WO2007021698A2 - Multi-channel video pump - Google Patents
Multi-channel video pump Download PDFInfo
- Publication number
- WO2007021698A2 WO2007021698A2 PCT/US2006/030926 US2006030926W WO2007021698A2 WO 2007021698 A2 WO2007021698 A2 WO 2007021698A2 US 2006030926 W US2006030926 W US 2006030926W WO 2007021698 A2 WO2007021698 A2 WO 2007021698A2
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- stream
- packet
- signals
- bit
- bit rates
- Prior art date
Links
- 230000005540 biological transmission Effects 0.000 claims description 29
- 238000000034 method Methods 0.000 claims description 9
- 238000010586 diagram Methods 0.000 description 11
- 239000000872 buffer Substances 0.000 description 10
- 230000007246 mechanism Effects 0.000 description 9
- 230000032258 transport Effects 0.000 description 6
- 238000005516 engineering process Methods 0.000 description 4
- 230000003111 delayed effect Effects 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 238000012546 transfer Methods 0.000 description 3
- 230000006835 compression Effects 0.000 description 2
- 238000007906 compression Methods 0.000 description 2
- 238000010276 construction Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000007493 shaping process Methods 0.000 description 2
- 230000006978 adaptation Effects 0.000 description 1
- 230000002411 adverse Effects 0.000 description 1
- 230000032683 aging Effects 0.000 description 1
- 238000007630 basic procedure Methods 0.000 description 1
- 210000004556 brain Anatomy 0.000 description 1
- 230000001186 cumulative effect Effects 0.000 description 1
- 125000004122 cyclic group Chemical group 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000005538 encapsulation Methods 0.000 description 1
- 239000000835 fiber Substances 0.000 description 1
- 238000003780 insertion Methods 0.000 description 1
- 230000037431 insertion Effects 0.000 description 1
- 230000002452 interceptive effect Effects 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000002093 peripheral effect Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 230000001105 regulatory effect Effects 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L47/00—Traffic control in data switching networks
- H04L47/10—Flow control; Congestion control
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04J—MULTIPLEX COMMUNICATION
- H04J3/00—Time-division multiplex systems
- H04J3/02—Details
- H04J3/06—Synchronising arrangements
- H04J3/062—Synchronisation of signals having the same nominal but fluctuating bit rates, e.g. using buffers
- H04J3/0632—Synchronisation of packets and cells, e.g. transmission of voice via a packet network, circuit emulation service [CES]
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04J—MULTIPLEX COMMUNICATION
- H04J3/00—Time-division multiplex systems
- H04J3/22—Time-division multiplex systems in which the sources have different rates or codes
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L47/00—Traffic control in data switching networks
- H04L47/10—Flow control; Congestion control
- H04L47/22—Traffic shaping
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L47/00—Traffic control in data switching networks
- H04L47/10—Flow control; Congestion control
- H04L47/28—Flow control; Congestion control in relation to timing considerations
- H04L47/283—Flow control; Congestion control in relation to timing considerations in response to processing delays, e.g. caused by jitter or round trip time [RTT]
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/20—Servers specifically adapted for the distribution of content, e.g. VOD servers; Operations thereof
- H04N21/21—Server components or server architectures
- H04N21/218—Source of audio or video content, e.g. local disk arrays
- H04N21/21815—Source of audio or video content, e.g. local disk arrays comprising local storage units
- H04N21/2182—Source of audio or video content, e.g. local disk arrays comprising local storage units involving memory arrays, e.g. RAID disk arrays
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/20—Servers specifically adapted for the distribution of content, e.g. VOD servers; Operations thereof
- H04N21/23—Processing of content or additional data; Elementary server operations; Server middleware
- H04N21/231—Content storage operation, e.g. caching movies for short term storage, replicating data over plural servers, prioritizing data for deletion
- H04N21/2312—Data placement on disk arrays
- H04N21/2318—Data placement on disk arrays using striping
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/20—Servers specifically adapted for the distribution of content, e.g. VOD servers; Operations thereof
- H04N21/23—Processing of content or additional data; Elementary server operations; Server middleware
- H04N21/234—Processing of video elementary streams, e.g. splicing of video streams or manipulating encoded video stream scene graphs
- H04N21/23406—Processing of video elementary streams, e.g. splicing of video streams or manipulating encoded video stream scene graphs involving management of server-side video buffer
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/20—Servers specifically adapted for the distribution of content, e.g. VOD servers; Operations thereof
- H04N21/23—Processing of content or additional data; Elementary server operations; Server middleware
- H04N21/238—Interfacing the downstream path of the transmission network, e.g. adapting the transmission rate of a video stream to network bandwidth; Processing of multiplex streams
- H04N21/23805—Controlling the feeding rate to the network, e.g. by controlling the video pump
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/20—Servers specifically adapted for the distribution of content, e.g. VOD servers; Operations thereof
- H04N21/23—Processing of content or additional data; Elementary server operations; Server middleware
- H04N21/238—Interfacing the downstream path of the transmission network, e.g. adapting the transmission rate of a video stream to network bandwidth; Processing of multiplex streams
- H04N21/2381—Adapting the multiplex stream to a specific network, e.g. an Internet Protocol [IP] network
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/20—Servers specifically adapted for the distribution of content, e.g. VOD servers; Operations thereof
- H04N21/23—Processing of content or additional data; Elementary server operations; Server middleware
- H04N21/238—Interfacing the downstream path of the transmission network, e.g. adapting the transmission rate of a video stream to network bandwidth; Processing of multiplex streams
- H04N21/2385—Channel allocation; Bandwidth allocation
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/60—Network structure or processes for video distribution between server and client or between remote clients; Control signalling between clients, server and network components; Transmission of management data between server and client, e.g. sending from server to client commands for recording incoming content stream; Communication details between server and client
- H04N21/63—Control signaling related to video distribution between client, server and network components; Network processes for video distribution between server and clients or between remote clients, e.g. transmitting basic layer and enhancement layers over different transmission paths, setting up a peer-to-peer communication via Internet between remote STB's; Communication protocols; Addressing
- H04N21/643—Communication protocols
- H04N21/64307—ATM
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04J—MULTIPLEX COMMUNICATION
- H04J2203/00—Aspects of optical multiplex systems other than those covered by H04J14/05 and H04J14/07
- H04J2203/0001—Provisions for broadband connections in integrated services digital network using frames of the Optical Transport Network [OTN] or using synchronous transfer mode [STM], e.g. SONET, SDH
- H04J2203/0089—Multiplexing, e.g. coding, scrambling, SONET
- H04J2203/0096—Serial Concatenation
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04J—MULTIPLEX COMMUNICATION
- H04J3/00—Time-division multiplex systems
- H04J3/02—Details
- H04J3/06—Synchronising arrangements
- H04J3/0635—Clock or time synchronisation in a network
- H04J3/0685—Clock or time synchronisation in a node; Intranode synchronisation
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L47/00—Traffic control in data switching networks
- H04L47/10—Flow control; Congestion control
- H04L47/43—Assembling or disassembling of packets, e.g. segmentation and reassembly [SAR]
- H04L47/431—Assembling or disassembling of packets, e.g. segmentation and reassembly [SAR] using padding or de-padding
Definitions
- the present invention is directed to streaming video signals and, more particularly, to an apparatus for simultaneously streaming user-specified video files encoded at varying bit rates over a single network.
- the role of streaming video in local area networks is expected to increase rapidly in the near future due to developments in video compression and deployment of transmission systems with increased bandwidth.
- MPEG Moving Pictures Expert Group
- This stream can be captured and stored on appropriate media such as a redundant array of independent disks (RAID) or a digital video (or versatile) disc (DVD).
- RAID redundant array of independent disks
- DVD digital video (or versatile) disc
- the MPEG compression standards are used worldwide for constant bit rate digital video encoding. Decoding of MPEG video so that each picture and each audio frame is played once and only once relies on the ability to deliver each bit from the encoder to the decoder with a constant delay. This constant bit rate delivery is generally termed "isochronous streaming.” In live broadcasts the encoder is responsible for generating the MPEG bit stream at the proper rate. However, when this information is stored for later playback another mechanism is required to "meter" the data from the storage media to the playback device. Normally, no feedback is provided to the sender by the receiver of MPEG video. The receiver depends on the transmission rate to be both smooth and accurate in order to decode MPEG video properly.
- MPEG audio and video content may be recorded at any arbitrary rate.
- Some examples are streams that are 3.282, 3.420, 6.144, and 6.000 megabits per second.
- Some conventional systems use a handshake protocol to inform the receiving device what is the bit rate of the video stream that will be sent. However, that requires the receiving device to be programmed to use the protocol to communicate with the sending device.
- Other systems distribute the MPEG data in large "chunks" of data (up to tens of kilobytes) that require the receiving device to have enough expensive memory to buffer the data for smooth display. This type of delivery precludes using the MPEG System Clock synchronization mechanism which is required for precise playback.
- Compressed video (e.g., MPEG-2) data is normally transmitted via satellite, cable, terrestrial digital broadcast and other transmission systems using serial bit stream mechanisms.
- the data is clocked one bit at a time into the transmission data stream using the bit level clock of the transmission system.
- the average data rate is regulated directly by the bit level clock and the jitter is only that present on the bit level clock (normally substantially less than one millisecond). Jitter is a measurement of how early or how late a specific bit arrives from its intended arrival time.
- the MPEG data streams that are sent on these type of transmission systems are multiplexed (with multiple programs and padding) so that the total MPEG stream rate is exactly the same as the network payload bit rate.
- An MPEG data stream cannot be as easily sent using an asynchronous transmission system or network like Ethernet or Asynchronous Transfer Mode (ATM), because such asynchronous networks operate on the packet (or cell) level rather than the bit level. Due to the use of packets to transmit data, it is more difficult to recover the input bit rate for a data stream from a network like Ethernet or ATM than from a multiplexed serial signal sent at a similar rate.
- a packet network will transmit an entire packet of MPEG data at the bit rate of the packet network, e.g., 155.52 Mega-bits per second (Mb/s) for ATM Optical Carrier - Level 3 (OC-3).
