JP3079208B2 - Method and apparatus for optimizing program transmission - Google Patents

Description

Description: BACKGROUND OF THE INVENTION The present invention relates to a system and method for optimizing the transmission of a program by a distribution system, and in particular to a system and method for providing video on demand of a subscriber over a cable television network. , But is not limited to this.

2. Description of the Prior Art In the following specification, the term "program" refers to visual, auditory, both, or other information that usually gives a continuous impression to a human through one or more of the five senses of the human. And shall be interpreted in the broadest sense.
The term "video program" also refers to visual information recorded in a reproducible format or transmitted "live"
Alternatively, it refers to a visual and auditory information program. In the current "information society," which focuses on the availability of information, it is frequently possible for two or more people to simultaneously access the same program.

Thus, for example, in a library in a large educational facility that stores lectures and other information in the form of visual and / or audio-video cassettes, the demand for certain programs may be very high at a particular time, and There is a need for a system that allows one student to watch or listen to a particular program at a time from the beginning, and that does not require everyone to start watching or listening at the same time from the beginning. Ideally, it would be desirable to be able to make a program available to everyone who needs it, on demand. In practice, however, this is not possible without the installation of many expensive devices and complicated electronic processing. A so-called demand-based video television is an example of this type of system that can simultaneously accommodate a plurality of users. For demand-based video systems, it is theorized that all subscribers have access to a particular video program at any time of the day.

A conventional demand responsive video system is disclosed in Walter U.S. Pat. In this system, video programs are pre-programmed in the storage device, and each program is selected by a central data station host computer in response to an address signal sent by a user. The host computer transmits the video program at a high non-real time rate over a fiber optic network to a data receiving station near the user. The data receiving station converts the received optical data into electrical data, which is then stored for transmission to the user's television in real time.

Walter's system has some major drawbacks. The biggest drawback is the existing television networks, especially
Incompatible with CATV and coaxial cable networks. To speed up the response time, Walter's system uses digital data equivalent to the entire program,
It must be sent to the receiving station in a very short time via a plurality of fiber optic networks. Even when compressing digital data, this system requires a fairly large bandwidth. For example, transmitting a 2 hour movie in about 31 seconds requires 16 optical data channels across 4 fiber optic lines. At present, most homes and buildings do not have access to fiber optic cables, and fiber optic networks are expensive to install.

Another disadvantage of Walter's system is that it cannot adequately handle the high demand for the same video program. According to a survey of libraries that rent videotapes, for example, assuming that a total of 5,000 tapes are stored, of which the demand is high within a certain time period is the 20-40 most popular It is limited to movies. In addition, a survey of the audience's behavior showed that the demographics of the audience changed over time throughout the day, so even the most popular videos changed. Walt
In the case of the er system, the central data station sends only the portion of the program selected by the first user to that user, and then sends the portion of the program selected by the second user to that user. They can do it, but they cannot meet the needs of several users at the same time for the same program. In such a case, later users must wait until the entire program has been sent to the user who placed the order before you. Obviously, a popular video program will have to wait for an unbearably long time.

SUMMARY OF THE INVENTION It is an object of the present invention to provide a method and system for optimizing program transmission by a distribution system for multiple users. More specifically, but not exclusively, it is an object of the present invention to provide a system and method for supplying video on demand, which can be used for existing video distribution systems such as CATV. I do. Throughout the specification, the term "distribution system" is to be interpreted in its broadest sense, including the use of ordinary radio and television networks, hotels, educational facilities, and, more recently, CATV and internal television, such as those used on airplanes and regular cruise ships. /video/
It shall include an audio distribution system and the like.

According to one aspect of the present invention, in a method of optimizing transmission when transmitting a program to a large number of users via a distribution system, at the head end of the distribution system, the program is divided into a plurality of program segments, Transmitting the program segment by a redundant sequence based on a scheduling algorithm; and storing, at a receiver side of the distribution system, the transmitted program segment in a buffer memory of the receiver for later reproduction. Wherein the program algorithm is adapted to receive all of the program segments at the user's receiver so that the program can be continuously played back in real time by the scheduling algorithm in use. That.

Preferably, the method includes the step of selecting the longest response time (MRT) corresponding to the longest time required before the user can start playing the ordered program from the beginning.

In the step of dividing the program, the program may be divided into a length such that at least one segment can be transmitted within one MRT time.

In a preferred embodiment, in the transmitting step, at least one of a first segment of the program, that is, a segment including a segment to be played first is transmitted within each MRT time based on the scheduling algorithm. In use, the first segment is always received by the receiver within 1 MRT, whereby the first segment
To play the segment immediately.

