TW571594B - Methods for providing video-on-demand services for broadcasting systems - Google Patents

Abstract

A method for sending data to a client to provide data-on-demand services comprises the steps of: receiving a data file, specifying a time interval, parsing the data file into a plurality of data blocks based on the time interval such that each data block is displayable during a time interval, determining a required number of time slots to send the data file, allocating to each time slot at least a first of the plurality of data blocks and optionally one or more additional data blocks, such that starting from any of the time slots, (i) the data file can be displayed by accessing the first of the plurality of data blocks; (ii) at a consecutive time slot, a next data block sequential to a prior displayed data block is available for displaying; and (iii) repeating step (ii) until all of the plurality of data blocks for the data file has been displayed, and sending the plurality of data blocks based on the allocating step.

Description

571594 V. Description of the invention (1) Related applications This application belongs to the title of ns vs terns and Methods for Providing Video-On-Demand Services for Broadcasting Systems', and was filed on May 31, 2000. Part of the parental application with application serial number 09 / 584,832 continues the application. SUMMARY OF THE INVENTION The present invention relates generally to on-demand data systems. The present invention is particularly related to on-demand video systems. BACKGROUND OF THE INVENTION The On-Demand Video (V0D) system is a certain type of On-Demand Fundamentals to provide video data files to a certain number of channels. In the traditional V0D architecture, there is a certain structure that makes people cry. Client, server-side material file (ie video data). The idiom Mingyi gives a server, and obtaining the server will request the 4 asset files;;: The client requests, the server, ... in the model, "" by Chuan Mou, two should = ^ The client's data file request. The client does not receive the server to satisfy a customer = any data file received. The capacity is based on the local communication infrastructure. However, the server-side model will ^ ㈡ = because of the existing: ί : ㈡ = An example of a νΛ-based basic-directional communication infrastructure can be constructed. Γτί;: Directional communication functions.ί Such as hybrid fiber with I ^

(HFC) ”, and at the same time, or the fiber-based foundation, the final service price is also $. The cost of replacing existing cables is extremely high. 3 It will not be affordable to most consumers. In addition, when a service provider (VOD service is provided to a large number of customers, the production = 1 cable company) attempts to provide it with many restrictions. This _quasi-quasi-client-server model will be limited to the service provider must; the end-to-server model of the other-listening and satisfaction comes from the machine! : Dagger :: Continuously pour: The number of customers receiving the service will be the same as this :: The capacity is two, such as 2 parallel computers, and a large and fast disk array for 2 locals. + Yes, even the fastest local feeding service can only deliver video data streams to about 1 0 0 to 2 0 0 clients at a time. So, to serve more clients, you must increase the local Number of servers. Adding local servers will require more high-end servers to control local servers. Another limitation of this standard client-server model is that each client will require its own bandwidth. In this way, the total bandwidth required will be directly proportional to the total number of subscribers. The G-fetch memory in the local server has been used to improve the bandwidth limitation, but cache memory cannot solve the problem because cache memory is also limited. At present, in order to make on-demand video services affordable for general customers, existing service providers are striving to increase the percentage of customers on each local server to exceed the capacity of the local 4-server. Usually, a local server that can provide services to 1,000 customers can actually serve up to 10,000 clients

Page 6 571594 V. Description of the Invention (3). And even if most clients do not subscribe to video services at the same time, this technology will work as usual. However, this technology is bound to fail, because most clients really tend to order video services at the same time (ie — night news and weekend holidays, etc.), which will cause overload problems on the local server. . 'Therefore, there is a need to provide a system that can provide a large number of customers with on-demand services through almost any transmission medium without having to replace the existing infrastructure. SUMMARY OF THE INVENTION In an exemplary embodiment, at the server end, a method for sending data to a client and providing on-demand data services includes the following steps: receiving a data file, calibrating a time interval, Time interval and _ analyze the data file into many data blocks so that each data block can be displayed in the time interval, determine the number of time slots required to send the data file, and at least one of the data blocks One or more additional blocks are allocated to each time slot as needed, so that these data blocks can be used for a client accessing the data file in sequence during any time slot, and according to the configuration steps Send out blocks of data. In a specific embodiment, the analyzing step includes the following steps: determining an estimated data block size, determining a memory cluster size in the channel server, and according to the estimated data block size and the size of the cluster Analyze the data file. In another specific embodiment, the determining step includes evaluating a resource allocation mode and a bandwidth availability. 4 In a specific embodiment, the method further includes selecting a set of prefetched data blocks from the data blocks and transmitting the prefetched data,

Page 7 571594 V. The steps of sending out the prefetched data blocks separately in the exclusive channel of the invention description (4), the program table, the advertisement, the firmware update, etc. In an exemplary embodiment, the steps for selecting the set of prefetched data blocks include the following steps: (1) determining the bandwidth reduction, the bandwidth allocation method of the prefetched data in the dedicated channel, and the delay time And (2) selecting the prefetched data blocks according to the bandwidth reduction, the bandwidth allocation method, and the delay time. In another specific embodiment, the method further includes the steps of receiving a preview request, randomly selecting a group of data blocks from the data blocks to synthesize the preview, and activating the steps of displaying the preview. In yet another specific embodiment, the method further includes sending a set of advertising data blocks in the exclusive channel, and starting to display the set of advertising data φ blocks at a predetermined time point. In an exemplary embodiment, the set of advertisement data blocks is sent in a dedicated channel in a continuous manner. In this embodiment, the step of displaying the group of advertisement data blocks includes the steps of receiving a user option price according to the frequency of display of the advertisements, and activating the group of advertisement data blocks according to the user options. In another specific embodiment, the method further includes the steps of checking the emergency bit of the data file packet header, switching to a dedicated channel to receive emergency information when the emergency bit is detected, and starting to display the emergency information. . In a specific embodiment, the method further includes determining whether the emergency information is directed to a relevant area, and displaying the emergency information if the emergency information is indeed directed to the relevant area.鄱 In an exemplary embodiment, at the client, a method for processing data received from a server, for a data-on-demand service, includes:

