file sharing, each being associated with a different quality of service. Alternatively, some or all flows can be multiplexed on the same quality of service, for example SIP / SDP together with VoIP.

Therefore, a packet filter preferably filters a stream of data packets with one or potentially multiple streams or service classes of potentially multiple applications or service components on separate streams, that is, associates packets with a particular class of service, or associate packages with a particular PDP context. A packet filter can be defined by a so-called Traffic Flow Template (TFT) that applies to uplink or downlink. For the downlink, 3GPP specifications also define a PCC (Policy Charging and Control) filter that can be used instead of a downlink TFT.

The flow between the user equipment and the edge node of the core network, for example a GGSN, is mapped and potentially multiplexed together with other flows over a dedicated logical tunnel. In the non-access layer between the user equipment and the core network, the tunnel is represented by a PDP context while in the access layer between the user equipment and the radio access network is represented by a radius bearer (RB - Radio Bearer). The quality of service for a flow in the tunnel is specified by the class of service that is associated with each PDP context and the corresponding radio bearer, that is, there is a one-to-one relationship between the PDP context and the bearer of radio. Although a tunnel only refers to a single class of service, it can accommodate multiple flows that each have different quality of service requirements if they are not distinguished by packet filters. For example, the "Internet Access" service can carry streams of various applications such as Skype and FTP (File Transfer Protocol).

Inside the tunnel, packet markings do not need to be considered. Mapping a flow over a tunnel represented by the PDP context pair and the radio bearer with the associated service class to associate a service quality flow, packet marking between the user equipment and the GGSN to associate the flow With the quality of service is not required. This allows smooth infrastructure migration deployed based on existing 3GPP specifications. However, packet marking can still be used outside the tunnel to signal the specific quality of service requirements for a packet, for example to the user equipment or an edge node of the core network. Accordingly, between classification networks that are interconnected from different operators, or between the GGSN and the service entity, packet classification and marking functions could be used.

Packet marking is an option to point to the edge node of the core network whose service class will be used for data packets. The edge node can then select the service class according to this. The edge node can also carry out other functions related to quality, for example speed policy or admission control in order to avoid congestion in the mobile telephone network.

Service differentiation allows an operator to control the distribution of network resources among the services provided. The differentiation of services is, for example, achieved through package planning based on the priority between packages of different service classes. In contrast, user differentiation allows you to control the allocation of network resources to a specific subscriber. For example, if two users have contracted both the "Internet Access" service, one may have hired a "standard option" with a lower bit rate (for example uplink / downlink = 200/100 kb / s) while the other subscriber may have contracted a "Premium option" with a higher bit rate (for example uplink / downlink = 500/250 kb / s). User differentiation is achieved, for example, by means of PDP or radio bearer context rate policies on an uplink and / or a downlink. In general, user differentiation can also be carried out by reserving service classes for different groups of users and attributing users to service classes depending on their contract alternately from or considering a requested quality of service.

The proposed method provides an effective but simple solution to carry out service differentiation by introducing a fixed number of service classes, each being associated with a defined quality of service, that is, pre-configuring the quality of service that is associated with each service class, for example through operator policies, and to control the distribution of network resources among the service classes of the services provided.

Preferred embodiments do not require changes in the protocols and attributes defined in the current 3GPP specifications. It is therefore possible to use existing protocols and attributes to allow an operator to provide service differentiation based on service classes. The proposed method can also reduce the time to market for the deployment of the operator of new services because today a new service with a defined bit rate requires a new radio access bearer if the required speed is not standardized. Using service classes, operators can use reserved service classes to test new services without standardization.

The current 3GPP specifications do not provide service classes so that in view of the number of quality of service attributes and the corresponding value ranges, the number of service classes that can be defined is huge. This makes it difficult for an operator to control the distribution of network resources among the service components of all services provided and constitutes an obstacle to providing service differentiation.

