US20010028789A1 - Method of and apparatus for data recording, method of and apparatus for data reproduction, recording medium, and method of and apparatus for data transmission - Google Patents

Abstract

Disclosed are a method of and apparatus for data recording, a method of and apparatus for data reproduction, a recording medium, and a method of and apparatus for data transmission in which compression-coded moving picture signals can be continuously reproduced irrespective of the amount of information generated. When recording data obtained by compression-coding a moving picture signal of a fixed period of time on a recording medium, only writing to an FIFO buffer memory is effected and reading to the recording medium is inhibited for a fixed period of time from the output start of the data and, after the fixed period of time has elapsed, the writing of the compression-coded data to the FIFO buffer memory is continued and, at the same time, the data written to the FIFO buffer memory is read out at a data rate which is riot lower than the data rate for the writing to packetize the data, the data thus packetized being recorded on the recording medium.

Description

BACKGROUND OF THE INVENTION
• 1. Field of the Invention[0001]
• The present invention relates to a data recording apparatus, a data reproducing apparatus, etc. in which moving picture signals are compression-coded for recording or reproduction.[0002]
• The present invention further relates to a data transmission apparatus in which moving picture signals are compression-coded for transmission to remote sites.[0003]
• 2. Description of the Related Art[0004]
• Generally speaking, in systems for recording and reproducing digital moving picture signals and systems for transmitting them to remote sites, an image signal compression-coding method is adopted in order to efficiently utilize the capacity of the recording medium, the transmission line, etc. In such a method, it is general practice to perform the compression coding of moving picture signals by utilizing in-frame correlation or inter-frame correlation. Examples of the compression coding formats include MPEG1 and MPEG2. Further, coding methods in conformity with the H. 261 recommendation, etc. are available.[0005]
• In compression coding apparatuses, etc. operating in accordance with such coding formats, coding is to be effected such that the amount of moving picture signal data generated during a fixed period of time is kept constant.[0006]
• However, the amount of data generated by compression coding is usually not constant. Further, the amount of data thus generated differs from frame to frame.[0007]
• If the amount of data generated per unit time is not constant, it is very difficult to directly record the generated data on a recording medium or directly transmit it.[0008]
• This is generally due to the fact that the data recording rate or the data transmission rate is constant.[0009]
• In view of this, a buffer memory is usually provided between the moving picture image compression coding section and the medium recording section or between the moving picture image compression coding section and the transmission section so that the data rate of the variable amount of coded data output to the recording apparatus, the transmission line, etc. may be kept constant. In the moving picture image compression coding section, the amount of coded data generated is adjusted such that the buffer memory will not suffer failure.[0010]
• Thus, the average rate of data amount generated per unit time in the moving picture image compression coding section when performing moving picture image compression coding is the same as the data rate of the coded data output from the buffer memory.[0011]
• The device for performing these operations is generally called an encoder, which includes the buffer memory.[0012]
• In the device for decoding the compression-coded data to the original moving picture signals, a buffer memory similar to the one mentioned above is provided between the recording apparatus or the output end of the transmission line and the expansion decoding section. The device for performing these operations is generally called a decoder, which includes the buffer memory.[0013]
• In this way, in the conventional encoder and decoder, compression coding processing is effected so that the amount of moving picture signal data per unit time may approach a constant value or it may approach a target generation amount of coded data.[0014]
• Here, a case will be considered in which coded data obtained by compression-coding a moving picture signal of a time length T is recorded on a recording medium at a writing rate R, the recorded data then being read out at reading rate R and decoded to thereby reproduce the moving picture signal. In this case, the target generation amount of coded data Ei is given by the following equation:[0015]
• Ei=R×T  (1)
• However, as shown by the following inequality, the actual amount of data generated Er is not the constant value R×T but varies as follows:[0016]
• R×T−B≦Er≦R×T+B  (2)
• where B represents the capacity of the buffer memory provided in the encoder. That is, it is possible for the amount of coded data Er to vary by the capacity B of the buffer memory in the encoder.[0017]
• Assuming that this data is recorded on the recording medium as it is and then read out, the reading time t is as shown by the following inequality:[0018]$\begin{array}{cc}T-\frac{B}{R}\le t\le T+\frac{B}{R}& \left(3\right)\end{array}$