According to another aspect of the present invention, in a system for optimizing transmission of a program when transmitting the program to a large number of users, a means for dividing a program into a plurality of program segments on a head end side of the system; Means for transmitting a segment by a redundant sequence based on a scheduling algorithm, wherein a buffer memory for storing the transmitted program segment for later reproduction by a receiver is provided on the receiver side of the system. Providing a system wherein all of the program segments are received by a user's receiver so that, in use, the scheduling algorithm allows for continuous playback of the program in real time.

According to yet another aspect of the present invention, a receiver for receiving a program supplied from a program transmission optimization system, wherein the buffer stores a plurality of program segments of the program received from a head end of the system based on a scheduling algorithm. Storage means, and processing means for processing the program segments stored in the buffer storage means and supplying the segments in the correct order for playback, whereby the scheduling algorithm is used in use. Thus, the receiver can receive all of the program segments so that the receiver can continuously reproduce the program in real time.

The processing means is provided with means for identifying a received program segment by a segment identification mark, wherein the segment identification mark can be identified by at least the segment number, so that the receiver is able to identify extra segments and continuous It is desirable to be able to distinguish segments required for reproduction.

According to another aspect of the present invention, there is provided a scheduling apparatus for a program transmission optimization system, wherein the apparatus includes means for dividing a program into a plurality of program segments, and a redundant sequence based on a scheduling algorithm. Means for scheduling a plurality of program segments, and means for determining a path for transmitting the scheduled program segments to one or more receivers of a user who has requested the program; Wherein the scheduling algorithm allows the receiver to receive all of the program segments so that the receiver can continuously reproduce the program in real time. I do.

The splitting means is adapted to transmit at least one segment within a maximum response time (MRT) corresponding to a longest time required before a user can start playing a program ordered from the beginning. Preferably, the program is divided into lengths.

If the count is a predetermined initial integer value that increases by one for each MRT, and a number from x = 1 to a number n equal to the number of segments obtained by dividing the program, the scheduling algorithm of the present invention The method includes a step of repeatedly calculating the result of the count modulo X = Y within the MRT time. In use, the X-th program segment is preferably transmitted every time Y = 0.

BRIEF DESCRIPTION OF THE DRAWINGS In order that the invention may be readily understood, a detailed description of a preferred embodiment of the program transmission optimization system and method is provided in the form of a demand-based video system and method based on the accompanying drawings. This is only an example.

FIG. 1 is a schematic diagram of a preferred embodiment of a demand-response video system, FIG. 2 is a more detailed block diagram showing functional blocks when the demand-response video system is applied to a CATV network, and FIG. FIG. 4 is a flowchart of the method steps used at the head end of the critical video system, FIG. 4 is a flowchart of the method steps used at the receiver side of the demand video system, and FIG. 5 represents a video segment transmission procedure based on a preferred scheduling algorithm. FIG. 6 is a graph showing the relationship between the longest response time and the required video transmission time.

DESCRIPTION OF THE PREFERRED EMBODIMENT FIG. 1 schematically illustrates a preferred embodiment of the demand-based video system of the present invention. In FIG. 1, external uncompressed data is sent into the system in the most basic format, such as 35 mm film, videotape, or over a communications link such as broadcast television transmission or satellite transmission. Uncompressed data is passed through a media compression system 10 for compressing audiovisual program data. The audiovisual program data may be compressed by a company that provides an external video compression service such as Intel. The externally compressed data may be sent directly from the storage distribution node 12 into the system. The storage distribution node 12 sends the compressed video data to a suitable storage medium.

There are three types of storage means in the system: long-term low-speed storage means 14, long-term high-speed storage means 16, short-term high-speed storage means 18.
There is. Construction of a mechanism for storing compressed video data in different types of storage means requires commercial considerations, for example, that a high-speed storage medium is more expensive than a low-speed storage medium. In addition, when selecting the type of storage means for storing various programs, it is necessary to predict what kind of video information demand will increase in the future. Daily news and the like should be stored in the short-term high-speed storage means 16, and movie classics such as "Gone with the wind" should be stored in the long-term high-speed storage means 16. Programs that are rarely requested, such as the silent movie that has become less popular, may be stored in the long-term low-speed storage unit 14. As the storage distribution node 12, a microcomputer or a mini computer that controls the flow of data between different storage devices is generally used.

Examples of the long-term low-speed storage unit 14 include a storage medium such as a magnetic tape and an optical disk. Since such a storage means stores a program which is rarely accessed, it may be one which requires human hand to retrieve it. As the long-term high-speed storage means 16, a jukebox type optical disk storage device is generally used. The optical disk storage means can perform high-density storage and random access, and can automatically call a program by employing a jukebox type access mechanism. Typical devices of this type currently available include the KODAK optical disk system 6800 drive /
Cabinets. As the short-term high-speed storage means 18, there is an IBM3380 type. This device allows quick random access to compressed video data stored in digital form, but is a relatively expensive storage medium so that it can be used only to store a group of popular video programs. Good.

