Patent Application
20050181782
The present invention relates to a method according to the preamble of claim 1. Devices and software programs embodying the invention are also described.
Both for mobile telecommunication networks, e.g. UMTS (Universal Mobile Telecommunication System) or GSM (Global System for Mobile communication) networks, and for fixed networks, the Bearer Independent Call Control (BICC) protocol is used to control the set-up and establishment of connections, e.g. telephone calls, independent of the underlying bearer technology for handling the payload of the connections.
BICC Capability Set 3 is introducing a priority service as specified in ITU-T Recommendation I.255.4, Integrated Services Digital Network (ISDN) General Structure and Service Capabilities, Priority Service. When congestion is encountered, priority connections shall be given preferential treatment within the network for the priority service. Such preferential treatment can especially include the functions of an extended waiting time before time-out of network units handling the connection and/or waiting for a line on the route to become free.
The priority of a connection can be indicated by a corresponding field in the initial message for the connection set-up, e.g. an initial address message (IAM).
For example, some or all bits of an octet in the IAM can be designated for this purpose.
Under normal conditions when the network is not congested and the nodes have the necessary resources to complete all connections, the connection establishment is processed without special treatments. At least one node processing, i.e. controlling or handling the connection attributes a timer to it, which ensures that resources reserved for said connection are released should the set-up fail. A node controlling the connection can be for example an MSC (Mobile Services Switching Center) or a transit-switching node (TSN) connecting a mobile network to a fixed network. The bearer can either be handled in the same node or in a separate bearer node, e.g. in a media gateway (MGW), which is at least partly controlled by the controlling node.
In case of network congestion when the nodes do not have sufficient resources to complete all incoming connection establishment requests, the nodes can give preferential treatments based on the priority level. The preferential treatment can especially include access to reserved network resources, e.g. only the highest priority connections are given access to the resources reserved for highest priority connections while the second priority connections are given only access to resources reserved for the second priority connections, excluding the resources reserved for the highest priority connections. Further priority levels are possible, connections for each level having only access to resources reserved for the same or a lower level.
Allocation of reserved network resources to specific priority levels is implementation specific. If a node cannot complete a high priority connection establishment request even after application of the preferential treatment, it clears the connection.
Nodes with interaction to the served user, especially the originating control or bearer node in a connection, can receive the connection priority information from the user. The originating node then forwards the information to further nodes in the connection. For example in BICC, when an outgoing protocol entity instance of a bearer node receives the connection priority information along with a request for a new connection from a nodal function, the connection priority parameter is included in the ERQ message (Establish Request). If a connection priority parameter is received in an ERQ message by an incoming protocol entity, the incoming protocol entity instance passes the connection priority information to the nodal function along with the request for a new connection.
The priority information can be used in the nodes in different ways. For example, when the nodal functions select a route at the originating endpoint of the bearer, priority information can be used to select a route that has sufficient resources to the succeeding bearer node or directly to the terminating user. For bearer nodes without interaction to the originating user of a connection, the nodal functions can also check the availability of resources in the incoming bearer path. The connection priority information can further be used to allocate internal resources of the bearer node to establish an internal path for the connection from the incoming bearer path through the bearer node to the outgoing bearer path, either to a subsequent bearer node or to the terminating user of the connection.
The above specifications for a telecommunication network allow to handle connections with the corresponding priority in all nodes. However, this does often not ensure a successful establishment of priority connections. Especially, it is possible that different nodes use different mechanisms for ensuring connection priority, especially if nodes in different parts of a network are updated at different times, e.g. by different operators. In this case it is possible that a connection is queued in a first node, i.e. waiting for free resources to become available in said or another node, while a timer for said connection expires in a still further node. Even if each node handles the connection with priority, it is therefore still possible that a connection is not successfully established. It is also possible that the user requesting the connection terminates the request if the establishment takes an excessive time due to queuing.
It is an object of the present invention to obviate the above disadvantages and provide a method, a communication system and a node, which allow a safe establishment of priority connections.
According to the invention, the method described in claim 1 is performed. Furthermore, the invention is embodied in a communication network, a node and a software program as described in claims 9 to 11. Advantageous embodiments are described in the further claims.
