Patent Application
20100293583
The present invention relates to the field of television (TV) distribution systems, and more particularly to load control in a networked personal video recorder (nPVR) system.
A simultaneous point-to-multipoint distribution of media content is called broadcast or multicast. In today's multicast scenarios generally several hundred television channels are offered to the end-user simultaneously. The channels are provided via satellite, terrestrial radio, via cable and/or via Internet. The television user selects a channel from all offered channels, or from a subset thereof, using an electronic program guide (EPG), digit sequence input, or by zapping. However, neither EPG nor zapping helps to start the selected program at the beginning of the program.
US2005/0183120A1 describes a media distribution technique for facilitating a more individual and a more interactive use of the multicast service. Therein, the user may receive and watch a complete program at any time after the start of the multicast of the program.
A multi-user personal video recording (PVR) media server is positioned at a television distribution network. A plurality of user equipments is coupled to the PVR-media server. The PVR-media server receives a plurality of television signals, which may be acquired from different sources. The PVR-media server includes at least a receiver and a permanent storage, e.g. a hard disk drive, for storing the television signals. The users may individually pause the watching of the television program. Various time-shift and trick-play functions as “pause”, “playback at a later time”, “catch-up”, “instant replay”, “fast forward”, “rewind”, and “fast rewind” may be facilitated by providing a means to the user for intelligent remote control of the individual playback process. The time-shift and trick-play functions require a user-specific program distribution, which is called unicast.
The simultaneous access of a large number of users to a multitude of pre-recorded programs may cause load and coordination problems. Live programs—such as live-shows or football matches—typically contain temporal highlights, which motivate large numbers of users to trigger a flash back at a same time. Each flash back requires a change from the resource-efficient multicast mode to the resource-intensive unicast mode. Such bursts of flash backs generally result in serious load and coordination problems.
Therefore, there may be a need for a method and for a corresponding equipment, which helps to avoid overload caused by many simultaneous accesses to flash back functionality of the users, while at the same time facilitating an efficient use of central resources and reduced requirements to local resources (in particular with respect to access bandwidth and storage capacity), and while at the same time providing a special service, as trick play functions, for a running program.
The discussed need may be met by the subject matter according to the independent claims. Advantageous embodiments of the present invention are described by the dependent claims.
According to a first aspect of the invention a method for distribution of media content of a television program is provided. Therein, first data of the media content is recorded on a storage medium. Second data of said media content is distributed by a first point-to-multipoint distribution and received by a first user equipment. The first user equipment sends a demand for a special service to a service control means. A load status of at least one network element is detected, e.g. by the service control means. Here, the term “network element” is understood as a network component which may have a load status, wherein the load status of the network component can be determined directly, or at least indirectly. In dependence on the detected load status, the service control means executes a load control action appropriate for changing the load in the at least one network element.
The at least one network element is employed for providing the special service. In a preferred embodiment the demand for the special service takes place when the first point-to-multipoint distribution is still in progress. Generally, the second data increases as long as a broadcast or point-to-point distribution of a program is running. In a further embodiment the demand takes place after or even before the first point-to-multipoint distribution takes place. The invention may also be employed in an application where the first point-to-multipoint distribution is performed as a quick flash-like high-speed download of a media content file.
The invention is based on the idea that a distribution or limitation of traffic at an actual bottleneck in the distribution system may facilitate an optimization of a routing and a scheduling of unicast and multicast streams. Thus, an over-all throughput for the special service may be increased.
In the following there will be described exemplary embodiments of the present invention with reference to a method for a load control within a television distribution system. It has to be pointed out that of course any combination of features relating to different subject matters is also possible. In a preferred embodiment the second data is completely or at least partially different to the first data.
In a further preferred embodiment the demand includes at least one of a pointer to the television program and of a pointer to the first point-to-multipoint distribution of the second data of the media content.
In a further preferred embodiment the load control action is performed in dependence on an actual or forecasted load impact, wherein the load impact is caused by a provision of the special service for a second user equipment.
In a further preferred embodiment the load control action for changing the load status of the at least one network element comprises a generation of a load control message and a trans-mission of the load control message from the service control means to the first user equipment, wherein the load control message causes the first user equipment to output, in particular to display, a service offering information to a first user.