  
• The reproduction time for the moving picture signal which is obtained by decoding the data read out is also T. This shows that image reproduction time and the requisite time for reading out the data is not the same.[0019]
• When, as in the case of an AV server, an arbitrary number of items of data obtained by compression-coding moving picture signals of a fixed period of time (they will be hereinafter referred to as “clips”) are written to a medium, and an arbitrary number of clips are read out from the medium to continuously reproduce the image signals, the following problem is involved:[0020]
• When an arbitrary number of clips obtained by compression-coding moving picture signals of a fixed period of time are written to a medium and an arbitrary number of clip image signals are continuously reproduced therefrom, the requisite time for reading out data differs for each item of data, so that, when the time for reading out data is longer than the time for reproduction, it is impossible to continuously reproduce the actual clip image signals without interruption.[0021]
SUMMARY OF THE INVENTION
• It is an object of the present invention to provide a method of and apparatus for data recording, a method of and apparatus for data reproduction, a recording medium, and a method of and apparatus for data transmission which enable a clip image of an arbitrary length of time to be reproduced without interruption.[0022]
• To achieve the above object, in accordance with the present invention, when recording compression-coded moving picture signals of a fixed period of time or outputting them to the transmission line, the compression-coded data is only written to an FIFO buffer memory for a period of time from the output start of the compression-coded data, and the reading thereof to the recording medium or to the transmission line is inhibited.[0023]
• When a fixed period of time has elapsed from the output start of the compression-coded data, the compression-coded data is written to the FIFO buffer memory, and, at the same time, the compression-coded data which has been written to the FIFO buffer memory is read out at a data rate exceeding that at the time of data writing and packetized. This packetized data is recorded on the recording medium, or output to the transmission line.[0024]
• When packet data is read from a recording medium on which compression-coded data, obtained by compression-coding moving picture signals of a fixed period of time, is recorded after being converted to packet data whose size is constant independently of the amount of data generated, or when it is received from a transmission line, the packet data is read at a data rate not lower than the rate at which it was written to the recording medium or the rate at which it was output to the transmission line and is then written to an FIFO buffer memory. After that, the written compression-coded data is read from the FIFO buffer memory at a rate that is the same at which it was compression-coded and is then decoded.[0025]
• Generally speaking, when compression-coding a moving picture signal, the amount of data generated fluctuates. However, regardless of this fluctuation, the compression-coded data when it is recorded on a recording medium or output to a transmission line is converted to packet data of a fixed size. Thus, it is possible for the moving picture signal to be recorded or transmitted within a fixed period of time regardless of the amount of data generated.[0026]
• Further, it is possible for the moving picture signal to be reproduced or received, whether on the reproduction or reception side, within a fixed period of time regardless of the amount of data generated.[0027]
BRIEF DESCRIPTION OF THE DRAWINGS
• FIG. 1 is a block diagram showing the basic construction of a data recording apparatus and of a data reproducing apparatus according to the present invention;[0028]
• FIG. 2 is a flowchart illustrating a recording operation by the data recording apparatus;[0029]
• FIG. 3 is a flowchart illustrating a reproducing operation by the data reproducing apparatus;[0030]
• FIG. 4 is a schematic diagram showing the relationship between data generation amount and processing time in recording;[0031]
• FIG. 5 is a schematic diagram showing the relationship between data generation amount and processing time in reproduction;[0032]
• FIG. 6 a block diagram showing the basic construction of a data recording apparatus and of a data reproducing apparatus according to another embodiment of the present invention;[0033]
• FIG. 7 is a flowchart illustrating a recording operation by the data recording apparatus;[0034]
• FIG. 8 is a schematic diagram showing the relationship between data generation amount and processing time in recording;[0035]
• FIG. 9 is a schematic diagram showing the relationship between data generation amount and processing time in reproduction;[0036]
• FIG. 10 is a block diagram showing an example of the construction of an AV server system; and[0037]
• FIG. 11 is a block diagram showing another example of the construction of an AV server system.[0038]
DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS OF THE INVENTION
• (1) The Principle on Which Recording and Reproduction are Performed[0039]
• To write an arbitrary number of items of data obtained by compression-coding a moving picture image of a fixed period time, that is, an arbitrary number of clips, to a recording medium and to continuously reproduce an arbitrary number of image signals from this recording medium without interruption, the recording apparatus of this embodiment adopts an arrangement in which, after writing compression-coded data to an FIFO buffer memory for a fixed period of time, the data is read out at a data rate higher than the rate of writing to record packetized data on the recording medium.[0040]
• Further, in the reproducing apparatus of this embodiment, data is extracted from a packet read from a recording medium at a rate that is the same or higher than the recording rate, and this data is written to an FIFO buffer memory having a capacity not smaller than that for the recording. Then, the data is read out at a rate that is the same as the coding rate to effect expansion decoding.[0041]
• (2) Basic System Configuration[0042]
• (2-1) Construction of Recording Apparatus and Reproducing Apparatus[0043]
• FIG. 1 shows a hardware construction example of a[0044]recording apparatus1 for executing this recording method and a reproducingapparatus2 for executing this reading method.
• First, the construction of the[0045]recording apparatus1 will be described. Therecording apparatus1 is composed of anencoder1A, anFIFO buffer memory1B, apacketizing section1C and astuffing section1D.
• The[0046]encoder1A is composed of a compression coding section1A1 and an encoding buffer memory1A2. The compression coding section1A1 effects variable-amount compression coding of a moving picture signal S1 at an average rate of Re, and successively outputs coded data S2 to the encoding buffer memory1A2 having a capacity Be.
• Further, when a fixed amount of coded data S[0047]2 has been accumulated in the memory, the encoding buffer memory1A2 outputs coded data S3 (S2) at the rate Re. Data indicating the buffer remainder Be′ of the encoding buffer memory1A2 is fed back to the compression coding section1A1 and the processing of the compression coding section1A1 is controlled so that the buffer may not suffer failure.
• The[0048]FIFO buffer memory1B is a buffer memory with a capacity Bw provided for data rate conversion. TheFIFO buffer memory1B accumulates the coded data S3 written at the rate Re for a predetermined period of time Tw. After that, it outputs the data to thepacketizing section1C at a reading rate Rw higher than the writing rate Re. Here, the accumulation time Tw, the reading rate Rw and the buffer capacity Bw are set as follows:
• Tw≧(Re×T+Be)/Re−T=Be/Re  (4)
• Rw≧(Re×T+Be)/T=Re+Be/T  (5)
• In the[0049]packetizing section1C, coded data S4 time-compressed by theFIFO buffer memory1B is successively input and stored in one or a plurality packets before being output. In this process, thepacketizing section1C writes the requisite information, such as the packet size, the kind of data, and time information, to a packet header, and records it on arecording medium3 at the fixed rate Rw. Strictly speaking, the rate at which the recording on the recording medium is effected must be somewhat higher than Rw because of the increase in data amount due to the packet header. However, this is negligible since it is a small amount as compared with the coded data.