Operator control over service differentiation can be carried out by means of network-controlled procedures to control the establishment of tunnels, the assignment of a class of service per tunnel and the multiplexing of flows over a tunnel in the direction of uplink and downlink In the prior art, only the versions controlled by the user equipment of these procedures are specified. As described above, there are two mechanisms for an operator to control the distribution of network resources among different classes of service. The operation support system can configure the quality of service attributes that apply to all users. In addition, the context-setting request may specify attributes that are specific to both the service class and the subscriber, for example the 'RAB Assignment' to control the uplink GBR, the downlink GBR and the MBR of uplink In current mobile telephony systems, the 'RAB Assignment' for a radio access bearer activates admission control and results in session management signaling to the user equipment. This prevents the possibility of the network pre-establishing a GBR radio access bearer without reserving resources to reduce establishment delays although this is an option to reduce establishment delays. It is also not possible to activate the admission control without session management signaling associated with the user equipment. These options are allowed by the proposed method. Finally, the invention is also applicable for service control in the service layer (for example, IMS) and convergence in fixed and mobile telephony.

The proposed method is especially suitable for shared channels on the radio link (for example in HSPA and Super3G). Some operators provide each service from a separate access packet network. In this case, multiplexing is facilitated in the radio access network of PDP contexts with the same kind of service but from different packet access networks in the radio link. However, the use of dedicated channels is also possible for the proposed method. Preferably, the radio access network is free to decide whether to make a radio bearer using a shared or dedicated channel if the required quality of service can be ensured in both ways. It is possible that a user team has different contexts of the same kind of service. Preferably, a planner in the radio access network then maps all the context flows over a single link layer flow (MAC - Medium Access Control) in the radio link.

In a preferred embodiment, combinations of service quality attribute values ​​define the service class represented by the service class identifier (ID). Since there is no service quality attribute that defines the service class identifier in the current 3GPP specifications, combinations of existing service quality attribute values ​​define the new service class identifier of service quality attributes. In other words, the values ​​of quality of service attributes are re-interpreted as encoders of a service class.

This means that individual attribute values ​​for quality of service are not used directly to specify packet handling but are reinterpreted as a constituent part of the service class definition. When a request for radio bearer assignment is received, the access node, for example the RNC, initially ignores those quality of service attributes that do not define the service class. Thus, arbitrary values ​​for, for example, uplink GBR (guaranteed bit rate), downlink GGBR and uplink MBR (minimum bit rate) can be used in the allocation request if these attributes are not evaluated. However, if the attributes that define the service class indicate a service class in which user-specific attributes are considered, attributes that do not define the service class are further evaluated to specify the specific quality of service for the user.

As an example, for a PDP context a class of service can be an indicator of the quality of service received by the flows that are multiplexed in the core network. In the same way, a radio bearer is associated with the class of service. Within the radio access network, the radio bearer represents a tunnel per user equipment that is associated with a corresponding PDP context. A user equipment can have multiple tunnels simultaneously, that is, combinations of PDP and radio bearer context.

In an example shown in Table 1, the second column indicates an adequate coding of the service class. For example, the combination of the "interactive", "signaling indication = Yes" and "THP = 1" attributes is interpreted to specify the service class ID "3". Consequently, the quality of service attribute 'Traffic Handling Priority (THP) is not interpreted as a planning priority. On the contrary, the Priority of Planning Defined for the Operator is read from the line of the table corresponding to the service class ID “3” and used to control the handling of packages. Likewise, the service quality attribute 'Signaling Indication' is not interpreted as specified in 3GPP TS 23.107, but is used to define the service class.

Table 1: Example of encoding Service Class IDs and associating radio access bearers with QoS attributes and a radio bearer embodiment.

<dl><dt>ID of </dt><dd>Codification Type Priority of GBR of MBR of HS-</dd></dl>

<dl><dt>Kind of </dt><dd>n of the ID of from Planning link link DSCH / E-</dd></dl>

<dl><dt>Service </dt><dd>Class of Service using existing QoS attributes Flow Class defined for the Operator (ODSP) as assigned from the OSS uplink and downlink GBR ascend entity DCH RB Performance (Association to Service class ID assigned from OSS </dd></dl>

<dl><dt>1 </dt><dd>“Online conversation” GBR two Assigned through 'RAB Assignment' Assigned or through 'RAB Assignment' (Option to) “RLC / UM” + “optimization for VoIP” (if necessary) </dd></dl>

<dl><dt>2 </dt><dd>"Real-time broadcast" GBR 3 Assigned through 'RAB Assignment' Assigned or through 'RAB Assignment' (Option ”RLC / AM” + optimization for Video ”(if necessary) </dd></dl>

<dl><dt>to the) </dt>&lt;dd /&gt;</dl>

<dl><dt>3 </dt><dd>"interactive" S.I.G 1 N / A Assign “RLC / AM” </dd></dl>

<dl><dt>+ “Signaling indication = Yes” + “THP = 1” </dt><dd>or through 'RAB Assignment' + “Optimization for GIS (if necessary) </dd></dl>