An advantageous method for establishing a priority connection in a telecommunication network with a plurality of nodes for processing the connection is described. The processing can be a handling or a controlling of the connection. The nodes can accordingly be control nodes or bearer nodes or nodes handling both control signaling and bearer establishment. The nodes comprise resources, which can be attributed for handling or controlling the connection. For each connection request, resources are attributed in the node. The resources can for example comprise circuitry to process the connection, allocated memory or processing intervals or software identifiers like addresses.
At least one timer in the node supervises the connection establishment. It is possible that more than one timer is attributed to a connection, e.g. for supervising different aspects of the establishment of said connection relating to the control signaling and the bearer establishment. An expiry of any of said timers releases the resources attributed to the connection.
For every timer, an expiry value is specified to avoid the blocking of resources if the request for connection establishment fails. When a first of said nodes receives a request to set up a priority connection, the first node selects a first of said timers according to an expiry value specified for priority connections, generally being different from the expiry value of the corresponding timer for regular connections. In a preferable embodiment of the method, the period for timer expiry is increased compared to a regular connection, i.e. a later timer expiry is specified for a priority connection. It is possible, that different timers attributed to a priority connection have different expiry values.
The first node sends an indication of the connection priority and a queuing indication to a further node controlling the connection. The queuing indication is a notification that at least one special expiry value was selected for this connection. The queuing indication can be for example an indication, that a queuing of priority call attempts can be accommodated by the preceding nodes, a flag that at least one timer was extended, or an information specifying which timers were selected, optionally including also an indication of the amended expiry value. The further node checks whether an indication of a connection priority and a queuing indication is received. The further node selects, according to the result of the check, at least one further of said timers according to an expiry value specified for priority connections, i.e. the expiry value for priority connections is selected if a queuing indication and a priority indication is received. For both the first and the further node it is possible, that the selection of the timers and corresponding expiry values depends on the value of the connection priority indication.
The proposed method enables to extend network timers without the danger of an unintended release of the connection. The method is easy to implement and can flexibly be adapted to different queuing requirements. The method is especially suitable for use within a BICC network.
In a preferable embodiment, the request is the initial message to set up a connection and said initial message comprises a data field indicating a priority class for the connection. The initial message is for example an IAM. The data field or specified values or selected bits of the data field can be used to indicate different levels of connection priority, e.g. an emergency call.
Said plurality of nodes can comprise control nodes controlling the connection, e.g. an MSC server, and bearer nodes adapted to establish a bearer for the connection, e.g. an MGW. It is, however, also possible that a node both controls the connection and establishes bearers like a customary MSC.
It is correspondingly possible that the first and the further node both control the connection, that both nodes handle the bearer for the connection or that one node controls the connection while the other node handles the bearer. Therefore different queuing indications in the control level, in the bearer level and between both levels for the connection can be distinguished. In an advantageous embodiment, the queuing indication is contained in a group comprising a first queuing indication for messages between nodes controlling the connection, a second queuing indication for messages between nodes handling the connection and a third queuing indication for a message between a node handling the connection and a node controlling the connection. It is either possible to amend existing messages in the used protocols to comprise these indications or to define new messages for this purpose.
In order to avoid a premature release of the connection by the user, the further node replies to the queuing indication with a connection progress notification. The connection progress notification can be forwarded by the nodes processing the connection in the direction of the party requesting the priority connection. It is possible that the sending of this message is delayed by a specified period of time. Especially, the connection progress notification can initiate the sending of a tone or an announcement to the party requesting the priority connection.
Often different nodes will handle or control a connection, wherein some of the nodes are adapted to the queuing of connections while others of the nodes are not. In this case, the further node sends a first request to set up the connection and the indication of the connection priority and the queuing indication to a third node processing the connection, preferably all in a single message. The third node then rejects the queuing indication as unsupported to the further node. The further node waits for a confirmation of a bearer establishment for said connection and sends a further request to set up the connection and a further indication of the connection priority to the third node, if a release message is received from the third node for said connection. In this way, an establishment of priority connections can also be ensured if a node in the connection is not adapted to the queuing of connections.