• As a result, packet data S[0050]5 can be written without fail to therecording medium3 within a period of time corresponding to the clip reproduction time T regardless of the amount of information generated.
• Due to the relation ship: Rw>Re, when the[0051]FIFO buffer memory1B runs out of data during the packet processing, thepacketizing section1C inserts meaningless data that is not directly related to the data supplied from thestuffing section1D (i.e., stuffing data) in the packet for the purpose of filling, whereby failure in data transmission form is prevented. Theencoder1A, thepacketizing section1C, theFIFO buffer1B and thestuffing section1D are controlled by acontrol section1F.
• Next, the construction of the reproducing[0052]apparatus2 will be described. The reproducingapparatus2 is composed of adata extracting section2A, anFIFO buffer memory2B and adecoder2C.
• The reproducing[0053]apparatus2 reads out recording data recorded on therecording medium3 at a fixed rate Rw, and reads it in thedata extracting section2A as reproduction data S6.
• The[0054]data extracting section2A extracts coded data S7 corresponding to the moving picture signal from the packet on the basis of header information, etc., and transmits it to theFIFO buffer memory2B for reading out. In this process, when stuffing data is extracted from the packet, thedata extracting section2A scraps it so that it may not be decoded.
• Assuming that the reading rate at which the[0055]data extracting section2A reads the recording data S6 from therecording medium3 is the same as the writing rate, i.e., Rw, the requisite time for reading Td can always be kept within the clip display time as shown by the following formula:
• Td=(Re×T+Be)/Rw≦T  (7)
• In this way, the coded data S[0056]7 extracted by thedata extracting section2A is written to theFIFO buffer memory2B for reading, and the data rate is restored to the original value. The capacity Br of theFIFO buffer memory2B satisfies the following condition:
• Generally speaking, the capacity of the decoding buffer memory[0057]2C1 contained in thedecoder2C is not smaller than that of the encoding buffer memory1A2, so that, in this embodiment, the capacity Br of theFIFO buffer memory2B is the same or larger than the capacity Bw of theFIFO buffer memory1B for writing, as shown by the following inequality:
• Br≧Bw  (8)
• Finally, the[0058]decoder2C will be described. Thedecoder2C is composed of a decoding buffer memory2C1 and an expansion decoding section2C2.
• The decoding buffer memory[0059]2C1 reads coded data S8 from anFIFO buffer memory2B at a fixed rate Re and supplies it to the expansion decoding section2C2. The expansion decoding section2C2 decodes the coded data S8 at the average decoding rate Re, and outputs the decoding result as a moving picture signal S10.
• The extracting[0060]section2A, theFIFO buffer2B and thedecoder2C are controlled by acontrol section2F.
• (2-2) Procedures for Recording and Reproduction[0061]
• Recording and reproducing operations executed in the[0062]recording apparatus1 and the reproducingapparatus2 of the above construction at the time of recording and reproduction will be described one by one.
• First, the recording operation when a plurality of clips are to be continuously reproduced without interruption, that is, the control operation of the[0063]control section1F, will be described with reference to FIG. 2.
• When the recording operation is started, the procedure for the[0064]recording apparatus1 advances fromstep SP1 to stepSP2, and designates the position for the recording of the moving picture signal on therecording medium3 by means of a pointer. This pointer is used as the reading pointer for reading out the recording data of the moving picture signal at the time of reproduction.
• When the designation with the writing pointer has been completed, the procedure for the[0065]recording apparatus1 advances to stepSP3, in which the moving picture signal S1 is input to theencoder1A, and variable amount encoding is started.
• When the encoding is started, the[0066]recording apparatus1 encodes the moving picture signal S1 by theencoder1A at the average rate Re and, as shown instep SP4, writes the coded data S3 obtained by the encoding from the encoding buffer memory1A2 to theFIFO buffer memory1B.
• As shown in the[0067]next step SP5, therecording apparatus1 checks the period of time elapsed from the writing start time and makes a judgment as to whether the period of time elapsed is larger than the accumulation time Tw as given by formula (4).
• When the result of the judgment is in the affirmative, the reading of the coded data S[0068]4 from theFIFO buffer memory1B to thepacketizing section1C is started in therecording apparatus1, as shown instep SP6. The data rate Rw in this reading is, as described above, a value larger than the data rate Re when the data is written to theFIFO buffer memory1B.
• After this, as shown in[0069]steps SP7 throughSP9, therecording apparatus1 packetizes the coded data S4 in thepacketizing section1C and records it on therecording medium3 as packet data S5 until it is made certain that the buffer remainder B′w of the FIFO buffer memory has become empty result ofstep SP9 is in the affirmative and it is made sure that theFIFO buffer memory1B has become empty, the procedure for therecording apparatus1 advances to stepSP10, in which a judgment is made as to whether the writing to theFIFO buffer memory1B by theencoder1A has been completed.
• When the result of the judgment is in the negative, it is determined that the[0070]FIFO buffer memory1B has become temporarily empty due to the fact that the rate Rw at which reading from the FIFO buffer memory is effected is higher than the writing rate Re and, in steps SP11-SP7-SP8-SP9, operation is effected such that the shortage in packet data is compensated for by filling with stuffing data. This processing routine is continued until the reading from theFIFO buffer memory1B is started again.
• When the reading from the[0071]FIFO buffer memory1B is started again, the processing routines of steps SP9-SP7-SP8-SP9 are repeatedly executed.
• When the result obtained in[0072]steps SP9 andSP10 is in the affirmative, the completion of the recording of a series of moving picture signals S1 is confirmed instep SP12, and in thenext step SP13, a judgment is made as to whether the moving picture signal S1 of the next clip is to be recorded or not.
• As long as the result of the judgment is in the affirmative, the[0073]recording apparatus1 repeatedly executes the procedures ofsteps SP2 throughSP13. When the result obtained is in the negative, the recording of a plurality of clips is completed.
• Next, the reproducing operation effected when reproducing the thus recorded recording data from the[0074]recording medium3, that is, the control operation of thecontrol section25, will be described with reference to FIG. 3.
• When the reproducing operation is started, the procedure for the reproducing[0075]apparatus2 advances fromstep SP21 to stepSP22 and the position on therecording medium3 for the reading of the recording data is designated with the reading pointer.
• When the designation with the reading pointer has been completed, the reading of data from the[0076]recording medium3 is started as shown instep SP23.
• When reproduction data S[0077]6 has been read to thedata extracting section2A, the reproducingapparatus2 extracts, as shown instep SP24, stuffing data from the reproduction data S6 read out at the fixed rate Rw, and, instep SP25, operates such that only the portion of the coded data S7 corresponding to the moving picture signal is supplied to theFIFO buffer memory2B.
• In the[0078]next step SP26, the reproducingapparatus2 supplies coded data S8 rate-converted by theFIFO buffer memory2B to thedecoder2C, and the reproduction of the moving picture signal S10 is started again by thedecoder2C.
• Then, in the next step SP[0079]28, the reproducingapparatus2 makes a judgment as to whether the data reading has been entirely completed or not. As long as the result of the judgment is in the negative, the procedures of the steps: SP29-SP24-SP25-SP26-SP27-SP28 are repeated.