<dl><dt>4 </dt><dd>"interactive" BE 4 N / A Assign “RLC / AM” </dd></dl>

<dl><dt>+ “Signaling indication = No” + “THP = 1” </dt><dd>or through 'RAB Assignment' + “Non-specific optimization ions” (main </dd></dl>

<dl><dt>TCP / IP mind) </dt>&lt;dd /&gt;</dl>

<dl><dt>5 </dt><dd>"interactive" BE 5 N / A Assign See </dd></dl>

<dl><dt>+ “Signaling indication = No” + “THP = 2” </dt><dd>or through 'RAB Assignment' Flow Class 4 </dd></dl>

<dl><dt>6 </dt><dd>“Interactive” + “indication of BE 6 N / A Assigned or through See Flow Class 4 </dd></dl>

<dl><dt>signaling = No ”+“ THP = 3 ” </dt><dd>'RAB Assignment' </dd></dl>

<dl><dt>7 </dt><dd>"background" BE 7 N / A Assigned or through 'RAB Assignment' See Flow Class 4 </dd></dl>

The other columns in table 1 specify the parameters of the radio bearer for the access node for the specific service class. Correspondingly, table 1 consists of two sub-tables, specifying the columns on the left encoding the service classes and specifying the other columns

5 the adjustment of the attributes according to the service classes. Table 2 shows another example of coding service classes. Here, another parameter (Allocation / Retention Priority ARP) is used in the definition of the service class to allow a greater number of service classes to be defined. A corresponding table also defines the attributes related to each class

10 of service but it is omitted here for simplicity.

Table 2: Another example of encoding Service Class IDs.

<dl><dt>Service Class Identifier </dt><dd>Service Class Coding </dd></dl>

<dl><dt>0 </dt><dd>Reserved for signaling </dd></dl>

<dl><dt>1 </dt><dd>“Interactive” + “signaling indication = Yes” + “THP = 1” </dd></dl>

<dl><dt>2 </dt><dd>“Conversational” + “ARP = 1” </dd></dl>

<dl><dt>3 </dt><dd>“Conversational” + “ARP = 2” </dd></dl>

<dl><dt>4 </dt><dd>“Conversational” + “ARP = 3” </dd></dl>

<dl><dt>5 </dt><dd>“Real-time transmission” + “ARP = 1” </dd></dl>

<dl><dt>6 </dt><dd>“Real-time transmission” + “ARP = 2” </dd></dl>

<dl><dt>7 </dt><dd>“Real-time transmission” + “ARP = 3” </dd></dl>

<dl><dt>8 </dt><dd>"Interactive" + "signaling indication = No" + "THP = 1" + "ARP = 1" </dd></dl>

<dl><dt>9 </dt><dd>"Interactive" + "signaling indication = No" + "THP = 1" + "ARP = 2" </dd></dl>

<dl><dt>10 </dt><dd>"Interactive" + "signaling indication = No" + "THP = 1" + "ARP = 3" </dd></dl>

<dl><dt>11 </dt><dd>"Interactive" + "signaling indication = No" + "THP = 2" + "ARP = 1" </dd></dl>

<dl><dt>12 </dt><dd>"Interactive" + "signaling indication = No" + "THP = 2" + "ARP = 2" </dd></dl>

<dl><dt>13 </dt><dd>"Interactive" + "signaling indication = No" + "THP = 2" + "ARP = 3" </dd></dl>

<dl><dt>14 </dt><dd>"Interactive" + "signaling indication = No" + "THP = 3" + "ARP = 1" </dd></dl>

<dl><dt>15 </dt><dd>"Interactive" + "signaling indication = No" + "THP = 3" + "ARP = 2" </dd></dl>

<dl><dt>16 </dt><dd>"Interactive" + "signaling indication = No" + "THP = 3" + "ARP = 3" </dd></dl>

<dl><dt>17 </dt><dd>“Background” + “ARP = 1” </dd></dl>

<dl><dt>18 </dt><dd>“Background” + “ARP = 2” </dd></dl>

<dl><dt>19 </dt><dd>“Background” + “ARP = 3” </dd></dl>

Fig. 3 also illustrates the setting of attributes for the quality of service in an access node. In a first stage of receiving (31), the access node receives a set of attributes to establish a context. The access node then determines (32) then the service class encoded by the attributes using a function

5 of mapping. More specifically, the service class is an indicator of a different set of attributes, which is selected (33) from a plurality of sets (34) specified above during an access node configuration procedure. The access node then carries out a control (35) on the transmission of data packets associated with the context according to the set of

10 selected attributes. If the setting request for a context comprises both an indication of the service class and attributes related to the subscriber, those attributes related to the service class are used to select the set of attributes while other user-related attributes can be used to modify the set, that is to carry out user differentiation.