- jitter increases as the packet size increases and as the difference increases between the input data stream rate and the network transmission rate.
- the bit rate is 155.52 Mb/s and the cell size is 53 bytes (roughly 48 payload and 5 overhead bytes).
- one ATM cell transmitted at 155.52 Mb/s typically contains 384 bits of MPEG data.
- the packet size is typically closer to 1000 bytes. So, for roughly 1000 bytes, the MPEG data stream created at 3.42 Mb/s, is sent at 100 Mb/s to produce a maximum jitter of 2.26 ms.
- a video distribution system using a packet network like 100 Mb/s Ethernet or ATM to transmit MPEG data to devices that are not programmed to use a handshake protocol must strike a balance between the size of receiving video buffers and the packet size, so as to keep overhead at an acceptable level and still provide a smooth flow of data.
- the absolute minimum size of the receiving video buffer will be determined by the best-case jitter of the last bit, calculated using equation (1).
- each stream is well-behaved to maximize network efficiency.
- Well-behaved implies that each stream is as nearly isochronous as possible within the packet structure of the network. This is referred to as packet isochronous transmission.
- bursty transmission of MPEG video streams in the network will result in congestion and network failure much more quickly than constant bit rate transmission.
- the more closely the individual data streams are maintained at a constant bit rate the higher the total aggregate of such streams that can be carried on the network while maintaining a desired quality of service.
- a system for transmitting multiple streams of stored signals to receiving devices including at least one playback device to access recordings, each recording containing stored signals encoded at one of a plurality of bit rates; a streaming device, coupled to the at least one playback device, to receive a request to reproduce a specified recording, to detect the one of the bit rates used to encode the stored signals on the specified recording, and to output packet isochronous signals based on the stored signals on the specified recording at the one of the bit rates; and a network, coupled to the streaming device and the receiving devices, to deliver the packet isochronous signals to the receiving devices.
- the streaming device includes a plurality of timer circuits, each including a base counter to count a truncated period for transmission of packets; and a dithering circuit to indicate transmission of one of the packets one clock pulse later than the truncated period, a predetermined number of times within a dither cycle.
- Figures IA and IB together provide a timing diagram showing differences between desired MPEG data timing and the timing provided by a packet network.
- Figure 2 is a block diagram of a digital media retrieval system using the present invention.
- FIG. 3 is a flowchart of the operation of a video streaming device according to the present invention.
- Figure 4 is a functional block diagram of a video streaming device according to the present invention.
- FIG. 5 is a block diagram of the hardware architecture of a video streaming device according to the present invention.
- Figure 6 is a block diagram of a channel timing module according to the present invention.
- Figure 7 is a block diagram of one set of channel timing counters in a video streaming device according to the present invention.
- Figure 8 is a block diagram of a digital media retrieval system using multiple video streaming devices to produce video streams with a total of 480 megabits per second.
- Figure IA illustrates the bit pattern desired for the MPEG stream (isochronous transmission).
- Figure IB illustrates the bit pattern for the same MPEG data stream as it is transmitted in a packet network, i.e., packet isochronous transmission.
- packet isochronous transmission.
- the precise release point of a given packet will not have much effect on jitter as long as the magnitude of the error of actual release versus desired release is small compared to the jitter of the last bit caused by the difference between the network and stream bit rates, which will be referred to as the best case jitter of the last bit.
- the first bit can be released several bits early or late without changing the order of magnitude of the best case jitter of the last bit.
- Another factor to be controlled is average bit rate. If a small packet is used to reduce the best case jitter of the last bit, the packets will be released at a higher frequency than required to transmit the same MPEG data stream using larger packets, hi addition, the smaller the packet, the shorter the interval between packets, assuming the release of each packet is delayed so that the first bit of each packet is released at the correct time. A unidirectional error (i.e., always early or always late) in the period between releases is multiplied by the number of releases per second and thus, an accurate inter-packet time is key to maintaining a constant average bit rate.
- the packet size could be large to reduce the frequency and the absolute timing tolerance per release. For example, if one packet were released per second, a one bit per second accuracy could be obtained using a 100 KHz clock and a 17-bit count down timer set for 100000. If this timer is off by one clock cycle, the error is only 1 / 100000. But, having a release at one second intervals using the example above creates a jitter of 966 ms for a 3.353 Mb/s stream.
- the packets must be released 419.125 times per second. It is more difficult to precisely achieve this rate. Again using a 100 KHz clock, the counter must count 238.5923054 clock pulses between the release of each packet. If the period is 238 clock pulses, 420.1680672 packets will be released per second for a bit rate of 3,361,344 bits per second. If the period is 239 clock pulses, 418.4100418 cells will be released per second for a bit rate of only 3,347,280 bits per second.
- the amount of error resulting from a period of 238 or 239 clock pulses will preclude locking the decoder clock to the program clock reference (PCR) embedded in the MPEG data stream, since it would place the ratio of the decode clock to the display clock outside National Television Standard Committee (NTSC) limits.
- PCR program clock reference
- NTSC National Television Standard Committee
- Using a period of 239 clock pulses results in losing 5719 (3,353,000-3,347,280) bits per second which is a bit rate error of only .17% (5719/3,353,000).
- a 3.353 Mb/s MPEG data stream has an average frame size of 111,766 bits, leaving the receiving device one frame short every 19.54 (111766/5719) seconds.
- a frame hold would need to occur about every 20 seconds.
- increasing the MPEG data rate reducing the packet size or reducing the clock frequency used for timing packet releases will cause the error from a single stage counter to go up.
- FIG. 2 Illustrated in Fig. 2 is a block diagram of digital media retrieval system 10 using a two-stage timer mechanism which allows much higher counter resolution than a single stage counter with the same clock frequency.
- Digital media retrieval system 10 provides interactive distribution of video, text, graphics, and Internet content over a high speed digital network.
- Video pump 12 is a key component in system 10. The purpose of video pump 12 is to retrieve MPEG audio/video streams from various storage devices, such as RAID array 14 and DVD jukebox 16 and place this data into high speed digital network 18 for distribution to set top devices 20 at the specific rate required for each stream. Channels are opened in the system illustrated in Fig. 2 to transport data from the storage devices 14, 16 to set top devices 20 via ATM network 18.
- Video pump 12 responds to system commands from system control server 22 for the retrieval and distribution of isochronous data including both audio and video. For simplicity's sake, this data will subsequently be referred to as either video or simply as data. Although illustrated as separate components in Fig. 2 that may be networked, it is possible to construct video server 10 such that video pump 12 and system control server 22 are in the same computer system.
- PVC Permanent Virtual Circuit
- SVC Switched Virtual Circuit
- CBR Constant Bit Rate
- ATM network 18 is able to establish end-to-end connections with guaranteed bandwidth availability and requires that data is introduced to ATM network 18 in such a way that the established connection rate is not exceeded. If the bit rate of a specific connection exceeds that agreed to when the connection was established ATM network 18 may discard the excess data.
- the timing mechanism of the present invention used in conjunction with the ATM interface provides MPEG data streaming that meets all required specifications for bit rates between roughly 1 and 20 megabits per second.
- Video pump 12 may be strictly a server, with commands received via a command protocol over TCP/IP, such as real time streaming protocol (RTSP). These commands open and close video streams, assign video streams to specific PVC/SVC channels, and perform actions on these video streams, such as pause, play, stop, fast forward, rewind, etc.
- Video pump 12 receives via the commands, the start and stop addresses of the data within a given file that is to be streamed through ATM network 18.
- Video pump 12 provides timing to allow each individual channel to be streamed at unique arbitrary rates, m this embodiment, a maximum of 60 channels may be streamed, with a maximum total aggregate bit rate of 120 Mb/s.
- the timing for each channel may be specified via the application program interface (API) executing on system control server 22 or set top device 20, or determined directly from the stream itself by video pump 12 using the program clock references (PCRs) that are stored in MPEG transport stream data at least every 100 milliseconds.
- Video primp 12 can determine the bit rate of the signal by the number of bits between PCRs and the difference in time between the PCRs.
- Video pump 12 can operate on blocks of data as small as two MPEG transport packets (376 bytes) to minimize jitter imposed by the distribution of video within the system and to comply with ATM Forum requirements for MPEG-2 transmission.
- Fig. 3 The basic procedure for generating a stream of packet isochronous signals according to the present invention is illustrated in Fig. 3.
- Stored signals are read 24 from a recording that has been specified for playback in a conventional manner.
- the bit rate used to encode the stored signals on the recording is detected 26 by detecting the PCRs.
- the stream of packet isochronous signals is output 28 with timing more precise than the clock signals in the device, by using a two stage timer, as described below.
- Video pump 12 has four main functional components that operate as individual processes.
- RAID streaming logic 30 issues and receives control signals to and from ultra Small Computer System interface (SCSI) interface 31 and sends status information to control and status logic 32.
- Real-time pump 34 receives data from Dynamic Random Access Memory (DRAM) buffer 35 and data rate information from RAID streaming logic 30 and outputs the data and a channel identifier to ATM adapter 36.
- Control and status logic 32 also receives status information from realtime pump 34 and ATM adapter module 36 and issues commands to RAID streaming logic 30, real-time pump 34 and ATM adapter module 28.
- DRAM Dynamic Random Access Memory
- start and stop addresses and start and stop commands are sent to RAID streaming logic 30, open and close channel commands and parameters including a priority list of counter values are sent to real-time pump 34 and open and close channel commands and parameters indicating channel bandwidth are sent to ATM adapter module 36.
- RAID streaming logic 30 fetches data from RAID array 14. This data is placed in DRAM buffer 35 where it is read by real-time pump 34. RAID streaming logic 30 receives start and stop commands, as well data addresses from video pump control and status logic 32. RAID streaming logic 30 preferably reads data including PCRs from the video file to determine the encode rate, and passes this rate on to real-time pump 34. The encode rate is the rate at which the set top device decoder will use the data, and it is therefore the rate at which video pump 12 must send the data to the decoder, as described in more detail below. RAID streaming logic 30 also ensures that the data being read from RAID array 14 is transport packet aligned. This is crucial to the operation of video pump 12, and any errors are immediately reported to control and status logic 32.