In a further preferred embodiment the load control action for changing the load status of the at least one network element comprises a generation of a load control message and a trans-mission of the load control message from the service control means to the first user equipment, wherein the load control message causes the first user equipment to output, in particular to display, a service revocation information to a first user.
In a further preferred embodiment the service offering respectively revocation information is displayed for visual reception by the first user.
The load control action for changing the load status of the at least one network element may comprise an output of price information at the first user equipment.
In a further preferred embodiment the generation of the load control message includes a determination of an increased price level at a high load status and a determination of a reduced price level at a low load status.
In a further preferred embodiment the price levels are determined user-specifically.
In a further preferred embodiment the special service comprises a time-shift function.
In a further preferred embodiment the special service is at least one of a trick-play function, a pause function, a play-back-at-a-later-time function, a catch-up function, an instant-replay function, a fast-forward function, a rewind function, and a fast-rewind function. If a user has missed the beginning of a program he may use the catch-up function. With the catch-up function the video is displayed with a higher speed than normal speed, for example with double speed. The purpose is to make up time until the playback reaches the scene which is currently presented on the multi-cast channel. Then, in a further advanced embodiment, the unicast channel is released, and the subscriber equipment may switch back to the multicast channel. In a further embodiment, to keep the audio output understandable, the audio channel is presented at normal speed during catch-up time even. That is when the video output is displayed with higher speed than normal speed. Then, the missing time for audio presentation may at least partially be gained by abbreviation of such audio pauses, which do not contain important audio information. At times, when this method does not provide sufficient alignment then—in a further advanced embodiment—the audio information will be only presented to the customer as far as the audio information remains almost synchronous to the displaying of the video information. Then, all other portions of the audio information will not be output, but cut off until a new piece of audio information can be presented almost synchronously to the user.
In a further preferred embodiment the special service is a storage and accessibility of media content or parts thereof as personal video recorder item.
In a further preferred embodiment only those demands for the special service are served which result in a time shift that starts at the beginning of the media content. Thereby, a number of unicast streams is limited.
In a further preferred embodiment one multicast serves a number of demands for the special service, in order to limit a number of unicast streams.
In a further preferred embodiment a waiting period is employed for collecting the demands.
In a further preferred embodiment each user equipment of a plurality of first user equipments transmits a demand for the special service during a waiting period, wherein all these demands transmitted during the waiting period are answered by one single provision of the special service. A number of similar special service requests (for example within a geographical region or domain) will be answered with a modified special service. For instance a number of users demanding the special service of jumping back by 10 seconds, and restarting the program from that time, and generating this request in the time interval [t1, t2] will get the replay started at time t1-10 s, and will be serviced by multicast.
Prior art television distribution systems deny a portion of special service requests if the number of requests to be served increases over at certain capacity level. Therefore there may be a need to increase the capacity level without substantially adding any additional hardware to the system. This object is solved by a further preferred embodiment, wherein a distribution of the recording is controlled in dependence on the load status of the at least one network element. The distribution of the recording may be optimized in dependence on an availability of the at least one network element or in dependence on a load on the at least one network element. This feature is based on the idea that the routing and scheduling of a subset of the unicast streams may have a strong impact on the remaining choices for the routing and scheduling of other unicast streams. Because of this mutual influencing between the routing and scheduling of different unicast streams, an optimization of the distribution of the recording may lead to a higher overall throughput, and consequently, to a higher cost-efficiency, of the system, without substantially adding additional hardware to the system. For a better understanding following example is given: or storing the recordings (see first step of the inventive method), a distributed system of storage media may be employed. Then, for each specific recording a specific storage medium has to be selected from the set of all storage media of the distributed system. This may be done in advance, i.e. statically, or may be done dynamically, for example in dependence on a load status of a network element. In a preferred embodiment, that storage medium is selected which—of all available storage media—provides the best connection to the access network. In a further preferred embodiment that storage medium is considered as providing the best connection to the access network which has the highest bandwidth to that service area which—for the specific recording to be performed—probably produces the highest number of demands for the special service.
