WO2007113516A1

WO2007113516A1 - Routing communications to devices with likely presence of user

Info

Publication number: WO2007113516A1
Application number: PCT/GB2007/001175
Authority: WO
Inventors: Robert John Collingridge; Philip Martyn Clarke
Original assignee: British Telecommunications Public Limited Company
Priority date: 2006-03-30
Filing date: 2007-03-30
Publication date: 2007-10-11

Abstract

A method of selecting a communications device associated with a address identifier for receiving a communication addressed to the address identifier, the communications device being one of a plurality of devices associated with said address identifier, the method comprising the steps of: retrieving device activity information associated with each one of said plurality of devices when a communication for said address identifier is received; extrapolating said device activity information for each device using one or more resource availability rules to determine if one or more of said plurality of devices meets one or more criteria for the availability of one or more resources on the device for receiving said communication; and, in the event at least one criterion is met, selecting one or more of said communications devices according to said extrapolated device activity information, wherein said extrapolated device activity information provides an indication of the probability of one or more parameters being availability on one or more of said communications devices for the communication.

Description

ROUTING COMMUNICATIONS TO DEVICES WITH LIKELY PRESENCE OF USER

The present invention relates to a routing scheme for a communications system. In particular but not exclusively, the invention relates to a routing scheme for a communications system in which communications are addressed to an address-identifier which is associated with a plurality of different communications devices.

The association of a single address-identifier (such as, for example, a user or subscriber identity) with one or more ongoing communication sessions is already known in the art. It is possible for communication devices to operate in a communications system. in a session context such as that supported by SIP, for example, VoIP communications as well as other data communications can be supported by SIP. When a communication is addressed to an address-identifier (e.g. a unique user identity or other form of unique address) and not to a particular device address, it can be problematic to control which devices associated with that address-identifier respond to the communication. Known ways to address this problem include the user of the device configuring each device or alternatively network equipment can be configured to ensure that the communication is selectively delivered to a only one or more of the communications devices associated with that address-identifier.

In such communications systems, a calling party has no ability to control how their communication is delivered in such communication systems as they are only aware of the user's "identity address". This means that unless intelligent routing is implemented, if a user

■ has configured communications devices at a plurality of locations to respond to incoming communication requests, the network will utilise more network resources when signalling to each communications device the incoming communication than is necessary as the user can only be contacted at one location at that point in time.

If the calling party wants to initiate a communications call with a particular individual, to cause only the mobile communications device of that individual to ring and not other devices associated with the same address-identifier such as, for example, the land-line of the phone at that individual's home or their work phone, the network apparatus which resolves the "address-identifier" must be provided with information which allows it to implement intelligent routing to select the mobile communications device address and not resolve the address- identifier to any of the other device addresses associated with that address-identifier.

Baker in United States Patent Application number US 2005/0068166, entitled "Method and Apparatus for routing a communication to a user based on a predicted presence" describes how presence patterns can be used to provide a prediction of when a user is still present at a device even when the device has been idle for a period of time. Once a presence pattern is detected, a call that is destined for a user during the associated time interval can be automatically routed to the user at the given communication device. The presence of a user can be detected explicitly by requiring, for example, a user to affirmatively register their presence information or implicitly, for example by observing user activity, or by a combination of explicit and implicit detection means. When the user changes locations or devices, the presence patterns indicate the probability of the user being on one device decreasing whilst the probability of the user being on another device increases.

Horvitz et al in European Patent Application EP1271371 , entitled "Methods for and applications of learning and inferring the periods of time until people are available or unavailable for different forms of communication, collaboration, and information access", describes using scheduling information obtained using calendar and appointment applications collated over a long period of time to determine when a user is available.

The presence of an individual to receive a call is only one factor which determines whether a call routed to a particular communications device is going to be successfully completed on that device. Other factors include whether the device can be determined to have sufficient resources such as electrical power (for example, is the device likely to be powered by a battery or connected to mains power such as when the battery is charging) and/or processing power (as many communications devices are capable of running other applications which utilise considerable processing resources limiting their availability to the communications application running on the device) and/or any other sub-system resources (such as for example, an audio sub-system or video sub-system). Some communications device utilise network services to process communications, whereas others may be capable of receiving a communication without accessing a central service, for example, in some instances a user may use a conferencing service to receive a communications call or require a higher bandwidth connection (a network service). If the service required by the device to receive the communication is not available, then the device may not be able to receive the call or may only provide an audio communication component and not a video component of that communications call.

The present invention seeks to provide a method of determining from a communication device's activity pattern whether a communication addressed to a plurality of communication devices should be selectively routed or not routed to that communication device. The device's activity pattern is used to determine the probable availability of one or more selection criteria which are dependent on a number of parameters which affect the likely success of the communications call. Examples of factors which can impact whether a communication can be successfully completed or not include whether the communications device which receives the call has sufficient power for complete delivery of the communication, whether the communications device has sufficient processing resource free to support the call (which may be an audio and/or video call and/or which may be an encrypted call etc), as well as whether an intended recipient for the call is co-located with the device.

To implement a selective routing scheme in a communications network the communications devices themselves must be capable of reporting information relating to the parameters which impact each device's ability to process an incoming communication to the network routing system. As some communications devices are not always connected to a communications network, it is particularly useful for such communications devices if forecasts of the parameters determining routing are provided. Although such forecasting can be performed by the network rather than the device itself, if forecasts can be locally generated, it enables each device having a network presence to provide the network routing system with the current parameters which are used by the routing system to select one or more devices for receiving an incoming call. Accordingly, it is useful if the forecasts can be generated in a manner which is not processor intensive or and which does not require a high processor capability.

The invention thus seeks to mitigate and/or obviate some of the problems of the prior art by providing a routing scheme for a communications system which is capable of selecting one or more communications devices for receiving communications which are all associated with the same unique address identifier. The routing scheme seeks to selectively route communications to one or more of the plurality of devices associated with the address- identifier based on the probable availability of a local selection parameter determined by the activity recorded on device. A selection parameter provides an indication of the probable ability of the communications device to successfully deliver the communication to the intended recipient(s). The address identifier is resolved within the network to a format suitable for receipt by each device as appropriate for the communication attached. Examples of selection parameters determined by each communications device include presence and availability information for the intended recipient of the communication (who may comprise one or more individual(s) (e.g. in the case of an organisation being associated with the . address-parameter more than one individual) and/or the probable availability of one or more device resources and/or services which the device may need to access.

A device resource includes any component or sub-system (whether implemented in hardware or software) which the device uses to process the received communication, including resources such as memory, processing capacity, battery power, network connection bandwidth etc as well as those previously mentioned herein above. The term device resource here includes device service resources such as applications which are necessary to process the received communication (for example, if a user is listing to a music track on a mobile, voice calls should be diverted). Where a particular service is required to process a communication, it is helpful if the communication is forwarded only to devices which have a likelihood of the service being available. This is particularly useful where a service may become congested. The service may be accessed directly by a calling party (e.g., when a calling party dials a call-centre having an automated call response system), or indirectly (e.g., when a calling party sends someone a link to a url to see an image, or needs to use a video streaming service application which is currently in use on that device).

The invention also seeks to provide a configuration scheme for a device to enable it to select which parameters should be provided to the network to implement a routing scheme according to the invention. The aspects and embodiments of the invention are provided by the accompanying claims, and as set out below.

A first aspect of the invention seeks to provide a method of selecting a communications device associated with a unique address identity for receiving a communication addressed to the unique address identity, the communications device being one of a plurality of devices associated with said unique address identity, the method comprising the steps of: retrieving device activity information associated with each one of said plurality of devices when a communication for said unique address identity is received; extrapolating said device activity information for each device using one or more resource availability rules to determine if one or more of said plurality of devices meets one or more criteria for the availability of one or more resources on the device for receiving said communication; and, in the event at least one criterion is met, selecting one or more of said communications devices according to said extrapolated device activity information, wherein said extrapolated device activity information provides an indication of the probability of the resource availability on said communications devices for the communication.

In one embodiment, the method further comprises the step of: extrapolating said device activity information for a communications device using one or more resource availability rules which are dependent on device activity information for one or more other communications devices.

In one embodiment, an availability rule for a resource of a communications device is hierarchically dependent on one or more other availability rules for the same or one or more other resources.

In one embodiment, a device resource availability is network dependent.

In one embodiment, a device resource availability rule is dependent on one or more of the following: device activity; application activity; user activity associated with the presence of a user at the communications device; and resource activity.

In one embodiment, said extrapolated device activity information provides probability information which is time-dependent. In one embodiment, said probability information for a communications device is linearly extrapolated from historical device activity information associated with said communications device.

In one embodiment, said probability information is non-linearly extrapolated from historical device activity information associated with said communications device.

In one embodiment, said probability information is extrapolated from historical device activity information associated with said service's usage of said communications device as a function of time of day.

In one embodiment, the probability information is extrapolated as a function of calendar information associated with the service.

In one embodiment, the extrapolation is as a function of day of the month calendar information associated with the user and/or as a function of time-of-day calendar information associated with the service.

In one embodiment, for a predetermined duration of time, device activity or inactivity is recorded to provide said device activity information.