Page 8 571594 V. Description of the invention (5) The following steps: (a) receiving the data / options in the data file during the first time slot-the second time slot; (b) in the second time slot Process data; block; then display a certain data area of the data file 0 = 22 = shell material block, all the certain adapters :: to perform this item-, in a non-standard embodiment, A data file will be cut into multiple ^ blocks, and a scheduling matrix will be generated based on these data blocks. The ^ ^ yi ^ ^ matrix can be provided as a transmission sequence for sending data blocks. In the non-target specific embodiment, a method for generating ^ 丨 ΐϊ "and" square f "for a certain data file includes the following steps: (a) receiving data blocks [X] of a data building; (b) ) Set a first variable [j] to zero value; (c) Set a first variable [i] to zero value, (d) Clear each item in a reference array; (e) 〃 If a reference array has not If the data block is included, at least one data block in the column position [(i + j) modulo X] stored in the matrix is written into the reference array; (f) If a reference array does not already contain The data block [1] 'writes the data block [1] into the reference array and the matrix position [(i + j) modulo x, j] in the matrix; (g) increments the second variable one by one [i] and repeat step (e)-until the first-number "丨] is equal to the number of data blocks [X]; and (h) recursive one by one; change =; (step (C) until the first A variable [j] is equal to the number of data blocks [χ]. In an exemplary embodiment, each data in a certain level data file may be

571594 V. Description of the invention (6) The file generates a schedule matrix and applies a round-robin method to it. According to the schedule matrix, a transfer matrix for generating the set of data files is generated. An on-demand data system includes a first set of channel servers, a central control server for controlling the first set of channel servers, and a first set of uplink converters coupled to the first set of channel servers A combiner / amplifier coupled to the first set of uplink converters, and a combiner / amplifier adapted to transmit data through a transmission medium. In an exemplary embodiment, the on-demand data system may further include a channel monitoring module for monitoring the system, a switching matrix, a second group of channel servers, and a second group of uplink converters. The channel monitoring module is configured to report to the central control server when the system fails. The central control server responds to the report of the channel monitoring module by φ, which can cause the switching matrix to replace an error channel server in the first group of channel servers with a channel server in the second group. The channel server replaces an incorrect uplink converter in the first group of uplink converters with an uplink converter in the second group of uplink converters. A method for providing an on-demand data service includes the steps of calculating a transfer matrix of a data file, and sending the data file according to the transfer matrix, so that a large number of clients can view the data file in an on-demand manner. In a specific embodiment, the data file includes a video file. BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1A is an exemplary DOD system according to a specific embodiment of the present invention. ❿ FIG. 1B is an exemplary DOD system according to another embodiment of the present invention.

Page 10 571594 V. Description of the invention (7) '------ 一 ~-= ί —root ί ί Exemplary embodiment of the exemplary channel server. Lί-fiiii is an exemplary video adapter of the specific embodiment. FIG. 4 is a method for processing irregularities according to a specific embodiment of the present invention. Detailed description of the invention of the torque matrix FIG. 1A is an exemplary DOD system 10 according to the present invention and with little effort & . In the f μ octant embodiment, the DOD system 100 can be broadcasted according to the information on demand, video rights, and so on. No, you are not limited to 陆 f Ϊ ί ί files. You can also provide other information files. They are provided on-demand. Video file. This official account = lla l0V \ ^ ^ Big 5§ins 11 Interview with Shi Xi on the transition state 106a-106n and a combiner / amplifier ^. u Central control server 1 (Γ2 can control channel servers) 04 ^ 2 θ threat storage device 103 can store data files in digital format. In the specific example of target 1 shown, it can be used by any authorized A computer connected to the network, such as the central control server 丨 〇2, accesses the data stored in the central storage device 10 3 / through a standard network interface (such as an Ethernet connection). Each channel server 丨 〇4 will be assigned a channel, and is connected to a certain uplink converter 106. Channel servers 104 can follow the instructions of the central control server 〇2, The data file obtained from the central storage device 103 is provided. The output of each channel server 〇 04 is modulated by the f-amp amplitude modulation (qam) and has an uplink converter corresponding to it. 1/0 6 Intermediate frequency (IF) signal of suitable frequency. QAM modulation IF signal is related to the adopted standard. "Data_over-cabie-SyStems- 571594 V. Description of the invention (8) interface-specification (D0CSIS) "is the standard currently used in the U.S. IF frequency of 43 · 75 MHz. The uplink converter 106 can convert the IF signals received from the channel servers into radio frequency signals (RF signals). The rf signal includes frequency and bandwidth, It depends on the desired channel and the standard adopted. For example, under the current standard of 80 cable television channels in the United States, the RF signal has a frequency of about 5 5 · 25 MHz, and the bandwidth is about 6 MHz. The output of the up-converter 106 is applied to the combiner / amplifier 08. The combiner / amplifier 108 can amplify, adjust and combine the received RF signal, and then The signal is output to the transmission medium 1 1 0. In a non-expendable specific embodiment, the central control server 丨 0 2 includes a user interface (not shown) ′ so that the service provider can 〇2 ϋ ^ It% ΛΠ *. F Ϊ 1 m ^ Start or stop operation. Frequency trunk two services two 4 = data, dT data Λ fetch: channel, and calculated to provide data on a selected frequency Data = matrix of transfer. The central control server 102 can 102 can be updated by briefly Bright). The central control server centrally stores the device 103. The data slot and database inside it are used to control the median, = continued mil jj: medium '-the existing cable TV system 12 provides non-DOD service items to the \ merger / amplifier 108, and to the clients. Thus, the DOD system according to the present invention is not

