GB2480820A

GB2480820A - Multicasting video to multiple wireless terminals with adaptive modulation and coding and parallel multimedia carousel channel

Info

Publication number: GB2480820A
Application number: GB1009133A
Authority: GB
Inventors: James Chung-How; Andrew Nix; David Bull
Original assignee: GLOBAL PROVISION Ltd; PROVISION COMM TECHNOLOGIES Ltd; GI Provision Ltd
Current assignee: GLOBAL PROVISION LIMITED
Priority date: 2010-06-01
Filing date: 2010-06-01
Publication date: 2011-12-07
Also published as: GB201009133D0

Abstract

A system for multicasting video data to portable wireless terminals wherein the terminals feed back quality information (eg video quality, packet loss/error rates) 360 and the transmitter adapts modulation and coding (AMC) 36 parameters in response (eg video en/transcoding rates, cross packet FEC rates) 362. The feedback may be from one, some or all terminals. A parallel multimedia carousel 34,346 may be provided which includes data related to the video 342 and/or repetitive video portions that have been removed from the video stream for compression purposes. It may also include alerts about video content which are generated dependent upon client preferences 349. The system may be used to deliver video to spectators at sports events or in similar crowd/stadia situations.

Description

ADAPTIVE MULTICAST TRANSMISSION SYSTEM

The present invention relates to a multicast transmission system for the wireless transmission of data. In particular, the present invention relates to an adaptive multicast transmission system for the wireless transmission of multimedia data and video streaming data to a plurality of fixed and mobile clients.

Organisers of large venue events, such as stadium events including pop concerts and sporting events are continually looking for "added-value" entertainment features which will attract attendees and maintain consumer interest in a crowded leisure market. The atmosphere of a live event can often be unparalleled. There are now a large range of live events occurring regularly, often at conflicting times.

Not only do such events compete against each other for attendees, but as home entertainment system quality has improved, the live events must also compete against the, often live, broadcasting of the event into the comfort of peopl&s homes. For example, many top flight football and baseball games are available to view live on a subscription basis from a television broadcasting company. For viewers who pay the subscription, the game is available in real time, with expert commentary. Many of the television companies have multiple cameras simultaneously filming the game, from a variety of different angles and viewpoints, including close up footage of the game. Depending on the television package being used, some viewers can select interactively which camera footage they wish to watch. However, attendees at the game are restricted to a single viewpoint from their seat, which may be a considerable way from the pitch itself.

In an effort to provide more consumer value at such large stadium events, large display screens have been used for some time, with live close up footage of the event being displayed in near to real time, along with replays of key action points interjected into the live feed as and when they arise. Until recently these systems used traditional analogue transmission techniques. However, Sony's Emirates Stadium digital high definition (HD) LCD display screens display live action in

I

real time which is recorded, encoded and streamed over the Emirates Stadium Local Area Network (LAN). This live streamed video data can also be customised with add-on graphics and split screen display. In addition, prior to and after the main event, pre-recorded footage including behind the scenes footage and interviews, or post event analysis, can be shown on the screens. Whilst this system provides a great deal more entertainment to the audience, the output displayed to the audience is determined by an operator, with no audience interaction or choice in the video or data being viewed.

More recently, a system called Kangaroo TV has been developed which provides users with a handheld television system which enables them, at events where Kangaroo TV is being transmitted, to view live footage of the event from one of several cameras. This provides a multi-channel mobile TV experience but lacks user interactivity and provides no accompanying data service. Along similar lines is YinzCam which, at chosen live events, provides live footage which can be viewed on an attendee's individual hand held device or on touch screen in-suite displays provided around the stadium. Whilst these systems both provide users with entertainment options approaching those available to home viewers, there are limitations on the services provided by these systems.

In particular, such existing video distribution systems have been developed based on unicast transport protocols and/or cross packet forward error correction (FEC) codes (i.e. erasure codes), and fixed video bit rates. These fixed systems are not adaptive, do not scale for multicast delivery, and must be designed for the worst case environment and crowd scenario. As they do not trade off the cross packet FEC rate against the video rate dynamically based on the client packet loss seen for a given installation and at a given time, they are not able to provide the best video quality to viewers. These fixed solutions also fail to maximise the number of video channels that can be sent since they cannot adapt the video rate to the available wireless multicast throughput rate. Furthermore, they cannot adjust to deliver a given number of video streams by reducing the bit rate per stream and are unable to guarantee coverage and performance as they do not adapt if packet loss, or FEC decoding errors, are observed by the client.