In one embodiment, said device activity or inactivity is associated with operation of a resource of the device.

In one embodiment, the operation of a resource is derived from a session-related activity of a user of the device.

In one embodiment, a resource availability rule is dependent on one or more media types and/or communications protocols.

In one embodiment, a communication comprises one of the following: a telephony call, a video telephony call, an electronic email including one or more of the following: video data, audio data, text-based data, an Instant Message™, a short message service (SMS) message including one or more of the following: video data, audio data, text-based data.

In one embodiment, said selection of one or more of said communications devices according ,to said extrapolated device activity information to provide an indication of the probability of the resource availability on said communications devices for the communication is further dependent on the required usage of said resource for said communication. In one embodiment, a resource is one or more of the following: a power resource; a processing resource; a data storage resource; an application resource; user interface resource; a media display resource; an audio resource.

Another aspect of the invention seeks to provide apparatus arranged in use to select a communications device associated with a unique address identity for receiving a communication addressed to the unique address identity, the communications device being one of a plurality of devices associated with said unique address identity, the apparatus comprising: means to retrieve device activity information associated with each one of said plurality of devices when a communication for said unique address identity is received; means to determine if one or more of said plurality of devices meets one or more criteria for being associated with the availability of a resource on the device from device activity information for each device extrapolated using one or more resource availability rules; and, in the event at least one criterion is met, means to select one or more of said communications devices according to said extrapolated device activity information, wherein said extrapolated device activity information provides an indication of the resource availability on said one or more communications devices.

In one embodiment, the apparatus further comprises means to extrapolate said device activity information for a communications device using one or more resource availability rules which are dependent on device activity information for one or more other communications devices.

In one embodiment, a said resource availability rule for a communications device is hierarchically dependent on one or more other resource rules for the same or one or more other communications devices.

In one embodiment, the resource availability is network dependent.

In one embodiment, the one or more resource availability rules are dependent on one or more of the following: device activity; application activity; user activity associated with the presence of a user at the communications device; and resource activity.

In one embodiment, the apparatus further comprises means arranged to route said communication to said selected one or more communications devices.

In one embodiment, said apparatus comprises one of said communication devices.

In one embodiment, said apparatus is implemented on remote communications processing platform arranged to communicate with any one of said communication devices using one or more communications network.

Another aspect of the invention seeks to provide apparatus arranged in use to store received device activity information collated from a plurality of communications devices, the data store comprising: means to receive device activity information from each of said plurality of communications devices; means to provide stored device activity information to the apparatus aspect.

In one embodiment, said device activity information is periodically sent to the apparatus by each communications device.

In one embodiment, said device activity information is collated by the apparatus when a communications device determines a change in a device activity state has occurred.

In one embodiment, any one of the above two apparatus aspects requests device activity information from each one of said plurality of communications devices upon receiving a communication for the unique address identity of said plurality of communications devices.

Another aspect of the invention seeks to provide a communications device arranged in use to receive a communication associated with a unique address identity, the device being selected from a plurality of devices associated with said unique address identity using the method aspect, the apparatus comprising: means arranged to receive a communications call for said unique address identity; means record device activity information; and means to provided said 'device activity information to any one apparatus aspect.

In one embodiment, the device further comprises means for extrapolating said device activity information recorded by the using one or more resource availability rules, and provides said extrapolated device activity information to any apparatus aspect.

Another aspect of the invention relates to a suite of one or more computer programmes, which when collectively executed are arranged to implement appropriate steps in a method aspect. In one embodiment, the functionality of said suite is provided in a distributed processing environment.

Another aspect of the invention relates to a service operated by a service provider, the service providing a method according to any method aspect to a user having said unique address identity.

Thus a routing scheme is provided in which each communications device according to the invention is associated with a device profile having device availability rules for determining whether the device should be used to receive communications requiring a particular device resource. A device availability rule comprises a number of resource availability rules, i.e., a device is only available if all critical resources for operation of the device are also available. A device (or equivalent^ a resource) availability rule may be dynamic or static, and is based on historical information relating to activities involving the resource(s). The rule determines how the historical information is extrapolated to a subsequent point in time. The extrapolation may be linear, but usually follows an availability rule which provides a probability for the availability of the resource which declines non-linearly with passing time (both linear and non-linear availability rules are referred to herein as a "decay" rules). Activation of a decay rule is triggered by the resources inactivity on a particular device. Different resources on the same device can have different decay rules. User-related calendar information and any other appropriate usage patterns may be used to further modify the manner in which the historical information for resource activity is extrapolated to determine the current probability of the resource being available on a particular device.

It is possible to further refine the resource activity information stored so that a communications device provides an indication of the likelihood of the resource's availability according to one or more criteria associated with a mode of operation of the device. A resource decay rule can be weighted according to the remaining availability of the resource.

One embodiment of the invention seeks to predict the probable presence of a user to be determined from the user's usage of communications devices, and on the basis of the user's probable forward communications to a selection of said communications devices. The selection of a communications device based on probable user presence is in one embodiment of the invention, also dependent on the availability of one or more services which are required by the communication, in addition to being device resource dependent.

Effectively, the selection of a communications device from a group of communications devices all associated with the same user-identity communications address, requires determining both the user's proximity to the communications devices (whose physical locations may or may not be known) and the availability of any services and/or any device resources which are required by the communication. In this way, in addition to determining service availability, a method of forwarding a communication to a communications device for which resource activity information has been collated to determine a likelihood of the resources required to implement the service on the device is provided. In this way, an indication is also provided of whether the service is likely to be available for receiving a communication on the device on the basis of whether the device has sufficient resources to receive the communication. In addition, the device should not be already being used in a way which would prevent the device from receiving the communication.

The aspects of the invention and those set out in the accompanying independent claims and the embodiments described above and by the dependent claims may be combined in any appropriate manner apparent to those skilled in the art. The preferred embodiments of the invention will now be described with reference to the accompanying drawings which are by way of example only and in which:

Figure 1 shows a number of communications devices associated with a user identity;

Figure 2 shows discrete activity information for a user identity on a communications device as known in the art;

Figure 3 shows how device activity information may be determined as the number of occurrences of certain durations of activity and/or inactivity by the user and the cumulative totals for certain durations of activity and/or inactivity;

Figure 4 shows in table form how the user's probable presence information can be determined;

Figure 5A shows how the users probable presence on a device varies with the duration of user inactivity; and

Figure 5B shows per device how the user's presence information can be extrapolated in a linear time-dependent manner to enable the likelihood of the user's probable presence on a device to be determined from the information reported by a plurality of communication devices;

Figure 6 shows steps in a presence and availability based routing system according to the invention;

Figure 7 shows service resource activity information for a communications device for services; and

Figure 8 shows per device how the service resource activity information can be extrapolated in a linear time-dependent manner to enable the likelihood of the user's probable presence on a device to be determined.

Figure 9 shows steps in a routing scheme according to another embodiment of the invention;

Figure 10 shows schematically steps in a routing scheme according to yet another embodiment of the invention;

Figure 11 shows schematically how decay rules can be determined; Figure 12 shows how a routing scheme using forecasting rules can be implemented to determine current probability information according to an embodiment of the invention;

Figure 13 shows schematically a hierarchical decay rule routing scheme which is configurable according to an embodiment of the invention;

Figure 14 shows a hierarchical routing scheme as configured according to another embodiment of the invention;

Figure 15 shows a configuration scheme for a routing scheme according to another embodiment of the invention; and

Figure 16 shows steps associated with a reporting mode for a configuration scheme according to the invention.

The best mode of the invention as currently contemplated by the inventors will now be described. Where their presence is apparent to those of ordinary skill in the art, features of the invention apparent to those skilled in the art may be omitted for clarity from the description and/or drawings. Where any elements of the invention have functional equivalents which are apparent to those of ordinary skill in the art, the description should be read as implicitly incorporating such elements, such that the described elements are synecdoches for any appropriate equivalents, unless explicitly excluded.

The invention relates to routing communications and a communication in this respect can comprise a telephone-type call or non-telephone type communication, comprise text, voice, or video or still image communications, and includes fax or small message service (SMS, also know as a "text" or "Txt") messages as well as e-mail, and instant messaging, and may include file transfer and other types of communication in addition to voice communications. In the best mode of the invention, a communication comprises real-time content (including streamed live content such as voice and/or audio and/or video content) and/or recorded content and/or text. In the best mode of the invention contemplated by the inventors a communication conveys a real-time streamed data comprising a "conversation" between a calling and called parties. A configuration scheme according to another embodiment of the invention is also described.

The term communications mode refers to any media mode, for example, video, audio or data (text-based etc.). The term communications protocol refers to any suitable communications protocol arranged to support the communication , including any broadband or narrowband protocol capable of supporting a wired and/or wireless connection between the calling and called parties (for example 802.11b, 802.11 g, Bluetooth, GSM, Ethernet, ADSL, xDSL, VDSL etc), including session-based communications and their modes and protocols for communication are also (e.g. communications using the session-initiation protocol SIP).