- Real-time pump 34 is the heart of video pump 12. It is here that the data for each channel is pulled from the DRAM buffers 35 for each channel at the specified rate. Data for each channel is passed from real-time pump 34 to buffers in ATM adapter module 36 for insertion into ATM distribution network 18. hi an exemplary design described below, realtime pump 34 is capable of maintaining 60 separate video streams, each with arbitrary data rates, and processing the data flow in such a manner to minimize jitter as the data is placed in the stream, hi this embodiment real-time pump 34 is capable of maintaining an aggregate data flow bandwidth of 120 Mbps.
- ATM adapter module 36 receives the video data from real-time pump 34, packetizes this data into ATM cells, and passes this data stream on to ATM network 18 for distribution to set top devices 20.
- the data received from real-time pump 34 is in the form of MPEG transport stream packets, and the ATM encapsulation is performed according to ATM Adaptation Layer 5 (AAL5).
- ATM adapter module 36 is coupled to OC-3c (concatenated) fiber.
- a network interface device traffic shaper in ATM adapter module 36 is initialized so that for the current channel it will introduce data into the network at the closest rate to the required rate that is higher than the required rate.
- Channel timing module 40 provides a signal to transfer the data block to the network interface device traffic shaper in- ATM adapter module 36 each time the timer for a channel expires. The result is that each block of data is introduced to the network at a rate that is faster than desired. However, because only data that has been transferred to the network interface device can be sent, from time to time there will be no data available to the traffic shaper. This will result in no data being sent until the next block of data is made available. The resulting data stream will consist of a period when data is being sent too fast followed by a period in which no data is sent. Over time, the desired data rate will be achieved to the precision of the channel timing module 40, as described below.
- control and status logic 32 serves as the brains for video pump 12 by coordinating and directing all internal elements and processes.
- Control and status logic 32 provides the interface to the "outside world", receiving commands and passing status to other elements within digital media retrieval system 10.
- Control and status logic 32 processes these system level commands, generating local commands as required to the other functional elements of video pump 12.
- FIG. 5 A block diagram of the hardware architecture of a video streaming device (video pump) according to the present invention is illustrated in Fig. 5.
- the functional components illustrated in Figs. 2 and 4 correspond to the physical components illustrated in Fig. 5 as follows.
- Video pump 12 may be constructed using a standard "IBM PC” type Intel ® Pentium ® platform with processor 42 connected to Peripheral Component Interconnect (PCI) bus 44.
- PCI Peripheral Component Interconnect
- Channel timing module 40 which implements multiple two-stage timers may be a custom PCI interface card inserted into PCI bus 44.
- DRAM buffers A and B 35 are provided by Random Access Memory (RAM) 46.
- Hard drive 48 stores an operating system, such as Windows ® NT 4.0 and pump software executed by processor 42 to provide the functions of video pump control and status logic 32, real time pump 34, RAID streaming logic 30 and system control server 22.
- SCSI controller 50 provides an interface to devices, such as a RAID storage unit (not shown) connected via ultra SCSI 31.
- Network interface 52 corresponds to ATM adapter module 36.
- An Ethernet network interface (not shown) may also be included to provide for external control by a separate system control server 22.
- a separate system control server 22 may also be included to provide for external control by a separate system control server 22.
- the components illustrated in Fig. 5 can be replaced with different components capable of performing the functions illustrated in Figs. 2 and 4 with greater or less capacity, depending on the requirements of the system in which they operate.
- PCI interface 70 is provided by a PLX Technology PCI9050-1 PCI bus target interface, compliant to PCI Specification 2.1, to connect channel timing module 40 to PCI bus 44.
- PCI interface 70 is a PCI slave interface providing a bridge to local bus 71.
- PCI configuration registers (not shown separately) in PCI interface 70 are mapped to I/O space.
- AU resources in channel timing module 40 are preferably 32-bit accessible.
- local bus 71 preferably is clocked at 10 MHZ, enabling jitter to be reduced to a desired level. Timing of local bus accesses are determined by timer Field- Programmable Gate Array (FPGA) 72 in the manner described below.
- FPGA Field- Programmable Gate Array
- the initial configuration of channel timing module 40 is loaded from configuration EEPROM 74 attached to PCI interface 70.
- the following fields in the PCI configuration registers are loaded from configuration EEPROM 74 at power up: Device ED, Vendor ID, Class Code, Subsystem ID, Subsystem Vendor ID, and Interrupt Pin. These registers are reloaded at every instance of PCI Reset signal assertion.
- Configuration EEPROM 74 may be a Fairchild Semiconductor NM93CS46 which holds 1024 bits of information.
- the data within the device may be altered via registers within PCI interface 70, depending on the state of the protection register within EEPROM 74.
- FIG. 7 A partial block diagram of timer FPGA 72 is shown in Fig. 7.
- Each FPGA 72 contains 15 timers 80, one of which is illustrated in Fig. 7, interrupt controller 82 and an interface (not shown) to local bus 71.
- Each timer FPGA is configured upon reset via loader EPROM 84 (Fig. 6) which may be provided by Altera EPC 1441 devices each containing 400K x 1 bits of information.
- the dither cycle used by timer circuit 80 is 1024 and the clock rate is 10 MFfZ. Operation of each circuit timer 80 in channel timing module 40 is initiated by processor 42 (Fig.
- Each timer circuit 80 consists of two counters: base counter 87 and dither counter 88.
- Timer circuit 80 begins operation when the 22-bit count value and 10-bit dither value have been written in registers 85, 86 and the 22-bit count value is transferred to base counter 87.
- Dither counter 88 is initially set to all ones (1023 decimal).
- base counter 87 is decremented at each cycle of the 10 MHZ clock.
- base counter 87 reaches zero, a compare is performed between the 10-bit dither value in register 86 and dither counter 88. If dither counter 88 is less than or equal to the 10-bit dither value a timeout occurs.
- the timeout is delayed by one clock cycle when dither counter 88 is greater than the 10-bit dither value in register 86.
- Dither counter 88 is decremented at each timeout.
- Base counter 87 is reloaded with the 22-bit count value in register 85 at each timeout.
- Dither counter 88 is 10 bits wide and thus, automatically resets to its maximum value every 1024 timeouts.
- An interrupt for the timer circuit 80 is generated each time the timeout occurs if the interrupt has been enabled by setting the appropriate bit in the interrupt control register. Therefore, an average timeout period is defined by formula (2), where Base is the 22-bit count value in register 85 and Dither is the 10-bit dither value in register 86.
- a packet should be released 3,353,000/8000 or 419.125 times per second.
- 100,000/419.1241778 or 238.5923054 clock signals are counted for each time a packet is released.
- the fractional part is .5923054.
- Multiplying the fractional part by 1024 produces 606.52 which can be rounded to 607. This is the number of times out of 1024 that the count will be 239.
- the average count over 1024 releases in this example can be calculated by formula (3) as 238.59.
- the present invention is able to reduce the bits per second error from 5917 to 6.6 bits per second or about three orders of magnitude using the same clock frequency.
- the clock frequency can be increased. Increasing the clock frequency to 10 MHz reduces the error in the above example to .01 bit per second. Using the 10 MHz clock, the problem still requires a release of 8000 bits 419.125 times each second. The number of clock pulses per timeout is 10,000,000/419.125 which is 23,859.23054. The fractional part is .23054. Multiplying the fractional part by 1024 produces 236.07 which can be rounded to 236. This is the number of times out of 1024 that the count will be 23860. The number of counts out of 1024 at 23859 is 788 (1024-236).
- the average over 1024 timeouts (one dither cycle) is ((236*23860)+(788*23859)) / 1024 which is 23,859.23047.
- the number of releases per second is 10,000,000/23859.23047 or 419.1250012. This makes the average bit rate 419.1250012 * 8000 or 3,353,000.01.
- Timer 80 will start counting upon loading the base and dither counter values into base and dither counters 87, 88. This must be done as a single 32-bit write. In fact, the value loaded into base counter 86 is the desired value minus one, since the check for timeout occurs after each clock cycle. For simplicity in the description above this detail was ignored. Timer 80 may be stopped by writing all zeroes to 22-bit base count register 85.
- the local bus interface in each timer FPGA 72 provides the timing and address decode for accesses to resources of timer FPGA 72. The local bus is clocked from the same 10 MHz source that drives the timers.
- timing module 40 produces very close results for all combinations within a wide range of bit rates. For example, using a dither cycle of 1024 and a clock rate of 10 MHz, it is possible to control output of video data in 8000 bit packets onto an ATM OC-3 network for video data rates of 2 to 20 Mb/s with a maximum average data rate error that is less than one bit per second.
- each device receiving the data stream may have a video timing clock.
- a video timing clock may be used to drive an output of a video signal, such as an NTSC video signal or a Phase Alternating (or Alternation) Line (PAL) video signal.
- the devices that receive the data stream from the video pump 12 can obtain (i.e., recover) a clock that is embedded in the data stream.
- An example of the embedded clock is the PCR embedded in an MPEG data stream.
- the devices that receive the data stream may phase lock their respective video timing clock to the embedded clock obtained from the data stream.
- the video pump 12 When establishing phase lock of the video timing clock, ideally, the video pump 12 outputs the data stream at an average bit rate equivalent to the bit rate used to encode the data being streamed.
- receiving devices due to various factors (e.g., frequency inaccuracy, stability, and/or aging of components used to output the data stream), receiving devices have been developed such that when establishing phase lock of the video timing clock, the data stream from the video pump 12 can be output with an average transmission rate that falls within a range of average transmission rates.
- the range of average transmission rates may be based on (i) the bit rate used to encode the data being streamed, and (ii) upper and lower tolerances for the average transmission rate.
- the upper and lower tolerances for the average transmission rate may be based on a pe ⁇ nissible frequency deviation specified for a system clock frequency for devices that receive a data stream.
- a permissible frequency deviation for a 27 MHz clock signal may be specified as ⁇ 810 Hz.
- the system clock may operate in a range from 26,999,190 Hz to 27,000,810 Hz.