The scheduling and routing computer 20 receives requests for particular audiovisual programs from the user's receivers 22A, 22B, 22C via the interactive request and distribution network. The scheduling and routing computer 20 retrieves the selected video program, divides it into a plurality of parts, schedules a plurality of segments for transmission according to a scheduling algorithm, and sets the receivers 22A, 22B, 22 according to the schedule.
Decide the route to send to one or more of C. This allows each viewer who sends the request to watch the program immediately and continuously. This demand-based video system uses a combination of a frequency multiplexing method and a time division multiplexing method. The time division multiplexing of video segments is controlled by a scheduling and routing computer 20 according to a scheduling algorithm. Frequency multiplexing is performed by the scheduling and routing computer 20 through the subscriber distribution node.
Perform 24 by controlling. The processing capacity of the scheduling and routing computer 20 is required to be comparable to that of a computer for a bank cash dispenser. The scheduling and routing computer 20 can be any computer capable of processing 1.5-200 MIPS (x1 million instructions per second), depending on the size of the subscriber base and other loading factors. .

As the receiver 22 on the viewer side, a receiver having a frequency variable capability to adapt to frequency multiplexing used at the head end of the system is used. The receiver is provided with processing means for receiving data packets generated by time division multiplexing of video segments. Also, the receiver 22 is provided with buffer storage means for storing the received video segments, and is further provided with data decompression means for decompressing the compressed video data and displaying the video on a television screen or conventional television. .

Although the demand-based video system of FIG. 1 can be used with either analog or digital communication circuits, in the preferred embodiment described below, the video distribution system is a conventional cable television system, ie, an analog system. In the embodiment of the system described below, the modulated digital data is mainly transmitted via the CATV network, but it can be applied to future systems using a combination of analog and modulated digital signals.

The method of the present invention for optimizing program transmission realizes optimization of transmission of both digital and analog signals.

Since conventional CATV systems are mostly simplex (one-way) communication systems, there is no easy way to retransmit data when an error is found. Therefore, some kind of error correction means is required. Fortunately, however, television data is usually ad hoc, unlike computer data, which must be checked for correct transmission. Even if the TV screen is disturbed by a few frames, most viewers will not even notice that it has happened. Therefore, the bit error rate can be much higher. For example 1
One bit error per 000 data bits is within the allowable range. At this rate, the human eye and brain tissue are usually not even aware of the effect this error bit has on the video. The error rate of most digital modems is on the order of 1: 100,000,000 to 1: 1,000,000,000. Thus, for the demand-based video system of the present invention, an error rate of 1,000 to 10,000 times the error rate of most current computer data systems is acceptable. Of course, increasing the error rate can improve performance, but viewers will not notice such improvements.

2 is a block diagram showing a preferred embodiment when the demand-based video system is used in a CATV network. The head end of the demand responsive video system is provided with a means for sending the video program in a compressed state. This means constitutes a long-term high-speed storage section 16 and a short-term high-speed storage section 18, respectively, which can be written once and read many times (WO
RM) storage device and a magnetic disk storage device 28. In this embodiment, the compressed video data is stored in digital form in the storage means, and the video program may be divided in advance into video packets of a size conforming to the specifications of the system in the storage medium. The magnetic disk storage device 28 and the WORM device 26 are connected to a scheduling and routing computer 30 via a data bus 32.

In response to a subscriber's request for a particular program, the scheduling and routing computer 30 retrieves the video program from the corresponding storage medium and divides the video program into multiple parts.
As described above, the video program is stored in the storage medium in a state of being divided into a plurality of parts in advance according to the scheduling conditions of the system, and is stored in the computer 30 during the video program retrieval and division processing steps. Such a load may be reduced. Computer 30 then schedules the segments of the video program according to a scheduling algorithm, as described in more detail below, and sends the video to one or more receivers of the viewers who have requested the video program according to the scheduling. Submit the program. For a group of popular video programs (video programs that are continuously in demand from at least one subscriber for a time longer than the time to play a single video (VPT)), the scheduling algorithm runs only once. The load on the computer 30 may be further reduced by storing the scheduled packets in a serial recording device such as a tape drive (not shown).