The term "calling party" refers to the initiator (or equivalently the sender) of a communication, and may be an individual or an automated calling system. An intended recipient (also referred to herein as a called party) and comprises an individual or a plurality of individuals if the address-identifier is associated with an organisational entity.

Figure 1 shows schematically a communications system in which a routing scheme according to an embodiment of the invention is implemented. In Figure 1 , a plurality of different communications devices 16, 18, 20, 22, 24, 26, 28, 30, 32 associated with the same address- identifier are shown in different possible locations 10, 12, and 14.

As shown in Figure 1 , the address identifier is capable of being resolved to within a wireless communications enabled environment 10 to a laptop 16, a first mobile phone 18, second mobile phone 20, to within a work environment 12 to second mobile telephone 20, a phone with wired connectivity 24, and a work computer 22, and to within a home environment 14, communications devices such as a telephone 28 with wired connectivity; mobile telephone 20, and a television set top box 32. The invention can be applied to any communications- enabled device which can be addressed by a calling party to a called party using an address identifier in accordance with an appropriate communications protocol, for example, domestic appliances such as home refrigeration, cooking, heating and entertainment devices (not shown in Figure 1).

As shown in Figure 1 , a calling party operates a communications device 24 to generate and/or instigate a communication with a called party by addressing the communication to a unique address identifier. As the address identifier is associated with all of the plurality of devices shown in Figure 1 , even if for example, a calling party using communications device 24 shown in Figure 1 wants to communicate with the called party 38 shown located in the home environment 14 in Figure 1 , all of the plurality of devices 16, 18, 20, 22, 24, 26, 28, 30, 32 at all locations 10, 12, and 14 will receive the communication unless routing selection criteria are applied within the network by routing server 36 to limit the routing of the call to the devices associated with the home environment location 14.

In this context, an address identifier according to the invention comprises any suitable format known to those of ordinary skill in the art as appropriate for the relevant routing protocol and/or communications mode or medium, for example, it may comprise a string of characters (e.g., ASCII characters comprising alphabetical and/or numeric characters such as are well known in the art) which is used by the calling party acting as the sender or initiator of a communication with another entity (the called party) to address the communication to the intended recipient.

The address identifier is resolved within the communications system to each of the device addresses associated with the address identifier, as shown in Figure 1 by an appropriately configured routing/forwarding server 36 located in communications network 1 which then selectively routes (or equivalents forwards) the communication to one or more of the plurality of communications devices 16,18,20,22,24,25,28, 30, and 32 associabie with the address- identifier.

The term activity information (unless otherwise qualified) refers to any information indicating a intended recipient, service or resource of a device has performed some activity involving the communications device, including information which is indirectly derived from a related action. In this way, resource activity can be inferred indirectly by recording usage of the device by a intended recipient or service (i.e., from other types of activity information), or directly from event based information (e.g., available processing power can be determined directly).

Returning firstly to Figure 1 , some or all of the communications devices 16, 18, 20, 22, 24, 26, 28, 30, 32 are capable of operating in one or more communications networks of the communications system according to the invention. Each of the communications devices includes suitable means to record intended recipient and/or service and/or device resource activity and/or inactivity information as described in more detail herein-beiow.

The communication devices shown in Figure 1 are capable of receiving a call or session invitation, for example, of operating in accordance with the session initiation protocol (SIP). Where activity information is session-related it is collated by a collating device (which is routing server 36 as shown in Figure 1 , but may be provided by other apparatus in alternative embodiments of the invention). A communications device according to the invention is provided with suitable activity recording means which is arranged to record activity information. For example, intended recipient activity information. Intended recipient activity information includes activity associated with a particular mode of operation of the device, including key press (es), or use of a mouse/stylus/voice command usage etc. The intended recipient activity information which is recorded is processed to provide an indication of the probable presence of a intended recipient in proximity to the device. In most embodiments, a intended recipient-identity is permanently associated with a communications device, however, in alternative embodiments, the intended recipient-identity is not permanently associated with a device, such as, for example, in the case where a intended recipient is hot-desking. Service activity information includes session related information and any logical event related information indicating a service is utilising an element of a communications device. Resource activity information includes any usage of a resource which generates suitable events which can be captured to indicate activity and also indirect activity information, where this enables resource usage to be determined. For example, it is possible for battery life to be predicted from the number of times a intended recipient has had to recharge the device. If a communication when received by a communications device would use up a resource which was approaching a low or critical level, then the lack of availability for the resource would be indicated as a low probability of the resource being available which could result in a communication being instead routed to an alternative device.

Once it has been determined that the address-identifier is associated with a routing scheme according to the invention, apparatus (here routing server 36) processes activity information related to the address-identifier which has been received from a number of different communications devices 16, 18, 20, 22, 24, 26, 28, 30, 32 which are associated with the address-identifier.

The apparatus 36 processes the activity information in such a way as to determine the device which is most probably available for contacting the intended communication recipient. The activity information reported relates to one or more of: the likelihood of the intended recipient's presence; and/or the likelihood of the availability of the device, the likelihood of availability of a resource on the device; and/or the likelihood of any necessary services which the device must access to process the communication and present it to the intended recipient(described in more detail herein below).

Once one or more devices have been located for contacting the intended recipient, routing apparatus 36 forwards the received communication to the selected device(s) associated with the address-identifier.

In this way, it is possible to forward a communication from routing apparatus 36 only to selected devices associated with the actual presence of the intended recipient and the availability of the intended recipient on the device (e.g. device 20 in Figure 1), and/or for which sufficient device resources are available and/or for which a service is available etc for the successful delivery of the communication (e.g. for a communications call to be successfully completed with the intended recipient).

The routing scheme according to the invention determine which communications device(s) are "available" for the intended recipient to receive a communication and which communications device(s) are able alert the communication to the intended recipient by using "presence" and "availability" information derived from activity (and in-activity) information reported by each device. Network presence information is obtained from the network used by a device (for example, network traffic to/from non-presence-aware applications, WiFi POPs in use, mobile cells currently connected to, triangulation of radio sources etc). Availability information is determined by other factors, for example, if an intended recipient is engaged in a telephone call using their mobile telephone, then although the device is associated with the intended recipient's presence, the device isn't available to receive another telephone call (ignoring call waiting etc.) Similarly, if an intended recipient is working on a PC, the device may not have sufficient processing resources to also receive a video communications call. In such circumstances a intended recipient may not answer a communication if sent to the device, even though the device is technically able to receive the call, due to the impact such a call will have on other applications running on the device. In one embodiment, use of a communications device is determined, from example, by network traffic even when this provided to/from non-presence aware applications running on the device. Such applications impact the availability of the reporting device's resources and this is also reportable. It is possible to determine the likelihood of one or more device resources, or of a service being available based on the previous activity profile of the intended recipient and/or device resources and/or services used by the intended recipient and/or the device which would be required for the call to be successfully concluded.

The likely presence of an intended recipient and the likely availability of an intended recipient on a particular device from the perspective of events that the device is able to detect are determined by a number of factors. For example, is the intended recipient using another "device", is this other device likely to be in the proximity of the device in question, is the device in question also "busy" with some other task? For example, if a personal computer is "busy" performing a virus check/disk fragmentation program receiving a communication could interrupt this process? Accordingly, it is useful if a device can provide information which enables the router in the network to determine if the device is likely to have "available" resources and/or is a "service" is likely to be available on the device where these are required for the user on that device to successfully complete the incoming communications call.

The receiving of a communication, the generation of an alert, and the operation of any applications supporting the media of the communication will utilise resources on the communications device. Accordingly, to determine if resources are available, the device must be able to record and report activity (including inactivity) information which provides an indication of the likely availability of resources on a device to the network routing apparatus (server 36 shown in Figure 1). The information which enables a probability of the resource being available when a communication is subsequently received is referred to collectively herein as "activity" information even where this relates to "inactivity" unless specifically distinguished. Each communication device reports limited information of its user's activities as can be derived from the information the device can stored on those activities. Figure 2 shows how a personal computer type device located in a intended recipient's office environment (location 12) may be used to generate a profile of a intended recipient's activities. In Figure 2, a binary state information time-line is shown, in which the horizontal axis represents time, the upper state represents the detected device activity and the lower state represents the detected inactivity periods. The term "intended recipient activity information" is used generally to refer to data whose collation represents not just specific activity by the intended recipient, but also an indication of the inactivity of the intended recipient (i.e., information may indicate lack of activity), unless the context indicates otherwise. Intended recipient activity information includes information recording some physical or verbal actuation of the device by a intended recipient for a communication and information which is logically inferred from the processes run by certain software applications.

For example, if a communications device is temporarily wirelessly connected to another communications device, and the wireless connection is short range, the presence of the intended recipient may be inferred from the presence of the connection etc. Another example is when an intended recipient is generating session information whilst browsing the internet etc.