Internet Protocol television (IPTV) has also seen the development of a number of near-live TV systems. For example, transmission systems exist which enable a user to watch live baseball on their mobile phone using a unicast Wi-Fi link, Ii this case a TCP transport protocol is used to provide unicast delivery to the mobile terminal and packet errors are overcome via MAC layer (Wi-Fi) and transport layer (TCP) packet retransmission. However, this type of transmission system does not scale up well to provide a robust multicast delivery system since in the case of a multicast event, the lack of packet retransmission, especially over the wireless link, renders the transmitted video stream prone to very severe video distortion.

Furthermore, most Wireless Access Points (APs) fail to reliably deliver a smooth stream of multicast packets, especially at higher input data rates and for input streams with large amounts of timing jitter.

Current systems of this type are based on User Datagram Protocol (UDP) or Transmission Control Protocol (TCP), neither of which can support the scaling of transmission to reach tens of thousands of clients within a local venue. UDP guarantees low packet delivery latency, but this occurs at the expense of packet error rate. UDP is an unreliable protocol with no end-to-end handshaking which means features, such as transmission rate adjustment, need to be achieved using a higher layer proprietary protocol. UDP (often together with the Real Time Protocol, RTP) is however used for many real-time applications, with one well know example being SKYPE. TCP is very commonly used for video streaming and for almost all data distribution, i.e. File Transfer Protocol (FTP). TCP is very convenient for application developers to use as TCP insists on delivering all the packets to all the clients, therefore application developers do not need to worry about how to deal with missing packets. The problem with TCP is the unicast link to the wireless clients (which does not scale), and the throughput variations that are caused by unreliable wireless delivery channels. TCP insists on delivering all the packets to all the clients, and over poor wireless channels the retransmission rates and transmission backoff can severely lower the throughput to the point where the video flocks up', resulting in video rebuffering'.

For interactive services, where the clients interact regularly with the server, a TCP protocol is inappropriate. Instead, a UDP (for small numbers of clients) or multicasts (for a large number of clients) is necessary. In a stadium application live' video streams may typically be delayed by up to 15 seconds. However, even in this case it is not possible to use TCP in the server since there are no client return paths (for TCP packet retransmission and rate adaptation) over a multicast wireless link. One existing solution is to replace TCP with multicast delivery and to use cross packet erasure codes to recreate' the missing packets. This approach can work, but there are many other issues that also need to be addressed. These include video structure, packet flow into the wireless Access Point, packet buffering, video packetisation, FEC rate adaptation, modulation and coding rate adaptation, client quality feedback, channel metadata distribution, and video stream presentation in the client players.

An object of the present invention is to obviate or mitigate at least one of the aforementioned problems.

According to a first aspect of the invention there is provided a method of transmitting multicast video data to a plurality of client receivers, the method comprising transmitting video data to a plurality of client receivers simultaneously using an adaptive transmission scheme, receiving unicast feedback data from at least one client receiver, the feedback data including feedback information relating to received video data at the client receiver, updating the adaptive transmission scheme in dependence upon received feedback data, transmitting subsequent video data to the plurality of client receivers using the updated adaptive transmission scheme.

The provision of unicast feedback obtained from at least one client within the network enables adaptive video encoding or transcoding which results in optimisation of a variable data rate and resolution of the video for multicast distribution. This feedback also adopts the FEC erasure code rate for video independently of that for the multimedia data.

The adaptive transmission scheme may include at least one of an adaptive encoding/transcoding scheme, and an adaptive modulation scheme, and an adaptive cross packet forward error correction scheme.

Preferably, the unicast feedback is received from a predetermined subset of the plurality of client radio receivers. The unicast feedback may be received from each of the plurality of client radio receivers.

The provision of feedback from multiple clients enables refinement of the optimization of the multimedia data streams for transmission.