- Such a system clock frequency standard is specified in International Standard ISO/IEC 13818-1, Information technology - Generic coding of moving pictures and associated audio information: Systems, Second edition, which is incorporated by reference herein.
- a permissible frequency deviation for a recovered clock signal may be specified in the amount of error-per-Hz.
- the permissible frequency deviation may be specified as 0.00003% (810 Hz divided by 27 MHz).
- the permissible frequency deviation may be specified in "parts per million" (ppm) and thus a permissible frequency deviation of 0.00003% may be specified as 30 ppm.
- the permissible frequency deviation for a recovered clock signal may be translated to the upper and lower frequency tolerances for the average transmission rate of a data stream being output from the video pump 12.
- the upper and lower tolerances for outputting the data stream is the bit rate in Mb/s multiplied by 30.
- the upper and lower tolerances are 30 bits per million bits per second.
- the video pump 12 should output the data stream within a range of 1 Mb/s ⁇ 30 bits per second (b/s) (i.e., 999,970 b/s to 1,000,030 b/s).
- the video pump 12 should output the data stream within a range of 6 Mb/s ⁇ 180 b/s (i.e., 5,999,820 b/s to 6,000,180 b/s).
- a network that transports a data stream from the video pump 12 to a receiving device may cause some error in the amount of error-per-Hz of the data being streamed to a receiving device.
- the upper and lower frequency tolerances for the average transmission rate of the data stream being output from the video pump 12 may be reduced to upper and lower tolerances that are less than the maximum upper and lower frequency tolerances. For example, maximum upper and lower tolerances may be multiplied by a given factor, such as 0.8. For maximum upper and lower tolerances of 30 and a factor of 0.8, the reduced upper and lower tolerances is 24 ppm (30 ppm multiplied by 0.8).
- the video pump 12 should output the data stream within a range of 1 Mb/s ⁇ 24 bits per second (b/s) (i.e., 999,976 b/s to 1,000,024 b/s).
- b/s bits per second
- Other examples of the factor for reducing the upper and lower tolerances are also possible.
- Video pumps 12 may be connected to one or more storage devices, such as RAID array 14, DVD jukebox 16, and other devices, such as compact disc changers, not shown.
- the disclosed embodiment is used with an Asynchronous Transfer Mode (ATM) network that is able to support multiple constant bit rate streams per segment as well as the bursty traffic created by more traditional network traffic.
- ATM Asynchronous Transfer Mode
- the present invention is not limited to use with ATM networks, but could be used with any network that can deliver the required amount of data. The quality of delivery will be dependent on the quality of service provided by the network.
- New protocols for TCP/UDP over switched and gigabit Ethernet networks may eventually support a quality of service provided by ATM networks, but presently an Ethernet network will be able to transmit a smaller number of high quality video streams per network segment than ATM and will be adversely affected by other network traffic.
- the present invention will work very well on any network with high quality of service capabilities including IEEE 1394 and networks conforming to the Home Phone Network Alliance (PNA) V 2.0 specification. Further, since numerous modifications and changes will readily occur to those skilled in the art, it is not desired to limit the invention to the exact construction and operation illustrated and described, and accordingly all suitable modifications and equivalents may be resorted to, falling within the scope of the invention.
Landscapes
- Engineering & Computer Science (AREA)
- Signal Processing (AREA)
- Multimedia (AREA)
- Computer Networks & Wireless Communication (AREA)
- Computer Hardware Design (AREA)
- Databases & Information Systems (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
- Two-Way Televisions, Distribution Of Moving Picture Or The Like (AREA)
- Compression Or Coding Systems Of Tv Signals (AREA)
- Television Signal Processing For Recording (AREA)
Abstract
Video or other converted signals streamed from storage to receiving devices maintain signal streams at an encoded bit rate which is detected. A queue is set-up for outputting data at the detected bit-rate. A channel-timing module in the streaming device contains a two-stage dithered counter for each bit rate. The first stage of the counter counts one clock cycle longer than the second stage. A ratio of the first stage and second stage counters is adjusted in a fixed number of cycles to provide a precise average count. The average count is calculated to achieve the desired bit rate with a given packet-size. Every time the first or second stage counter times out, a packet of data is sent to the corresponding queue in the network interface, and the network interface is able to output packet isochronous signals with an average bit rate within one bit-per-second of desired bit rates.
Description
MULTI-CHANNEL VIDEO PUMP
CROSS-REFERENCE TO RELATED APPLICATION
This application is related to U.S. Provisional Application Serial Number 60/112,866, entitled Multi-Channel Video Pump, by Timothy W. Dygert, filed December 18, 1998, and is related to U.S. Patent No. 6,473,441, entitled Multi-Channel Video Pump, by Timothy W. Dygert, filed January 7, 1999, and is a continuation-in-part of U.S. Patent Application Serial No. 09/478,407, entitled Multi-Channel Video Pump, by Timothy W. Dygert, filed January 6, 2000, all of which are incorporated herein by reference.
FIELD OF THE INVENTION
The present invention is directed to streaming video signals and, more particularly, to an apparatus for simultaneously streaming user-specified video files encoded at varying bit rates over a single network.
DESCRIPTION OF THE RELATED ART
The role of streaming video in local area networks is expected to increase rapidly in the near future due to developments in video compression and deployment of transmission systems with increased bandwidth. Using the Moving Pictures Expert Group (MPEG) standards it is possible to compress an audio/video source in such a way that a constant bit rate stream is created. This stream can be captured and stored on appropriate media such as a redundant array of independent disks (RAID) or a digital video (or versatile) disc (DVD). The MPEG data stream then has to be reproduced at the encoding rate for use by playback devices.
The MPEG compression standards are used worldwide for constant bit rate digital
video encoding. Decoding of MPEG video so that each picture and each audio frame is played once and only once relies on the ability to deliver each bit from the encoder to the decoder with a constant delay. This constant bit rate delivery is generally termed "isochronous streaming." In live broadcasts the encoder is responsible for generating the MPEG bit stream at the proper rate. However, when this information is stored for later playback another mechanism is required to "meter" the data from the storage media to the playback device. Normally, no feedback is provided to the sender by the receiver of MPEG video. The receiver depends on the transmission rate to be both smooth and accurate in order to decode MPEG video properly.
MPEG audio and video content may be recorded at any arbitrary rate. Some examples are streams that are 3.282, 3.420, 6.144, and 6.000 megabits per second. Some conventional systems use a handshake protocol to inform the receiving device what is the bit rate of the video stream that will be sent. However, that requires the receiving device to be programmed to use the protocol to communicate with the sending device. Other systems distribute the MPEG data in large "chunks" of data (up to tens of kilobytes) that require the receiving device to have enough expensive memory to buffer the data for smooth display. This type of delivery precludes using the MPEG System Clock synchronization mechanism which is required for precise playback.
Compressed video (e.g., MPEG-2) data is normally transmitted via satellite, cable, terrestrial digital broadcast and other transmission systems using serial bit stream mechanisms. In those systems, the data is clocked one bit at a time into the transmission data stream using the bit level clock of the transmission system. As a result, the average data rate is regulated directly by the bit level clock and the jitter is only that present on the bit level clock (normally substantially less than one millisecond). Jitter is a measurement of how early or how late a specific bit arrives from its intended arrival time.
The MPEG data streams that are sent on these type of transmission systems are multiplexed (with multiple programs and padding) so that the total MPEG stream rate is exactly the same as the network payload bit rate. An MPEG data stream cannot be as easily sent using an asynchronous transmission system or network like Ethernet or Asynchronous Transfer Mode (ATM), because such asynchronous networks operate on the packet (or cell) level rather than the bit level. Due to the use of packets to transmit data, it is more difficult to recover the input bit rate for a data stream from a network like Ethernet or ATM than from a multiplexed serial signal sent at a similar rate. A packet network will transmit an entire packet of MPEG data at the bit rate of the packet network, e.g., 155.52 Mega-bits per second (Mb/s) for ATM Optical Carrier - Level 3 (OC-3). In the best case, i.e., when the first bit of a packet is received at precisely the correct time, the maximum jitter of MPEG data in a packet will be determined by how much too early the last bit is received. Ignoring packet overhead, the formula for maximum jitter of the last bit when the first bit is on time is set forth in equation (1):
(Packet Size / Stream Rate) - (Packet Size / Network Rate) (1)
As apparent from this formula, jitter increases as the packet size increases and as the difference increases between the input data stream rate and the network transmission rate.
For example, in ATM OC-3 the smallest unit of data that can be sent is called a "cell", the bit rate is 155.52 Mb/s and the cell size is 53 bytes (roughly 48 payload and 5 overhead bytes). As a result, one ATM cell transmitted at 155.52 Mb/s typically contains 384 bits of MPEG data. Assuming a 3.42 Mb/s stream, if the first bit of MPEG data in an ATM cell is received at the correct time, the last MPEG data bit in the cell will be received early by 384 / 3.42 Mb/s - 384 / 155.52 Mb/s, for a maximum jitter of about 0.11 ms. With 100 Mb/s Ethernet, the packet size is typically closer to 1000 bytes. So, for roughly 1000 bytes, the MPEG data stream created at 3.42 Mb/s, is sent at 100 Mb/s to produce a maximum jitter of
2.26 ms.
It is possible to use smaller packet sizes (with increased overhead) but the packets or cells will still be transmitted by ATM OC-3 and 100 Mb/s Ethernet far faster than the input MPEG data stream rate. To properly shape the data as it is introduced into the network, some network technologies, such as ATM, provide a traffic shaping mechanism. The specifics of how this mechanism works vary, but in general constant bit rates are metered to the network with some level of granularity. For a network interface running at OC-3 speed (roughly 155 megabits/sec) this granularity will be no better than about 40,000 bits/second. At one of the higher MPEG data rates of 6 megabits/second and assuming thirty frames per second with an average frame size of 200,000 bits, this granularity in the worst case would cause a full frame over-run or under-run every 5 seconds which is unacceptable for playback of high quality video. Thus, even in networks with robust "quality of service" mechanisms for providing constant bit rate transmission, it is not possible to rely on the inherent traffic shaping mechanism alone. Other network technologies such as IEEE 1394 have similar limitations.