The binary activity state information shown in Figure 2 is the activity view of the intended recipient as perceived by their office personal computer, for example, such as may be collated by key-stroke information, their calendar entries and/or journal entries (for example, such as the Microsoft Outlook™ application provides). ^•

Figure 3 shows how intended recipient activity information is recorded as durations of activity and inactivity. In this embodiment, predetermined periods of activity and inactivity are collated for the communications device. The left-hand side of Figure 3 shows the number of periods for which the intended recipient has been inactive according to the device is shown. For example, twelve periods of inactivity of less than two minutes, were recorded, and only three periods of longer than 10 minutes. The right-hand side of Figure 3 shows the cumulative number of minutes of inactivity. For example, the intended recipient has been inactive in total for more than twenty-nine periods of 1 minute and more, and has been inactive for seventeen periods of two minutes or more. This type of information can be analysed using appropriate techniques known to those skilled in the art to collate communications device activity information.

Where resource and/or device and/or service activity recorded for embodiments of the invention which base routing decisions on the likely availability of these additional parameters, similar records are generated by each device having the capability to capture this information.

Figure 4 shows how the device activity information can be used to determine the weighting to be assigned to activity information which is derived from the historical probability that the intended recipient still has a presence associated with the device. Based on the twenty-nine periods where the user has been inactive for a period of longer than one minute, and the device's view of "unavailable" events, the relative probability that a intended recipient will still be present at the device is- determined. For example, in the exemplary situation shown in Figure 4, the forecast probability applied decay the likely probabilities based on the activity information so that after a minute has elapsed the presence likelihood is 66%, however, after two minutes this presence likelihood falls to 42%, etc. These values are applied as a weighting to the time at which the activity information was captured (when a user is using the device it is assumed that the likelihood is 100%). The information captured is time-stamped using an appropriate technique so that it can be used for forecasting purposes. As shown in Figure 4, after ten minutes of inactivity, the forecasted probability weighting is tending towards 0, i.e., the existence of previous activity over ten minutes ago is not considered under this rule set to infer any current presence information for the user of the device after 10 minutes have elapsed since the device last recorded any user-related activity. The historical intended recipient activity information which the personal computer has detected has been collated into appropriate time-segments of a predetermined duration. Accordingly, this device is able to extrapolate from the recorded intended recipient activity information which has been stored to determine a probability for the intended recipient to still be available on the device at a future time (i.e., for activity to resume) after a certain period of in activity. This enables a routing server which receives the historical information to determine when a communications call is subsequently received whether that device should be selected to receive the communications call. By reporting information to a collation point accessible to all communications devices and the routing server, however, a collective view of the device activity for all devices associated with the same address-identifier can be determined. This can change the selection of a communications device as the probability of a user to be located at one device may be overridden if the user is more likely to be present at another device (if, for example, more recent activity is recorded on a different device).

Figure 5A shows the probability of an intended recipient having a presence at the device after ^• a certain period of inactivity declining in time since the last action by the intended recipient.

This information can be used to determine the how the user's presence information weighting changes with time. A comparison of the activity information reported by device 22 say shown ^■ in the communications system of Figure 1 with the information from one or more of the other devices, for example, with the mobile telephone device 24 of Figure 1 is shown schematically in Figure 5B.

Figure 5B shows how the user is in fact more likely to be at location 14 associated with cell phone 24 and not location 12 associated with an office PC at time "C". Accordingly, if a communication is received at time "C" it may be routed by the network to the mobile telephone device and not to the office PC. At time "D" however, the communication should be routed to the office PC.

In Figure 5B the intended recipient activity information gathered from a plurality of devices which individually provide probabilities of a intended recipient's presence is compared to determine an overall probability which is visible to any device able to access the collating apparatus, such as the network routing apparatus 36. Those of ordinary skill in the art will appreciate that the collating apparatus may simply comprise a suitably configured data-base which may be provided by the same platform as routing server 36.

In one embodiment of the invention a configuration scheme is provided which enables weighting to be applied, for example, according to the type of communications device. This can influence the probability of selecting one communications device relative to another type of communications device. In this way, for example, a mobile communications device normally carried on a intended recipient's person can be given a higher weighting than any other type of communications device.

Returning to Figure 5B the percentage probability of containing an intended recipient (designated "Phil") at any time is shown in graphic form. At point A, the intended recipient is highly likely to be contactable at the device 26 (denoted by the probability curve labelled "office PC"), and has a reasonable chance of being contacted using mobile phone 18 (denoted by the probability curve labelled "cell phone". At a subsequent time B, the phone 18 is more probably associated with its user's ("Phil's") presence, and the phone is highly likely at point C, where the device 26 is switched off or otherwise unable to make its presence known to the server. At time D, the user's presence at device 26 is determined to be more probable than at device 18. The activity information which is collated to generate such probability information can be obtained using calendar information etc stored on a device's calendar application as well as from the actual activity information recorded by the device. This enables the probability graph to be generated by determining when the intended recipient takes a lunch break and if they use their mobile during their lunch break etc.

As mentioned hereinabove, comparison of the probability of the intended recipient's presence at a plurality of devices requires the intended recipient activity information collated at each device to be made available to collating apparatus, for example, server 36. In alternative embodiments, information can be indirectly reported by another device which interfaces with the device recording the user activity (for example, using Bluetooth) in an ad-hoc network reporting mode. This device then relays the collating apparatus (e.g. routing server 36) with sufficient information for a communication to be forwarded to one or more selected devices at which a intended recipient is likely to be available for receiving the incoming communication and which are likely to be able to support successfully delivery of the entire communication. In the above embodiments, the intended recipient activity information has been collated at each device, and each individual device has processed the intended recipient activity information to determine a likelihood of the intended recipient being present on the device and available for receiving a communication prior to forwarding the intended recipient activity information to a collating device. In an alternative embodiment of the invention, intended recipient activity information is forwarded to the collating device and the collating device processes the received intended recipient activity information to determine for each device a likelihood of the intended recipient's presence. The processed intended recipient activity information for each device is then additionally processed to determine collectively which device should be selected for forwarding a communication to (i.e., to determine at which device(s) the intended recipient is overall most likely to be present). The selection of one or more destination device addresses for a communication can be made on the basis of the "most likely" device(s) at that point in time to be able to successfully deliver the communication to the user (e.g. complete the call), or to a number of devices whose probability exceeds a cut-off value (e.g., devices for which the intended recipient has more than a 75% probability of being present) at the time the communication is received by the routing server, rather than based on the actual information reported by each communication device which will have been recorded at some point in the past

Returning now to the embodiment of the invention shown in Figure 1 , it is assumed that when the devices generate intended recipient activity information and provide reports at fixed periods to the routing server 36 the reported activity information is subjected to forecasting probability rules which result in the probabilities of the user being present at a device "decaying" or "reducing" in time since the time the last activity was recorded on that device. These rules are also referred to herein as "decay" rules. For example, if a key on a keyboard associated with a intended recipient's computer 22 was pressed last 10 minutes ago, the likelihood of presence of the intended recipient at the computer associated with the keyboard is less than if the key was last pressedi 0 seconds ago. If the intended recipient is known to be operating the telephone 24, however, the probability that the intended recipient is still at the computer 22 (as these are co-located) is higher than if the key was last pressed 10 minutes ago and the intended recipient has not used any other device in the meantime.

The method of reporting intended recipient (i.e. user) activity information comprises obtaining activity information from a communications device indicating usage of the device by an operator. The information is then stored, either locally on the device or at a remotely located storage facility. The stored information is then used to determine intended recipient activity . information patterns of usage. These patterns of usage shown the durations of intended recipient inactivity followed by intended recipient activity, and from these probability information for the intended recipient to recommence activity on a device after a duration of inactivity can be determined. In this way, the probability information calculated from the usage pattern provides an indication of the likelihood of a intended recipient's presence at the device. It is also possible for collective patterns of usage for a plurality of said communications devices to be determined. In this embodiment, usage patterns from a plurality of devices are compared, and associations of devices can be determined. From the collective patterns of usage, durations of intended recipient inactivity followed by intended recipient activity can again be determined, and probability information for a intended recipient to recommence activity on one or more devices after a duration of inactivity on one or more devices calculated. It is a particular feature that by determining the likelihood of a intended recipient's presence at a device, the probability information shows a percentage of presence, i.e., instead of just determining an absolute values presence is indicated as a relative value either greater than 0% or less than 100%.

As mentioned above, a rule which determines how the intended recipient activity information (referred to also as historical data) should be processed to determine the likelihood of a intended recipient being available on a particular device is referred to herein as a "decay rule". In the embodiment shown in Figure 4, a decay rule comprising a probability estimate is determined for a device and is derived from historical intended recipient activity information for that device. Such a decay rule is dynamic rather than static, and changes as the intended recipient activity information for that device changes. It is also possible to "weight" the decay rule derived from historical data, so that certain devices will always indicated a relatively higher probability of a intended recipient presence than other devices. The weighting functions can be manually configured by a user of the device or within the network or using some other heuristic derived from the activity recorded by the collating device.

In the best mode of the invention, the historical activity information recorded by the device is processed so that, for example, a forecasting process is performed which extrapolates the probable availability and presence of a selection parameter (for example the presence of the intended recipient and/or the availability of a device resource) which was recorded in the past to a value associated with the time a communication is received by the routing server. In most cases the probability will decline if the activity information reported comprises an indication of the duration of inactivity recorded on the device. However, another rules are possible depending on the nature of the information which is derived from the device activity and/or inactivity. In the best mode of the invention, any complex processing of historical device activity data to implement a forecasting probability rule is performed at the collating apparatus or at some other apparatus. This prevents the communications device from using its resources (battery power/processing capacity etc).