The adaptive transmission scheme may include at least one of: adaptive video rate, cross-packet FEC rate, and video structure. The feedback information may include information relating to at least one of: packet loss rate and cross packet forward error correction decode error rate. Inclusion of these parameters in the feedback improves optimization and adaptation of the data to be transmitted to reflect current system performance.

Preferably, the multimedia data and video data are transmitted using a modulation scheme wherein the modulation scheme is modified in dependence upon received feedback data. The modulation scheme is preferably a wireless local area network (LAN) multicast modulation scheme.

Such a method may also include transmitting multimedia data, different to the video data, using a second adaptive transmission scheme, the second adaptive transntission scheme being adapted in dependence upon the received feedback data.

Such a method enables a separate data path to be provided in a multicast environment.

The second adaptive transmission scheme may include at least one of an adaptive encoding/transcoding scheme, and an adaptive modulation scheme, and an adaptive cross packet forward error correction scheme.

The second adaptive transmission scheme may include at least one of: adaptive video rate, cross-packet FEC rate, and video structure, and the feedback information may include information relating to at least one of: packet loss rate and cross packet forward error correction decode error.

According to a second aspect of the invention there is provided a wireless multicast video data transmission system comprising a transmitter operable to transmit video data to a plurality of client receivers simultaneously using an adaptive transmission scheme, and a receiver operable to receive unicast feedback data from at least one client receiver, the feedback data including feedback information relating to received video data at the client receiver concerned, wherein the transmitter is operable to update the adaptive transmission scheme in dependence upon received feedback data, and to transmit subsequent video data to the plurality of client receivers using such an updated adaptive transmission scheme.

The provision of unicast feedback obtained from a client within the network enables adaptive video encoding or transcoding which results in optimisation of at least one of: video data rate, cross packet forward error correction rate, wireless modulation and coding scheme, and video resolution for multicast distribution to the clients.

The adaptive transmission scheme may include adaptive video rate, cross-packet FEC rate, and video multimedia data structure, and the feedback information may include information relating to packet loss rate and cross packet forward error correction decode error rate. Preferably, the video data are transmitted using a modulation scheme, and wherein the modulation scheme is modified in dependence upon received feedback data.

The transmitter may also be operable to transmit multimedia data, different to the video data, using a second adaptive transmission scheme, the second adaptive transmission scheme being adapted in dependence upon the received feedback data.

Such a system enables a separate data path to be provided in a multicast environment.

The second adaptive transmission scheme may include at least one of: adaptive video rate, cross-packet FEC rate, and video structure, and the feedback information may include information relating to at least one of: packet loss rate, and cross packet forward error correction decode error rate.

According to another aspect of the present invention, there is provided a method of decoding a received wireless multicast video data stream. the method comprising receiving a wireless multicast video data stream, converting a received wireless multicast data stream to multicast video data, converting such multicast video data into unicast format video data, and decoding such unicast format video data into a video display driver signal.

According to another aspect of the present invention, there is provided a device for receiving a wireless multicast video data stream, the device comprising a receiver unit operable to receive a wireless multicast video data stream, and to output multicast video data, a data processor operable to receive multicast video data from the receiver unit and to output unicast format video data, a video decoder operable to receive unicast format data from the data processor, and to output a video display driver signal relating to such received unicast format data.

I

Such a method and device enables a standard unicast video decoder/display driver to be used with a multicast video stream transmission.

According to another aspect of the present invention, there is provided a method of transmitting a wireless multicast video data stream to a plurality of receivers, the method including removal of periodically repeated information from the video stream, and the transmission of such removed information separately from the video stream, According to another aspect of the present invention, there is provided a method of transmitting wireless multicast video data stream to a plurality of receivers, the method including transmitting multimedia data, different from the video data stream, to the receivers separately from the video data stream.

According to another aspect of the present invention, there is provided a method of receiving a wireless multicast video data stream transmitted in accordance with such a method, the receiving method including selecting multimedia data for display in dependence upon comparison of metadata relating to the multimedia data with preference information for the receiver concerned.