In this embodiment, the subscriber distribution node 24 comprises a plurality of modems 34 controlled by a scheduling and routing computer 30. Each modem 34 is a CATV
By modulating different carrier frequency signals corresponding to each channel of the network 36, the scheduling and routing computer 30 modulates the modem 34 via the bus 32.
To send the data packet of the video segment sent to. Each subscriber of the CATV network 36 has a receiver 40 for receiving the data packet of the video segment corresponding to the requested program and storing it for later viewing. Each receiver 40 processes a buffer memory 42 for storing a segment of the video program sent from the head end, and a video segment stored in the buffer memory,
It comprises video processing means for sending it to the subscriber's television in the correct order. The video processing means is a controller
52 and one or more modems 48 sent from the CATV network to the receiver
And a capture memory 46 for capturing the video segment data packet demodulated by the controller.
Under the control of the controller 52, the video processing means identifies a segment of the received program by a segment identifier such as a PKT ID, ignoring extra parts,
Overwrite on the capture memory 46. As the modem 48, a wideband modem having a frequency variable capability such as Fairchild M505 is preferable, but as described above, a low-grade digital modem having a lower bit error rate may be used. The compressed video data packet captured by the capture memory 46 is transferred to the buffer memory 42.
Is stored. Just take out the video segment from here and decompress the data with the decompressor 50,
You can see the program right away or later. A microprocessor-based controller 52 controls the data flow and video processing steps within the receiver 40.

Depending on the type of CATV system, a bidirectional decoder or receiver can be provided. In the case of this type of system, the receiver 40 is provided with an operation button 54 so that the subscriber can
Enable ordering via CATV network 36. However, most CATV systems are unidirectional (simplex),
The subscriber must place the order through the public switched telephone network (PSTN) 56. Subscriber orders through the PSTN 56 may be made verbally or with push-button keys such as those provided by other online subscriber network service companies.

There are several ways to prevent non-subscribers from viewing programs transmitted from the system without permission. Before transmission, the modem 34 may add a standard encryption algorithm to the program. In that case, each receiver 40 needs a key for decrypting the received data. The encryption and decryption keys may be distributed to the subscriber in a manner similar to the distribution of PINs by financial institutions for automated teller machines. Or, as another method, a person who tries to receive a specific video program without permission by putting a different receiver ID (identification mark) at the head of each data packet transmitted from the head end for each subscriber. May have to choose a matching recipient ID.

The demand-based video system shown in FIG. 2 is merely an example, and the method for supplying video according to demand and the hardware for implementing the system according to the present invention include:
Various types other than those described above are conceivable. For example, the receiver 40 can be provided with several modems to simultaneously receive data packets sent from several channels, and the video segments are stored in a buffer memory in a compressed state. If so, the capture memory can be omitted.
In this case, the decompression of the data of the video segment takes place when it is sent to the subscriber's television in the correct order. Further, certain portions of the headend device or receiver may be located at different locations. For example, given the architecture specific to CATV systems, the modem and buffer part of the receiver should be located at a location that would be part of the so-called subscriber outlet of the cable network, and the other part would be in the subscriber's premises. It is also possible to provide.

It is not necessary that the data packets of the video segment of a particular program be sent to all viewers over the same channel. Using a combination of time division multiplexing and multiple channels at the head end of the system to keep the data rate of each channel to a minimum requires less expensive hardware on the receiving side. The receiver 40 may scan the channels periodically to determine which channel the video program is being sent on. In addition to this,
Scheduling and routing computer
A dedicated control channel for transmitting data from 30 may be provided, and each receiver may be instructed which packet of which channel to receive. However, by attaching each video segment with the ID of the video title and the ID of each segment, and transmitting continuously from the headend according to the scheduling algorithm, each receiver can be assigned a video segment with the corresponding title ID. It's even better to receive all of the messages and discard or overwrite any that you have already received.

A key feature of the present invention is that by employing a redundant sequence for scheduling video segment transmissions, all video segments of the requested program can be re-arranged according to a schedule that allows the video program to be continuously played in real time on the receiver side. The point is to make sure that it can be received. The preferred form of the efficient scheduling algorithm and its execution method will be described in detail below.

In the following description, the term "longest response time" (MRT) refers to the longest time it takes for a subscriber to make a request for a video program before it can be viewed at the receiver. MRT refers to the longest time it takes for a system to respond to a request. Video playback time (VPT) refers to the time required to play a particular video program at normal playback speed. When dividing the data making up the video program into video segment data packets, each packet must be of a length that can be transmitted within 1 MRT. The playback time of the video segment, that is, the slot length of each data packet does not need to be less than 1 MRT, and how much bandwidth can be used within 1 MRT when transmitting via the transmission medium for data packet transmission Depending on the case, it may be longer than 1 MRT. The slot length may be variable to adjust the instantaneous load and data rate on the transmission medium or to adjust the size of the buffer storage space required for the receiver. However, the slot length and MRT are usually fixed for each system. In the following description, the slot length is set equal to the MRT for simplicity. So, for example, a video program
Assuming a length of 60 minutes and an MRT of 5 minutes, the video program is divided into 12 discrete data packets, each corresponding to 5 minutes of video segment data. Each data packet is numbered from 1 to n (where n is equal to VPT / MRT) in video viewing order. A preferred form of the scheduling algorithm is software controlled by a scheduling and routing computer 30. The basic flow of the scheduling program is shown below.