In embodiments where a intended recipient or other party is able to configure the forecasting probability rules (the decay rules), the rule configuration information is provided using a configuration application which is communicated to the collating apparatus which either operates on a communications device or permits the user to interface directly with the collating apparatus (for example, a web-portal which allows the intended recipient to configure the decay rules to be implemented by the collating device (or routing server 36)). Those of ordinary skill in the art will appreciate that a similar configuration application could be used to configure network wide forecasting probability rules by a network or service provider.

In addition to determining how fast the probability the intended recipient's presence as determined by any activity information is to decay, in some embodiments of the invention the intended recipient is able to configure the rate at which activity information is collated and/or whether any decay rules should be overridden if a communication having a high enough level of priority is to be routed. For example, this could enable all devices to receive an incoming emergency communication. Where a communications device has a limited amount of power, it is particularly useful for the device to be configured to ensure that it reports intended recipient activity at a suitable rate. This allows the amount of signalling and overhead of network communications traffic to be reduced and also helps the reporting devices manage their power consumption more efficiently. Naturally, the time between reporting intervals needs to be appropriate to ensure reliability of the decay rule.

In embodiments where a decay rule is determined automatically from historical activity information reported by the device, the information recorded and/or the decay rule may be further modified to indicate a weighting based on the type of communications device and/or the type of activity. For example, consider an embodiment where presence information is determined using one or more decay rules which depend not only on recorded intended recipient activity information (i.e., the duration of the time intervals for which intended recipient activity/in-activity is noted) but also on the type of intended recipient activity from a session/application context. In this way, if a intended recipient has opened a secure application/session on a device, there is a higher probability of the intended recipient still being available on the device than if, for example, the intended recipient is using a device for a non-secure application. In one embodiment, the forecasting rules which generate the weighting to apply to different devices' historical activity information is configured by the user of a device but in alternative embodiment they can be configured by the user and/or determined from historical intended recipient activity information (for example, the collating device may automatically determine when an additional weighting is appropriate). '

All recorded activity information is historic by definition and the forecasting probability rule(s) need to reflect the rate at which the recorded activity information becomes a less reliable indicator of each selection parameter upon which routing is to be based by the routing server.

For example, a selection parameter such as the intended recipient's likely presence on a communications device will decay as the time from the last recorded activity elapses. In one embodiment of the invention, similar forecasting rules which implement a decay rule type functionality are applied to the recorded data either by the recording or the collating apparatus. In addition to activity information recorded by the device, user presence information can also be derived from calendar or diary information to generate forecasting probability rules. Such rules are derived from calendar or diary information stored on the

, device or on other devices and enable probability information to take into user presence information associated with the user. For example, the time of day and/or the usual activity habits of the intended recipient weight the probabilities by an additional factor. Where a collating apparatus applies forecasting probability rules to collated data, the decay rule for one communications device may be affected by one or more other decay rules and/or activity information collated from other devices also associated with the intended recipient. In addition, calendar and/or diary information and/or intended recipient configured or service provider configured rules may override decay rules derived from historical intended recipient activity information.

In this way, the invention provides only a probable indicator of intended recipient presence, a "fuzzy" presence, which can be contrasted with a binary state ("present" or "not present") indicator. For example, binary intended recipient activity state information might indicate that intended recipient A was operating device 22 at the home of intended recipient A in the last ten minutes. This is reasonable for intended recipient A at 7am (they are still likely to be at home). At 10am, however, the information has less relevance as the intended recipient is probably now at work.

According to one embodiment of the invention, a forecasting probability rule, for example a decay rule, is configurable to indicate the likelihood of availability or presence of one or more of the following: the likelihood of an intended recipient's presence on a device, i.e., the likelihood of the intended recipient being available to receive a communication from a calling party on the device; the likelihood of an intended recipient being available to receive a particular mode of communication; the likelihood of an intended recipient being available to receive a communication having a certain communications protocol (for example, is the intended recipient on a broadband network type device); the likelihood of a media type and application available to a calling party on one or more devices associated with the intended recipient; and/or the likelihood of a device resource being sufficient to support the communication being successfully established between the called and calling party for the duration of the communication.

In this way, the invention seeks to determine the probable current likelihood of a communication being successfully completed if it is routed to a particular communication device associated with the address-identifier of the communication by extrapolating the collated activity/inactivity information. Each reporting communication device reports predetermined types of activity/inactivity information which relate to a selection parameter influencing the routing decision performed in the network. The selection parameters, for

' example, is a user present at a device and if so, is the device itself available in terms of its ability to allow the communication to be successfully completed, are time-stamped and then a forecasting process is applied. The forecasting process allows the probable accuracy of the recorded information to decrease as the time from when the information was recorded passes. Thus device recorded in-activity information is subjected to a simply "decay rule" which will cause the likelihood of the device recording activity information for that particular selection parameter. Any suitable means of recording device activity can be used, such as, for example, a key press logger such as are well known to those of ordinary skill in the art.

Where activity information is reported by a communications device to the collating device 36 without any forecasting probability (decay) rules having been applied, there is no need to communicate a decay rule to each reporting communications device. However, in another embodiment each communications device applies one or more decay rules to the intended recipient activity information prior to reporting the information to the collating device. In this embodiment, the forecasting probability rules must be configured on each communications device which is operating that reporting mode, for example, by using a configuration application running either on the device or on a different device which then interfaces with that communications device.

Figure 6 of the accompanying drawings shows schematically how routing server 26) according to the invention may respond to a communication received (step 40) which is determined as being for an intended recipient comprising a intended recipient who has subscribed to or otherwise implemented a routing scheme for their communications devices according to the invention (step 42). In Figure 6, the routing server 26 functions as a collating device and processes stored information which has been previously reported by communications devices associated with the address-identifier of the receiving communication. The information is stored in either a historical state (i.e., containing its original time-stamped information (44) (but alternatively one or more devices may have reported information which as already been subject to some forecasting processing (step 46). In the best mode of the invention, server 36 processes stored information which the communications devices have "pushed" without any forecasting to the server, and the server performs the forecast operation to determine how the probability values of one or more selection criteria have changed since the activity information was originally recorded. By appropriately processing the extrapolated (i.e., forecast) information for all reporting communications devices, the routing server 36 determines which device(s) are most appropriate according to the forecasting probability rules and routes the incoming communication to the most appropriate device(s) (step 48). Alternatively, the server may respond to incoming communications by pulling information from each communication device (to which decay rules may or may not have already been applied) and processes this information (46, 48) as above to determine the most likely device on which the intended recipient is available to receive the incoming communication. Once the most likely device has been determined, the incoming communication is routed to this device.

In another embodiment of the invention, if the incoming communication has been routed to a first device and the server becomes aware through more recently reported activity information that another device is now more suitable, the server may redirect the call to the other device and/or to the first selected device.

Although the above embodiments have been described from the context of a single individual being associated with an address-identifier, in alternative embodiments a plurality of individuals are associated with an address-identifier (each recipient being associated with one or more of the plurality of communications devices associated with that address-identifier). For example, a intended recipient's mobile telephone may be associated with a business sub- identity and a private home sub-identity to facilitate billing etc, and the invention can be adapted accordingly for each sub-set of devices associated with a intended recipient, i.e., the devices associated with each "sub-identity", with the addition that as the sub-identities are known to relate to the same intended recipient, decay rules can be applied across all devices so that calls can be forwarded as appropriate to the intended recipient. In one embodiment, device activity information is determined within the network, for example, from session event information in any appropriate manner known to those of ordinary skill in the art.

The type of activity information recorded is pre-determined. For example, each time a session records intended recipient activity, it may be automatically recorded in an appropriate form and forwarded to the collating device (e.g. server 36 in Figure 1 ). Alternatively, the intended recipient's device may be configured to either periodically report activity or inactivity or to simply report a change of activity state - e.g., activity has ceased or activity as recommenced (for example after a certain triggering period), or the communications device may be polled by the central server periodically to determine its current state of use. In practice, the nature of the device and the impact the collation process has on the network overhead is likely to determine the most appropriate mode of intended recipient activity data collation.

In an alternative embodiment, each device 16, 18, 20, 22, 24, 26, 28, 30, 32 associated with a intended address-identifier receives an incoming communication, but only those device(s) for which the intended recipient has associated decay rules with will respond to the incoming communication by generating an appropriate alert. In one embodiment of the invention in which additional weightings are to be applied to or incorporated into a forecasting probability (e.g. a decay rule), a hierarchical rule-set comprising one or more rules is applied to collated activity information. The hierarchical rule set enables activity information to be biased or overridden by rules which may be absolute (for instance where a device cannot easily detect a intended recipients presence, e.g. "intended recipient A will have medium likelihood of presence at his television set top box between 7pm and 10.30pm" and "intended recipient A's presence on his cell phone will tend to 0 between 11pm and 7.30am, whether he uses it to make outgoing calls or not") or apply additional weight-scaling factors. In one embodiment, a hierarchical rule weights the likelihood of presence by multiplying the likelihood of presence on preferred devices or subtracting an arbitrary value oh devices for which a intended recipient would prefer no communications to be received on (for example, on a mobile device in a non-home network).