These and other aspects of the present invention will be more clearly understood from the following description and, by way of example only, and with reference to the following figures, in which: Figure 1 is a schematic diagram of a server client adaptation according to an aspect of the present invention; Figure 2 is a schematic diagram of a wireless multicast data network according to a first embodiment of the invention; Figure 3 is a schematic diagram of a transmission part of the network of Figure 2; Figure 4 is a schematic diagram of a client device of the network of Figure 2;and Figure 5 is a schematic diagram of a forward correction error mechanism for use in the transmission system of Figure 2.

Figure 1 illustrates the concept of server-client adaptation for multicast distribution systems. Active client devices (which can be mobile or fixed) extract quality of service information from the received multicast streams, and send this information as feedback information back to the server as a unicast transmission. The feedback information is then used to form a statistical error surface, which is used in the adaptation of global stream parameters, such as video format structure, stream number, and wireless modulation and coding scheme. Local stream parameters can also be adjusted, such as video rate and resolution, and the cross packet FEC rate and block size. Parameters can be adjusted independently to allow quality to be mapped as required to particular video channels. Statistical multiplexing can also be supported, where video rates are set dynamically for each video data stream.

Figure 2 illustrates a wireless multicast network which embodies various aspects of the present invention, and comprises a server 12 to which are connected a plurality of video data sources 13a... 13n, an operator data input device 14, and a data source 15, such as a database of, for example, pre-recorded video, audio, or text.

The server 12 comprises a plurality of encoder units 16a... 16n connected to receive video data from respective ones of the video data sources 13a. ..13n. The server 12 also includes a controller 17 which is connected to receive encoded data from the encoders 1 6a. ..16, to receive control data from the input device 14, and multimedia data from the database 15.

The server 12 includes a wireless transceiver 19 connected to receive data from the controller 17, and operable to output that data, via an antenna 20, as radio frequency signals over an air interface 21. The wireless transceiver 19 may be provided by one or more wireless transceivers. Typically tens of transceivers (access points) will be used to cover a stadium or other venue.

A plurality of client devices 22a.. .22m, each of which is provided with a wireless transceiver 24a. ..24m, communicate with the server 12, and receive data transmitted from the wireless transceiver(s) In embodiments of the present invention, the data transmitted by the server 12 is multicast to all of the client devices 22a. . .22m using a single modulation and coding scheme (MCS), compressed at a target bit rate k1 bits/second. In cases where client, experiences different channel conditions to c1ient (i!=j), due to different packet error rates (PER), it may be advantageous to modify the MCS mode by changing the error control coding and/or modulation mode.

Figure 3 illustrates the transmission part of the system of Figure 2 in more detail.

The transmission part includes a video subsystem 32 (provided by the encoder units 16a.. .16n of Figure 1), a data subsystem (equivalent to the data unit 15 in Figure 1), and an adaption subsystem 36 (provided by the controller 17 in Figure 1). A multicast server 38 (provided by the controller 17 in Figure 1) is connected to provide an output data stream to the wireless transceiver 19.

The video subsystem 32 comprises a video capture unit 322, a plurality of video encoders 324a.. .324n, a plurality of first video data processing units 326a. . .326n, a plurality of second video data processing units 328a. . .328n.

The video capture unit 322 is connected to receive input video data from the video data sources 13a-13n. The video capture unit 322 then outputs that video data to corresponding video encoder units 324a, ..324n, Feedback data are also input into the video encoder units 324a. . 324n from the adaption subsystem 36 as will be described in more detail below.

Each video encoder unit 324a.. .324n may be a flexible video encoder, or may alternatively be a flexible video transcoder. Each video encoder unit 324a. . .324n implements adaptive video bit rate, resolution, and structure encoding on the arriving video stream data by creating multiple video data slices, in which each slice is of a fixed byte size, The fixed slice size takes account of the packet header overhead required for the various transport protocols.

Each video encoder unit 324a. . . 324n then passes encoded video data to a corresponding first video data processing unit 326a. . 326n which also receives feedback data from the adaption subsystem 36. A video encoder unit 326a. ..324n removes redundant data from the encoded video slice data, before undergoing packetisation, buffering and multiplexing. The redundant data removed by the first

I

video data processing unit 326a.. .326n can include periodically repeated information within the input video data streams.