Page 12 571594 V. Description of Invention (9) The current cable TV service project will be interrupted. FIG. 1B illustrates a specific embodiment of another exemplary DOD system 100 according to the present invention. In addition to the elements shown in Figure 1A, the DOD system 1 00 also includes a switching matrix 1 1 2, a channel monitoring module 1 1 4, a group of redundant channel servers: 116a-116b, and a group Backup uplink converters 118a-118b. In a specific embodiment, the switching matrix 1 12 system is located between the uplink converter 106 and the "combiner / amplifier 108". The switching matrix 1 12 is controlled by the central control server 1 02 Controlled. The channel monitoring module 1 1 4 includes many video adapters that have been configured. These can simulate many potential customers to monitor the operating status of the DOD system 100. The monitoring results will be monitored by the channel monitoring module.丨 14 It is transmitted to the central control server 102. If a channel fails (that is, if a channel server fails, an uplink converter fails, or a communication link fails), the central control server 1 02 can Remove the faulty component through the switching matrix 1 12 and connect a Jianjia backup component 1 1 6 and / or 1 1 8 to restore the service status.-In an exemplary embodiment, since The data files broadcasted at 100 of the DOD system will be included in the "Motion Picture Experts Group (MPEG)" file. Each MPEG file will be dynamically divided into a number of data blocks and sub-blocks that map to a specific part of the data y along the time axis. These data blocks and sub-blocks are sent in a predetermined time according to the three-dimensional intersection, matrix provided by the central control server 02. The DOD system 1000 does not need to feed back channels to provide DOD services. However, if a feedback channel becomes available ’it can be used for other purposes, such as accounting or providing Internet services.

Page 13 571594 V. Description of the Invention Fig. 2 shows an exemplary channel server 104 according to a specific embodiment of the present invention. The channel server 104 includes a server controller 202, a CPU 204, a QAM modulator 206, a local memory 208, and a network interface 21 °. The server controller 202 can instruct the CPU 204 to divide the data file into blocks (or further into sub-blocks and data packets) according to the handover provided by the server controller 202 The matrix selects data blocks for transmission, encodes the selected data, compresses the encoded data, and then delivers the pressurized data to the QAM modulator 206 to control the overall operation of the channel server 104. The QAM modulator 206 will receive data to be transmitted via a certain bus (ie, PCI, CPU local bus) or Ethernet connection. In an exemplary embodiment, the QAM modulator 206 includes a lower QAM modulator and a forward error correction amplitude modulation / positive phase shift key (QAM / QPSK) surge modulator and / or uplink modulator. The output of the QAM modulator 206 is an ip signal, which can be directly supplied to the uplink converter 106.

The network interface 2 1 0 connects the channel server 丨 〇4 to other channel servers 1 04 and the central control server 丨 02 to execute & from the central f system feeder 1 0 2 Schedules and control instructions, and report the status to the central control server 102 'while receiving data from the central storage device ι03. Any data received from the central storage device 103 will be stored in the channel server before processing the data file according to the instructions sent by the feeder controller 202. The local memory of 4 is 208. In an exemplary embodiment, the channel server 104 can be adjusted according to the bandwidth of the channel (eg, 6, 6, 5 or 8 MHz), and QAM modulation. (which is

Page 14 571594 V. Description of the invention (li) For example, QAM 64 or QAM 2 5 6) and the compression standard / bit rate (MPEG-1 or MPEG-2) of the DOD data stream to send the DOD data stream. FIG. 3 is an exemplary video adapter (STB) 300 according to an embodiment of the present invention. The STB 3 0 0 includes a QAM modulator 3 0 2, a CPU ° 304, a conditional access module 3 0 6 (such as a smart card system), a local memory 308, and a buffer memory. 309, an STB controller 31, a decoder 312, and a graphics overlay module 314. The STB controller 310 can control the CPU 304 and the QAM modulator 302 in response to a client request 9 to select data, decode the selected data, decompress the decoded data, reassemble the decoded data, and store the data. Decode the data in the local memory 308 or the buffer memory 309 and pass the existing data to the decoder 3 丨 2, so as to control the overall operation of the STB 300. In one In an exemplary embodiment, the gTB controller 3 1 0 can control the overall operation of the STB 300 according to the received data packet header ^ data packet header originating from the transmission medium 1 1 0. In one In a specific embodiment, the local memory 308 includes a non-volatile memory (such as a hard disk drive), and the buffer memory 309 includes a volatile memory. In a specific embodiment, the QAM modulator 302 includes a transmitter and receiver module 'and one or more of the following items: Private encryption / decryption module: forward error correction decoder / encoder, tuner control , Downstream and upstream processors, CPU and memory interface circuits. The QAM Variable is 3〇2 can receive both the modulated IF signal, the demodulated signal is sampled and reset the resource Shu material obtained 0, when both certified and / or after receiving Now that obtained by taking proper credits

Page 15 571594 V. Description of the invention (12) The Du Fetch module 306 is available to allow a solution method. The access stop is determined by the service provider. Jie Ji access strip 31 2 in the solvable Ding Fan: less with a certain body: in the example, when it can be accessed, the resolver is obviously dry; ^ t & a φ / block, 俾To convert that block of data into an image on the screen. The demapper 312 can support instructions from = read: user: come, such as broadcast & stop, rewind, fast forward, and so on. $ Τ7 Appropriate: ΠίίΠ14 can be provided by like ... or in-image, for example, when the service is provided by a person in charge 田田 姥姥 /, the embodiment chooses the service (顼 目 B #). The system to provide the game with the graphics overlay module 314 can be used as graphics acceleration during the game mode process T. In the embodiment, 'Although the data will be broadcast to be decoded and used for various data on-demand services, in an exemplary and consistent embodiment, the intellectual property in module 3 06 can be accessed through this condition. Pan type ^ / The vending machine set by the service provider recharges the smart card. ^ ^ = In a specific embodiment, a single pricing (f Ut f ee) is to allow subscribers unlimited access to all available data files. In an exemplary embodiment, the data on-demand interactive feature client selects available data targets at any time. The amount of time the client presses ^ 1 until the selected data file starts playing is called the response time. Resources configured to provide D0D services (ie, bandwidth, server

Page 16 571594 V. Description of the invention (13) The more capacity), the shorter the response time will be. In an exemplary implementation + example, the response time may be determined according to the resource allocation and the desired service quality evaluation results. In an exemplary embodiment, a selected response time may determine the length of the time slot. The time slot (TS) period is the time interval at which a client plays a data block at normal speed. In an exemplary embodiment, a data file, such as a video file, is divided into a plurality of data blocks, so that each data block can support playing the data in the time slot period. In a specific embodiment, the number of data blocks (NUM-OF_BLKS) of each data file can be calculated according to the following equations: |