In the proposed method, it is possible to specify an infinite expiry time for a priority connection. In this case, the selection of the first or the further timer can be a decision that no timer for releasing the resources is attributed to a priority connection. It is of course also possible to use extended timers for supervising one aspect of priority connections while supervising a different aspect only for regular connections, i.e. to attribute a reduced number of, optionally extended, timers to priority connections. If an infinite expiry time is specified, preferably a mechanism to detect and release unsuccessful connection establishment requests is implemented in the communication system to avoid the blocking of resources.
A preferable node for processing connections in a telecommunication network with a plurality of nodes comprises resources, which can be attributed for processing the connection and at least one timer for supervising the connection establishment. An expiry of the timer initiates a release of the resources attributed to the connection. The node comprises also interfaces to receive a request to set up a priority connection and to send messages to further nodes, i.e. transmission and reception units for sending and receiving messages. The node comprises a processing system which is adapted to select the at least one timer according to an expiry value specified for priority connections. The node is also adapted to send an indication of the connection priority and a queuing indication being indicative of the timer selection to a further node processing the connection.
The proposed method can also be embodied in a program unit adapted for execution in the processing system of a node for processing connections in a telecommunication network with a plurality of nodes. The program unit comprises code for attributing resources to the connection for processing the connection and at least one timer for supervising the connection establishment. An expiry of the timer initiates a release of the resources attributed to the connection. The program unit is adapted to select the at least one timer according to an expiry value specified for priority connections. The program unit is also adapted to send an indication of the connection priority and a queuing indication for the timer selection to a further node processing the connection. The steps performed by the processing unit can be executed by function calls to an operating system or commands to hardware units in the node.
The program unit as well as the node can perform any steps relating to the respective node in the method described above. A program unit according to the invention is for example stored on a data carrier or loadable into the operating system of a node, e.g. as a sequence of signals.
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.
According to ITU_T specifications different timers are used for the supervision of a connection establishment. All timers ensure, that resources reserved for a connection during the set-up and establishment are released in case of a failed connection attempt. Timer t1, also denoted Timer_ERQ, ensures that a reply to a bearer establishment request is received. A typical time-out value is 5-30 seconds. The timer t1 is started when an ERQ message is sent and stopped at the receipt of an ECF (Establish Confirm) message. If timer t1 expires, all resources and the connection are released and a release message (REL) is sent. Timer t1 is the bearer establishment supervision timer for the Q.2630.1 bearer control protocol as defined in ITU-T Recommendation Q. 2630.1 (12/99), AAL Type 2 Signaling Protocol (Capability Set 1). Other bearer control protocols have similar supervision timers that can initiate premature connection release. Throughout the text these timers are generically referred to as “bearer establishment supervision timers”.
A timer denoted T7 ensures that the network has enough routing information to route the connection to the called party. A typical time-out value is 20-30 seconds. Timer T7 is started in the control level when the latest address message is sent and stopped when the condition for normal release of address and routing information is met, e.g. the receipt of a message ACM (Address Complete Message) or CON (Connect). At expiry, a release procedure for the resources attributed to the connection is initiated.
In the connection control level, a timer denoted T8 ensures the supervision of the bearer path establishment. A typical time-out value is 10-15 seconds. Timer T8 is started when the CSF (Call Service Function) at a Serving Node (SN) receives an IAM indicating that a continuity message (COT) is to be expected and stopped at receipt of the COT. At expiry, a release procedure for the reserved resources is initiated.
In the following, AAL2 (ATM Adaptation Layer Type 2) bearer technology utilizing Q.2630.2 (ITU-T Recommendation Q. 2630.2 (12/2000), AAL Type 2 Signaling Protocol (Capability Set 2)) as bearer control protocol is used as an example for a connection bearer. However, it should be noted that any other suitable bearer protocol, e.g. the Internet Protocol IP, can be used and also a different bearer control protocol compared to the examples.