Each first video data processing unit 326a.. .326n then passes processed data to a corresponding second video data processing unit 328a. . ,328n, which also receives feedback data from the adaption subsystem 36. The received data undergoes cross packet adaptive forward error correction (FEC) using erasure encoding, buffering and further encoding. The further encoded data output from each second data refinement unit 328a.. .328n is then forwarded to the multicast server 38 which implements packet flow control upon such received data packets.

The server 38 then outputs data packets to the wireless transceiver 19 for transmission to client units via the antenna 20.

Content analysis and statistical multiplexing of input video streams within the video subsystem 32 maximises channel number, video quality and/or FEC rate for the data being transmitted by the server 18.

The data subsystem 34 operates in parallel to the video subsystem 32, and comprises a data capture unit 342, a data formatting unit 344, a data carousel unit 346, an encoder unit 348, and a client preference server 349.

The data capture unit 342 acquires multimedia data that is to be made available to the multicast clients. This multimedia data may include HTML files (for example, team information, stadium information etc), audio files, video clips, and game statistics. High bandwidth items are compiled to be sent via a data carousel, whilst timely low bandwidth information (for example, late breaking scores or in-game information) is sent to the first data processing units 326a.. ,326n of the video subsystem 32 for delivery to the clients via a parallel data stream.

The process of compressing and restructuring the data for transmission over the data carousel is performed by the data formatting unit 344. Tn practice, two or more data carousels may be used (one comprising the full data set, and others comprising updates or specific data subsets). Metadata (data about the data) is also created to allow the dataset to be search manually, via a client browser, or automatically via the client preference server 349. The combination of data and metadata allows information of interest to specific clients to be presented to their users. Data to be delivered using the data carousel method is transmitted to the client devices for local storage thereon. The client device is then able to access the data when required without the need for a unicast data request and data delivery scheme. The data carousel technique is particularly suitable for data that changes infrequently.

The data carousel unit 346 packetises the data generated by the data formatting unit 344, into a form suitable for cross packet FEC encoding.

The encoder unit 348 is independent of the video FEC encoders, and operates with flexible coding rate and block size parameters. These parameters are defined manually, or via the adaption subsystem 36, based on channel, environment and latency issues. For more challenging radio channels, a lower FEC rate andlor a larger block size may be used.

The client preference server 349 maintains personal profile information for all active clients on the system (i.e. connected to any distribution AP). The information may include name and address, current location (i.e. seat number), billing information (for product purchases), and personal preferences (favourite teams, players etc.). Metadata from the data formatting unit 344 is cross referenced periodically against the information in the client preference server to determine if client specific alerts or information should be provided. The client preference server 349, in combination with the data formatting unit 344 allows the system to provide a personalisect service to all clients over a multicast distribution network.

The inclusion of per-client preference information with video and database metadata automatically enables relevant content to be displayed on a client device 22a-22n. The availability of personalised content can be indicated in a number of ways at the client device. For example, the user of a client device may be alerted that the content has become available, the content may be displayed automatically, or the content may be presented to the user in response to a specific user request.

The client user interface provides the user with the ability to publish their preferences and interests to the client preference server 349, which can then cross reference these interests against metadata. If a match is found, an alert can be sent, via the data carousel, or via a parallel data stream (using a proprietary session announcement protocolsession announcement protocol) in the video subsystem 32 to inform the user of the relevant update, data, or video stream. This provides the user with the appearance of a personalised service.

The adaption subsystem 36 comprises a quality of service feedback processing unit 360, which receives short unicast packets from the active client device group. This data is generated periodically by each client device 22a.. .22n, and provides feedback information including items such as signal quality, packet loss rate, and FEC decoder loss rate. The data from all the clients is combined to form an adaptive error surface. This is then used to determine key parameter changes, such as FEC rate, block size, and AP MCS mode. The use of unicast group feedback combined with multicast distribution provides a robust, self-adapting and scaleable video and data distribution solution for tens of thousands of fixed or mobile clients.

An adaptive system controller 362 interfaces with the encoder unit 324a. . 324n, and the wireless transceiver 19 adjust system parameters dynamically "on-the-fly, based on the quality of service feedback data processed by the processing unit 360.