Estimated_BLK_Size = (DataFile—Size * TS) /

DataFile—Length (1) BLK SIZE = (Estimated BLK Size + CLUSTER—SIZE-lByte) / CLUSTER_SIZE (2) BLK—SIZE—BYTES = BLK—SIZE * CLUSTER—SIZE (3) NUM_OF_BLKS = (DataFile— Size + BLK_SIZE_BYTES-lByte) (4) / BLK_SIZE_BYTEs In equations (1) to (4), Est imated_BlK — Size is estimated: Cang M wide block size (in bytes) ); DataFile_size is the size of the data file; (TS) is the time slot period (in seconds); DataFile—Length ^ bit rH; n'EHs 1ZE is the cluster required for each data block. The channel server 104 is in local memory 20; the size of the family (such as 64 Kbytes); BLK_SIZE_mES is a certain 7

Page 17 571594 V l. Description of the Invention (14) — Size of byte. In this specific embodiment, the data area (NUM_OF — BLKS) will be equal to the data file size (in bytes) plus the data block size of the group minus 1 byte, and then divided by the 彳Li Fan's data block size. Equations (丨) to (4) illustrate a specific and 70-group example. For those who are familiar with this skill, it should be clear and easy to implement other methods to calculate the number of data blocks in the data file. For example, the data block from which ^^ is formed may be roughly a function of an estimated block size and the local memory 208 cluster size of the channel 2 ^ partition 104. As such, the server should be limited to the specific embodiments described above. As shown in FIG. 4 is an exemplary processing method for generating a transmission data file according to a specific embodiment of the present invention. In an exemplary process matrix, the present invention uses "time-division multiplexing" (TDM) and rhyme and evening embodiment (FDM) technology to compress and schedule data secondary jobs on the server and server side. "An exemplary specific implementation For example, one job can be delayed for each data file. In. In a specific embodiment, each data file is ^: a scheduling matrix block, and a scheduling matrix is generated according to the number of data blocks. === Evening data arrays can provide a method for transmitting certain data files from a server = Second, the sending order of the moment end of the schedule, so that any intentional access to the data = block to the client can access the client at a random time in a sequential manner at various times. At step 40, 2 receives Many data blocks of a data file. A variable j is set to a zero value (step 4 0 4). Then, master, ... block (x). Step 4 0 6). The reference array will continue to track the data area = f test array (step. Then, compare j and X with each other (step 40 8). The second variable i for internal management is set to zero (step 412). If \ is less than X, then i and x are compared with each other

571594 I '' ~~ ___________________ V. Description of the invention (15) ^ Compare (step 4 1 4). If i is less than χ, the data block stored in the column [(i + j) mod (X)] in the scheduling matrix will be written into the reference array / Nu 4 1 8). If the reference array already has these data blocks, do not write the duplicates. Initially, this step can be skipped because the reference array does not yet have any terms ^. Next, it can be checked whether the reference array contains a binary block i (step 420). Initially, since all items in the reference array have been cleared at step 406, there will be no data in the reference array. If the reference array does not contain data block i, the data block i will be added to the column in the scheduling matrix [(i + j) m 〇d (X), · place and the reference Inside the array (step 4 2 2). After the data block i is added to the L / Li scheduling matrix and neutralized in the reference array, the value of i is increased by one so that i = i + l (step 424), and then the procedure is repeated at step 414. , ^ Ii = X. If the reference array contains the data and material block i, the value of i will be added to make i = i + l (step 424), and then the process is repeated at step 414 until i = x. When i = x, the value of j is increased by one, so that j = j + 1 (step-

4 1 6), and then repeat the procedure at step 4 6 until j =: X. When ~ ends the whole program (step 4 1 0). X In an exemplary embodiment, if a data file is divided into six data blocks (X = 6), the scheduling matrix and reference array may be as follows: Scheduling Matrix (SM)

TS0 TS1 TS2 TS3 TS4 TS5 [0,0] blkO [l, 0] blkl [2,0] blk2 [3,0] blk3 [4,0] blk4 [5,0] blk5 [〇, 1] [l, l] blk0 [251] "3,11 [4,1] [5, 〇]" 〇, 2] _ [1,2] [2,2] blkO [3,2] blkl [4,21 [5, 1] [〇, 3] [1,3] [2,3] [3,3] blkO [4,3] [5,2] blk2 [〇, 4] [l, 4] blk3 [2,41 [ 3,4] [4? 4] blkO [5,3] blkl 1 * 〇, 5] [1,5] [2,51 [3,5] blk4 [4,5] [5,4] blkO 19th Page 571594 V. Description of the invention (16) Reference array (RA) spaceO space1 space2 space3 space4 space5 TSO blkO blkl blk2 blk3 blk4 blk5 TS1 blkl blkO blk2 blk3 blk4 blk5 TS2 blk2 blkO blk3 blkl blk4 blk5 TS3 blk3 blk4 blkl blk blkO blk5 blk2 blkl blk3 TS5 blk5 blk2 blkl blkO blk3 blk4 In an exemplary embodiment, according to the scheduling matrix as above, the six data blocks of the data file will be sent out in the order: TSO = > blkO TS1 II > blkO, blkl, blk3 TS2 blkO, blk2 TS3 II > blkO, blkl, blk3, blk4 TS4 = > blkO, blk4 TS5 = > blkO, blkl, blk2, blk5 is an exemplary implementation In the example, a look-ahead processing method can be used to calculate a look-ahead scheduling matrix, so that a predetermined number of data blocks of a data file are sent before a predicted access time. For example, if a predetermined look-ahead time is For a time slot period, for any time slot higher than or equal to the fourth time slot, the STB 300 located at a subscriber client will receive the data block of the tribute slot at or before TS3 4 (b 1 k 4) 'However, blk4 will not be played until TS4. Except that the look-ahead schedule matrix in this specific embodiment will schedule an earlier transmission sequence based on a look-ahead table, the look-ahead row is generated The method steps of the process matrix are roughly similar to the method steps of FIG. 4. Hereby