Preferably, all entities that control packet handling, for example the radio access node, SGSN, GGSN, or a PCRF (Policy and Charging Rules Function) are adapted to handle classes of service . In one option, the PCRF (15) has interfaces to the service entity (AF) and the GGSN (GGSN1). Accordingly, you can point out the attributes that encode the service class and the GGSN when the edge node transparently sends the encoded service class. Alternatively, the PCRF (15) only indicates the numerical value of a service class ID and the GGSN then maps the ID to the corresponding coding. The service layer, for example the PCRF, defines the mapping between the service components provided by the operator to the service classes, and the mapping of service classes to the service class types (for example GBR, SIG or BE ). Admission control may be limited to GBR flows of service class type.

Figure 4 shows a method according to the invention. The method specifies the quality of service in a transmission of data packets between a service entity and a user equipment over a mobile telephone network comprising a core network node and an access node that is adapted to control the handling. of packets data sent between the service entity and the user equipment.

In a first stage (41) a transmission is initiated between a service entity and a user equipment. In a selection (42), a service class is selected for the transmission of a plurality of service classes relative to the quality of service that is preconfigured. After selection (42), the selected service class is communicated to the core network node. The establishment (43) of a first transmission context is initiated between the core network node and the user equipment. The first transmission context is associated with a first set of attributes to define a first quality of service for the exchange of data packets with the user equipment in the transmission. The first set is determined in a mapping operation (44) using a single mapping function that maps the service class to the first set of attributes. After the establishment procedure (45) is completed, the transmission control (46) is carried out in accordance with the transmission context. In this way, the selected service class defines the handling of the data packets as they are controlled by the core network node, that is, the control (46) of the transmission is carried out using the first set of attributes that It is specified in the mapping operation (44) according to the service class.

Second establishment (47) of a second transmission context between the access node and the user equipment is also initiated. Typically, the core network node activates the second establishment (47) after the first establishment has ended although this order of establishment is not essential. The second transmission context is associated with a second set of attributes that define the quality of service for the exchange of data packets with the user equipment in the transmission as controlled by the radio access node. In another mapping operation (48), the second set of attributes is determined in a single mapping function from the selected service class. After the context establishment (49) is completed, the selected service class accordingly defines the handling of the data packets controlled by the access node in the transmission control (50).

Fig. 5 shows a device according to the invention, for example a Serving General packet radio service support node SGSN - Radio Service Support Node in General Server packages, a Gateway General packet radio service support node GGSN - Radio Service Support Node In General Gateway packets, an enhanced general packet radio service support node, an RNC radio network controller - Radio Network Controller, a base station controller, a base transceiver station or a B node. The device has a memory unit (MEM) to store different sets of service attributes. It also has an input / output unit (IO - Output Input) to receive and send data packets and receive messages specifying a service class selected for a carrier. A processing unit (PU - Treatment Unit) is adapted to carry out a method as described above. Especially, the processing unit can determine the service class selected from a received message and establish a corresponding bearer to another device that handles the packets. According to the class

Once the selected service is selected, the treatment unit obtains a corresponding set of attributes from the memory (MEM) and initiates a control unit (CU - Control Unit) adapted to handle packets sent in the tunnel according to the attributes. The control unit (CU - Control Unit) then handles the packets according to the attributes obtained from the memory, that is to say according to the selected service class. In a preferable embodiment of the invention, the processing unit (PU) determines the selected service class of a set of attributes in the received message. The output input unit (IO) also allows the exchange of messages with an operation support system (OSS1 - System

10 Operation Support 1) to configure the device, especially to define and update the attribute sets in memory (MEM). The units of the control device can be made as electronic or optical circuits or as software executed in such circuits.

The above embodiments admirably reach the objects of the

fifteen invention. However, it should be noted that experts can make variations without departing from the scope of the invention that is limited only by the claims.