Set MRT to value equal to selected longest response time Set counter to initial value 0 Take out video segment data packets (PKT1, PKT2 ... PKTn) with size corresponding to MRT playback time Loop to MRT Add 1 to the count value to wait for the remaining time of the same time. If (count modulo 1) = 0, send PKT1. If (count modulo 2) = 0, send PTK2. (Count modulo 3) = 0 Send PKT3 if ...
...
If (count modulo n) = 0, send PKTn Return to the loop Note: (x modulo y) = rest of (x 残 り y) For the above scheduling algorithm, the video segment data packet is based on the redundant sequence One or more data packets are transmitted during each MRT time. Each transmission starts at an increment time n · MRT, and in most cases most of the MRT is spent on the actual transmission. For the above scheduling algorithm, PKT1
Is always transmitted, but other packets may or may not be transmitted when the count has the corresponding value. Therefore, some receivers may not be able to continuously receive packets. For example, MRT = 5, VPT = 60
, Packets are sent in the following order: The above means are only examples of many packet supply procedures that can be performed based on algorithms. Figure 5 shows each MR
5 graphically depicts the flow of a video segment data packet transmitted within a T time interval. The vertical axis represents the number of each video segment, and the horizontal axis represents the MRT interval. In the case of FIG. 5, the maximum value of the number of video segments is 30, and the maximum value of the number of MRT intervals is 49, but these values are arbitrarily determined, and both axes can be continued indefinitely. .
In practice, the number of video segments that divide a video program is limited, but the number of MRT intervals is a function of the amount of continuous demand for a particular program. Casually looking at the continuous output of the algorithm, the order of the packets appears to be random. But while the procedure seems pseudorandom, it is not. Because it never happens that a packet is not being sent to the receiver when it should be playing, and the packet is being sent to the receiver so that the video segments can be viewed immediately in the correct order . Thus, by the time the receiver is ready to play a particular packet, that packet should be in the buffer memory or otherwise received at that moment.

The column of “received packets” in the above table does not include extra packets. This is because these packets are actually discarded or overwritten by the receiver. In the above sequence example, all 12 packets have been received within 8 MRT time, and a packet, for example, PKT12, was received by the receiver long before it should be played. Packets received earlier than PKT12 or other playback time are stored in the buffer memory until the time to be played back. The scheduling algorithm ensures that the receiver receives the packet whenever it should be played or before.

In FIG. 5, at a specific point in time, for example, the MRT interval is 1
At 2, 24 and 36, it can be seen that more packets are received than at other times. In such a case, the load on the transmission medium and the buffer memory of the receiver increases. Therefore, the capacity of the buffer memory is desirably large enough to store all data packets of a specific program. This also allows the receiver to store the program for later viewing if necessary.

When determining the scheduling algorithm and the amount of buffer memory required for the receiver, the required response time (MRT), the bandwidth required to service the request,
Consideration must be given to the placement space of the buffer storage means in the receiver. An advantage of the scheduling algorithm described above is that the transmission medium can be used efficiently. That is, for example, when the MRT is set to 5 minutes, without the scheduling algorithm, the entire video program must be continuously transmitted every 5 minutes from the beginning. Therefore, if the playback time of the program is 60 minutes, the entire program must be transmitted 12 times. Using the above scheduling algorithm, the number of data packs that need to be transmitted is only 3.12.

The relationship between MRT and the amount of data that must be sent is
It is expressed by the following equation of “best fit curve”.

The total amount of data is the value associated with VPT. Therefore, the data amount 3 is equal to three times the VPT, that is, in the case of a program with an MRT of 5 minutes and a program of 60 minutes, the data amount is equivalent to 180 minutes. FIG. 6 is a graph showing the relationship between the MRT and the required video transmission time, that is, the total transmission data amount.

Although the above scheduling algorithm is most preferable, the algorithm can be changed in consideration of the balance between the transmission efficiency and the required capacity of the buffer storage device. In this case, the cost balance of each component of the system must be considered. If expensive buffers are used for the receiver, the transmission line may be low cost.

The scheduling algorithm can be changed so that PKT1 is not always sent. In this case, PKT1
And the number of transmissions of other specific packets may be reduced, and these may be stored in the space of the storage unit of the receiver until a request is issued. For example, the following method can be used.