Page 20 571594 V. Description of the invention (17) A certain data file is divided into six data blocks. According to a look-ahead scheduling matrix, an exemplary transmission sequence of look-ahead time with two time slot periods It can be expressed as follows: TSO = > blkO TS1 = > blkO, blkl, blk3, blk4 TS2 = > blkO, blk2 TS3 two > blkO, blkl, blk3, blk4, blk5 TS4 = > blkO, blk5 TS5 = > blkO, blkl, blk2 In this way, according to the scheduling matrix of each data file in a set of data slots, a three-dimensional submission matrix can be generated for sending the set of data files.在 这 三 In the three

In the dimensional submission matrix, a third dimension including the ID of each data file in the set of data files can be generated. The three-dimensional delivery matrix is calculated to efficiently use the available bandwidth in each channel to deliver multiple data streams. In an exemplary embodiment, a convolution method known in the industry may be used to generate a three-dimensional submission matrix for scheduling effective submissions of the set of data files. For example, 'the following principles can be included in a roundabout method: (1) the total number of data blocks transmitted in any time slot (TS) period should be kept at the minimum energy number; and (2) if it is relative to the principle (丨) There are many parts ^ ^ 'Then, the best solution can be to sum up the data blocks that are to be in any reference time slot period f, and wait (relative to the reference time slot) for the previous time slot. Data block and the data block to be sent (relative to the reference time slot) in the next paragraph, but does have the smallest data block, for example, 'assuming an exemplary system sends two short information

Page 21 571594 V. Description of the invention (18) Each of the data files is divided into six blocks of data. 1 The transmission sequence according to a scheduling moment L can be as follows: TSO = > blkO TS1 = > bl kO, blkl, blk3 TS2-> bl k 0, blk2 TS3 = > b 1 k 0, blkl, blk3, blk4 TS4 = > bl k 0, blk4 TS5 = > blkO, blkl, blk2, blk5 applies this exemplary method of convolution in the manner described above, the possible combinations of handover matrices are as follows: Option 1: Send video file N at offset 0 TS Total data block · TSO = > M0, NO 2-2 TS1 = > M0, Ml, M3, NO, N1, N3 6 TS2 = > M0, M2, NO, N2 4 TS3 two > M0, Ml, M3, M4, NO, Nl, N3, N4 8 TS4 = > M0, M4, NO, N4 4 TS5 = > M0, Ml, M2, M5, NO, Nl, N2, N5 8 Option 2 etc. Send video file at displacement 0 TS N 'Total data block TSO = > M0 , NO, Nl, N3 4 TS1-> M0, Ml, M3, NO, N2 5 «1 TS2 two > M0, M2, NO, N1 , N3, N4 6 TS3 = > M0, Ml, M3, M4, NO, N4 6

Page 22 571594 V. Description of the invention (19)

TS4 = > MO, M4, NO, N1, N2, N5 TS5 = > MO, Ml, M2, M5, NO Option 3 · Send video file N at displacement 2 TS Total data block TS0 2> M0 , NO, N2 3 TS1 II > M0, Ml, M3, NO, Nl, N3, N4 7 TS2 II > M0, M2, NO, N4 4 TS3 = > M0, Ml, M3, M4, NO, Nl , N2, N5 8 TS4 2 > M0, M4, NO 3 TS5 M0, Ml, M2, M5, NO, Nl, N3 7 Option 4: Send video file N at shift 3 TS Total data block TS0 M0, NO , Nl, N3, N4 '5 TS1 two > M0, Ml, M3, NO, N4 5 TS2 = > M0, M2, NO, Nl, N2, N5 6 TS3 two > M0, Ml, M3, M4, NO 5 TS4 = > M0, M4, NO, Nl, N3 5 TS5 = > M0, Ml, M2, M5, NO, N2 6 Option 5 • Send video slot N at shift 4 TS Total data block TS0 = > M0, NO, N4 3 TS1 = > M0, Ml, M3, NO, Nl, N2, N5 7 TS2 = > M0, M2, NO 3 TS3 M0, Ml, M3, M4, NO, Nl, N3 7

Page 571594

V. Description of the invention (20) TS4 = > M0, M4, NO, N2 4 TS5 Second > M0, Ml, M2, M5, NO, Nl, N3, N4 8 Option 6: Send video slot at displacement 5 TS N total lean block TS0 2 > M0, NO, N1, N2, N5 5 TS1 = > M0, Ml, M3, N0 4 TS2 2 > M0, M2, NO, N1, N3 5 TS3 = > M0, Ml, M3, M4, NO, N2 6 TS4 = > M0, M4, NO, N1, N3, N4 6 TS5 = > M0, Ml, M2, M5, NO, N4 6 Application principle (1), Options 2, 4 and 6 will have the smallest maximum number of data blocks (ie 6 data blocks) sent in any time slot. Application principle (2): The optimal delivery matrix in this embodiment is option 4, because this option 4 has any reference time slot data block plus the data area with the smallest number of adjacent time slot data blocks The total number of blocks (ie 16 data blocks). In this way, for the most suitable embodiment, the transmission sequence of the data file N should be shifted by three time slots. In an exemplary embodiment, a three-dimensional submission matrix can be generated for each channel server 104. When the data blocks of each data file have been sent according to the submission matrix, a large number of subscribers can The data file is accessed at random times, and the appropriate data blocks in the data file will be available to each subscriber in a timely manner. In the above example, assuming that the time slot period is equal to 5 seconds, the D0D system 100 will send the data blocks of the data files M and N (that is, Data file N shift submission sequence): Page 24 571594 V. Description of the invention (21)

Time 00 ·· 00: 00 = > MO NO N1 N3 N4 time 00:00:05 2> MO Ml M3 NO N4 time 00:00] 0 = > MO M2 NO N1 N2 N5 time 0 0: 0 0: 1 5 = > M0 Ml M3 M4 NO time 0 0: 0 0: 2 0 = > M0 M4 NO N1 N3