The upper parts of all figures indicate nodes for processing the connection while the lower part of the figures shows examples of messaging sequences between the nodes. Corresponding reference signs in different figures relate to corresponding features. All figures in this specification show gateway serving nodes 1, 2 (GSN) comprising both connection control entities designated CSF and bearer handling units denoted BCF (Bearer Control Function). However, it is also possible that control and bearer handling entities are implemented in different nodes.
In a BICC network, several network timers need to be extended in order to allow a priority handling of connections by queuing.
Corresponding timers T7 and T8 are started in the control level at times ST7, ST8, respectively. Both nodes 1,2 send also ADD requests to the bearer layer to initiate the bearer establishment. In the bearer level, establishment requests indicating also a priority connection are used to initiate the bearer establishment. A congestion CG is encountered at an intermediate AAL2 bearer switching node 3 with a Bearer Control Function (BCF) designated BCF-R. If the connection establishment request is queued waiting for resources to become free in BCF-R, the timer t1 in BCF-G (z) of node 2 will expire at time TO1 and the connection will be cleared by a corresponding notification message for BNC (Backbone Network Connection) release to the control level in node 2 and a subsequent release message (REL) to all preceding nodes controlling the connection. This is obviously an undesirable result for a connection that has high priority. If either timer BICC T7 or BICC T8 expired instead of timer t1 because they may be set to shorter timeout values, the result would be the same, i.e. the connection is released.
Several approaches can be taken to avoid that network timeouts occur. Either the BICC T7, T8 and bearer establishment supervision timers are not started for priority connections. Alternatively, the expiry values or BICC T7, T8 and bearer establishment supervision timers are extended or the timers are replaced by timers having longer time periods for priority connections. The proposed method can use either alternative or both if several timers are selected. When timers are inhibited, the connection establishment is not being time supervised and hanging resources must be avoided by an appropriate procedure. Therefore, extended timers are generally preferable.
If queuing is supported in the originating node, then the timers are extended or inhibited and a queuing indication is set. In all further nodes that support queuing, the corresponding timers should also be extended or inhibited. This is shown in
For this reason, there needs to be also an indication that timers have been extended or inhibited. This indication tells subsequent nodes involved in the connection establishment that queuing of priority connections can be accommodated.
When node 2 in
The changes required for the extension or inhibition of network timers and the queuing indication need to be introduced into all bearer control protocols that are to support the queuing of bearer establishment requests and especially into the BICC protocol. Either new or a modified messages in the bearer control protocols can indicate that a connection request has been queued. Also the protocol used between the bearer level and the connection level must be updated with a new event to notify the connection level that queuing of the bearer establishment request has occurred.
With these enhancements, a BICC network is able to tolerate queuing of priority connection attempts without premature release. However, subscriber behavior must also be taken into account. If the calling subscriber does not receive any progress information within a reasonable time during the connection establishment, i.e. if there is an excessive post-dialing delay, the calling subscriber might get impatient and hang up.
The calling subscriber is expecting to receive a connection-progress notification. This can be for example a tone, an announcement or a message. When a node decides to queue a priority connection, a mechanism is required to trigger the sending of this connection progress information. In BICC networks, connection progress information can be sent using out-of-band signaling between the controlling nodes. In order to know when to send this information, the control layer needs an information whether the establishment attempt is being queued.
If the control layer queues the establishment request as in
It is also possible that the bearer layer queues the establishment request as shown in
Instead of mapping the indication 6 that queuing has occurred to a tone in node 2, a corresponding indication can be passed backwards toward the calling party. This indication can then be either passed to the terminal of the user or mapped to a tone or announcement by any of the preceding nodes that have the capability to do so. Also, a notification does not need to be sent immediately when the connection is first queued, but instead it can be delayed by a defined time, e.g. X seconds after the queuing is performed. Of course, the latter alternatives are possible for any congestion, e.g. for a congestion encountered on the control level as described with respect to
The above embodiments admirably achieve the objects of the invention. However, it will be appreciated that departures can be made by those skilled in the art without departing from the scope of the invention which is limited only by the claims.
| Filing Document | Filing Date | Country | Kind |
|---|---|---|---|
| PCT/EP02/01613 | 2/15/2002 | WO |