In a further preferred embodiment a user-specific provision of the special service to a second user equipment is employed for the provision of the special service to the first user equipment. With prior art television distribution systems a long-term storing of the television programs is cost-intensive, because a fulfilling of all individual special service requests requires large processing and storage capacities. Therefore there may be a need to improve the cost-efficiency of such a system with respect to costs. This object is solved by a further preferred embodiment, wherein the recording is performed at a second user equipment selected in dependence on the load status of the at least one network element. The location of the decentralized recording may be determined in dependence on availability of a system component or in dependence on a load on at least one network element. This feature is based on the idea that recordings which are of a remaining interest to a large number of users shall not only be provided at a central server, due to cost efficiency, whereas recordings which are of interest to only a small number of users cannot be provided cost-efficiently at the central server. Advanced set top boxes comprise significant storage space for storing recordings for one single user. Recordings stored in an advanced set top box of a second user may be cost-efficiently made available for the first user, if a sufficient large bandwidth for uploads from the set top box to the network can be provided. As well sufficient bandwidth for generation of recordings by streaming or download to the set top box of the second user is assumed. This embodiment is advantageous for home networking with several users. All or some functions of the coordinating unit (BE) may also be distributed to the user equipments.
In a further preferred embodiment a user-specific provision of the special service to a second user equipment is employed for the provision of the special service to the first user equipment.
According to a further aspect a service control means comprises a controller for the load control according to one of above embodiments.
In a further preferred embodiment the service control means assesses a predefined usage of the distribution service as a demand or as a revocation of the demand of the special service. This solution is based on the idea that user actions, even if not explicitly addressing the special service, may contain information about whether and to which extent a future use of the special service with respect to a specific television program may be expected. Taking this information into account, an unexpected unavailability of the special service—caused by a stop of the processing of a special service demand for another user—can be avoided.
According to a further aspect a user equipment is designed for generating and submitting the demand for the special service according to one of the above embodiments.
According to a further aspect a means for bandwidth management is designed to acquire knowledge about the load status of the at least one network element, and to influence the load status at the at least one network.
In a further preferred embodiment the influencing of the load status of the at least one network element is performed by providing load information to a service control means as described above.
In a further preferred embodiment the influencing of the load status of the at least one network element is performed by providing information to a user equipment as described above.
In a further preferred embodiment the influencing of the at least one network element is performed by providing information to the at least one network element.
In a further preferred embodiment the means for bandwidth management is included in the service control means as described above.
It has to be noted that embodiments of the invention have been described with reference to different subject matters.
In particular, some embodiments have been described with reference to apparatus type claims whereas other embodiments have been described with reference to method type claims. However, a person skilled in the art will gather from the above and the following description that, unless other notified, in addition to any combination of features belonging to one type of subject matter also any combination between features relating to different subject matters, in particular between features of the apparatus type claims and features of the method type claims is considered to be disclosed with this application.
The aspects defined above and further aspects of the present invention are apparent from the examples of embodiment to be described hereinafter and are explained with reference to the examples of embodiment. The invention will be described in more detail hereinafter with reference to examples of embodiment but to which the invention is not limited.
The illustration in the drawings is schematically. It is noted that in different figures, similar or identical elements are provided with the same reference signs or with reference signs, which are different from the corresponding reference signs only with respect to the first digit.
The television distribution system 101 supports a special service for recording and playing back of portions of media content 103. The special service may be a time-shift or a trick-play function as pause, playback at a later time, catch-up, instant replay, forward, fast forward, rewind, and fast rewind. The media content 103 comprises a first 105 and a second 107 portion of a television program 109.
In a preferred embodiment the demand for the special service takes place when the first point-to-multipoint distribution 113 is still in progress, and the second data 107 increases as long as a broadcast 113 of a program is running. In a further embodiment the demand takes place after or even before the first point-to-multipoint distribution 113 takes place.
The invention may also be employed in an application where the first point-to-multipoint distribution 113 is performed as a quick flash-like high-speed download of a video file.
The method 201 is performed in following steps: A first step 203 of the method 201 is a recording 111 of the first data 105 of the media content 103 on a storage medium 112. A second step 205 of the method 201 is a first point-to-multipoint distribution (multicast) 113 of the second data 107 of the media content 103. A third step 207 of the method 201 is a reception 115 of the second data 107 of the media content 103 using a first user equipment 117. A fourth step 209 of the method 201 is a transmission of a demand 119 for the special service from the first user equipment 117 to a service control means 121, wherein the demand 119 typically includes at least implicitly one of a pointer to the media content 103 and a pointer to the first point-to-multipoint distribution 113.