Provide a low-capacity buffer memory device for each receiver,
Power is always supplied to receive a specific packet of the video program. For example, if there are 10 popular video programs, the receiver should be able to store the PKT1 of the above 10 video programs, and, as in the case of the above scheduling algorithm, PKT1 for each MRT.
The scheduling algorithm is set so that 1 is not transmitted but transmitted only at predetermined intervals. In this way, when a request for a popular program is made, the first packet is already stored in the buffer memory of the receiver, so that the program can be immediately viewed, and the scheduling algorithm is used in the meantime. Is executed. This can slightly increase the cost of the receiver, but can significantly reduce the required transmission bandwidth and provide the viewer with a popular program instantly.

Further, since it is not necessary to start transmission of all requests within the longest response time MRT, peak transmission load can be reduced. Further smoothing of the transmission load is possible if only a little transmission delay is tolerated.

Next, a typical software control procedure at the head-end computer and the receiver will be described with reference to FIGS.
It will be described based on. Upon receiving the request from the subscriber, the scheduling and routing computer at the head end records the subscriber ID, the ID of the requested program, and the time of the request. Computers are based on a scheduling algorithm
Manages the request and the procedure to complete the processing of the request.

Usually, several programs are constantly transmitted. As mentioned above, the scheduling algorithm generates different amounts of data over time. By making the input values of the counts for the different program flows shifted from each other, the total data amount in the transmission medium can be made relatively constant. Depending on the count value at the time the particular request entered the algorithm, from the 1MRT until the transmission of the requested program was completed
It takes time within 1VPT.

When recording the subscriber ID, title ID, and the time of the request, the head-end computer determines whether the requested program is currently running,
If so, start the scheduling procedure at the end of the current MRT time interval. If the requested program is not currently running, the initial value of the count of the internal counter of the computer (which may be a software counter) will be different from the input value of the count for each different program flow. Set to. Therefore, for example, if requests are issued simultaneously to programs A, B, C, and D,
Processing of the four requests begins simultaneously. However, when entering the algorithmic scheduling procedure, the count value of program A is zero, the count value of B is one,
Assuming that the count value of C is 2 and the count value of D is 3, a different number of video segments are transmitted simultaneously to each program during each MRT time interval, instead of the same number of video segments. Will be.

At the start of the next MRT interval, the computer starts the above-mentioned scheduling algorithm program procedure, and schedules the data packet of the requested title based on the scheduling algorithm. The computer also assigns a title ID and a packet ID to each data packet. The computer then selects an available channel and sends a data packet to the modem corresponding to that channel, which in turn sends it to the receiver that made the request. The head-end computer executes the above procedure from when the last request is issued for this title until transmission of all packets having this title is completed. If the computer determines that all requests have been processed, it stops sending data packets for that program.

On the receiver side, when a subscriber sends a request, the receiver scans the transmission channel and searches for a title ID. When the receiver finds the title ID, it searches for the packet ID, and stores all the packets not yet stored in the buffer storage unit. Packets that have already been received are discarded, and the receiver continues to search for remaining data packets until it has received all data packets for the requested program. The data packet sent to the buffer storage unit can be sent directly to the receiver so that the program can be viewed immediately, or stored for later viewing. Although not shown in FIG. 4, the receiver may be able to search for an address ID for preventing a non-subscriber from receiving data without permission.

As will be apparent from the above description of the preferred embodiment of the transmission optimization system and method for providing video on demand, the use of a scheduling algorithm allows for efficient response to a large number of subscriber requests. The program can be transmitted and the reproduction of the program can be started within a predetermined maximum response time. Those skilled in the electronics, television, and communications arts will appreciate that various modifications and variations can be made to the above-described system and method for optimizing program transmission without departing from the basic concepts of the present invention. Seem. For example, as a modification of the above, an optical fiber network may be used in the distribution system to provide a program in response to, for example, a request from an airline passenger. Further, the system and method can be used for analog communications, digital communications, or a mixture of both. In the case of the demand-based video system described above, the video program segments were transmitted in a compressed state, but it will be clear that this is not a key feature of the present invention. This is because transmission efficiency can be greatly improved only by the scheduling algorithm. All such modifications and variations are considered to be within the scope of the present invention. The features of the invention are defined by the appended claims.