A video distribution system using a packet network like 100 Mb/s Ethernet or ATM to transmit MPEG data to devices that are not programmed to use a handshake protocol must strike a balance between the size of receiving video buffers and the packet size, so as to keep overhead at an acceptable level and still provide a smooth flow of data. Once the packet size has been determined, the absolute minimum size of the receiving video buffer will be determined by the best-case jitter of the last bit, calculated using equation (1).
Furthermore, as the number of concurrent video streams in a given network segment increases, it is essential that each stream be well-behaved to maximize network efficiency. Well-behaved implies that each stream is as nearly isochronous as possible within the packet structure of the network. This is referred to as packet isochronous transmission. In a network with a large percentage of streaming traffic, bursty transmission of MPEG video streams in
the network will result in congestion and network failure much more quickly than constant bit rate transmission. The more closely the individual data streams are maintained at a constant bit rate, the higher the total aggregate of such streams that can be carried on the network while maintaining a desired quality of service.
SUMMARY OF THE INVENTION
It is an object of the present invention to provide a video streaming device that can output video signals at an average rate within 30 bits per million bits per second of the rate at which the signal was encoded.
It is another object of the present invention to provide a video streaming device that can output signals, with different signal rates, each having a jitter of less than two milliseconds.
It is a further object of the present invention to provide a video streaming device capable of outputting multiple video signals at various rates using close to full maximum payload of the network that receives the video signals.
It is yet another object of the present invention to provide a video streaming device capable of outputting video signals to display devices with a minimal amount of buffer memory and without using a handshake protocol.
It is a still further object of the present invention to transmit multiple MPEG data streams, each in a near isochronous, or packet isochronous manner, such that appropriate decoders can properly recover the embedded system clock, decompress the MPEG data streams and re-create the original audio and video content.
The above objects can be attained by a system for transmitting multiple streams of stored signals to receiving devices, including at least one playback device to access recordings, each recording containing stored signals encoded at one of a plurality of bit rates; a streaming device, coupled to the at least one playback device, to receive a request to reproduce a specified recording, to detect the one of the bit rates used to encode the stored signals on the specified recording, and to output packet isochronous signals based on the stored signals on the specified recording at the one of the bit rates; and a network, coupled to the streaming device and the receiving devices, to deliver the packet isochronous signals to
the receiving devices.
Preferably, the streaming device includes a plurality of timer circuits, each including a base counter to count a truncated period for transmission of packets; and a dithering circuit to indicate transmission of one of the packets one clock pulse later than the truncated period, a predetermined number of times within a dither cycle.
These together with other objects and advantages which will be subsequently apparent, reside in the details of construction and operation as more fully hereinafter described and claimed, reference being had to the accompanying drawings forming a part hereof, wherein like numerals refer to like parts throughout.
BRIEF DESCRIPTION OF THE DRAWINGS
Figures IA and IB together provide a timing diagram showing differences between desired MPEG data timing and the timing provided by a packet network.
Figure 2 is a block diagram of a digital media retrieval system using the present invention.
Figure 3 is a flowchart of the operation of a video streaming device according to the present invention.
Figure 4 is a functional block diagram of a video streaming device according to the present invention.
Figure 5 is a block diagram of the hardware architecture of a video streaming device according to the present invention.
Figure 6 is a block diagram of a channel timing module according to the present invention.
Figure 7 is a block diagram of one set of channel timing counters in a video streaming device according to the present invention.
Figure 8 is a block diagram of a digital media retrieval system using multiple video streaming devices to produce video streams with a total of 480 megabits per second.
DESCRIPTION OF THE PREFERRED EMBODIMENTS
Figure IA illustrates the bit pattern desired for the MPEG stream (isochronous transmission). Figure IB illustrates the bit pattern for the same MPEG data stream as it is transmitted in a packet network, i.e., packet isochronous transmission. When the packet is released at precisely the correct time, as illustrated in Figs. IA and IB, each bit after the first bit in each packet arrives early by an increasingly larger amount. The maximum early jitter occurs on the last bit of the packet. Ideally, the release of the next packet will be delayed, so that the first bit of the next packet will be received precisely "on time" by the receiving devices.
As should be apparent from the description of the related art above, the precise release point of a given packet will not have much effect on jitter as long as the magnitude of the error of actual release versus desired release is small compared to the jitter of the last bit caused by the difference between the network and stream bit rates, which will be referred to as the best case jitter of the last bit. As indicated in Figs. IA and IB, the first bit can be released several bits early or late without changing the order of magnitude of the best case jitter of the last bit.
Another factor to be controlled is average bit rate. If a small packet is used to reduce the best case jitter of the last bit, the packets will be released at a higher frequency than required to transmit the same MPEG data stream using larger packets, hi addition, the smaller the packet, the shorter the interval between packets, assuming the release of each packet is delayed so that the first bit of each packet is released at the correct time. A unidirectional error (i.e., always early or always late) in the period between releases is multiplied by the number of releases per second and thus, an accurate inter-packet time is key to maintaining a constant average bit rate.
If the size of the receiving MPEG data buffer was unlimited, the packet size could be
large to reduce the frequency and the absolute timing tolerance per release. For example, if one packet were released per second, a one bit per second accuracy could be obtained using a 100 KHz clock and a 17-bit count down timer set for 100000. If this timer is off by one clock cycle, the error is only 1 / 100000. But, having a release at one second intervals using the example above creates a jitter of 966 ms for a 3.353 Mb/s stream.
If a packet size of 8000 bits instead of 3.353 Mbits as in the previous example is used for the same 3.353 Mb/s MPEG data stream, the packets must be released 419.125 times per second. It is more difficult to precisely achieve this rate. Again using a 100 KHz clock, the counter must count 238.5923054 clock pulses between the release of each packet. If the period is 238 clock pulses, 420.1680672 packets will be released per second for a bit rate of 3,361,344 bits per second. If the period is 239 clock pulses, 418.4100418 cells will be released per second for a bit rate of only 3,347,280 bits per second.
The amount of error resulting from a period of 238 or 239 clock pulses will preclude locking the decoder clock to the program clock reference (PCR) embedded in the MPEG data stream, since it would place the ratio of the decode clock to the display clock outside National Television Standard Committee (NTSC) limits. Using a period of 239 clock pulses results in losing 5719 (3,353,000-3,347,280) bits per second which is a bit rate error of only .17% (5719/3,353,000). However, assuming a frame rate of 30, a 3.353 Mb/s MPEG data stream has an average frame size of 111,766 bits, leaving the receiving device one frame short every 19.54 (111766/5719) seconds. Thus, a frame hold would need to occur about every 20 seconds. For a fixed network speed, increasing the MPEG data rate, reducing the packet size or reducing the clock frequency used for timing packet releases will cause the error from a single stage counter to go up.
Illustrated in Fig. 2 is a block diagram of digital media retrieval system 10 using a two-stage timer mechanism which allows much higher counter resolution than a single stage
counter with the same clock frequency. Digital media retrieval system 10 provides interactive distribution of video, text, graphics, and Internet content over a high speed digital network. Video pump 12 is a key component in system 10. The purpose of video pump 12 is to retrieve MPEG audio/video streams from various storage devices, such as RAID array 14 and DVD jukebox 16 and place this data into high speed digital network 18 for distribution to set top devices 20 at the specific rate required for each stream. Channels are opened in the system illustrated in Fig. 2 to transport data from the storage devices 14, 16 to set top devices 20 via ATM network 18. These channels may be Permanent Virtual Circuit (PVC) or Switched Virtual Circuit (SVC) channels, such as Constant Bit Rate (CBR) PVC 6 Mb/s channels. Video pump 12 responds to system commands from system control server 22 for the retrieval and distribution of isochronous data including both audio and video. For simplicity's sake, this data will subsequently be referred to as either video or simply as data. Although illustrated as separate components in Fig. 2 that may be networked, it is possible to construct video server 10 such that video pump 12 and system control server 22 are in the same computer system.
ATM network 18 is able to establish end-to-end connections with guaranteed bandwidth availability and requires that data is introduced to ATM network 18 in such a way that the established connection rate is not exceeded. If the bit rate of a specific connection exceeds that agreed to when the connection was established ATM network 18 may discard the excess data. The timing mechanism of the present invention used in conjunction with the ATM interface provides MPEG data streaming that meets all required specifications for bit rates between roughly 1 and 20 megabits per second.
Video pump 12 may be strictly a server, with commands received via a command protocol over TCP/IP, such as real time streaming protocol (RTSP). These commands open and close video streams, assign video streams to specific PVC/SVC channels, and perform
actions on these video streams, such as pause, play, stop, fast forward, rewind, etc. Video pump 12 receives via the commands, the start and stop addresses of the data within a given file that is to be streamed through ATM network 18. Video pump 12 provides timing to allow each individual channel to be streamed at unique arbitrary rates, m this embodiment, a maximum of 60 channels may be streamed, with a maximum total aggregate bit rate of 120 Mb/s. The timing for each channel may be specified via the application program interface (API) executing on system control server 22 or set top device 20, or determined directly from the stream itself by video pump 12 using the program clock references (PCRs) that are stored in MPEG transport stream data at least every 100 milliseconds. Video primp 12 can determine the bit rate of the signal by the number of bits between PCRs and the difference in time between the PCRs. Video pump 12 can operate on blocks of data as small as two MPEG transport packets (376 bytes) to minimize jitter imposed by the distribution of video within the system and to comply with ATM Forum requirements for MPEG-2 transmission.
The basic procedure for generating a stream of packet isochronous signals according to the present invention is illustrated in Fig. 3. Stored signals are read 24 from a recording that has been specified for playback in a conventional manner. Then, the bit rate used to encode the stored signals on the recording is detected 26 by detecting the PCRs. Finally, the stream of packet isochronous signals is output 28 with timing more precise than the clock signals in the device, by using a two stage timer, as described below.