The multicast server 38is responsible for sending the video and multimedia data multicast packets to the client devices via the wireless transceiver 19, and includes intelligent packet flow control to ensure that packets sent to the transceiver 19 via Ethernet are not lost in the transceiver's limited input buffer. The transceiver 19 must support nmlticast and unicast traffic. To achieve this, multicast transmissions are limited to specific signalling periods. Since the transmission of the multicast packets is inherently bursty in nature, careful packet flow and smoothing is required to avoid dropped packets and to achieve smooth video playback.

The server 38 is able to connect to any number of wireless transceivers 19 (one is shown in Figure 2 for the sake of clarity), as determined by the required coverage and capacity. Each transceiver sends the same set of multicast packets to the client devices. The transceivers support a mixture of unicast and multicast data. Unicast is used for standard wireless LAN services, including over the top applications like betting and electronic shopping. To ensure full functionality, the wireless transceivers allow the modulation and coding scheme (MCS) for multicast traffic to be set remotely via Ethernet (or equivalent) by the adaption subsystem 36.

Figure 4 shows a block diagram of a client device 22a.. .22n, which includes the wireless transceiver 24. In addition, the client device includes a multicast client unit 70, a video subsystem 72, an adaption subsystem 74, a data subsystem 76, and a local application 78.

The video subsystem 72 includes a first data processor 720 which is operable to reinsert the redundant data (required by the standard video player) that was removed (to save bandwidth) by the first data processing unit 326a...326n of the video subsystem 32 of the transmission system. The first data processor 720 also extracts information conveyed in the parallel data stream. A schematic representation of the mechanism which implements the FEC encode and redundant data removal and replacement is shown in Figure 5.

A decoder unit 722 extracts and buffers received FEC symbols, and then performs cross packet FEC decoding. Depending on the FEC rate and block size, which is dynamically controlled via group client feedback, only a subset of the transmit packets are required in order to successfully recover the original FEC block. The use of cross packet FEC overcomes the lack of packet retransmission in the wireless multicast system.

A UDP server 724 acts as a unicast packet server to bridge the received multicast video stream into a video decoder unit 726. The decoder unit 726 may be a standard unit which includes video decoding, digital rights management (DRM) and display driving. Alternatively, the decoder unit may be a higher performance unit that includes a low-latency video decoder, digital rights management, error concealment, and display driving.

Since standard mobile video players typically do not support operation over a multicast link, and instead rely on unicast signals, the UDP server 724 imitates such a uriicast data stream for the player concerned. It will be appreciated that a video player that supports multicast transmission could be provided, and that the UDP server 724 would then not be required.

The video decoder unit 726 also receives overlay data from a local video overlay unit 728. The overlay unit 728 supplies data to be incorporated into the display to the user, and such data is provided by the local application 78. The local application receives data from the data subsystem 76 (described below), and generates the data to be overlaid on the video images received via the video subsystem 72.

The client data subsystem 76 comprises a carousel unit 762, and a database control unit 764. The carousel unit 762 receives and processes the incoming multicast packets from a chosen data carousel being transmitted by the transmission system.

On request from the local application 78, the carousel unit 762 performs FEC decoding for a specified data carousel stream. A list of available carousels is included as proprietary data in the parallel data stream (using a proprietary session announcement protocol). Received data is stored until the entire carousel has been received. Once all the data has been received, it is passed to the database control unit 764.

The database control block 764 extracts the multimedia data from the received carousel and updates the necessary local databases and file systems. This data is then available to the local application 78.

The client adaption subsystem 74 comprises a quality of service extraction unit 742 and a feedback server 744. The quality of service extraction unit 742 computes parameters such as the packet loss and FEC decoder block error rates. This information, together with the received signal level, is then passed to the feedback server 744.

The feedback server 744 intermittently sends back standard unicast data packets from the client to the processing unit 360 of the adaptation subsystem 36 of the transntission system shown in Figure 2. These data are combined with information from other clients to drive the adaptive system controller 362.

A schematic representation of the mechanism which implements the FEC encode and redundant data removal and replacement is shown in Figure 4.