Continuation of front page (58) Field surveyed (Int.Cl. ⁷ , DB name) H04H 1/00 H04N 7/173

Claims (24)

(57) [Claims] 1. A method for optimizing the transmission of a program to a plurality of receivers via a program distribution system, wherein one or more receivers request a specific program simultaneously or at different timings. The program distribution system minimizes the transmission means required to respond to the request while minimizing each request regardless of whether or not a receiver other than the one or more receivers has already received the particular program. (A) dividing a program into a plurality of program segments and executing a scheduling algorithm for organizing the program segments into a predetermined sequence; And (b) required to reproduce a program at a normal speed in the plurality of receivers. Within a time period, at least a portion of the program segment is transmitted from the head end to the plurality of receivers at the same time many times without having to use the head end of the program distribution system only for a specific receiver. Transmitting a program segment in accordance with a scheduling algorithm to be performed; and (c) storing the transmitted program segment in a buffer means of the receiver for later reproduction, and When the program segment is received many times by the receiver, the program segment is stored only once in the buffer means and, if desired, is smaller than the capacity required to hold the entire program if desired by the scheduling algorithm. Has flexibility to compose buffer means A method for optimizing transmission, the method comprising the steps of enabling all receivers to receive all program segments so that the program can be played continuously at the normal playback speed of the program. 2. The method of claim 1, further comprising the step of selecting a longest response time (MRT) corresponding to the longest time required to begin playing the program requested by the viewer from the beginning of the program. 3. In the step (a), when dividing the program into program segments, a length of the program segment is selected so that at least one program segment can be transmitted in one MRT. 3. The method according to 2. 4. The method according to claim 1, wherein the step (b) includes a first program segment corresponding to a first segment of a program reproduction time according to the scheduling algorithm.
4. The method according to claim 3, wherein one or more program segments are transmitted in each MRT, and the first program segment is always received by the receiver in one MRT when used, and can be immediately reproduced. Method. 5. The program distribution system according to claim 4, wherein in the head end of the program distribution system, numbers from 1 to n are allocated to n program segments obtained by dividing the program in the order of reproduction at the time of reproduction at normal speed. The described method. 6. The scheduling algorithm comprises the following steps: count modulo X = Y (count = a predetermined initial integer value that increases by 1 at the end of each MRT, x = 1 to nn = a segment obtained by dividing a program. ) Iteratively calculating the result in each MRT, organizing the program segments into the predetermined sequence based on the result of the calculation, and using the number X in each use when Y = 0. 6. The method according to claim 5, wherein the transmission of the program segment is performed. 7. A method for recording a viewer identification number, a program title identification number, a time when a request is made by a viewer, and a step of monitoring the progress of each request and its completion. 2. The method according to 1. 8. An apparatus for optimizing the transmission of a program when transmitting the program to a plurality of receivers, wherein when one or more receivers request a specific program at the same time or at different times, the program distribution system may , While minimizing the transmission means required to respond to the request, responding to each request regardless of whether the receiver other than the one or more receivers has already received the particular program. Means for dividing a program into a plurality of program segments on a head end side of the program distribution system and executing a scheduling algorithm for organizing the program segments into a predetermined sequence; and a program in the plurality of receivers. Less program segments within the time it takes to play Some also follow a scheduling algorithm that allows the program distribution system's headend to be transmitted from the headend to multiple receivers many times at the same time without having to use the headend exclusively for a particular receiver. Means for transmitting, and buffer means for storing the transmitted program segment for later reproduction on the receiver side of the program distribution system, wherein the same program segment is repeatedly received by the receiver. The program segment is stored only once in the buffer means, and according to the scheduling algorithm,
If desired, assign all program segments to allow continuous playback of the program at the normal playback speed of the program, while providing the flexibility to configure a buffer that is smaller than required to hold the entire program. The transmission optimizing device comprising buffer means that can be received by a receiver. 9. A head end of the program distribution system,
Means for allocating numbers from 1 to n to n program segments obtained by dividing the program in the order of reproduction at the time of reproduction at normal speed, wherein the number allocating means comprises the program preparing means and the program segment 9. Apparatus as claimed in claim 8, wherein the apparatus is connected to cooperate with the transmitting means to transmit the program segments according to a scheduling algorithm. 10. A head end of a program distribution system, comprising means for assigning an identification mark to each program segment before transmission, wherein said identification mark allows a program segment to be identified at least by its number. 10. The apparatus according to claim 9, wherein in cooperation with said number assigning means, a number assigned to each program segment is determined and included in said identification mark. 11. The method according to claim 1, wherein the transmitting means transmits one or more of the program segments during each of a maximum response time (MRT) corresponding to a longest waiting time until the reproduction of the program requested by the viewer starts. 9. The device according to claim 8, adapted to transmit. 