A functional block diagram of video pump 12 is shown in Fig. 4. Video pump 12 has four main functional components that operate as individual processes. RAID streaming logic 30 issues and receives control signals to and from ultra Small Computer System interface (SCSI) interface 31 and sends status information to control and status logic 32. Real-time pump 34 receives data from Dynamic Random Access Memory (DRAM) buffer 35 and data rate information from RAID streaming logic 30 and outputs the data and a channel identifier
to ATM adapter 36. Control and status logic 32 also receives status information from realtime pump 34 and ATM adapter module 36 and issues commands to RAID streaming logic 30, real-time pump 34 and ATM adapter module 28. For example, start and stop addresses and start and stop commands are sent to RAID streaming logic 30, open and close channel commands and parameters including a priority list of counter values are sent to real-time pump 34 and open and close channel commands and parameters indicating channel bandwidth are sent to ATM adapter module 36.
RAID streaming logic 30 fetches data from RAID array 14. This data is placed in DRAM buffer 35 where it is read by real-time pump 34. RAID streaming logic 30 receives start and stop commands, as well data addresses from video pump control and status logic 32. RAID streaming logic 30 preferably reads data including PCRs from the video file to determine the encode rate, and passes this rate on to real-time pump 34. The encode rate is the rate at which the set top device decoder will use the data, and it is therefore the rate at which video pump 12 must send the data to the decoder, as described in more detail below. RAID streaming logic 30 also ensures that the data being read from RAID array 14 is transport packet aligned. This is crucial to the operation of video pump 12, and any errors are immediately reported to control and status logic 32.
Real-time pump 34 is the heart of video pump 12. It is here that the data for each channel is pulled from the DRAM buffers 35 for each channel at the specified rate. Data for each channel is passed from real-time pump 34 to buffers in ATM adapter module 36 for insertion into ATM distribution network 18. hi an exemplary design described below, realtime pump 34 is capable of maintaining 60 separate video streams, each with arbitrary data rates, and processing the data flow in such a manner to minimize jitter as the data is placed in the stream, hi this embodiment real-time pump 34 is capable of maintaining an aggregate data flow bandwidth of 120 Mbps.
ATM adapter module 36 receives the video data from real-time pump 34, packetizes this data into ATM cells, and passes this data stream on to ATM network 18 for distribution to set top devices 20. The data received from real-time pump 34 is in the form of MPEG transport stream packets, and the ATM encapsulation is performed according to ATM Adaptation Layer 5 (AAL5). The output of ATM adapter module 36 is coupled to OC-3c (concatenated) fiber.
A network interface device traffic shaper in ATM adapter module 36 is initialized so that for the current channel it will introduce data into the network at the closest rate to the required rate that is higher than the required rate. Channel timing module 40 provides a signal to transfer the data block to the network interface device traffic shaper in- ATM adapter module 36 each time the timer for a channel expires. The result is that each block of data is introduced to the network at a rate that is faster than desired. However, because only data that has been transferred to the network interface device can be sent, from time to time there will be no data available to the traffic shaper. This will result in no data being sent until the next block of data is made available. The resulting data stream will consist of a period when data is being sent too fast followed by a period in which no data is sent. Over time, the desired data rate will be achieved to the precision of the channel timing module 40, as described below.
If real-time pump 34 is the heart of a video streaming device according to the present invention, then control and status logic 32 serves as the brains for video pump 12 by coordinating and directing all internal elements and processes. Control and status logic 32 provides the interface to the "outside world", receiving commands and passing status to other elements within digital media retrieval system 10. Control and status logic 32 processes these system level commands, generating local commands as required to the other functional elements of video pump 12.
A block diagram of the hardware architecture of a video streaming device (video pump) according to the present invention is illustrated in Fig. 5. The functional components illustrated in Figs. 2 and 4 correspond to the physical components illustrated in Fig. 5 as follows. Video pump 12 may be constructed using a standard "IBM PC" type Intel® Pentium® platform with processor 42 connected to Peripheral Component Interconnect (PCI) bus 44. Channel timing module 40 which implements multiple two-stage timers may be a custom PCI interface card inserted into PCI bus 44. DRAM buffers A and B 35 are provided by Random Access Memory (RAM) 46. Hard drive 48 stores an operating system, such as Windows® NT 4.0 and pump software executed by processor 42 to provide the functions of video pump control and status logic 32, real time pump 34, RAID streaming logic 30 and system control server 22. SCSI controller 50 provides an interface to devices, such as a RAID storage unit (not shown) connected via ultra SCSI 31. Network interface 52 corresponds to ATM adapter module 36. An Ethernet network interface (not shown) may also be included to provide for external control by a separate system control server 22. As will be apparent to one of ordinary skill in the art, the components illustrated in Fig. 5 can be replaced with different components capable of performing the functions illustrated in Figs. 2 and 4 with greater or less capacity, depending on the requirements of the system in which they operate.
An example of channel timing module 40 on a PCI timer card is provided in Fig. 6. In this example PCI interface 70 is provided by a PLX Technology PCI9050-1 PCI bus target interface, compliant to PCI Specification 2.1, to connect channel timing module 40 to PCI bus 44. PCI interface 70 is a PCI slave interface providing a bridge to local bus 71. PCI configuration registers (not shown separately) in PCI interface 70 are mapped to I/O space. AU resources in channel timing module 40 are preferably 32-bit accessible. As described in the example below, local bus 71 preferably is clocked at 10 MHZ, enabling jitter to be
reduced to a desired level. Timing of local bus accesses are determined by timer Field- Programmable Gate Array (FPGA) 72 in the manner described below.
The initial configuration of channel timing module 40 is loaded from configuration EEPROM 74 attached to PCI interface 70. The following fields in the PCI configuration registers (not shown) are loaded from configuration EEPROM 74 at power up: Device ED, Vendor ID, Class Code, Subsystem ID, Subsystem Vendor ID, and Interrupt Pin. These registers are reloaded at every instance of PCI Reset signal assertion. Configuration EEPROM 74 may be a Fairchild Semiconductor NM93CS46 which holds 1024 bits of information. The data within the device may be altered via registers within PCI interface 70, depending on the state of the protection register within EEPROM 74.
Several Altera EPF6024A FPGAs 72 are included in channel timing module 40. A partial block diagram of timer FPGA 72 is shown in Fig. 7. Each FPGA 72 contains 15 timers 80, one of which is illustrated in Fig. 7, interrupt controller 82 and an interface (not shown) to local bus 71. Each timer FPGA is configured upon reset via loader EPROM 84 (Fig. 6) which may be provided by Altera EPC 1441 devices each containing 400K x 1 bits of information. hi the embodiment described below, the dither cycle used by timer circuit 80 is 1024 and the clock rate is 10 MFfZ. Operation of each circuit timer 80 in channel timing module 40 is initiated by processor 42 (Fig. 5) loading count data, consisting of a 22-bit count value and a 10-bit dither value, into corresponding registers 85, 86 in channel timing module 40 via PCT bus 44. Upon timeout (explained below) an interrupt is generated for processor 42 to read interrupt status registers in channel status logic (not shown) to determine which timer circuit 80 has timed out.
Each timer circuit 80 consists of two counters: base counter 87 and dither counter 88. Timer circuit 80 begins operation when the 22-bit count value and 10-bit dither value have
been written in registers 85, 86 and the 22-bit count value is transferred to base counter 87. Dither counter 88 is initially set to all ones (1023 decimal). During operation, base counter 87 is decremented at each cycle of the 10 MHZ clock. When base counter 87 reaches zero, a compare is performed between the 10-bit dither value in register 86 and dither counter 88. If dither counter 88 is less than or equal to the 10-bit dither value a timeout occurs. The timeout is delayed by one clock cycle when dither counter 88 is greater than the 10-bit dither value in register 86. Dither counter 88 is decremented at each timeout. Base counter 87 is reloaded with the 22-bit count value in register 85 at each timeout. Dither counter 88 is 10 bits wide and thus, automatically resets to its maximum value every 1024 timeouts. An interrupt for the timer circuit 80 is generated each time the timeout occurs if the interrupt has been enabled by setting the appropriate bit in the interrupt control register. Therefore, an average timeout period is defined by formula (2), where Base is the 22-bit count value in register 85 and Dither is the 10-bit dither value in register 86.
Period = (Base *(1024-Dither) + (Base +1) * Dither) / (Clk*1024) (2)
In the above example of a 3.353 Mb/s MPEG data stream transmitted over an appropriate network having a transmission rate higher than the MPEG data rate in packets of 1000 bytes (8000 bits), a packet should be released 3,353,000/8000 or 419.125 times per second. Using a 100 IcHz clock (to be consistent with the earlier example), 100,000/419.1241778 or 238.5923054 clock signals are counted for each time a packet is released. The fractional part is .5923054. Multiplying the fractional part by 1024 (the number of packets released during one dither cycle of timer circuit 80) produces 606.52 which can be rounded to 607. This is the number of times out of 1024 that the count will be 239. The number of counts at 238 is (1024-607)= 417. The average count over 1024 releases in this example can be calculated by formula (3) as 238.59.
((607 * 239)+ (417*238))/1024 = 238.59277 (3)
Thus, channel timer circuit 40 will release an 8000 bit packet of MPEG data on the average of 419.1241778 times per second (100,000/238.59277). Therefore, the average data rate (over an integral number of 1024 releases ) is 419.1241778 x 8000 = 3,352,993.4 bits per second. In this example, the present invention is able to reduce the bits per second error from 5917 to 6.6 bits per second or about three orders of magnitude using the same clock frequency.
To further reduce the error in average bit rate, e.g., to within one bit per second, the clock frequency can be increased. Increasing the clock frequency to 10 MHz reduces the error in the above example to .01 bit per second. Using the 10 MHz clock, the problem still requires a release of 8000 bits 419.125 times each second. The number of clock pulses per timeout is 10,000,000/419.125 which is 23,859.23054. The fractional part is .23054. Multiplying the fractional part by 1024 produces 236.07 which can be rounded to 236. This is the number of times out of 1024 that the count will be 23860. The number of counts out of 1024 at 23859 is 788 (1024-236). The average over 1024 timeouts (one dither cycle) is ((236*23860)+(788*23859)) / 1024 which is 23,859.23047. The number of releases per second is 10,000,000/23859.23047 or 419.1250012. This makes the average bit rate 419.1250012 * 8000 or 3,353,000.01.