For performing the special service for the first user equipment 117 the service control means 121 needs information about which user equipment 117 has submitted the demand 119, which one of special services shall be performed, and on which one of available media contents 103 the special service shall be applied. This information may be included in message fields of the demand 119. In an alternative embodiment the service control means 121 receives knowledge about this information by a data communication being different to the demand 119. An explicit addressing may be employed, for example, if the service control means 121 is designed to provide for all user equipments 117 only one same kind of special service and for only one media content 103 at a time. In such a case, explicit information about which one of special services shall be performed, and on which one of available media contents 103 the special service shall be applied, is dispensable. Nevertheless, the demand 119 must include at least implicitly one of such a pointer to the media content 103 and a pointer to the first point-to-multipoint distribution 113. A fifth step 211 of the method 201 is an execution of the special service, wherein the execution may be coordinated by the service control means 121. An execution of the special service comprises at least one of a transmission 125 of a portion of the recorded first data 123 of the media content 103 to the first user equipment 117, of a setting of a user-specific start marker, and of a setting of a user-specific stop marker. The transmission 125 of the media content 103 to the first user equipment 117 may employ a unicast or a multicast.
The method is characterized in that for load control a load control action is performed in dependence on a load of at least one network element 127. The at least one network element 127 is employable for the special service, in dependence on a load status 133 of the at least one network element 127. The load control action comprises at least one of following actions: assigning a resource to a special service (which temporarily requires more resources than already being assigned to it), assigning of traffic to a resource, enabling of a resource, limiting of traffic on a resource, bundling/concentrating of traffic onto a resource, and temporarily closing down a special service for all or a subset of users.
For example, a high traffic may temporarily occur in a cable or in a Digital Subscriber Line Access Multiplexer (DSLAM) 127 serving a residential area, because too many users within this area are currently using a time-shift service. Generally, a time shift service causes the occupation of one unicast channel per user. Each further user ordering a time shift service needs a further unicast channel. Finally, it may happen that there is no spare multiplexer-capacity left, respectively that there is no spare cable-capacity left for performing any new service order.
The load control may observe the load status in the network permanently and thus recognize such a critical load situation on the cable or on the DSLAM 127, respectively. Then the load control may take an appropriate action to reduce the load on the cable or on the DSLAM 127, respectively. For example, the load control may inhibit those new unicast transmissions 125 which have to be routed via the overloaded cable, respectively which have to be routed via the overloaded DSLAM 127. Where appropriate, the load control may even be configured to forcibly stop an ongoing unicast transmission 125. This may be, for example, reasonable if a non-served multicast trans-mission having a high priority cannot be served otherwise. This measure may be considered as a temporarily closing down of the special service which requires unicast transmission. When the traffic load is reducing again, the load control may re-enable the resource respectively special service.
Further, the load control may cause the service control means to set a limitation of the number of simultaneously performed unicast transmissions via a load-critical network element.
If via the same network element 127 a number of unicast transmissions are performed or shall be performed carrying a same or almost a same content with a same or almost a same time shift this unicast traffic may be rearranged or assigned to a single multicast channel. Then, for the traffic a single multicast channel is employed instead of a plurality of unicast channels. Such a traffic rearrangement respectively such traffic assignment may be considered as a bundling or concentrating of traffic.
In addition or alternatively, the load control action may comprise an output of price information at the first user equipment 117. The service control means 121 coordinates an assignment 135 of the recording demands 119 of the users to recording units taking into account an accessibility of the users and an availability of resources 127. In particular, the criteria of a sufficiently-likely accessibility of the user to a storage medium 112 and of a network load resulting thereof are considered. Further criteria are the available bandwidth and the storage capacity (notification 137).
After program x has begun to be output, the user at the STB1 may start ad-hoc a late complete recording. The user may also start the complete recording from the EPG without watching program x. A special benefit that the user at the STB1 has is the ability to decide about the recording during the output of program x. He needs not make the decision in advance without having any knowledge about the content of program x. This allows to record only programs which the user really likes to watch for a second time.