The above embodiments of the invention have described presence and availability information in the context of a intended recipient's activities on a device. However, if a caller wishes to contact a intended recipient at a device using a certain media type of communication or communication protocol, the likelihood of presence may instead relate to the presence of the intended recipient of a communication at a device with that media type and/or communication protocol enabled, or to the presence of a intended recipient in a particular communications network. A communications device such as an answer machine may be assigned as a default communications device in the event that no other device indicated the presence of the intended recipient (and/or the availability of the intended recipient).

Whereas the above description of Figures 2 to 5B has referred to concepts related to determining the likely the presence and availability of a intended recipient on a communications device by extrapolating known intended recipient presence and activity information using a decay rule, it is possible to apply a forecasting probability type of rule (or even a "decay" rule) to other activity information.

A hierarchical routing scheme provides an interdependency on the forecasting probability rules that it implements and enables user presence and availability information to be used in conjunction with other types of device-reportable activity information such as, for example, service activity information and device resource (e.g., battery resource, processing resource, display resource, audio resource (e.g. if a device is already playing an audio track/file etc then the audio resource is not likely to be available), resolution resource, interface resource (e.g. haptic resources).

Figure 7 shows a view similar to that shown in Figure 2 from a device's perspective in which a device notes a number of services are used by a intended recipient. Figure 7 shows when the services are actively using the device (which is often but not always when the intended recipient is using the service). From the devices perspective, a service is actively "using" its resources for a number of periods of time, and a service may not be "active" for other periods of time. A service session which is persistent may be transferred between devices, and may be maintained despite the fact a intended recipient is no longer using a particular device, even after the device has been switched off. It is therefore possible to determine for a service session to be still running even when a device from may or may not persist between period in which in addition to intended recipient availability, device availability (in terms of the resources). Thus it is possible to determine the likelihood of a service session being resumed (i.e., which becomes active) after a period of inactivity. Even when a service is not "session related", it is also possible to determine the likelihood of the service being used by the device by extrapolating historical service usage information using an appropriate decay rule (for example, by determining a previous service usage pattern). A benefit of using historical information stored locally at the routing server and extrapolated to provide a "likelihood" is to reduce overhead when seeking to determine what services are likely to be in use on a device.

As is well known to those skilled in the art, some services must be "available" e.g., already running on a device for a communication to be alerted to a intended recipient, (e.g., unless an email application is running a intended recipient won't know they have "email"), etc. In other cases, a intended recipient may be using a "service" but this may be using sufficient resources on the communications device to prevent another communication from being received using that service. The use of a device's resources also generates activity information which can be logged and extrapolated

Figure 8 shows a similar view to that shown in Figure 5B, in which a the relative availability of services on different communications devices is compared. In Figure 8, however, up to a point A in time, a video service is shown as more likely to be available on cell phone than an office PC, point B shows a later point in time when this situation has reversed, at point C the service is again more likely to be available on the PC, and at point D, it is more likely to be available on the cell phone.

Figure 9 shows steps in an another embodiment of the invention, in which a communication is directed to an address-identifier which is capable of resolving to a plurality of individuals (referred to as "intended recipients" in Figure 9) (for example, if the address-identifier is associated with an organisation whose service(s) is(are) provided by a plurality of service agents). For each communications device selected as probably available (step 910), the probability of the service being available to a calling party can be based on the likely availability of a user at that device (step 50). For each of a plurality of communications devices which meet one or more selection criteria (step 52), different users will be associated and their presence determined (steps 54). For each selected communications device (step 56), the respective intended recipient's presence and availability on the communications device is determined (step 58). If the intended recipient's presence and availability on the communications device meet the required selection criteria, the communication is delivered to the intended recipient via the communications device (step 48).

In a similar way, to the way in which intended recipient presence and availability and service availability can be determined, the likelihood of a device resource being available on a device can be calculated from historical information.

The above routing schemes require each the forecasting probability (decay) rule to be configured appropriately and in one embodiment of the invention this is performed by a user associated with the address-identifier. Within each type of routing scheme, a forecasting rule can be combined for each type of service or device resource on a per device or multiple device basis, and collating apparatus/ routing apparatus can be used to collate the probability information and determine which device(s) are likely to be the most appropriate. The different routing schemes can also be configured by a user to operate in combination, so that if, for example, the user's presence and availability indicate a device is suitable but the resource and/or service routing indicates the device is unsuitable, another device can be selected at which the intended recipient is likely to have a intended recipient presence.

The forecasting rules for a device are dynamically adjusted in one embodiment of the invention automatically by a device to increase the perceived likelihood that a particular resource will not be available to "force" the device not to be selected. This is similar in concept to the device indicating that the device itself is not "presence" - i.e., if the device isn't available as a resource isn't available, the device can be considered not to be "present" . Thus, if a device resource such as battery power is critically low, to conserve battery power, the device may indicate it has a very low power resource likelihood to automatically bias how the network routing server "perceives" its presence in the network to be sufficiently low to ensure another device receives incoming communications. An absolute "no presence" indication cannot be "forced" as this would not permit emergency calls^' to be routed to the device and could block all communication with the user even if other devices have a lower probable value of being available to receive the call and deliver it successfully to the user no network presence. Resource monitoring can be achieved by any known means, for example, it is well known for mobile telephone devices to internally have battery monitoring means and to display this to a intended recipient.

Third generation (3G) telephone devices which have certain capabilities/are able to access certain services only when located in a particular network. If a 3G telephone moves into a network which only supports voice/GPRS/SMS and not 3G video the likelihood of video media being present for communications will fall for that device, and so calls requiring this service are diverted to other devices. Other examples include communications devices such as PC's, set-top boxes and mobile telephone that can detect network bandwidth or quality of service (QoS) characteristics of a network connection, or even if a port usually used to establish a particular type of connection is being used for a different purpose preventing use of the connection or certain characteristics of a connection on that port (e.g., a device such as a PC, or games console or set-top box may find itself disconnected from a broadband connection but still provided with a land-line connection).

Figure 10 provides more detail of a routing scheme according to one embodiment of the invention. In Figure 10, on the left-hand side of Figure 10 steps in the routing scheme are shown which are implemented when the server 36 shown in Figure 1 receives a communication (step 60) containing an address-identifier associated with an intended recipient of the communication. The server 36 determines from the address-identifier a plurality of associated communications devices and their respective address information (step 62) to enable the communication to be delivered to the intended recipient via one or more of the communications devices.

The server 36 also determines if the address-identifier is associated with a routing scheme according to the invention in which one or more decay rules are configured (step 64). If not, then the communication is not delivered using the invention, i.e., routing is not selective (step 66). If the address-identifier is determined by router 36 to be associated with a routing scheme according to an embodiment of the invention (step 68), the server will apply the forecasting rules to stored activity state information for each communications device associated with the address-identifier to select one or more communications device(s) to route the communication to (step 78).

Step 68 is expanded within the dotted rectangle on the right-hand side of Figure 10 to show more steps in the implementation of a routing scheme according to the invention. As shown in Figure 10, for each communications device 16, 18, 20, 22, 24, 26, 18, 30, 32, the server 36 retrieves the latest reported activity state information, (step 70), applies one or more decay rules (step 72), to determine for each communications device the probability of the intended recipient's presence (72). The server 36 then applies one or more selection criteria (step 74), and in the event that the probability of the intended recipient's presence fulfils the one or more selection criteria for the respective communications device selects that respective communications device (step 76) as a delivery address for the communication.

Figure 11 of the accompanying drawings shows schematically how information is reported and used to provide forecast probabilities for one or more device selection criteria which are applied by the routing server 36 when a communications call is received having an address- identifier which resolves to one or more communication devices. In Figure 11 , the probable future values of selection factors such as user presence, device resource etc are determined using rules generated from recorded activity information, Firstly, each communications device 16, 18, 20, 22, 24, 26, 28, 30, 32 records activity information (step 82) from which activity information decay rules are determined. Two embodiments for determining decay rules are shown in Figure 10. One the left hand side, a communications device locally processes the recorded activity information (step 84) to determine a local decay rule (step 86), and at certain intervals sends the server 36 the last known activity state and the last decay rule determined (step 88). The intervals may be predetermined, for example, be periodic, or the communications device may send activity information each time an activity is recorded. Alternatively or in addition, the communications device sends the last known activity state to the server 36 (step 90) and server 36 then processes the activity information derived from historical activity state information to determine the decay rule to apply (step 92).

In some embodiments, the server 36 additionally determines decay rules based on a plurality of. devices activity information and last known state information (step 94). Optional steps 90,92 and 94 are shown outlined in a dot-dash form in Figure 11.

Regardless of how the specific embodiment for reporting/generating the last known activity state and associated decay rule(s) for each device, the last known activity state and associated decay rule(s) are stored so as to be accessible by the routing/forwarding server 36 (step 96).