In use, the wireless modulation and coding mode for multicast transmission is selected together with the video structure and the number of video streams based on latency, coverage and video quality needs. The values assigned to each of these parameters are then dynamically adjusted for the entire system, based on quality of service statistics gathered from the feedback received by wireless transceiver 19 from a group of active wireless clients 22a. . .22n. In this case, the video transmission bit rate and resolution of each stream of video data are adapted based on the signal issued by the adaptive system controller 362. The signal generated by adaptive system controller 362 is based on analysis, by the processing unit 360, of at least one of the following parameters: content analysis, statistical multiplexing, and the level of required cross packet FEC. These per-stream parameters are optimised in real-time, based on quality of service feedback statistics generated by data processing unit 360 based on latency, coverage and video quality targets data which are set by the system operator.

Adaptation of the video, wireless and error correction parameters based on the output from adaptive subsystem 36, is performed based on a closed loop approach to ensure optimum multicast delivery to all clients within the footprint of the wireless transceiver (access point or base station). This dynamic approach ensures that the system is able to self adapt and configure to changing environments and crowd levels. Quality of service statistics generated by the data processing unit 360 can also be used for diagnostic and maintenance processes which may be carried out directly or remotely via a wired or wireless network.

The feedback data provided by a group of active clients 22a. . 22n enables adaptation of a variety of parameters including video encoder bit rate and resolution, cross packet block size and FEC rate, video structure and wireless multicast modulation and coding mode to optimise the trade-off between video quality, wireless coverage and end-to-end latency.

A network embodying one or more aspects of the present invention can enable video quality, data robustness and signal coverage to be optimised without operator involvement. Such a network is adaptable to environmental changes and also to changing crowd positions when in use. The techniques embodying aspects of the present invention use significant levels of cross layer interaction to optirnise performance. For example, video data is packetised intelligently and key parameters are sent separately from the video data over the unreliable wireless multicast channel. Video transcoding or encoding is used to adjust the video rate to the FEC rate and available wireless rate as well as to restructure the video data to adjust dynamically end-to-end latency and to support mobile devices where short video data structures are desirable.

Embodiments of the present invention can implement intelligent generation and packaging of compressed video information into self-contained chunks suitable for transmission over wired and wireless networks such that a single packet loss will only impact a single slice of the compressed video, and unrecovered packets in an FEC block will only affect a single portion of video data. In addition, such an embodiment can facilitate joint wireless/video adaptation which operates on a "self-healing" basis for multicast video data streams being transmitted to overcome outages in the reception of the transmitted signal caused by issues such as crowd formation and motion which can occur in stadium (or other) environments.

The ability to adjust and scale the video rate to the available wireless throughput achieves robust wireless video data delivery. The system automatically adapts cross packet FEC parameters to any environment and crowd level and this simplifies installation and maintenance issues. As dedicated encoding or transcoding is required for mobile devices, which place specific constraints on the video structure, adaptive transcoding or encoding for analogue video inputs is applied adaptively prior to wireless multicast distribution. The performance of a stadium based system will vary significantly when crowds of people are present and the system 10 is robust enough to self-adapt to crowd levels to guarantee reception quality and stadium wide coverage.