After the start of program x a user at the STB2 joins the first point-to-multipoint distribution of program x. Then he may want to flash back to the beginning of program x or to another point of time in program x which is before the actual time. The service control means BE can fulfill this request 307 by individually outputting 309 program x for STB2 beginning with the specified time. Contrary to the user at STB1, the user at STB2 does not necessarily want to have a permanent access to program x after its end. For the user it is beneficial, that he may extend the previous channel-spanning one-dimensional zapping into the backward time of each channel. Thus, the zapping is extended into a second dimension without requiring additional local resources for this feature. If a user detects a program x he is interested in, he may zap back into a virtual past, in particular to the begin of program x.
nPVR recordings and/or cPVR recordings and/or flash-backs for single users may be included in the offer for subsequent complete recordings of a running or of a terminated program x. To avoid an overload by unicast 505, demands 119 may be denied ad-hoc. Or the number of demands 119 may be reduced in advance by withdrawal of the offer or by a restriction of time shift functions. The number of streams may be reduced by allowing only those time shifts which start at the beginning of program x or by serving a number of time-shift demands by one multicast. For collecting the time-shift demands for a multicast, a waiting period may be introduced.
The user may still use the EPG as usual to select a desired program x, for example by navigation to the desired program x using cursor buttons and by a selection using the OK button on the remote control. This selection method may also be employed when at the same time a recording in the network is being performed by a background process.
The end-user is in the full screen mode and watches a running program x. The operator has started an nPVR recording. A possibility is offered to the user for receiving a unicast stream to provide an access to a missed portion of program x. To indicate this possibility to the user corresponding symbols are provided or overlaid on the screen, for example, in an upper corner of the screen or in an info banner (“Mini EPG”). For this following functions may be offered simultaneously, wherein each function is being indicated by a dedicated symbol and/or indicator:
If a switching to unicast 505 is not possible for additional users, because of a network load or because of a saturation of the nPVR systems, the symbols respective indicators are extinguished. Thereby, the feature is temporarily no longer available. A demand 119 for the feature will be answered with denial respectively with indication of an adapted fee.
Normally, the user programs the recordings of the desired TV programs in advance. Therefore, a timer in the system is set for the programmed recording. As soon as the timer expires—which, in the optimal case, coincides exactly with the begin of the desired program x—the recording will be started, and will be, from this time on, available for play back as a video in the nPVR library. Even if the recording is still running an output of the recorded portions may be started.
The option “backward recording” may be offered with the start of the recording via the EPG. The option may also be offered at the start of an “instant recording”, which is started by pressing the recording button of the user's remote control or set top box STB in the television full screen mode.
From the user's perspective following features of an Internet Protocol Television (IPTV) provider result:
The figure contains following simplification. The recordings 703, 705 may be stored as sets of data blocks or data segments. Therefore, a recording of separated data streams is not necessary. In practice, all blocks or segments, respectively, will be stored and assigned to the presented complete recordings. Thus, blocks or segments, respectively, of overlap periods will be assigned to at least two recordings.
For elucidating the method of recording and the method of deletion of a recording,
1. The operator tags a first point-to-multipoint distribution 503 as nPVR-capable and as backward-recordable. Appropriate video servers 301, 303 will be manually or automatically assigned to the recording 703 initiated by the operator.
2. The central BE starts the recording 703 at a VoD server (e.g. with forward 601 and backward 603 guard times).
3. The first point-to-multipoint distribution 503 starts (inaccurate start time).
4. The user zaps to the appropriate channel or switches on a user terminal (e.g. a television set).
5. The user decides to watch a first point-to-multipoint distribution 503 from the beginning (switch-over to unicast).
6. Previously recorded portions of the first point-to-multipoint distribution 503 are available to the user (unicast 505). During the first point-to-multipoint distribution 503 the user may use “trick play”, in particular he may flash back.
7. The user may tag the first point-to-multipoint distribution 503 as recorded. Thereby, the recording of the first point-to-multipoint distribution 503 is also available after the first point-to-multipoint distribution 503. At the end of the broadcast 113 of the program, the first point-to-multipoint distribution 113 generally covers the whole program, because of the advancing time. Consequently, generally also the “recording by tagging” covers the whole program. This complete recording 703 remains available to the user until the user deletes 701 it, or until the recording 703 is automatically deleted 701 by a restriction configured by the operator, if applicable.