Referring now to Figure 12, where elements relating to the same features retain the numbering scheme of previous figures, steps in a routing scheme according to an embodiment of the invention are shown. In Figure 12 when a communication is received containing an address-identifier (step 60), the router 26 determines what communications devices are associated with the intended recipient (step 62) and determines if decay rules are to be applied according to a routing scheme (step 64). The server 36 retrieves the stored activity state from a data store (step 70a), and depending on what steps are implemented the communications device shown in Figure 11 , the server applies decay rule(s) determined either remotely by the communications device(step 72a) and/or locally by the server (step 72b). Alternatively, it is possible for the communications device to locally store state information and to apply decay rules locally so that server 36 is able to request a current decayed state probability from the communications device when a communication is received (step 70b). Those skilled in the art will appreciate that a intended recipient may be associated with a plurality of communications devices which may co-operate with server 36 in a combination of the above embodiments, in each case, however, once the current probability of the activity state is determined, selection criteria are applied in step 76 to determine which device(s) the communication should be connected to by the server.

Those skilled in the art will appreciate that server 36 may in practice comprise a number of different servers, arranged to implement the routing scheme for different types of communications and communication protocols etc, and that the functions provided by server 36 may be distributed over a number of different co-operating servers, for example, one server may collate and store state information and another apply decay rules and another may route communications calls.

All activity state information is associated with a time-stamp which indicates the time at which the state information was recorded and from which time the decay rule is applied to determine either the current likelihood of the same state perpetuating (or inversely, of it not perpetuating). In this way, each decay rule is able to extrapolate from activity state information stored at a previous point in time the "current" activity state probability at the point in time when the communication request is initiated by the calling party. Where the "current" activity state information indicates a intended recipient's presence, routing can be based on the inference that there is a likelihood of the intended recipient's presence being associated with each respective communications device which meets one or more selection criteria for the communications device.

A selection criterion may be automatically determined, and may include criteria derived from calendar information associated with the intended recipient, as well as the priority of the communication, and may be location specific for mobile or nomadic communications devices. For example, the communications device which has the highest, probable intended recipient presence) may be selected or a user may have configured the selection criteria to never send a communication to their mobile telephone if the likelihood of there being sufficient battery power to deliver the communication falls below a threshold value and/or to never send a communication to the user's home communications device within a particular time interval (e.g. don't send after a certain midnight unless there is a 99% probability of intended recipient presence on device unless high priority call).

HIERACHICAL ROUTING

Figure 13 of the accompanying drawings shows steps in a hierarchical routing scheme according to an embodiment of the invention in which more than one rule is applied to activity information and/or more than one type of activity information from one or more devices is used to determine a decay rule for one or more types of activity information. In the hierarchical routing scheme, a sequence of decay rules is determine, which may apply to different types of activity information or to the same activity information. Either having collectively processed the decay rules to determine a likely probability for an activity state, selection criteria are applied or selection criteria may be applied at an intermediate step to pre-select some devices to reduce the processing burden so that the hierarchical rules are used to further select appropriate communications devices from the initial selection. In Figure 13, activity information is retrieved (step 70), a primary decay rule applied (step 72), and a probability value determined (step 74a). If other decay rules are configured for the same activity information, then these are applied in turn (step 72b) and the resulting probability values determined (step 74a), until eventually selection criteria can be applied (step 76).

In one embodiment a hierarchical rule-set is applied to local and/or collated activity information. The hierarchical rule set enables activity information to be biased or overridden by rules which may be absolute (for instance where a device cannot easily detect a intended recipients presence, e.g. "intended recipient A will have medium likelihood of presence at his television set top box between 7pm and 10.30pm" and "intended recipient A's presence on his cell phone will tend to 0 between 11 pm and 7.30am, whether he uses it to make outgoing calls or not") or apply additional weight-scaling factors. In one embodiment, a hierarchical rule weights the likelihood of presence by multiplying the likelihood of presence on preferred devices or subtracting an arbitrary value on devices for which a intended recipient would prefer no communications to be received on (for example, on a mobile device in a non-home network).

In some of the above embodiments, a plurality of different types of activity information is , collated locally on a device before being collated with activity information generated by other devices. This assumes that the device has the capability to process the intended recipient activity information to determine a likelihood of the intended recipient being present on the device and available for receiving a communication prior to forwarding the intended recipient activity information to a collating device. Alternatively, activity information is forwarded to one communications device which functions as a collating point and this may processes the received activity information to determine for a plurality of devices a likelihood of the intended recipient's presence. The processed intended recipient activity information for each is then additionally processed to determine collectively which device should be selected for forwarding a communication to (i.e., to determine which device(s) the intended recipient is most likely to be present). The selection of a destination device for a communication can be made on the basis of the "most likely" device at that point in time, or to a number of devices whose probability exceeds a cut-off value (e.g., devices for which the intended recipient has more than a 75% probability of being present).

Figure 14 of the accompanying drawing shows another embodiment of the invention in which a hierarchical routing scheme is implemented. In Figure 14, a collating server 10 retrieves stored activity information, and applies decay rules to the intended recipient activity information to determine a intended recipient's probable presence on each communications device. Where a communications device has been associated with other decay rules, for example, resource decay rules as shown in Figure 13, these are then applied to activity information associated with the respective resource (step 100) to ensure that the respective resource for delivering the communication to the intended recipient is likely to be available. In Figure 14, three different types of probable resource availability are determined, the probability of a necessary service required by a resource (102), of a device resource 104), and of a network resource (106) are shown, however those skilled in the art will be aware of other resources whose availability can be determined either directly and indirectly using the invention.

CONFIGURATION SCHEME

Figure 15 of the accompanying drawings shows steps in a configuration scheme for a routing scheme as described hereinabove. The configuration scheme may be implemented in part on a communications device, in which case, step 110 is not necessary unless other devices are also to be configured using the communications device. For each communications device, configurable elements of the routing scheme according to an embodiment of the invention include: the sources of activity information (step 112), how and/or where the historical activity information is to be stored (step 11,4), how and/or where the decay rule is to be determined (step 116), a reporting mode (118), and one or more selection criteria (step 120) can be selected and configured using any appropriate means known to those skilled in the art. The reporting mode either causes historical information to be sent by the communications device automatically to the routing server at predetermined intervals for storage on the server, or the routing server requests historical activity information when a communication is received.

Figure 16 shows how the reporting mode can be configured in two different ways and how this affects routing scheme implemented by an embodiment of the invention. In Figure 16, If historical activity state data is stored on the server (122) in its native time-stamped form, then when a communication is received (124), the server simply applies the forecasting probability rules (i.e., the information is scaled to reduce the probable selection parameter associated with the activity information) to the stored historical activity state data to determine its value when a communication call is received that could be routed to that device. The routing server does this for each device associated with the communication's address-identifier and then determines overall which device is most probably suitable for receiving the call. If data is not pushed to the network by the communications device, then when a communication is received (124) by the routing server in the network, the activity data is processed locally on the communications device and decay rules applied when the server requests the "decayed" activity information from the communications device (130). In Figure 13, optional collective processing for a plurality of devices is then performed (132).

The routing schemes of the invention derive for each communications device a probable value by extrapolating a stored known state using a forecasting rule (which may simply comprise applying an exponential decay function to a previously determined parameter value) to reduces the probability of the information which influences its routing decision according to how out of date that information is. For example, in the case where a user's historical presence and/or availability was known, as time passes from the point at which the finite presence and/or available activity state was last known the actual likelihood of the user being present and/or available will decrease. As more activity records are generated, the forecasting rules may be updated so that dynamic rules are provided. In some embodiments, additional static rules and/or routing constraints are imposed through the configuration of one or more appropriate selection criteria.

One embodiment of the method of configuring a routing scheme according to the invention is implemented using an interface such as a web-portal which enables a user to configure their communications apparatus to implement the routing scheme. The interface provides means to select from a plurality of options for one or more communications devices, forecasting probability rules and/or activity information characteristics to control the selection criteria for a communications device and, for example, to configure what types of activity will generate activity information and/or how the activity information is to be collated and stored (e.g. what type of event generates a record of activity/inactivity (e.g. the passing of a predetermined period of time and/or a change of activity state), at what intervals the information is to be provided or what events should trigger the communications device to report the information to the collating apparatus or even if the device should only provide the information in response to receiving a request issued by the collating apparatus.

The configuration method enables default probable values to be assigned to activity information. In this way, it is possible to bias selection of particular communications devices according to rules which are based, for example, on time-of-day and/or calendar information. The configuration method enables the extrapolation rules, i.e., the decay rules to be configured by a intended recipient. In this way, it is possible for a intended recipient to weight how rapidly presence likelihood reduces as time passes. For example, a frequent mobile telephone intended recipient may configure their mobile telephone to have a constant probable value of being associated with their presence of 80% unless not utilised for a period of 24 hours, in which case they may want the probability of it being associated with their presence to drop to 10%. An infrequent intended recipient of a mobile device may want to let the automatic decay rule extrapolate from activity/inactivity information as this is more likely to reflect their actual presence, and to set as the default probable value for being present on the device to only 20%. For other devices, default presence probable values are not set as the recorded activity/inactivity information and automatically statistically determined probability decay rules are sufficient. However, a intended recipient may want to configure their mobile device power resource (i.e., their battery) to decay at the set rate determined by the manufacturer for certain usage, so that the probable value reflects the likely value for this device, and to set a threshold value of 20% so that if their mobile phone battery resource is likely to be at less than 20% of maximum, no incoming calls/other communications are received on the mobile device, but are instead automatically routed to other devices associated with the intended recipient's presence where available. This will ensure that a battery is not run down by receiving a call from someone which could be received on another device in the vicinity of the intended recipient automatically.