There is a cyclic jitter associated with this method that is very low. This jitter is nulled out each 1024 clock cycles (one dither cycle) and only amounts to a few bit times which is insignificant.
Timer 80 will start counting upon loading the base and dither counter values into base and dither counters 87, 88. This must be done as a single 32-bit write. In fact, the value loaded into base counter 86 is the desired value minus one, since the check for timeout occurs after each clock cycle. For simplicity in the description above this detail was ignored. Timer 80 may be stopped by writing all zeroes to 22-bit base count register 85. The local bus
interface in each timer FPGA 72 provides the timing and address decode for accesses to resources of timer FPGA 72. The local bus is clocked from the same 10 MHz source that drives the timers.
Note that it is important to look at a range of bit rates to see the various effects discussed above. For certain values, it is possible to get very low errors using low clock rates and single stage timers. However, timing module 40 produces very close results for all combinations within a wide range of bit rates. For example, using a dither cycle of 1024 and a clock rate of 10 MHz, it is possible to control output of video data in 8000 bit packets onto an ATM OC-3 network for video data rates of 2 to 20 Mb/s with a maximum average data rate error that is less than one bit per second.
It should also be clear from this example that attempting to do the fine grained timing required to get the data rate correct is not practical with conventional general purpose operating systems. The hardware embodiment described above maintains each counter precisely and independently of the processor. Characteristics of the operating system and speed of the host computer will determine how closely to scheduled time each packet is actually released. However, any error in release time of a given packet is not cumulative since the timers reset automatically at the end of each cycle and are completely independent of the main processor and operating system.
In another respect, each device receiving the data stream may have a video timing clock. A video timing clock may be used to drive an output of a video signal, such as an NTSC video signal or a Phase Alternating (or Alternation) Line (PAL) video signal. The devices that receive the data stream from the video pump 12 can obtain (i.e., recover) a clock that is embedded in the data stream. An example of the embedded clock is the PCR embedded in an MPEG data stream.
After obtaining (and/or while continuing to obtain) an embedded clock from a data
stream, the devices that receive the data stream may phase lock their respective video timing clock to the embedded clock obtained from the data stream. When establishing phase lock of the video timing clock, ideally, the video pump 12 outputs the data stream at an average bit rate equivalent to the bit rate used to encode the data being streamed. However, due to various factors (e.g., frequency inaccuracy, stability, and/or aging of components used to output the data stream), receiving devices have been developed such that when establishing phase lock of the video timing clock, the data stream from the video pump 12 can be output with an average transmission rate that falls within a range of average transmission rates.
The range of average transmission rates may be based on (i) the bit rate used to encode the data being streamed, and (ii) upper and lower tolerances for the average transmission rate. The upper and lower tolerances for the average transmission rate may be based on a peπnissible frequency deviation specified for a system clock frequency for devices that receive a data stream. For example, a permissible frequency deviation for a 27 MHz clock signal may be specified as ± 810 Hz. In this regard, the system clock may operate in a range from 26,999,190 Hz to 27,000,810 Hz. Such a system clock frequency standard is specified in International Standard ISO/IEC 13818-1, Information technology - Generic coding of moving pictures and associated audio information: Systems, Second edition, which is incorporated by reference herein.
In another respect, a permissible frequency deviation for a recovered clock signal may be specified in the amount of error-per-Hz. For example, with regard to the example 27 MHz clock having a permissible frequency deviation of 810 Hz, the permissible frequency deviation may be specified as 0.00003% (810 Hz divided by 27 MHz). For convenience, the permissible frequency deviation may be specified in "parts per million" (ppm) and thus a permissible frequency deviation of 0.00003% may be specified as 30 ppm.
The permissible frequency deviation for a recovered clock signal may be translated to
the upper and lower frequency tolerances for the average transmission rate of a data stream being output from the video pump 12. For a permissible frequency deviation of 30 ppm, the upper and lower tolerances for outputting the data stream is the bit rate in Mb/s multiplied by 30. In this regard, the upper and lower tolerances are 30 bits per million bits per second. As an example, for a permissible frequency deviation of 30 ppm and a video signal encoded at 1 Mb/s, the video pump 12 should output the data stream within a range of 1 Mb/s ± 30 bits per second (b/s) (i.e., 999,970 b/s to 1,000,030 b/s). As another example, for a permissible frequency deviation of 30 ppm and a video signal encoded at 6 Mb/s, the video pump 12 should output the data stream within a range of 6 Mb/s ± 180 b/s (i.e., 5,999,820 b/s to 6,000,180 b/s).
In yet another respect, a network that transports a data stream from the video pump 12 to a receiving device may cause some error in the amount of error-per-Hz of the data being streamed to a receiving device. In order to compensate for the error introduced by the network (or due to other factors), the upper and lower frequency tolerances for the average transmission rate of the data stream being output from the video pump 12 may be reduced to upper and lower tolerances that are less than the maximum upper and lower frequency tolerances. For example, maximum upper and lower tolerances may be multiplied by a given factor, such as 0.8. For maximum upper and lower tolerances of 30 and a factor of 0.8, the reduced upper and lower tolerances is 24 ppm (30 ppm multiplied by 0.8). In this regard, for a video signal encoded at 1 Mb/s, the video pump 12 should output the data stream within a range of 1 Mb/s ± 24 bits per second (b/s) (i.e., 999,976 b/s to 1,000,024 b/s). Other examples of the factor for reducing the upper and lower tolerances are also possible.
The present invention is scalable by combining multiple video pumps 12 connected to a single ATM switch 120, as illustrated in Fig. 8. Video pumps 12 may be connected to one or more storage devices, such as RAID array 14, DVD jukebox 16, and other devices, such as
compact disc changers, not shown.
The many features and advantages of the invention are apparent from the detailed specification and, thus, it is intended by the appended claims to cover all such features and advantages of the invention which fall within the true spirit and scope of the invention. For example, the disclosed embodiment is used with an Asynchronous Transfer Mode (ATM) network that is able to support multiple constant bit rate streams per segment as well as the bursty traffic created by more traditional network traffic. The present invention is not limited to use with ATM networks, but could be used with any network that can deliver the required amount of data. The quality of delivery will be dependent on the quality of service provided by the network. New protocols for TCP/UDP over switched and gigabit Ethernet networks may eventually support a quality of service provided by ATM networks, but presently an Ethernet network will be able to transmit a smaller number of high quality video streams per network segment than ATM and will be adversely affected by other network traffic. The present invention will work very well on any network with high quality of service capabilities including IEEE 1394 and networks conforming to the Home Phone Network Alliance (PNA) V 2.0 specification. Further, since numerous modifications and changes will readily occur to those skilled in the art, it is not desired to limit the invention to the exact construction and operation illustrated and described, and accordingly all suitable modifications and equivalents may be resorted to, falling within the scope of the invention.
Claims
1. An apparatus to output multiple streams of stored signals, each stream encoded at one of a plurality of bit rates and read from recordings, the apparatus comprising: a streaming device to detect which of the bit rates is used to encode each respective stream of the stored signals on the recording and to output each stream as packet isochronous signals at the one of the bit rates for that respective stream, wherein the streaming device outputs each stream of the packet isochronous signals with an average bit rate within 30 bits per million bits per second of the one of the bit rates used to encode that respective stream.
2. An apparatus as recited in claim 1, wherein the streaming device outputs each stream of the packet isochronous signals with an average bit rate within 24 bits per million bits per second of the one of the bit rates used to encode that respective stream.
3. An apparatus as recited in claim 1, wherein the streaming device outputs each stream of the packet isochronous signals with a jitter of less than two milliseconds.
4. An apparatus as recited in claim 1, wherein the streaming device comprises a plurality of timer circuits, each including: a base counter to count a truncated period for transmission of packets; and a dithering circuit to indicate transmission of one of the packets one clock pulse later than the truncated period, a predetermined number of times within a dither cycle.
5. A system for transmitting multiple streams of stored signals to receiving devices, the system comprising: at least one playback device to access recordings, each recording containing stored signal encoded at one of a plurality of bit rates; a streaming device, coupled to the at least one playback device, to receive requests to reproduce specified recordings, to detect the one of the bit rates used to encode the stored signals on each of the specified recordings, and to output a stream of packet isochronous signals based on the stored signals of the specified recordings at the one of the bit rates for each of the specified recordings, wherein the streaming device outputs each stream of the packet isochronous signals with an average bit rate within 30 bits per million bits per second of the one of the bit rates at which the corresponding recording was encoded; and a network, coupled to the streaming device and the receiving devices, to deliver each stream of the packet isochronous signals to the receiving devices.
6. An apparatus as recited in claim 5, wherein the streaming device outputs each stream of the packet isochronous signals with an average bit rate within 24 bits per million bits per second of the one of the bit rates at which the corresponding recording was encoded.
7. A system as recited in claim 5, wherein the streaming device outputs each stream of the packet isochronous signals with a jitter of less than two milliseconds.
8. A system as recited in claim 5, wherien the streaming device comprises a plurality of timer circuits, each including: a base counter to count a truncated period for transmission of packets; and a dithering circuit to indicate transmission of one ofhte packets one clock pulse later than the truncated period, a perdetermined number of times within a dither cycle.
9. A method of transmitting multiple streams of stored signals to receiving devices, the method comprising: reading the stored signal from recordings, each recording encoded at one of a plurality of bit rates; detecting the one of the bit rates used to encode the stored signals for each respective stream; and outputting to the reciving devices each stream of stored signals as packet isochronous signal at the one of the bit rates for that respective stream, wherein each stream of the packet isochronous signals is outputted with the average bit rate within 30 bits per million bits per second of the one of the bit rates used to encode that respective stream.
10. An apparatus as recited in claim 9, wherein each stream of the packet isochronous signals is outputted with the average bit rate within 24 bits per million bits per second of the one of the bit rates used to encode that respective stream.
11. A method as recited in claim 9, wherein the outputting outputs each stream of the packet isochronous signals with a jitter of less than two milliseconds.
12. A method as recited in claim 9, wherein the outputting of each packet of isochronous signals occurs on a timing pulse produced from clock pulses generated at a clock rate, the timing pulse produced with an average period of greater precision than the clock rate.