• When data reading corresponding to one clip has been completed by repeating this processing loop, the procedure for the reproducing[0080]apparatus2 advances from step SP28 to SP29, and a judgment is made as to whether there is a next clip to be continuously reproduced.
• As long as the result of the judgment is in the affirmative, the reproducing[0081]apparatus2 repeats the procedures ofsteps SP22 through SP28. When the result of the judgment obtained is in the negative, the procedure advances to step SP30.
• When, in step SP[0082]30, it is made sure that the buffer remainder of the decoding buffer memory2C1 has been reduced to zero, the procedure advances to step SP31, in which the reproducing operation is completed.
• FIGS. 4 and 5 show the relationship between data amount and processing time in an example of recording/reproducing operation in which two clips,[0083]clip1 andclip2, are continuously recorded.
• It will be assumed that the image periods of the[0084]clip1 andclip2, which are to be recorded continuously, are T and T′, as shown in FIG. 4(A). Generally speaking, it is possible for the amount of coded data generated in theencoder1A to fluctuate by an amount corresponding to the capacity Be of the encoding buffer memory1A2 of theencoder1A. For example, assuming that the amount of coded data is large, it takes T+ΔT and T′+ΔT′ to read the coded data S3 ofclip1 andclip2 from theFIFO buffer memory1B, as shown in FIG. 4(B).
• If this coded data S[0085]3 were recorded as it is on therecording medium3 as in the prior art, it would take T+ΔT and T′+ΔT′ to reproduceclip1 andclip2 from therecording medium3, which would be longer than the display time. Thus, in conventional reproducing apparatus, there is a fear of a blank time ΔT being generated if two clips,clip1 andclip2, are continuously reproduced
• In the[0086]recording apparatus1 of this embodiment, in contrast, data is read from theFIFO buffer memory1B at a data rate higher than the average encoding rate Re and in a condition satisfying formula (5), so that the dataportions including clip1 andclip2 can be completely written to therecording medium3 within the image periods T and T′ regardless of the buffer remainder Be′. This is shown in FIG. 4(C). The shaded portions in the drawing represent the stuffing data filling periods.
• At the time of completion of the recording on the recording medium, data is recorded on the[0087]recording medium3 in a data construction as shown in FIG. 4(C).
• In reproduction, the reproducing[0088]apparatus2 can read the recording data of the moving picture signals corresponding to clip1 andclip2 within the periods T and T′, respectively, as shown in FIG. 5(A).
• Furthermore, stuffing data is extracted from the reproduction data S[0089]6 read out, so that, as shown in FIG. 5(B), only the data really corresponding to the moving picture signals are read to theFIFO buffer memory2B.
• After this, coded data S[0090]8 is read out, as shown in FIG. 5(C), from theFIFO buffer memory2B to thedecoder2C at the original data rate Re and decoded. At this time, in FIG. 5(C), the apparent data reading periods are longer than the display periods ofclip1 andclip2 by an amount corresponding to the buffer remainder. However, since the decoding operation for each item of coded data is completed in periods T and T′, respectively, it is possible to continuously reproduce two clips,clip1 andclip2, as shown in FIG. 5(D).
• In the above-described construction, it is possible to read moving picture signals of two clips to be continuously reproduced from the[0091]recording medium3 within the respective display periods T and T′, whereby delay in clip display due to delay in decoding operation can be avoided. This makes it possible to realize arecording apparatus1 and a reproducingapparatus2 which can continuously reproduce a plurality of clips without involving any blank periods. Further, it is also possible to realize a recording medium which allows continuous reproduction.
• (3) Another Embodiment[0092]
• (3-1) Construction of Recording Apparatus and Reproducing Apparatus[0093]
• In a recording apparatus according to another embodiment, packetized data is written to an FIFO buffer memory for a fixed period of time after the compression coding, and then read out at a data rate not lower than the writing data rate and recorded on a recording medium.[0094]
• FIG. 6 shows a hardware construction example of a[0095]recording apparatus1′ for executing the recording method and a reproducingapparatus2′ for executing the reading method. In the following description, the components which are the same as those in the above-described embodiment will be indicated by the same reference numerals.
• First, the construction of the[0096]recording apparatus1′ will be described. Therecording apparatus1′ is composed of anencoder1A, apacketizing section1G, anFIFO buffer memory1H, and a padding packet processing section1I.
• The[0097]encoder1A is composed of a compression coding section1A1 and an encoding buffer memory1A2. The compression coding section1A1 effects variable-amount compression coding of a moving picture signal S1 at an average rate of Re, and successively outputs coded data S2 to the encoding buffer memory1A2 having a capacity Be.
• Further, when a fixed amount of coded data S[0098]2 has been accumulated in the memory, the encoding buffer memory1A2 outputs coded data S2 at the rate Re. The buffer remainder Be′ of the encoding buffer memory1A2 is fed back to the compression coding section1A1 so that the buffer may not suffer failure.
• The[0099]packetizing section1G successively inputs items of coded data S3 output from theencoder1A and stores them in one or a plurality of packets before outputting them. In this process, thepacketizing section1G writes the requisite information, such as the packet size, the kind of data and the time information, to the packet header, and outputs it to theFIFO buffer memory1H at a fixed rate Re which is the same as the encoding rate.
• The[0100]FIFO buffer memory1H is a buffer memory with a capacity Bw provided for data rate conversion. TheFIFO buffer memory1H accumulates the coded data S4′ written at the rate Re for a predetermined period of time Tw. After that, it reads the data at a reading rate Rw higher than the writing rate Re and records it on therecording medium3.
• As a result, packet data S[0101]5′ can be written without fail to therecording medium3 within a period of time corresponding to the clip reproduction time T regardless of the amount of information generated at the time of encoding.
• Here, the accumulation time Tw, the reading rate Rw and the buffer capacity Bw of the[0102]FIFO buffer memory1H at the time of this reading operation are calculated as follows:
• Tw≧(Re×T+Be)/Re×T=Be/Re  (9)
• Rw≧(Re×T+Be)/T=Re+Be/T  (10)
• Bw≧Tw×Re+total number of bytes of packet header  (11)
• Due to the relation ship: Rw>Re, it is possible for the[0103]FIFO buffer memory1H to run out of data during the data writing to therecording medium3. In view of this, the padding packet processing section1I is provided.
• The padding packet processing section[0104]1I or thecontrol section1F constantly detects the buffer remainder B′w of theFIFO buffer memory1H. When the buffer remainder B′w has become zero, the padding packet processing section1I outputs a meaningless packet that is not directly related to the data (i.e., padding packet) for the purpose of filling the blank period, whereby failure in data transmission form is prevented. Theencoder1A, thepacketizing section1G, theFIFO buffer1H, and the padding packet processing section1I are controlled by thecontrol section1F.
• Next, the construction of the reproducing[0105]apparatus2′ will be described. The reproducingapparatus2′ is composed of adata extracting section2A, anFIFO buffer memory2B and adecoder2C.
• The reproducing[0106]apparatus2′ reads out recording data recorded on therecording medium3 at a fixed rate Rw, and reads it to thedata extracting section2A as reproduction data S′6.
• The[0107]data extracting section2A removes the padding packet from the reproduction data S′6 read out and scraps it, and extracts only the encoding data S′7 corresponding to the moving picture signal and transmits it to theFIFO buffer memory2B for reading.
