Efficient encoding method, efficient code decoding method, efficient code encoding apparatus, efficient code decoding apparatus, efficient encoding/decoding system, and recording media

• US 6,104,321 A
• Filed: 05/01/1995
• Issued: 08/15/2000
• Est. Priority Date: 07/16/1993
• Status: Expired due to Term

• Current Assignee: Sony Corporation (Sony Group Corp.)
• Original Assignee: Sony Corporation (Sony Group Corp.)
• Inventors: Akagiri, Kenzo
• Primary Examiner(s): YOUNG, BRIAN K
• Application Number:
  US08/403,698
• Time in Patent Office: 1,933 Days
• Field of Search: 341/50, 341/51, 341/200, 395/2.38, 395/2, 395/2.14, 395/2.21
• US Class Current: 341/50
• CPC Class Codes: H04B 1/665 using psychoacoustic properties of the ear, e.g. masking effect
• CPC Class Codes: H04B 1/667 using a division in frequency subbands for TV signals H04N19/63

Abstract

Efficient encoding method for carrying out variable bit allocation between channels to samples in the time region or samples in the frequency region of a plurality of channels with respect to information signals of a plurality of channels (CH1˜CH8). This efficient encoding method decomposes bit allocation quantity to channels (CH1, CH3, CH6) in which bit quantity greater than fixed reference quantity (e.g., 147 kbps) is allocated into first bit allocation quantity which is not above 147 kbps and the remaining second bit allocation quantity (bit allocation quantity above 147 kbps) to quantize them. By using this efficient encoding method, it is possible to reproduce, in the state of high sound quality, by making use of inter-channel bit allocation, compressed signals in which sound quality is improved by using inter-channel bit allocation technology with respect to compression of multi-channel system. In addition, also with ordinarily frequently used decoders adapted for carrying out bit allocation within channels by using bit rate less than fixed value individually with respect to respective channels, such compressed signals can be reproduced without great degradation of sound quality.

Claims

1. An efficient encoding method of re-quantizing sample data of respective digital signals of a plurality of channels by a predetermined bit quantity to encode them,

the method comprising the steps of:

specifying a channel in which bit quantity above a fixed reference quantity determined in advance is allocated;

allocating a bit quantity allocated to the channel to a first bit quantity which is not above the fixed reference quantity at most and the remaining second bit quantity;

re-quantizing a portion of the sample data by using bits by the first bit quantity;

re-quantizing at least the other portion of the sample data by using bits by the second bit quantity; and

synthesizing the re-quantized one portion of the sample data and re-quantized the other portion of the sample data.

2. An efficient encoding method as set forth in claim 1, wherein sum total at all channels of bit quantities of bits allocated to the respective channels is substantially fixed.

3. The step of requantizing the sample data of the respective channels of the efficient encoding method as set forth in claim 1, further including the steps of:

normalizing a plurality of sample data by a common scale factor; and

limiting word lengths of respective sample data normalized by the common scale factor.

4. An efficient encoding method as set forth in claim 3,

wherein data by the first bit quantity and data by the second bit quantity are positioned in different areas within sync block.

5. An efficient encoding method as set forth in claim 1,

wherein the second bit quantity is bit allocation in which no subsidiary allocation bit is included, and the second bit allocation quantity is a difference between bit allocation in which subsidiary allocation bits are included and a bit allocation in which the first bit allocation quantity is not included.

6. An efficient encoding method as set forth in claim 1, wherein the second bit quantity is less than sum total bit quantity of excess bits.

7. An efficient encoding method as set forth in claim 1, wherein the step of carrying out re-quantization re-quantizes sample data every small blocks obtained by subdividing an input signal with respect to a time base and a frequency base.

8. An efficient encoding method as set forth in claim 7,

wherein sample data within the small blocks subdivided with respect to the time base and the frequency base is caused to undergo analysis of non-blocking frequency characteristic to allow an output of the analysis of the non-blocking frequency characteristic to undergo analysis of blocking frequency characteristic.

9. An efficient encoding method as set forth in claim 8, wherein, in the analysis of the blocking frequency characteristic, block size is adaptively altered in dependency upon time characteristic of the input signal.

10. An efficient encoding method as set forth in claim 1, wherein the fixed reference quantity determined in advance of the step of specifying a channel is at least two reference quantities of a reference quantity of bit quantity used for data and a reference quantity in which a bit quantity for sub information is taken into consideration.

11. An efficient encoding method as set forth in claim 1, wherein the fixed reference quantity determined in advance of the step of specifying a channel is a reference quantity in which a bit quantity for the sub information is taken into consideration.

12. An efficient encoding method as set forth in claim 1, wherein the allocation of bits is based on independent allocation in which bits are independently allocated every respective channels and dependent allocation in which bits are correlatively allocated between the respective channels.

13. An efficient encoding method as set forth in claim 12,

wherein the first bit quantity is based on a bit allocation by the independent allocation, and

wherein the second bit quantity is based on a bit allocation by the dependent allocation.

14. An efficient encoding method as set forth in claim 13, wherein alterations of block size are independently carried out every output bands of analysis of at least two non-blocking frequency characteristics.

15. An efficient encoding method as set forth in claim 1, wherein sum of the first bit allocation portion and the second bit allocation portion of the respective channels changes by maximum value of scale factors or sample data of the respective channels.

16. An efficient encoding method as set forth in claim 12, wherein the dependent allocation is changed by changes in point of time of amplitude information of energy values, peak values or mean values of signals of the respective channels.

17. An efficient encoding method as set forth in claim 12, wherein the dependent allocation is changed by changes in point of time of scale factors of the respective channels.