13. A method as recited in claim 11, further comprising producing timing pulses by: couting the clock pulses for a truncated period based on the average period; and generating the timing pulses one clock pulse later than the truncated period, a predetermined number of times within a dither cycle.
Priority Applications (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2008526146A JP2009505500A (en) | 2005-08-12 | 2006-08-04 | Multi-channel video pump |
EP06800985A EP1922829A4 (en) | 2005-08-12 | 2006-08-04 | Multi-channel video pump |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US11/202,592 US20060262813A1 (en) | 1998-12-18 | 2005-08-12 | Multi-channel video pump |
US11/202,592 | 2005-08-12 |
Publications (2)
Publication Number | Publication Date |
---|---|
WO2007021698A2 true WO2007021698A2 (en) | 2007-02-22 |
WO2007021698A3 WO2007021698A3 (en) | 2007-06-07 |
Family
ID=37758101
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/US2006/030926 WO2007021698A2 (en) | 2005-08-12 | 2006-08-04 | Multi-channel video pump |
Country Status (4)
Country | Link |
---|---|
US (1) | US20060262813A1 (en) |
EP (1) | EP1922829A4 (en) |
JP (1) | JP2009505500A (en) |
WO (1) | WO2007021698A2 (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108377428A (en) * | 2018-02-07 | 2018-08-07 | 深圳市亿联智能有限公司 | A kind of high efficiency safety monitoring equipment audio, video data flow control methods |
Families Citing this family (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20030200548A1 (en) * | 2001-12-27 | 2003-10-23 | Paul Baran | Method and apparatus for viewer control of digital TV program start time |
KR100784874B1 (en) * | 2002-07-16 | 2007-12-14 | 마쯔시다덴기산교 가부시키가이샤 | Content transmitting apparatus |
US8572665B2 (en) * | 2004-03-03 | 2013-10-29 | Cisco Technology, Inc. | Selective distribution of cell based video streams over packet based networks |
US9137502B2 (en) * | 2004-08-25 | 2015-09-15 | Broadcom Corporation | Method and system for fast digital channel change utilizing time-stamp management |
US10105441B2 (en) | 2007-08-16 | 2018-10-23 | The Schepens Eye Research Institute, Inc. | Method for inhibiting or reducing dry eye disease by IL-1Ra |
US8854964B2 (en) * | 2007-12-14 | 2014-10-07 | General Instrument Corporation | Method and apparatus for determining a transport bit rate for a Multiprogram transport stream |
US10158585B2 (en) | 2013-02-21 | 2018-12-18 | Intel Corporation | Packet processing with reduced latency |
Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6473441B1 (en) | 1998-12-18 | 2002-10-29 | Escient Convergence Corp | Multi-channel video pump |
Family Cites Families (24)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5486864A (en) * | 1993-05-13 | 1996-01-23 | Rca Thomson Licensing Corporation | Differential time code method and apparatus as for a compressed video signal |
CA2135681C (en) * | 1993-12-30 | 2000-01-18 | Srinivas V. Makam | System and method for directly accessing long-term memory devices |
CA2146801C (en) * | 1994-05-27 | 1999-11-02 | Barin Geoffry Haskell | Timing recovery for variable bit-rate video on asynchronous transfer mode (atm) networks |
US5603058A (en) * | 1994-09-08 | 1997-02-11 | International Business Machines Corporation | Video optimized media streamer having communication nodes received digital data from storage node and transmitted said data to adapters for generating isochronous digital data streams |
JP3575100B2 (en) * | 1994-11-14 | 2004-10-06 | ソニー株式会社 | Data transmission / reception apparatus and method, and data recording / reproduction apparatus and method |
US5561791A (en) * | 1995-04-10 | 1996-10-01 | Digital Equipment Corporation | Method and apparatus for conditioning timed program independent of transport timing |
US5646676A (en) * | 1995-05-30 | 1997-07-08 | International Business Machines Corporation | Scalable interactive multimedia server system for providing on demand data |
US5828670A (en) * | 1995-06-06 | 1998-10-27 | Symmetricom, Inc. | Distribution of synchronization in a synchronous optical environment |
US6138147A (en) * | 1995-07-14 | 2000-10-24 | Oracle Corporation | Method and apparatus for implementing seamless playback of continuous media feeds |
US5966387A (en) * | 1995-09-25 | 1999-10-12 | Bell Atlantic Network Services, Inc. | Apparatus and method for correcting jitter in data packets |
TW319942B (en) * | 1995-11-14 | 1997-11-11 | Nippon Bickter Kk | |
US5862450A (en) * | 1995-12-14 | 1999-01-19 | Sun Microsytems, Inc. | Method and apparatus for delivering simultaneous constant bit rate compressed video streams at arbitrary bit rates with constrained drift and jitter |
US6292621B1 (en) * | 1996-02-05 | 2001-09-18 | Canon Kabushiki Kaisha | Recording apparatus for newly recording a second encoded data train on a recording medium on which an encoded data train is recorded |
US5892535A (en) * | 1996-05-08 | 1999-04-06 | Digital Video Systems, Inc. | Flexible, configurable, hierarchical system for distributing programming |
US5881245A (en) * | 1996-09-10 | 1999-03-09 | Digital Video Systems, Inc. | Method and apparatus for transmitting MPEG data at an adaptive data rate |
US6208655B1 (en) * | 1996-11-27 | 2001-03-27 | Sony Europa, B.V., | Method and apparatus for serving data |
JP2000511383A (en) * | 1997-03-12 | 2000-08-29 | コーニンクレッカ フィリップス エレクトロニクス エヌ ヴィ | Apparatus and method for recording a digital information signal on a record carrier |
US5974503A (en) * | 1997-04-25 | 1999-10-26 | Emc Corporation | Storage and access of continuous media files indexed as lists of raid stripe sets associated with file names |
US6181711B1 (en) * | 1997-06-26 | 2001-01-30 | Cisco Systems, Inc. | System and method for transporting a compressed video and data bit stream over a communication channel |
JP3433071B2 (en) * | 1997-10-29 | 2003-08-04 | 富士通株式会社 | Clock frequency synchronizer |
UA57812C2 (en) * | 1997-11-04 | 2003-07-15 | Джорджія Тек Ресерч Корпорейшн | System and method for transmitting digital video signals and data over a communication link |
US6493832B1 (en) * | 1999-03-17 | 2002-12-10 | Sony Corporation | Communication apparatus which handles a time stamp |
JP2004328574A (en) * | 2003-04-28 | 2004-11-18 | Sony Corp | Data transmitting apparatus, and data transmitting method |
US7068686B2 (en) * | 2003-05-01 | 2006-06-27 | Genesis Microchip Inc. | Method and apparatus for efficient transmission of multimedia data packets |
-
2005
- 2005-08-12 US US11/202,592 patent/US20060262813A1/en not_active Abandoned
-
2006
- 2006-08-04 JP JP2008526146A patent/JP2009505500A/en active Pending
- 2006-08-04 WO PCT/US2006/030926 patent/WO2007021698A2/en active Application Filing
- 2006-08-04 EP EP06800985A patent/EP1922829A4/en not_active Ceased
Patent Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6473441B1 (en) | 1998-12-18 | 2002-10-29 | Escient Convergence Corp | Multi-channel video pump |
Non-Patent Citations (1)
Title |
---|
See also references of EP1922829A4 |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108377428A (en) * | 2018-02-07 | 2018-08-07 | 深圳市亿联智能有限公司 | A kind of high efficiency safety monitoring equipment audio, video data flow control methods |
Also Published As
Publication number | Publication date |
---|---|
WO2007021698A3 (en) | 2007-06-07 |
US20060262813A1 (en) | 2006-11-23 |
JP2009505500A (en) | 2009-02-05 |
EP1922829A2 (en) | 2008-05-21 |
EP1922829A4 (en) | 2010-08-18 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US7379653B2 (en) | Audio-video synchronization for digital systems | |
US6954469B1 (en) | Multi-channel video pump | |
JP3789995B2 (en) | Video server system and operation method thereof | |
US20060262813A1 (en) | Multi-channel video pump | |
US7613381B2 (en) | Video data processing method and video data processing apparatus | |
US6618396B1 (en) | Data transmitting device, data receiving device, and data recording device | |
EP1614291B1 (en) | Data requesting and transmitting devices and processes | |
AU708139B2 (en) | Methods of transmitting and receiving compressed television signals | |
US7756233B2 (en) | Data receiving device and data receiving method | |
JP4193297B2 (en) | COMMUNICATION DEVICE AND METHOD, COMMUNICATION SYSTEM, AND RECORDING MEDIUM | |
US20040086000A1 (en) | Communication protocol for controlling transfer of temporal data over a bus between devices in synchronization with a periodic reference signal | |
CA2528608A1 (en) | Simultaneously transporting multiple mpeg-2 transport streams | |
WO2007044563A1 (en) | Method and apparatus for rtp egress streaming using complementary directing file | |
WO1996034477A1 (en) | Data transmitter, data receiver, and data transmission controller | |
JPH0879705A (en) | Method and apparatus for transferring audio/video/data through system bus | |
CN1669310A (en) | Method and device for video data transmission for implementing special modes | |
US7433069B2 (en) | Image transmission method and its apparatus | |
US6735223B1 (en) | Method of controlling offset of time stamp and apparatus for transmitting packet using the same | |
CN108124183B (en) | Method for synchronously acquiring video and audio to perform one-to-many video and audio streaming | |
JP3856135B2 (en) | Stream converting / sending device | |
KR100962083B1 (en) | Method and system for converting a first data stream into a second data stream | |
JP2004088480A (en) | Image pickup device and method for controlling data transmission | |
JPH10313448A (en) | Moving image transmitter and receiver | |
CA2210375C (en) | Multiplexer for multiple media streams | |
JPH11234634A (en) | Data transmitter and data multiplexer |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application | ||
WWE | Wipo information: entry into national phase |
Ref document number: 2008526146 Country of ref document: JP |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
WWE | Wipo information: entry into national phase |
Ref document number: 2006800985 Country of ref document: EP |