• Assuming that the reading rate at which the[0108]data extracting section2A reads the recording data S′6 from therecording medium3 is the same as the writing rate, i.e., Rw, the requisite time for reading Td can always be kept within the clip display time as shown by the following formula:
• Td=(Re×T+Be)/Rw≦T  (12)
• In this way, the coded data S′[0109]7 extracted by thedata extracting section2A is written to theFIFO buffer memory2B for reading, and the data rate is restored to the original value. The capacity Br of theFIFO buffer memory2B satisfies the following condition:
• Generally speaking, the capacity of the decoding buffer memory[0110]2C1 contained in thedecoder2C is not smaller than that of the encoding buffer memory1A2, so that, in this embodiment, the capacity Br of theFIFO buffer memory2B is the same or larger than the capacity Bw of theFIFO buffer memory1H for writing, as shown by the following inequality:
• Br≧Bw  (13)
• (3-2) Procedures for Recording and Reproduction[0111]
• Recording and reproducing operations executed in the[0112]recording apparatus1′ and the reproducingapparatus2′ of the above construction at the time of recording and reproduction will be described one by one.
• First, the recording operation when a plurality of clips are to be continuously reproduced without interruption, that is, the control operation of the[0113]control section1F, will be described with reference to FIG. 7.
• When the recording operation is started, the procedure for the[0114]recording apparatus1′ advances from step SP41 to step SP42, and designates the position for the recording of the moving picture signal on therecording medium3 by means of a writing pointer. This pointer also is used as the reading pointer for reading out the recording data of the moving picture signal at the time of reproduction.
• When the designation with the writing pointer has been completed, the procedure for the[0115]recording apparatus11 advances to step SP43, in which the moving picture signal S1 is input to theencoder1A, and variable amount encoding is started.
• When the encoding is started, the[0116]recording apparatus1′ encodes the moving picture signal S1 by theencoder1A at the average rate Re and, as shown in step SP44, successively packetizes the coded data S3 obtained by the encoding. At the same time, therecording apparatus1′ starts to write the packetized packet data S4 to theFIFO buffer memory1H.
• As shown in the next step SP[0117]45, therecording apparatus1′ checks the period of time elapsed from the writing start time and makes a judgment as to whether the period of time elapsed is larger than the accumulation time Tw as given by formula (4).
• When the result of the judgment is in the affirmative, the[0118]recording apparatus1′ starts the reading of the packet data S′5 from theFIFO buffer memory1H, as shown in step SP46. The data rate Rw in this reading is, as described above, a value larger than the data rate Re when the data is written to theFIFO buffer memory1H.
• After this, as shown in step SP[0119]47, therecording apparatus1′ returns from step SP47 to step SP46 until it is made sure that the remainder B′ of theFIFO buffer memory1H has become zero, and continues to write the packet data S′5 read from theFIFO buffer memory1H to therecording medium3.
• When the result of step SP[0120]47 is in the affirmative and it is made sure that theFIFO buffer memory1H has become empty, the procedure for therecording apparatus1′ advances to step SP48, in which a judgment is made as to whether the writing of the packet data S4 to theFIFO buffer memory1H has been completed or not.
• When the result of the judgment is in the negative, it is determined in the[0121]recording apparatus1′ that theFIFO buffer memory1H has become temporarily empty due to the fact that the reading rate Rw is higher than the writing rate Re and the procedure advances to step SP48, where the shortage in packet data is compensated for by filling with padding packet.
• After this, the procedure for the[0122]recording apparatus1′ returns to step SP47, in which a judgment is made again as to whether the buffer remainder B′ is zero or not. If it is not zero, the writing to therecording medium3 is started again. When it is zero, a judgment is made in step SP48 as to whether the writing to theFIFO buffer memory1H has been completed or not.
• When the result obtained in step SP[0123]48 is in the affirmative, the completion of the recording of a series of moving picture signals S1 is confirmed in step SP50, and in the next step SP51, a judgment is made as to whether the moving picture signal S1 of the next clip is to be recorded or not.
• As long as the result of the judgment is in the affirmative, the[0124]recording apparatus1′ repeatedly executes the procedures of steps SP42 through SP51. When the result obtained is in the negative, the recording of a plurality of clips is completed.
• The reproducing operation effected when reproducing the thus recorded recording data from the[0125]recording medium3 is basically the same as that in the above-described embodiment. In this embodiment, the padding packet is removed from the reproduction data S′6 instep SP24.
• FIGS. 8 and 9 show the relationship between data amount and processing time in an example of recording/reproducing operation in which two clips,[0126]clip1 andclip2, are continuously recorded.
• It will be assumed that the image periods of the[0127]clip1 andclip2, which are to be recorded continuously, are T and T′, as shown in FIG. 8(A). Generally speaking, it is possible for the amount of coded data generated in theencoder1A to fluctuate by an amount corresponding to the capacity Be of the encoding buffer memory1A2 of theencoder1A. For example, assuming that the amount of coded data is large, it takes T+ΔT and T′+ΔT′ to read the coded data S3 ofclip1 andclip2 from theFIFO buffer memory1H, as shown in FIG. 8(B).
• If this coded data S[0128]3 were recorded as it is on therecording medium3 as in the prior art, it would take T+ΔT and T′+ΔT′ to reproduceclip1 andclip2 from therecording medium3, which would be longer than the display time. Thus, in conventional reproducing apparatus, there is a fear of a blank time ΔT being generated if two clips,clip1 andclip2, are continuously reproduced.
• In the[0129]recording apparatus1′ of this embodiment, in contrast, data is read from theFIFO buffer memory1H at a data rate higher than the average encoding rate Re and in a condition satisfying formula (5), so that the dataportions including clip1 andclip2 can be completely written to therecording medium3 within the image periods T and T′ regardless of the buffer remainder Be′. This is shown in FIG. 8(C). The shaded portions in the drawing represent the padding packet filling periods.
• At the time of completion of the recording on the recording medium, data is recorded on the[0130]recording medium3 in a data construction as shown in FIG. 8(C).
• In reproduction, the reproducing[0131]apparatus2′ can read the recording data of the moving picture signals corresponding to clip1 andclip2 within the periods T and T′, respectively, as shown in FIG. 9(A).
• Furthermore, data unrelated to the coded data, such as padding packet, is extracted from the reproduction data S[0132]6 read out, so that, as shown in FIG. 9(B), only the data really corresponding to the moving picture signals is read to theFIFO buffer memory2B.
• After this, coded data S[0133]8 is read out, as shown in FIG. 9(C), from theFIFO buffer memory2B to thedecoder2C at the original data rate Re and decoded. At this time, in FIG. 9(C), the apparent data reading periods are longer than the display periods ofclip1 andclip2 by an amount corresponding to the buffer remainder. However, since the decoding operation for each item of coded data is completed in periods T and T′, respectively, it is possible to continuously reproduce two clips,clip1 andclip2, as shown in FIG. 9(D).
• (4) Application System[0134]
• Next, a system construction example to which this basic system is applied will be described. Here, the[0135]recording apparatus1 will be described with reference to an integral-type system in which therecording medium3 and the reproducingapparatus2 are connected through a communication channel and to with reference to a separation type system in which there is a connection through cable between therecording apparatus1 and therecording medium3 or between therecording medium3 and the reproducingapparatus2.
• (4-1) Integral-Type System[0136]