8. The user selects from the menu “Recorded Broadcasts” the corresponding first point-to-multipoint distribution 503 and watches the corresponding recording 703.
9. The user deletes 701 the first point-to-multipoint distribution 703. From this results a notification to the BE, and the reference to the first point-to-multipoint distribution 703, which possibly covers the whole already ended program will be deleted.
10. The central BE deletes 701 the recording 703 at the video server 301, 303, when the last reference to the first point-to-multipoint distribution 503 has been deleted or is being deleted or is going to be deleted. There may be further pre-conditions for the deletion 701 of the recording 703 on the video server 301, 303, for example a minimal provision period of first point-to-multipoint distribution 503/multicast programs.
1. The received video signal will be encoded, for example in an encoder 901, and will be transmitted to an encryption server 903.
2. The encryption server 903 encrypts the signal and deposits 905 the keys on the key server 907.
3. The encrypted signal will be distributed 909 (to a number of STBs and to video servers 301, 303).
4. The user decides to watch the running first point-to-multipoint distribution 503 (respectively the recording of a first point-to-multipoint distribution 503) from the beginning of program x.
5. Authorizations for use of the features will be queried 911 at the BE.
6. After a positive response 913 from the BE, the nPVR server 301, 303 starts the unicast 505 of the recording 703 to the STB (user).
7. The necessary keys 907 will be queried 915 at the key server.
8. The key server queries 917 the authorizations of the users at the BE.
9. After a positive response 919 from the BE, the necessary keys will be sent 921 to the STB.
10. The STB decrypts the video and outputs the video on the screen.
As shown, the method may also operate with encrypted channels. As the keys for a channel are very small in size, they may be stored for a long time on the key server 907. Each key contains supplementary data about the start time and the duration of its validity. Using this time information the STB may fetch necessary data from the encryption server 903. For authentication of the STB and for administration of the necessary rights (specifying which user may watch which programs/recorded first point-to-multipoint distributions 503) the key server 907 has an interface to the IPTV Middleware (BE). Thereby the key server 907 may check in addition, whether a STB possesses the necessary rights, and the key server 907 may deny the key request 915, if appropriate.
RACF is a system that derives knowledge about the load situation in the network by evaluating acquired load information, thereby taking knowledge about the network topology into account. The load information may be acquired by periodically scanning/querying the network elements (DSLAM, Edge Router, Router). The RACF determines the network load, for example, by periodical polling of the DSLAM and (Edge-)Router components (for example via SNMP-based status queries). To facilitate at peak times of real-time traffic more unicast streams 505 than in the standard configuration, advanced RACF systems may control a dynamical switching of the traffic in the network elements 127. This may be accomplished by reduction of the data traffic and by allowance of more real-time traffic in network elements 127.
RACF fulfills the resource management function discussed above. A great advantage is the interface to the BE for the decision of the switch-over from multicast 503 to unicast 505.
For the previously described new zapping feature a scenario could be as follows:
1. An end-user wants to watch a running program x from the begin of program x. The corresponding option is offered at the user interface (UI) of the end-user.
2. The end-user presses the “PLAY” button to initiate the play “starting at begin of the program”.
3. The STB sends a policy message to the IPTV Middleware (BE) for checking the current traffic load.
4. Using the RACF, the Middleware (BE) checks the current traffic situation in view of the actual user demand and returns a corresponding response to the STB client.
5. Dependent on the received response the switching-over from multicast-live-television to unicast-retrospective-television will be initiated.
RACF provides 133 knowledge about the network load. This knowledge is employed for updating those features offered to the user, which imply a switch-over to unicast 505. Spontaneous messages 951 from RACF to BE or cyclical queries 953 from BE to RACF may be used therefore.
It should be noted that the term “comprising” does not exclude other elements or steps and the “a” or “an” does not exclude a plurality. Also elements described in association with different embodiments may be combined. Reference signs in the claims should not be construed as limiting the scope of the claims.
Summarized a method for distribution of media content of a television program is provided, wherein the method comprises following steps:
| Number | Date | Country | Kind |
|---|---|---|---|
| 07017520.3 | Sep 2007 | EP | regional |
| Filing Document | Filing Date | Country | Kind | 371c Date |
|---|---|---|---|---|
| PCT/EP2008/061638 | 9/3/2008 | WO | 00 | 6/8/2010 |