Time 00:00:25 = > MO Ml M2 M5 NO N2 time 00:00:30 2> MO NO N1 N3 N4 time 00: 00:35 = > MO Ml M3 NO N4 time 00:00:40 = > MO M2 NO N1 N2 N5 time 0 0: 0 0: 45 = > MO Ml M3 M4 NO time 00: 00: 50 = > MO M4 NO N1 N3 time 0 0: 0 0: 5 5 2> Μ 0 M1 M 2 M 5 N 0 N 2 If at time 0 0: 0 0: 0 0, a client A selects the movie M, the STB 3 0 0 at the client A will receive, store and return as follows Rejected data block · Time 0 0: 0 0: 00 = > Receive M0 = > Play M0, store M0 time 0 0: 0 0: 05 = > Receive Ml, M3 = > Play Ml, store M0, M1, M3 time 0 0: 0 0: 10 = > Receive M2 II > Play M2, store M0, Ml, M2, M3 time 0 0: 0 0: 15 = > Receive M4 = > Play M3 M4, Store M0, Ml, M2, M3, time 0 0: 0 0: 20 play M4 M4 without receiving, store M0, Ml, M2, M3, time 0 0: 0 0: 25 = > receive M5 = > play M5 , Store M0, Ml, M2, M3,

Page 25 571594 V. Description of the invention (22) M4, M5 If the client M selects the movie M at time 00: 00: 10, the STB 300 at the client B will receive as follows: Save and reject the data block:, Time 00:00:10 = > Receive M0, M2 = > Play M0, save M〇, M2 time 〇〇 ·· 00: 15 = > Receive M1, M3, M4 = > Play M1, store m〇, m1, m2, M3, M4 time 0 〇: 〇〇 ·· 2 〇 = > No reception = > Play M2, store MO, M1, M2, M3, M4 time 0 0 ·· 〇〇: 2 5 = > Receiving M 5 = > Playing M 3, storing M 0, M 1, M 2, ^ M3, M4, M5 ^ 0 0 · · 00 · · 30 = > No reception => Play M4, store M0, M1, M2, M3, M4, M5 'time 00:00:35 => No reception = > Play M5, store M0, M1, M2, M3, M4, Μ5 And if at time 0 0: 0 0: 1 5, a client c selects the movie ν, the STB 300 at the client C will receive, store and reject the data block as follows: 'Time 00: 00:15 = > Receive N0 = > When playing N0 and storing no 0 0: 0 0: 2 0 => Receive N1, N3 = > Play N1, store N0, N1, N3 time 00:00:25 => Receive N2 => Play N2, store N0, N1, N2 N3 time 00:00:30 = > Receive N4 = > Play N3, store Ν〇, Ν1, Ν2 N3 N4 ,, 'Time 00:00:35 = > No reception = > Play N4, store n〇, n1 , N2, N3,

Page 26 571594 V. Description of the invention (23) & N4 time 00:00:40 = > Receive N5 for # μ〉 Play N5, store NO, N1, JV2 J \ f3 N4, N5 '' and if Time 0 0 ·· 0 0: 3 0 See, the STB 3 0 0 at the client D will select the movie N as follows: Block 0, then the material block: X receives, stores and returns the time of capital rejection 00 as follows: 0 0: 3 0 = > Receive 1 ^ 0, call 11 \ ^^,, call, N3, N4 = > Play No, store No, N1, time 0: 0 0: 3 5 => No Receive N3, N4 = > Play N1, store N〇, N1, time 0 0: 0 0: 4 0 = > receive N2, Ν5 time 0 0: 0 〇 · 4 5 = > no receive time 0 0 : 0 0: 5 0 => no reception time 0 0: 0 0: 5 5 = > no reception N3, N4 '= > play N2, store NO, N1, (N2, N3, N4, N5' = > Play N3, store N0, N1, N2, N3, N4, N5 => play N4, store NO, N1, N2, N3, N4, N5 = > play N5, store NO, N1, as shown in the above example Indicate that any N2, N3, N4, N5 can be selected and played in an independent way. Two = can be selected at random moments, according to the material. Any information provided by the service provider operates normally. (Service providers can make mistakes in broadcast values, such as ig% W, recognition 0 :, and delivery: first schedule the bins, and then output I, such as video files) to the channel. , Λ a ^ ^ sentence service line 1 0 4. The central control server 丨 〇 2

Page 27 571594 V. Description of the invention (24) Calculate the three-dimensional delivery matrix (ID, time slot, and data and send it to the channel server 104. In the broadcast, a second, transmission order) and server 1 04 Consult the 3D handover matrix to press ^ ^ the data block served by the channel. Each data file will be divided into ^ sequence ^ out of the appropriate number of subscribers can be at a random moment, in a bundle ’

Sequentially: Reward the information broadcast. The data block of the data file :: JJ ΐ 'In the mpeg data stream of the rate, each data block 二 has two, two, and small points. ”Size (according to MBytes) = BitRates (Mb / s: xTS (sec ) / 8 (1) From the exemplary embodiment, the size of the data block will be adjusted to the next higher multiple of the size of the memory cluster in the local memory 208 of the channel server 104. For example, if the length of the data block calculated according to equation (丨) is 720 Kbytes, and if the cluster size of the local memory 208 is 64 Kbytes, the length of the resulting data block should be 76 8 Kby ^ es. In the present embodiment, the 'data block should be further divided into a number of sub-blocks, each of which has the same size as the cluster size. In the present embodiment, the' data block can be There are 12 sub-blocks of 64 Kbytes. The sub-blocks can be further split into many data packets. Each data packet contains a packet header and packet data. The length of the packet data is based on the CPU of each channel server The maximum transmission order of the physical layer towards which the data is delivered Bit (MTU). In a preferred embodiment, the total size of the packet header and packet data should be lower than the M TU. However, for maximum benefit, the packet data length should be as long as possible.