• Here, as an example of the integral-type system, a server system in which AV signals (audio and video signals) are distributed to a number of terminal devices through a communication circuit (hereinafter referred to as an “AV server system”) will be described. FIG. 10 shows the overall construction of an[0137]AV server system11.
• This[0138]AV server system11 is composed of adata supply source12, anencoding section13 and aserver section14. The system is used to distribute AV signals from theserver section14 to a number of terminals15A1 through15AN. Each section is constructed as described below.
• The[0139]data supply source12 consists of reproducingapparatuses12A, such as video tape recorders (VTR) or magneto-optical disks. Each reproducingapparatus12A supplies AV data signals SD1, SD2 . . . to theencoding section13 through a transmission line or a signal line. In this example, a plurality ofdata supply sources12 are provided.
• The[0140]encoding section13 is a section corresponding to therecording apparatus1 of the basic system, and is composed of anencoder13A, an FIFO-buffer/packet-processing section13B and acontrol section13C.
• Of these, the[0141]encoder13A encodes the AV data signals SD1, SD2 . . . input from thedata supply sources12 into signals of a data format in conformity with theMPEG 2 standards, converting them to data streams D1, D2, . . . .
• The FIFO-buffer/packet-[0142]processing section13B stores these data streams D1, D2 . . . temporarily in the FIFO buffer memory for rate conversion, and the rate-converted coded data is converted to packet data Dw in the packet processing section.
• The[0143]control section13C controls thedata supply source12, theencoder13A and theserver section14 by a control signal S_CTL, and manages the operating condition of each section.
• The[0144]server14 is composed of a recording/reproducingsection14A and adecoding section14B. The recording/reproducingsection14A records coded data obtained by further compressing compression-coded data supplied from theencoding section13 in time-axis direction on the recording medium, and, when there is an instruction from the user, reads corresponding coded data to output it to thedecoding section14B. The instruction from the user is input as a request signal S_request.
• The recording/reproducing[0145]section14A is composed of a media control section14A1 and a plurality of media units14A2 through14A8. Here, the media control section14A1 is a unit for controlling the recording/reproducing operations of the plurality of media units14A2 through14A8. Further, each of plurality of media units14A2 through14A8 contains a plurality of hard disks so that they can record and reproduce a plurality of signals simultaneously and in parallel, whereby multi-access from a number of terminals15A1 through15AN can be realized. Further, coded data supplied from theencoding section13 are recorded on these hard disks as clips.
• The[0146]decoding section14B is composed of a data-extracting-section/FIFO-buffer-memory14B0 and a plurality of decoders14B1 through14BM. Here, the number of the decoders14B1 through14BM provided in thedecoding section14 is determined by the number of terminals15A1 through15AN finally connected and the number of items of coded data that can be simultaneously reproduced by the recording/reproducingsection14A.
• The[0147]decoding section14B decodes coded data read out by the decoders14B1 through14BM through the media units14A2 through14A8 and outputs it to the terminals15A1 through15AN.
• It is not always necessary for this[0148]AV server system11 to be provided in one room; it also allows connection through LAN (local area network) or the like.
• (4-2) Separation-Type System[0149]
• Next, as an example of the separation-type system, an AV server system, which distributes AV signals (audio and video signals) to a number of terminal devices through physical delivery of a recording medium, will be described. FIG. 11 shows the construction of an[0150]AV server system21.
• This[0151]AV server system21 is a system composed of aproduction station22, akey station23, anartificial satellite24, acable station25, abranch station26, ahome27, etc., which are geographically spaced apart from each other.
• In this example, the[0152]recording apparatus1 described with reference to the basic system corresponds to anencoder22A provided in theproduction station22, and the reproducingapparatus2 corresponds to adecoder26A provided in thebranch station26.
• The[0153]production station22 is a producing company or a producing office which edits image source to produce commercials (CM), programs, etc. The moving picture of a program or a CM is compression-coded here and recorded on therecording medium3. Therecording medium3 is physically delivered to thekey station23, thecable station25, etc.
• The[0154]key station23 is used when image data reproduced from therecording medium3 is transmitted to theartificial satellite24 in the form of compression-coded data and transferred to thebranch station26, which is at a remote site. This data transfer may be regarded as a radio transfer between therecording medium3 and the reproducingapparatus2 in the basic system.
• The[0155]cable station25 is used when image data reproduced from therecording medium3 is transmitted in the form of compression-coded data to thebranch station26, which is at a remote site, through a cable. Apart from this, data transmission between thecable station25 and thebranch station26 can also be effected through physical delivery of therecording medium3.
• The[0156]branch station26 is provided with adecoder26A and afile server26B and is used when decoding received coded data at the original data rate and transmitting it to thehome27 as a moving picture signal.
• The basic system allows such an application, too. Further, while in this example the[0157]encoder22 is provided in theproduction station22, it is also possible for the encoder to be provided in thekey station23.
• Further, while in this example the[0158]decoder26A is provided in thebranch station26, it is also possible to realize an example in which the decoder is provided in eachhome27.
• (5) Still Another Embodiment[0159]
• While in the above-described embodiment the[0160]encoder1A, theFIFO buffer memory1B, thepacketizing section1C and thestuffing section1D (or the padding packet processing section1I) are united as therecording apparatus1, the present invention is not restricted to this construction. It is also possible for these components to be contained in separate casings.
• Similarly, while in the above-described embodiment the[0161]data extracting section2A, theFIFO buffer memory2B, and thedecoder2C are united as the reproducingapparatus2, the present invention is not restricted to this construction. It is also possible for these components to be contained in separate casings.
• Further, while in the above-described embodiment a hard disk, a video tape or the like is used as the[0162]recording medium3, this should not be construed restrictively. The present invention is also applicable to other recording media. For example, it is also applicable to an optical disk or a so-called optical card, such as a video disk, a magneto-optical disk or a phase change type disk. Further, it is also applicable to a magnetic recording medium, such as a floppy disk.
• Further, while in the above-described embodiment the reading of data from the[0163]recording medium3 is effected at the rate Rw which is the same as the rate at which the writing of data to the recording medium is effected, the reading rate may be higher than this.
• When compression-coding a moving picture signal, the amount of data generated usually fluctuates. However, as described above, in accordance with the present invention, conversion of data to packet data of a fixed size has been effected when the data is recorded on a recording medium or when it is output to a transmission line, so that, in spite of this fluctuation, it is possible to complete the recording or transmission of a moving picture signal within a fixed period of time regardless of the amount of data generated, whereby it is always possible, in reproduction or reception, to reproduce or receive a moving picture signal within a fixed period of time regardless of the amount of data generated, making it possible to continuously reproduce expansion-decoded moving picture signals.[0164]