A user can configure their device(s) so that certain communications are only received when they are in a network which has a likelihood of having appropriate resources, for example, so certain communications are only be received when the intended recipient is connected to a communications network having certain characteristics. This prevents a intended recipient from receiving communications requiring a characteristic such as high-bandwidth or low latency on a device which using a network connection/session having differing characteristics such as low bandwidth and high latency.

Those skilled in the art will recognise that several equivalents to the features described in each embodiment exist, and that it is possible to incorporate features of one embodiment into other embodiments. Accordingly, the spirit and scope of the invention is not to be confined to the specific elements recited in the description but instead is to be determined by the scope of the claims, when construed in the context of the description, bearing in mind the common general knowledge of those skilled in the art.

Claims

1. A method of selecting one or more of a plurality of communications devices for receiving a communication having an address-identifier which is resolvable by routing apparatus in a communications network to the device address of each communications device, the method comprising: comparing the current relative probabilities of one or more selection parameters derived forecasted from previous values of said selection parameters, wherein each said communication device is arranged to provide information indicative of said selection parameters for comparison; routing the communication to one or more of said communication device in accordance with said selection parameters meeting one.or more selection criteria.

2. A method as claimed in claim 1 , wherein at least one selection parameter is derivable from activity information recordable by a plurality of said communications devices.

3. A method as claimed in claim 2, wherein a said selection parameter comprises activity information indicative of a device resource being available on said communications device for receiving the communication.

4. A method as claimed in any previous claim, wherein a said selection parameter comprises activity information indicative of a user's availability on said communications device for receiving the communication.

5. A method as claimed in any previous claim, wherein a said selection parameter comprises activity information indicative of a service being available to said communications device for receiving the communication.

6. A method of selecting one or more communications devices associated with a address identifier for receiving a communication addressed to the address identifier by a calling party which is deliverable using one or more communication networks, the selected one or more communications device being one of a plurality of devices associated with said address identifier in said one or more communication networks, the method comprising the steps of: retrieving device activity information associated with each one of said plurality of devices when a communication for said address identifier is received; extrapolating said device activity information for each device using one or more resource availability rules to determine if one or more of said plurality of devices meets one or more criteria for the availability of one or more selection factors required for the device for receiving said communication; and, in the event at least one criterion is met, selecting one or more of said communications devices according to said extrapolated device activity information, wherein said extrapolated device activity information provides an indication of the probability of said one or more selection factors having availability on said communications devices for the communication, at least one selection factor comprising a resource availability comprising one or more device resources required for a device to receive the communication.

7. A method as claimed in claim 6, wherein the resource availability comprises one or more device resources required to generate an alert that the communication has been received.

8. A method as claimed in claim 6, wherein the resource availability comprises one or more device resources required to establish the communication with the intended recipient.

9. A method as claimed in any one of clams 6 to 8, wherein the communication requires a real-time dialogue between a calling party initiating the communication and an intended recipient of the communication request.

10. A method as claimed in any one of claims 6 to 9, wherein at least one selection factor comprises the availability of the intended recipient.

11. A method as claimed in any one of claims 6 to 10, wherein at least one selection factor comprises the availability of a service associated with the address identifier.

12. A method as claimed in any one of claims 6 to 11 , further comprising the step of: extrapolating said device activity information for a communications device using one or more resource availability rules which are dependent on device activity information for one or more other communications devices.

13. A method as claimed in any one of claims 6 to 12, wherein an availability rule for a resource of a communications device is hierarchically dependent on one or more other availability rules for the same or one or more other resources.

14. A method as claimed in any one of claims 6 to 13, wherein a device resource availability is network dependent.

15. A method as claimed in any one of preceding claims 6 to 14, wherein a device resource availability rule is dependent on one or more of the following: device activity; application activity; user activity associated with the presence of a user at the communications device; and resource activity.

16. A method as claimed in any one of preceding claims 6 to 15, wherein said extrapolated device activity information provides probability information which is time- dependent.

17. A method as claimed in claim 16, wherein said probability information for a communications device is linearly extrapolated from historical device activity information associated with said communications device.

18. A method as claimed in preceding claim 16, wherein said probability information is non-linearly extrapolated from historical device activity information associated with said communications device.

19. A method as claimed in any one of preceding claims 6 to 18, wherein said probability information is extrapolated using one or more of the following: historical device activity information associated with said service's usage of said communications device as a function of time of day; as a function of calendar information associated with the service;

as a function of day of the month calendar information associated with the user and/or as a function of time-of-day calendar information associated with the service;.

20. A method as claimed in any on of preceding claims 6 to 19, wherein for a predetermined duration of time, device activity or inactivity is recorded to provide said device activity information.

21. A method as claimed in any one of preceding claims 6 to 20, wherein said device activity or inactivity is associated with operation of a resource of the device.

22. A method as claimed in claim 21 , wherein the operation of a resource is derived from a session-related activity of a user of the device.

23. A method as claimed in any preceding claim, wherein a resource availability rule is dependent on one or more media types and/or communications protocols.

24. A method as claimed in any one of claims 1 to 21 , wherein said selection of one or more of said communications devices according to said extrapolated device activity information to provide an indication of the probability of the resource availability on said communications devices for the communication is further dependent on the required usage of said resource for said communication.

25. A method as claimed in any previous ones of claims 6 to 25, wherein a resource is one or more of the following: a power resource; a processing resource; a data storage resource; an application resource; user interface resource; a media display resource; an audio resource.

26. Apparatus arranged in use to select one or more communications devices associated with an address identifier for receiving a communication addressed to the address identifier, the communications device being one of a plurality of devices associated with said address identifier, the apparatus comprising: means to retrieve device activity information associated with each one of said plurality of devices when a communication for said address identifier is received; means to determine if one or more of said plurality of devices meets one or more criteria for being associated with the availability of a resource on the device from device activity information for each device extrapolated using one or more resource availability rules; and, means operable in the event at least one criterion is met to select one or more of said communications devices according to said extrapolated device activity information, wherein said extrapolated device activity information provides an indication of the resource availability on said one or more communications devices.

27. Apparatus as claimed in claim 26, wherein the apparatus further comprises means arranged to route said communication to said selected one or more communications devices.

28. Apparatus as claimed in claim 26, wherein said apparatus comprises one of said communication devices.

29. Apparatus as claimed in any one of claims 26 or 27, wherein said apparatus is implemented on remote communications processing platform arranged to communicate with any one of said communication devices using one or more communications network.

30. A suite of one or more computer programmes, which when collectively executed are arranged to implement appropriate steps in a method as claimed in any one of claims 1 to 25.

31. A computer readable medium carrying program steps which when executed on apparatus according to any one of claims 26 to 29 causes the apparatus to carry out appropriate steps in the methods of any of claims 1 to 25.

32. A method as claimed in any one of claims 1 to 25, wherein the address-identifier is associated with a user of said devices.

33. A method as claimed in any one of claims 1 to 25, wherein the address-identifier is associated with a service provided by a plurality of individual users, each individual user being associated with one or more devices.

34. A method of configuring a routing scheme to perform steps in the method of selecting one or more communications devices as claimed in any one of claims 1 to 25, 32 or 33 whereby the routing scheme selects one or more communications devices associated with an address-identifier for receiving a communication addressed to an intended recipient, the communications device being one of a plurality of devices capable of receiving said communication and associated with said address-identifier, the method of configuring the routing scheme comprising the steps of: configuring one or more rules to extrapolate stored activity information recorded by each communications device to predict a probable value of each stored activity information when said communication is received, whereby said routing scheme is able to forward said communication to one or more communications devices whose extrapolated stored activity information probable value meets one or more selection criteria.

35. A method as claimed in claim 34, further comprising the step of: manually configuring at least one of said selection criteria on a said communications device.

36. A method as claimed in any one of preceding claims 34 or 35, wherein the method further comprises the step of: for at least one of said plurality of communications devices, configuring one or more characteristics of activity information to be recorded.

37. A method as claimed in any one of preceding claims 34 to 36, wherein the method further comprises configuring a mode of storage for storing said recorded activity information, said mode of storage for said recorded activity information enabling information to be retrieved on demand by apparatus arranged to generate forecast values for said activity information.

38. A method as claimed in any one of claims 34 to 37, wherein said activity or inactivity information includes session-related activity information.

39. Apparatus arranged to provide means for performing one or more steps in configuring a routing scheme according to any one of previous method claims 34 to 38.

40. A computer readable medium carrying program steps which when executed on apparatus according to any one of claims 39 causes the apparatus to carry out appropriate steps in the methods of any of claims 34 to 38.