12. The apparatus according to claim 11, further comprising scheduling means for scheduling said plurality of program segments, wherein said scheduling means comprises a scheduling algorithm: count modulo X = Y (count = each MRT). Means for repeatedly calculating the result of (x = 1 to nn = the number of segments formed by dividing the program) in each MRT, and a result of this calculation. Means for organizing the program segments into the predetermined sequence based on the above, and executing the transmission of the X-th program segment each time Y = 0 when used, wherein the organizing means comprises: The program segment for executing transmission of the program segment based on the program segment 12. The apparatus according to claim 11, wherein said apparatus cooperates with a transmitting means. 13. A receiver for receiving a program supplied to a plurality of receivers by a system for optimizing the transmission of the program, wherein one or more receivers request a particular program at the same time or at different times. The program distribution system responds to each request regardless of whether a receiver other than the one or more receivers has already received the particular program, while minimizing transmission means required to respond to the request. Within the time required to play the program at normal speed on the plurality of receivers, at least a portion of the program segments may be used to set the head end of the program distribution system only for a particular receiver. Scheduling algorithm that allows multiple simultaneous transmissions from the headend to multiple receivers without having to use A buffer means for accumulating a plurality of program segments transmitted from a head end of a system for optimizing the transmission of a program according to a rhythm, wherein when the same program segment is received by a receiver many times, the same Buffer means for accumulating a program segment only once; processing means for processing the program segment accumulated in the buffer means and supplying the program segment in a correct sequence for performing reproduction, wherein the scheduling algorithm All program segments so that the program can be played continuously at the normal playback speed of the program, while providing the flexibility to configure a buffer means smaller than required to hold the entire program if desired. Throat A receiver comprising processing means for enabling a receiver to receive a message. 14. The processing means includes means for identifying a received program segment by a segment identification mark, wherein the segment identification mark can identify a segment by at least a segment number.
14. Apparatus according to claim 13, whereby the receiver is able to distinguish between segments needed for playback and extra segments. 15. The apparatus according to claim 13, wherein said processing means comprises decompression means for decompressing a compressed program segment transmitted from a head end of the program transmission optimization system. 16. A scheduling apparatus for a system for optimizing the transmission of a program from a headend to a plurality of receivers, wherein the program is divided into a plurality of program segments, and the program segments are organized into a predetermined sequence. Means for executing a scheduling algorithm for at least a portion of the program segment within the time required to play the program at normal speed on a plurality of receivers, identifying the head end of the program distribution system A means for scheduling said plurality of program segments according to a scheduling algorithm that allows multiple simultaneous transmissions from the headend to the plurality of receivers without having to use only for the receiver of the requested program. The program segment scheduled to one or more receivers of the listener is stored in the buffer means of the receiver only once when the same program segment is received by the receiver many times. And the above-mentioned scheduling algorithm allows continuous reproduction of the program at the normal reproduction speed of the program, while having the flexibility to configure a buffer means smaller than the capacity required to hold the entire program if desired. A scheduling device that enables all receivers to receive all program segments. 17. The program division means can transmit at least one program segment within a longest response time (MRT) corresponding to a longest waiting time from the beginning of a program requested by a viewer to the start of reproduction. 17. The scheduling device according to claim 16, wherein the device is divided into pieces. 18. The program distribution system according to claim 17, wherein in the head end of the program distribution, numbers from 1 to n are assigned to n program segments obtained by dividing the program in the order of reproduction at the time of reproduction at normal speed. A scheduling device as described. 19. The scheduling algorithm may include: count modulo X = Y (count = predetermined initial integer value incremented by one at the end of each MRT x = 1 to nn = part of a segment obtained by dividing a program ) Iteratively calculating the result in each MRT, organizing the program segments into the predetermined sequence based on the result of the calculation, and using the number X in each use when Y = 0. 19. The scheduling apparatus according to claim 18, wherein transmission of the program segment is executed. 20. The scheduling according to claim 16, wherein the program segmenting means takes out a program segment from a means for accumulating a compressed program, and accumulates the program in a compressed segment format corresponding to the plurality of segments. apparatus. 21. The method of claim 1, wherein in step (b), program segments are transmitted more frequently at an earlier stage of the program than at a later stage of the program. 22. In the step (b), a plurality of program segments are transmitted according to a scheduling algorithm modified so as to smooth a transmission load and to reduce a bandwidth required for transmission. 2. The method according to 1. 23. The method according to claim 23, wherein the plurality of program segments include a first program segment, and in step (b),
The one or more first program segments are transmitted at predetermined time intervals, thereby accumulating indefinitely in a buffer means of the receiver, and a plurality of program segments except the one or more first program segments are stored according to a modified scheduling algorithm. 23. The method according to claim 22, characterized in that the first program segment is always available for immediate viewing at the receiver when transmitted and used. 24. The method of claim 1, wherein prior to step (b), the program is compressed and a plurality of program segments are stored in a compressed format.