Claims (30)

What is claimed is:

1. A data recording method in which a moving picture signal of a fixed period of time T is compression-coded and recorded on a recording medium, comprising the steps of:

writing the compression-coded data to a first buffer memory and inhibiting recording the same on the recording medium for a fixed period of time Tw from the input start of compression-coded data;

after the fixed period of time Tw has elapsed, continuing the writing of the compression-coded data to the first buffer memory and reading out the compression-coded data written to the first buffer memory at a data rate Rw which is not lower than the data rate Re of the writing to the first buffer memory;

packetizing the read compression-coded data of a fixed size; and,

recording the packetized data on the recording medium.

2. A data recording method according to

claim 1

, wherein the fixed period of time Tw is set to be a value (Be/Re) which is not smaller than the value of the minimum requisite time for reading an amount of data corresponding to the total capacity (Be) of a second buffer memory, which is provided in an encoding section for encoding the moving picture signal into compression-coded data and outputting it, from this second buffer memory.

3. A data recording method according to

claim 1

, wherein the data rate at which the compression-coded data is read from the first buffer memory after the fixed period of time Tw has elapsed is a data rate sufficient for making it possible for compression-coded data corresponding to the moving picture signal of the fixed period of time T and data corresponding to the total capacity Be of a second buffer memory, which is provided in an encoding section for encoding the moving picture signal into compression-coded data and outputting it, to be read within the fixed period of time T.

4. A data recording method according to

claim 1

, wherein, when the first memory has become empty during the reading of compression-coded data from the first memory, meaningless stuffing data is added to thereby generate the packet data of a fixed size.

5. A data recording apparatus comprising:

an encoding section for compression-coding a moving picture signal of a fixed period of time T and outputting it;

a first buffer memory to which compression-coded data is input from the encoding section and which, for a fixed period of time Tw from the input start of the compression-coded data, only allows the writing of the compression-coded data and inhibits reading and, after the fixed period of time Tw has elapsed, reads the compression-coded data that has been written at a data rate which is not lower than the data rate for the writing while continuing the writing of the compression-coded data; and

a packetizing section for packetizing the compression-coded data read from the first buffer memory and recording it on a recording medium always as packet data of a fixed size.

6. A data recording apparatus according to

claim 5

, wherein the fixed period of time Tw is set to be a value which is not smaller than the value of the minimum requisite time for reading an amount of data corresponding to the total capacity Be of a second buffer memory, which is provided in the encoding section, from this second buffer memory.

7. A data recording apparatus according to

claim 5

, wherein the data rate at which the compression-coded data is read from the first buffer memory after the fixed period of time Tw has elapsed is a data rate sufficient for making it possible for compression-coded data corresponding to the moving picture signal of the fixed period of time T and data corresponding to the total capacity of a second buffer memory, which is provided in the encoding section, within the fixed period of time T.

8. A data recording apparatus according to

claim 5

, further comprising means for generating stuffing data, wherein, when the first memory has become empty during the reading of compression-coded data from the first memory, the packetizing section adds the stuffing data to generate the packet data

9. A data reproducing method in which compression-coded data obtained by compression-coding a moving picture signal of a fixed period of time T is accumulated in a first buffer memory for a fixed period of time Tw from the input start and, after that, while the writing to the first buffer memory is being continued, data is read at a rate higher than the writing rate to packetize the data to thereby obtain packet data, which is recorded on a recording medium and reproduced therefrom, the reproducing method comprising the steps of:

reading out the packet data at a data rate not lower than the data rate at which the packet data was written to the recording medium;

writing the read out packet data to a second buffer memory;

reading out the data written to the second buffer memory at a data rate which is substantially the same as the data rate of writing the compression-coded data to the first buffer memory; and

decoding the data.

10. A data reproducing apparatus comprising:

a data extracting section in which compression-coded data obtained by compression-coding a moving picture signal of a fixed period of time T is accumulated in a first buffer memory for a fixed period of time Tw from the output start and, after that, while the writing to the first buffer memory is being continued, data is read at a rate higher than the writing rate to packetize the data to thereby obtain packet data, which is recorded on a recording medium and read therefrom, the compression-coded data being extracted from this packet data;

a second buffer memory to which the compression-coded data is sequentially input and written from the data extracting section, wherein, after the writing, the compression-coded data is read out at a data rate which is lower than the rate at which the compression-coded data is read from the data extracting section and which is substantially the same as the data rate at which the compression coding is effected; and

a decoding section for decoding the compression-coded data read from the second buffer memory to obtain the moving picture signal.

11. A recording medium on which compression-coded data obtained by compression-coding a moving picture signal of a fixed period of time T is recorded as packet data of a fixed size regardless of the amount of compression-coded data generated,

wherein said packet data is of the following two kinds:

first packet data consisting of header data and the compression-coded data; and

second packet data consisting of header data, the compression-coded data and dummy data.

12. A data transmission method in which data obtained by compression-coding a moving picture signal of a fixed period of time T is transmitted through a transmission line, comprising the steps of:

writing the compression-coded data to a first buffer memory and inhibiting data output to the transmission line for a fixed period of time Tw from the input start of the compression-coded data;

after the fixed period of time Tw has elapsed, continuing the writing of the compression-coded data to the first buffer memory and reading out the compression-coded data that has been written to the first buffer memory at a data rate Rw which is not lower than the data writing rate;

packetizing the read compression-coded data of a fixed size; and

outputting the packetized data to the transmission line.

13. A data transmission apparatus comprising:

an encoding section for compression-coding a moving picture signal of a fixed period of time T and outputting it;

a first buffer memory to which compression-coded data is input from the encoding section and which, for a fixed period of time Tw from the input start of the compression-coded data, only allows the writing of the compression-coded data and inhibits outputting and, after the fixed period of time Tw has elapsed, reads the compression-coded data that has been written at a data rate which is not lower than the data rate for the writing while continuing the writing of the compression-coded data; and

a packetizing section for packetizing the compression-coded data read from the first buffer memory and outputting it to a transmission line always as packet data of a fixed size.

14. A data transmission method in which compression-coded data obtained by compression-coding a moving picture signal of a fixed period of time T is accumulated in a first buffer memory for a fixed period of time Tw from the input start and, after that, while the writing to the first buffer memory is being continued, data is read at a rate higher than the writing rate to packetize the data, which is received through a transmission line, the data transmission method comprising the steps of:

reading out the packet data at a data rate not lower than the data rate at which the packet data was output to the transmission line;

writing the read out packet data to a second buffer memory;

reading out the data written to the second buffer memory at a data rate which is substantially the same as the data rate of writing the compression-coded data to the first buffer memory; and

transmitting the data.

15. A data transmission apparatus comprising:

a data extracting section in which compression-coded data obtained by compression-coding a moving picture signal of a fixed period of time T is accumulated in a first buffer memory for a fixed period of time Tw from the output start and, after that, while the writing to the first buffer memory is being continued, data is read at a rate higher than the writing rate to packetize the data to thereby obtain packet data, which is received through a transmission line, the compression-coded data corresponding to the moving picture signal being extracted from this packet data;

a second buffer memory to which the compression-coded data read from the data extracting section is sequentially input and written from the data extracting section, wherein, after the writing, the compression-coded data is read out at a data rate which is lower than the data rate at which the compression-coded data is read from the data extracting section and which is substantially the same as the data rate at which the compression coding is effected; and

a decoding section for decoding the compression-coded data read from the second buffer memory to obtain the moving picture signal.

16. A data recording method in which a moving picture signal of a fixed period of time T is compression-coded and recorded on a recording medium, comprising the steps of:

packetizing compression-coded data output from an encoding section into packet data of a fixed size;

writing the packet data to a first buffer memory and inhibiting recording thereof on the recording medium for a fixed period of time Tw from the output start of the packet data;

after the fixed period of time Tw has elapsed, continuing writing of the packet data to the first buffer memory; and,

reading out the packet data that has been written to the first buffer memory at a data rate which is not lower than the data rate of the writing to the first buffer memory to record the data on the recording medium.