Claims

CLAIMS1. A method of transmitting multicast video data to a plurality of client receivers, the method comprising: transmitting video data to a plurality of client receivers simultaneously using an adaptive transmission scheme; receiving unicast feedback data from at least one client receiver, the feedback data including feedback information relating to received video data at the client receiver; updating the adaptive transmission scheme in dependence upon received feedback data; transmitting subsequent video data to the plurality of client receivers using the updated adaptive transmission scheme.
2. A method as claimed in claim 1, wherein the adaptive transmission scheme includes at least one of an adaptive encoding/transcoding scheme, and an adaptive wireless modulation and coding scheme, and an adaptive cross packet forward error correction scheme.
3. A method as claimed in claim 1 or 2, wherein such unicast feedback is received from a predetermined subset of the plurality of client radio receivers.
4. A method as claimed in claim 1 or 2, wherein such unicast feedback is received from each of the plurality of client radio receivers.
5. A method as claimed in any one the preceding claims, wherein the adaptive transmission scheme includes at least one of: adaptive video rate, cross-packet FEC rate, and video structure, and the feedback information includes information relating to at least one of: video quality, video channel number, and latency, packet loss rate, and cross packet forward error correction decode error rate.
6. A method as claimed in any one of the preceding claims, further comprising transmitting multimedia data, different to the video data, using a second adaptive transmission scheme, the second adaptive transmission scheme being adapted in dependence upon the received feedback data.
7. A method as claimed in claim 6, wherein the second adaptive transmission scheme include at least one of an adaptive encoding/transcoding scheme, and an adaptive modulation scheme, and an adaptive cross packet forward error correction scheme.
8. A method as claimed in claim 6 or 7, wherein the second adaptive transmission scheme includes at least one of: adaptive video rate, cross-packet FEC rate, and video structure, and the feedback information includes information relating to at least one of: latency, packet loss rate, and cross packet forward error correction error rate.
9. A wireless multicast video data transmission system comprising: a transmitter operable to transmit video data to a plurality of client receivers simultaneously using an adaptive transmission scheme; and Y) a receiver operable to receive unicast feedback data from at least one client receiver, the feedback data including feedback information relating to received video data at the client receiver concerned, wherein the transmitter is operable to update the adaptive transmission scheme in dependence upon received feedback data, and to transmit subsequent video data to the plurality of client receivers using such an updated adaptive transmission scheme.
10. A system as claimed in claim 9, wherein the adaptive transmission scheme includes at least one of an adaptive encoding/transcoding scheme, and an adaptive wireless modulation and coding scheme, and an adaptive cross packet forward error correction scheme.
11. A system as claimed in claim 9 or 10, wherein such unicast feedback is received from a predetermined subset of the plurality of client receivers.
12. A system as claimed in claim 9 or 10, wherein such unicast feedback is received from each of the plurality of client receivers.I
13. A system as claimed in any one of claims 9 to 12, wherein the adaptive transmission scheme includes at least one of: adaptive video rate, cross-packet FEC rate, and video data structure, and the feedback information includes information relating to at least one of: video quality, video channel number, and latency, packet loss rate, and cross packet forward error correction decode error rate.
14. A system as claimed in any one of claims 9 to 13, wherein the transmitter is operable to transmit multimedia data, different to the video data, using a second adaptive transmission scheme, the second adaptive transmission scheme being adapted in dependence upon received feedback data.
15. A system as claimed in claim 14, wherein the second adaptive transmission scheme includes at least one of an adaptive encoding/transcoding scheme, and an adaptive modulation scheme, and an adaptive cross packet forward error correction scheme.
16. A system as claimed in claim 14 or 15, wherein the second adaptive transmission scheme includes at least one of: adaptive video rate, cross-packet FEC rate, and video structure, and the feedback information includes information relating to at least one of: video quality, video channel number, latency, packet loss rate and cross packet forward error correction decode error rate.
17. A method of decoding a received wireless multicast video data stream, the method comprising: receiving a wireless multicast video data stream; converting a received wireless multicast data stream to multicast video data; converting such multicast video data into unicast format video data; decoding such unicast format video data into a video display driver signal.
18. A device for receiving a wireless multicast video data stream, the device comprising: a receiver unit operable to receive a wireless multicast video data stream, and to output multicast video data; a data processor operable to receive multicast video data from the receiver unit and to output unicast format video data; a video decoder operable to receive unicast format data from the data processor and to output a video display driver signal relating to such received unicast format data.
19. A method of transmitting a wireless multicast video data stream to a plurality of receivers, the method including removal of periodically repeated information from the video stream, and the transmission of such removed information separately from the video stream.
20. A method of transmitting wireless multicast video data stream to a plurality of receivers, the method including transmitting multimedia data, different from the video data stream, to the receivers separately from the video data stream.
21. A method of receiving a wireless multicast video data stream transmitted in accordance with a method as claimed in claim 20, the receiving method including selecting multimedia data for display in dependence upon comparison of metadata relating to the multimedia data with preference information for the receiver concerned.
22. A method of transmitting personalized data to a receiver using mulitcast transmission, the method comprising: transmitting content data to a plurality of receivers using a multicast transmission scheme; receiving a unicast data transmission from a receiver, which unicast data transmission includes preference data for the receiver; and transmitting, in parallel to the content data, an alert to the receiver when the content data includes items indicated by the preference data.
23. A method as claimed in claim 2, wherein the content data include video data and/or multimedia data.
24. A method as claimed in claim 22 or 23, wherein the alert is in the form of a real-time session announcement protocol message.