571594 V. Description of the invention (25) In the exemplary embodiment, the data in the packet header contains information, and the STB 3 0 0 for the subscriber will decode any received data and determine the data. Whether the packet belongs to a related data file (such as agreement signature, version, ID, or packet type information). The packet header may also contain other information, such as block / sub-block / packet number, packet length, cyclic redundancy check (CRC) and shift value within the sub-block, and / or encoding information. Once received by the channel server 104, the data packet will be sent to the QAM modulator 206, where another header will be added to the data packet to generate a QAM-modulated IF output signal. The maximum bit rate output by the Q AM modulator 20 6 is related to the available bandwidth. For example, for a 6 MHz QAM modulator 206, the maximum bit rate is 5.05 (bit / symbol) X 6 (MHz) = 30.3 Mbit / sec ° The QAM-modulated if output signal will be sent upstream A converter i06 for converting to an rf signal suitable for a specific channel (ie, CATV channel 80, 559.250 MHz and 6 MHz bandwidth). For example, if a cable TV network has this bandwidth (or bit rate), each channel can be used to provide more than one data stream, and each data stream occupies a virtual channel. For example, QAM modulation technology can be used to assemble three MPEG 1 data streams into one channel. The output of the up-converter 106 is applied to the combined gain / amplifier 108, whereupon the combined signal is sent to the transmission medium 110. In an exemplary embodiment, the total system bandwidth (BW) used to transmit "N (the required bandwidth may be M = Nxbw 'where bw is the respective data stream, for example, at the same time, By having 3〇 \ 3

Page 571594 V. Description of the Invention (26)

Mbi t / sec system bandwidth of the DOCSIS cable channel to transmit three-"-data streams, because each MPEG-data stream will occupy 9 Mbi ts / sec system bandwidth. Generally speaking ' Regardless of the number of subscribers who actually access the DOD service item, bandwidth will be consumed. In this way, even if the concurrent users use the DOD service item early, they will still consume their bandwidth companion system In an exemplary embodiment, the data block can be used to reduce the total number of channels, ^ ^ each-a certain channel in the material file to continuously transmit prefetch See you in the data area. You can send certain data individually, specifically, and sequentially in a group. In a specific embodiment, you can reduce the pre- and data blocks by sending pre-fetched data blocks. Borrow 丨The total system bandwidth (BW) required for the block. In the first decision, it is used to deliver the number of desired prefetched data blocks transmitted by the remaining data areas.俾 With the prefetch k ', the rest of the data blocks are transmitted according to this adjustment. The shell material block will not be transmitted. In addition, for example, according to the above-mentioned exemplary schedule for starting obstacles, basin ^ is six data blocks, and if the first ~ a twists: the data broadcast is divided into blocks "b 1" both are Pre-fetch data ^ ^ block "b 0" and the second remaining data block (b2-b5), and add a repair such as ° block. 'The scheduling matrix should be based on the remaining TSO = > [None] τ 4 TS1 = > blk3 TS2 two > blk2 TS3 = > blk3, blk4

571594 V. Description of the invention (27) TS4 = > blk4 TS5 II > blk2, blk5

In the above example, if only b 0 is a prefetched data block, the total bandwidth used to send the remaining data blocks (b b b 5) in the data file will be reduced by 37.5% [ie Remove six data blocks from a total of sixteen data blocks]. Then, if the data block "b 1" is also a prefetched data block, the total bandwidth used to send the remaining data blocks (b 2-b 5) in the data file will be reduced by 1 2 again. 5%. In this way, the added bandwidth reduction according to the b 1 prefetch will not be higher than the bO prefetcher. This is because the additional bandwidth reduction will decrease as more data blocks are pre-fetched, and the optimal number of pre-fetched data blocks for each data file can be calculated based on the desired bandwidth reduction. In an exemplary embodiment, the bandwidth saved by pre-fetching data blocks X to y in the data file can be estimated as follows: Σ (1 / η); where η = (χ + 1) to ( y + 1) In addition, as the number of prefetched data blocks increases, its prefetch delay time also increases. The prefetch delay time is based on the size of the data blocks in each data file, the number of prefetched data blocks in each data file, the number of data files sent, and the prefetched bandwidth configured in the dedicated pilot. decided. In an exemplary embodiment, all the pre-fetched data blocks in each data file are transmitted in a dedicated frequency sequence in a sequential, continuous manner, with one data block per time slot. For those skilled in this technology, it should be clear that as the number of prefetched data blocks increases, the prefetch delay time will increase. Therefore, when deciding on the optimal number of prefetched data blocks for each data file, the acceptable prefetch delay time should also be considered.

Page 31 571594 V. Description of the invention (28) For example, 'If the data blocks b and ^ of the data broadcast M and N are prefetchers to be added in the exclusive pilot', these prefetched data can be transmitted as follows Block: MO Ml NO N1 MO Ml N〇N1 ... Let us assume that the cheek width PRF-BW (Mb / s) is pre-fetched for a given configuration, then in an exemplary embodiment, 'can be calculated as follows The prefetch delay time-prefetch delay time-material block size (Mbytes) X number of prefetched data blocks x (number of data files to be transmitted + 1) X8] / PRF-BW (Mb / s). In an exemplary embodiment, the prefetch cycle time (p RF _ τ IME), which is the time required to send a complete prefetched data block to all data files to be transmitted, can be calculated as follows To get: Prefetch cycle time = [data block size (Mbytes) x number of prefetched data blocks x number of data files to be added X 8] / PRF_Bw (Mb / s :). In a specific embodiment, in order to reduce the pre-fetch delay time, the pre-fetched data blocks of the data file are compared with the old data file (the pre-fetched data area is continuously transmitted and the data file has passed a certain amount of time). J ^ f

J ^ :: If false, for the old capital system and N is the new factory's way to transfer these pre-fetched data blocks ... ⑴ 可 妆 卜 夕 J NO N1 NO N1 MO M1 N0 N1 N〇 The STB 300 is in After opening it, the program list in the local memory 3 0 8 will be continuously = M1 ... 3 0 0. It will also be received and updated and stored in the STB. The STB 300 will contain the latest programs. Example on the TV screen. Μ If it is like "" ^ 讯 显 ^ in video eaves information" in movie ID, movie title, plot description (press more, go to _ Tianbei News may include electricity but 50 &), type (such as action