17. A data recording method according to

claim 16

, wherein the fixed period of time Tw is set to be a value which is not smaller than the value of the minimum requisite time for reading an amount of data corresponding to the total capacity of a second buffer memory, which is provided in an encoding section for encoding the moving picture signal into compression-coded data and outputting it, from this second buffer memory.

18. A data recording method according to

claim 16

, wherein the data rate at which the packet data is read from the first buffer memory after the fixed period of time Tw has elapsed is a data rate sufficient for making it possible for compression-coded data corresponding to the moving picture signal of the fixed period of time T and data corresponding to the total capacity of a second buffer memory, which is provided in an encoding section for encoding the moving picture signal into compression-coded data and outputting it, to be read within the fixed period of time T.

19. A data recording method according to

claim 16

, wherein, when the first memory has become empty during the reading of the packet data from the first memory, a meaningless padding packet is generated and recorded on the recording medium.

20. A data recording apparatus comprising:

an encoding section for compression-coding a moving picture signal of a fixed period of time T and outputting it;

a packetizing section to which compression-coded data is input from the encoding section and in which the compression-coded data is packetized; and

a first buffer memory to which packet data is input from the encoding section and which, for a fixed period of time Tw from the input start of the packet data, only allows the writing of the packet data and inhibits the reading thereof and, after the fixed period of time Tw has elapsed, reads the packet data that has been written at a data rate which is not lower than the data rate for the writing while continuing the writing of the packet data and writes it to a recording medium.

21. A data recording apparatus according to claim20, wherein the fixed period of time Tw is set to be a value which is not smaller than the value of the minimum requisite time for reading an amount of data corresponding to the total capacity of a second buffer memory, which is provided in the encoding section, from this second buffer memory.

22. A data recording apparatus according to

claim 20

, wherein the data rate at which the packet data is read from the first buffer memory after the fixed period of time Tw has elapsed is a data rate sufficient for making it possible for compression-coded data corresponding to the moving picture signal of the fixed period of time T and data corresponding to the total capacity of a second buffer memory, which is provided in the encoding section, within the fixed period of time T.

23. A data recording apparatus according to

claim 20

, further comprising means for generating a meaningless padding packet, wherein, when the first memory has become empty during the reading of the packet data from the first memory, the padding packet is recorded on the recording medium.

24. A data reproducing method in which packet data that is obtained by packetizing compression-coded data obtained by compression-coding a moving picture signal of a fixed period of time T is accumulated in a first buffer memory for a fixed period of time Tw from the output start and, after that, while the writing to the first buffer memory is being continued, the packet data is read at a rate higher than the writing rate and recorded on a recording medium and reproduced therefrom, the reproducing method comprising the steps of:

reading out the packet data at a data rate not lower than the data rate at which it was written to the recording medium;

writing the read out packet data to a second buffer memory;

reading out the packet data written to the second buffer memory at a data rate which is substantially the same as the data rate for the packetizing; and

decoding the packet data read out of the second buffer memory.

25. A data reproducing apparatus comprising:

a data extracting section in which packet data that is obtained by packetizing compression-coded data obtained by compression-coding a moving picture signal of a fixed period of time T is accumulated in a first buffer memory for a fixed period of time Tw from the output start and, after that, while the writing to the first buffer memory is being continued, the packet data is read from a recording medium at a rate higher than the writing rate to extract the compression-coded data from the packet data;

a second buffer memory to which the packet data is sequentially input and written from the data extracting section, wherein, after the writing, the packet data is read out at a data rate which is lower than the rate at which the packet data is read from the data extracting section and which is substantially the same as the data rate at which the packetizing of the compression-coded data is effected; and

a decoding section for decoding the packet data read from the second buffer memory to obtain the moving picture signal.

26. A recording medium on which compression-coded data obtained by compression-coding a moving picture signal of a fixed period of time T is recorded always as packet data of a fixed size regardless of the amount of compression-coded data generated,

wherein said packet data is of the following two kinds:

first packet data consisting of header data and the compression-coded data; and

second packet data consisting of header data and dummy data.

27. A data transmission method in which data obtained by compression-coding a moving picture signal of a fixed period of time T is transmitted through a transmission line, comprising the steps of:

packetizing the compression-coded data output from the encoding section into packet data of a fixed size;

writing the packet data to a first buffer memory and inhibiting data output to the transmission line for a fixed period of time Tw from the output start of the packet data;

after the fixed period of time Tw has elapsed, continuing the writing of the packet data to the first buffer memory; and

reading out the packet data that has been written to the first buffer memory at a data rate which is not lower than the data writing rate to transmit the packet data to the transmission line.

28. A data transmission apparatus comprising:

an encoding section for compression-coding a moving picture signal of a fixed period of time T and outputting it;

a packetizing section for packetizing the compression-coded data read from the first buffer memory and outputting it as packet data; and

a first buffer memory to which the packet data is input and which, for a fixed period of time Tw from the input start of the packet data, only allows the writing of the packet data and inhibits the outputting thereof and, after the fixed period of time Tw has elapsed, reads the packet data that has been written at a data rate which is not lower than the data rate for the writing while continuing the input of the packet data.

29. A data transmission method in which packet data that is obtained by packetizing compression-coded data obtained by compression-coding a moving picture signal of a fixed period of time T is accumulated in a first buffer memory for a fixed period of time Tw from the output start and, after that, while the writing to the first buffer memory is being continued, the packet data is read at a rate higher than the writing rate and recorded on a recording medium and read therefrom, the data transmission method comprising the steps of:

reading out the packet data at a data rate not lower than the data rate at which it was written to the recording medium;

writing the read out packet data to a second buffer memory;

reading out the packet data written to the second buffer memory at a data rate which is substantially the same as the data rate for the packetizing; and

decoding the packet data read out of the second buffer memory.

30. A data transmission apparatus comprising:

a data extracting section in which packet data that is obtained by packetizing compression-coded data obtained by compression-coding a moving picture signal of a fixed period of time T is accumulated in a first buffer memory for a fixed period of time Tw from the output start and, after that, while the writing to the first buffer memory is being continued, the packet data is read at a rate higher than the writing rate and received form a transmission line for data transmission, the compression-coded data corresponding to the moving picture signal being extracted from this packet data;

a second buffer memory to which the packet data read from the data extracting section is sequentially input and written, wherein, after the writing, the packet data is read out at a data rate which is lower than the data rate at which the compression-coded data is read from the data extracting section and which is substantially the same as the data rate for the packetizing of the data; and

a decoding section for decoding the packet data read from the second buffer memory to obtain the moving picture signal.

Images