Page 32 571594 V. Description of the invention (29), children), grading (such as R, P G 1 3), cable company statement (price, free trial length), subscription period, movie posters and movie trials. In an exemplary embodiment, the data file information is transmitted through a dedicated channel such as a channel reserved for firmware update, advertising, and / or emergency information. In other exemplary embodiments, the information is transmitted through the physical channel shared by other data streams. In an exemplary embodiment, when the STB 300 does not play the acknowledgement and data file time interval, the STB 300 will be transferred to the dedicated channel and prepared to receive and update pre-fetched data that has not yet been received. Block. In an exemplary embodiment, the test strip includes randomly selected data blocks in a data stream of a data file. In this way, a user selects a data file test strip multiple times and will not see the same test strip. The advantage of the randomly composed test strip method is that the DOD system 100 does not require additional bandwidth to broadcast a predetermined test strip program. Conversely, after the user requests a data file for viewing a test piece, the DOD system 100 may randomly select certain data blocks in the normal data stream of the data file. In a preferred embodiment, certain data blocks cannot be part of any test strip. For example, if a data file provides movie viewing, the second half of the movie cannot be selected as a randomly selected test piece. ’A subscriber can view a list of available profiles that have been categorized and displayed on the TV screen. When the client selects an item from the list of available data files, the STB 300 will control its hardware to tune to the corresponding physical channel and / or virtual subchannel to start receiving the data file. Data packet. The STB 3 0 0 looks at the header of each data packet,

Page 33 571594 V. Description of the invention (30) ^^ ~-The shell material is decoded 'before deciding whether to keep the received data packets. If STB 300 decides that the received data packet should not be retained, it discards the data packet. Otherwise, the data packet is stored in the local memory 308 for subsequent retrieval, or temporarily stored in the buffer 309 until it is sent to the decoder 3 12. In order to improve the execution efficiency by avoiding frequent reading / writing to the local memory 308, in an exemplary embodiment, the local memory 308 uses the Unwilling to be loyal to the knees μ ^ ^ ^, 月 月 "Expected technology to bring the expected data area 疋" within 5 g of the memory buffer. If a broken engine 'appears in the expected sliding window, the data block will be directly transferred from the memory buffer 309, and the decoder will be at 3.12. However, if there is no tap in the expected sliding window, then ΐ ί ί blocks are transferred from the decoder 31 2 to the memory buffer 3009, where == Jn from the local memory 3 0 8 Read the data block into the memory buffer β 3 0 9. $; No, in the specific embodiment, the STB 300 can be through the infrared, ^ control single 70 keys, IR keyboard or include a front panel for pause, slow playback between 1 play and single-frame broadcast, etc. Press the key to respond to the order. In an exemplary embodiment, if a subscription is used, any action (such as scrolling a program list, L., f, or movie, etc.) is automatically played within a predetermined period of time. Scheduled one or two, and when the side subscriber provides a certain action (such as pressing the remote control button 栌-key), the scheduled advertisement can also automatically stop. In another exemplary example, the STB 3 0 0 can be automatically placed into an advertisement when a video program is being played. Service providers (such as cable companies) can set a price

Page 34 571594 V. Description of Invention (31) The strategy shall indicate the frequency with which the advertisement may interrupt the film being shown. In an exemplary embodiment, a cable television company using the DOD system 100 may preset a price list according to the number of advertisement interruptions. In a specific embodiment, the data area of the advertisement is continuously broadcasted through a dedicated channel, and the channel can also broadcast the program table, emergency bits and any firmware update items. The user can select an acceptable balance between the price and the advertisement in the price list. In an exemplary embodiment, the DOD system 100 can implement the user option by maintaining an internal clock, which can be automatically allowed to be set in accordance with the pricing rule selected by the user. Advertising data blocks are inserted at predetermined intervals. If an emergency information bit is found in the header of the data packet, the STB 300 can suspend any data receiving operation and control its hardware to tune to the channel reserved for receiving data file information, and obtain and decode any pending information Emergency information displayed on the output screen. In an exemplary embodiment, when the STB 300 is idle, it will tune to the channel reserved for receiving data file information, and is always ready to receive and display any Urgent information. In a specific embodiment, when the STB 300 is displayed as idle (if a user does not apply the system at the moment), a warning may be issued to inform the user to turn on the output device to view emergency information. In another specific embodiment, when the STB 300 can distinguish emergency information in different regions. For example, urgent information about an unrelated area does not interrupt the data file being played or trigger an alert. In contrast, in existing systems, cable companies must manually interrupt the broadcast of emergency information.

Page 571594

Page 36

Claims (1)

571594 VI. Scope of Patent Application 1. A method for sending data to customers and providing on-demand data services, including the following steps: (a) receiving data files; (b) calibration time interval; (c) according to the time interval Analyze the data file into a plurality of data blocks, (d) select a set of prefetched data blocks from the plurality of data blocks, (e) send the set of prefetched data blocks in a dedicated channel; (f) transmitting the remaining data blocks of the plurality of data blocks within the transmission channel. 2 · The method according to item 1 of the patent application scope, wherein the step of selecting a set of prefetched data blocks includes the following steps: determining the bandwidth reduction and the prefetch delay time; and according to the bandwidth reduction and the prefetch delay Time to select the prefetched data block. 3. —A method for sending data to customers and providing on-demand data services, which includes the following steps: (a) receiving a data file; (b) calibrating a time interval; (c) based on the time interval, the data file Analyze into a plurality of data blocks so that each data block is displayable in the time interval; (d) send the plurality of data blocks in the transmission channel; (e) receive a test strip request, Page 38 571594 VI. Scope of patent application (f) Randomly select a group of data blocks from the plurality of data blocks to form the test strip; and (g) Promote the display of the test strip. 4. The method according to item 3 of the scope of patent application, further comprising the following steps: sending a set of advertisement data blocks in the exclusive channel; and promoting the display of the set of poor data blocks at a predetermined time. 5. The method of claim 4 in the scope of patent application, wherein the step of facilitating display comprises: receiving a user option price according to the frequency of display of the advertisement; and causing the group of advertisement data blocks to be displayed according to the user option. 6. The method of claim 3, further comprising the steps of: checking the packet header of the data file as an emergency bit; and causing an emergency message to be displayed when an emergency bit is detected. 7. The method according to item 6 of the patent application scope, further comprising the following steps: determining whether the emergency information is directed to a relevant area; and if the emergency information is indeed directed to the relevant area, displaying the emergency information 0