18. A recording medium adapted so that encoded signals formed by the efficient encoding method as set forth in any one of claims 1 to 17 are recorded thereon or therein.

19. An efficiently encoded signal recording method for recording data of a plurality of channels into a single sync block,

the method comprising the steps:

recording data of a channel in which a bit quantity smaller than a fixed reference quantity determined in advance is allocated; and

recording data of a channel in which a bit quantity greater than a fixed reference quantity determined in advance is allocated,

the step of recording data further comprising the steps of;

allocating a bit quantity allocated to the channel to a first bit quantity which is not above the fixed reference quantity at most and the remaining second quantity;

re-quantizing a portion of the sample data by using bits by the first bit quantity to record the re-quantized data; and

re-quantizing at least the other portion of the sample data by using bits by the second bit quantity to record the re-quantized data.

20. An efficiently encoded signal recording method as set forth in claim 19, wherein the re-quantized data are alternately recorded within the sync block.

21. A decoding method for an efficiently encoded signal by an encoding method of re-quantizing sample data of respective digital signals of a plurality of channels by a predetermined bit quantity to encode them,

the efficient encoding method comprising the steps of:

specifying a channel in which a bit quantity greater than a fixed reference quantity determined in advance is allocated;

allocating a bit quantity allocated to the channel to a first bit quantity which is not above the fixed reference quantity at most and the remaining second bit quantity;

re-quantizing a portion of the sample data by using bits by the first bit quantity;

re-quantizing at least the other portion of the sample data by using bits by the second bit quantity; and

synthesizing a portion of the re-quantized sample data and the other portion of the re-quantized sample data,

the synthesized sample data being recorded or transmitted, sample data at least by the first bit quantity of the recorded or transmitted sample data being caused to be sample data of the corresponding channel to obtain a digital signal of at least one of the plurality of channels.

23. The step of re-quantizing the sample data of the respective channels of the decoding method for efficiently encoded signal as set forth in claims 21 and 22,

further including the steps of:

normalizing a plurality of sample data by a common scale factor; and

limiting word lengths of the respective sample data normalized by the common scale factor.

24. A decoding method for efficiently encoded signal as set forth in claim 23, wherein scale factor for sample data relating to the second bit allocation quantity is determined from scale factors and word lengths for sample data relating to the first bit allocation quantity.

22. A decoding method for an efficiently encoded signal by an encoding method of re-quantizing sample data of respective digital signals of a plurality of channels by a predetermined bit quantity to encode them,

the encoding method comprising the steps of:

specifying a channel in which a bit quantity greater than a fixed reference quantity determined in advance is allocated;

allocating a bit quantity allocated to the channel to a first bit quantity which is not above the fixed reference quantity at most and the remaining second bit quantity;

re-quantizing a portion of the sample data by using bits by the first bit quantity;

re-quantizing at least the other portion of the sample data by using bits by the second bit quantity;

synthesizing a portion of the re-quantized sample data and the other portion of the re-quantized sample data,

the synthesized sample data being recorded or transmitted, only sample data by the first bit quantity of the recorded or transmitted sample data being caused to be sample data of the corresponding channel to obtain a digital signal of at least one of the plurality channels.

25. An efficient encoding apparatus adapted for re-quantizing sample data of respective digital signals of a plurality of channels by a predetermined bit quantity to encode them,

the efficient encoding apparatus comprising:

means for specifying a channel in which a bit quantity greater than a fixed reference quantity determined in advance is allocated;

means for allocating a bit quantity allocated to the channel to a first bit quantity which is not above the fixed reference quantity at most and the remaining second bit quantity;

means for re-quantizing a portion of the sample data by using bits by the first bit quantity;

means for re-quantizing at least the other portion of the sample data by using bits by the second bit quantity; and

means for synthesizing a portion of the re-quantized sample data and the other portion of the re-quantized sample data.

26. An efficient encoding apparatus as set forth in claim 25, wherein sum total at all channels of bit quantities of bits allocated to the respective channels is substantially fixed.

27. Means for re-quantizing the sample data of the respective channels of the efficient encoding apparatus as set forth in claim 25 further including:

means for normalizing a plurality of sample data by a common scale factor; and

means for limiting word lengths of respective sample data normalized by the common scale factor.

28. An efficient encoding apparatus as set forth in claim 25, wherein data by the first bit quantity and data by the second bit quantity are positioned in different areas within sync block.

29. An efficient encoding apparatus as set forth in claim 25, wherein the second bit quantity is bit allocation in which no subsidiary bit is included, and the second bit allocation quantity is a difference between bit allocation in which subsidiary allocation bits are included and bit allocation in which the first bit allocation quantity is not included.

30. An efficient encoding apparatus as set forth in claim 25, wherein the second bit quantity is less than sum total bit quantity of excess bits.

31. An efficient encoding apparatus as set forth in claim 25, wherein the means for carrying out re-quantization re-quantizes sample data every small blocks subdivided with respect to the time base and the frequency base.

32. An efficient encoding apparatus as set forth in claim 31,

wherein sample data within the small blocks subdivided with respect to the time base and the frequency base is caused to undergo analysis of non-blocking frequency characteristic to allow an output of analysis of the non-blocking frequency characteristic to undergo analysis of blocking frequency characteristic.

33. An efficient encoding apparatus as set forth in claim 32, wherein, in the analysis of the blocking frequency characteristic, block size is adaptively altered in dependency upon time characteristic of an input signal.

34. An efficient encoding apparatus as set forth in claim 25, wherein the fixed reference quantity determined in advance is at least two reference quantities of a reference quantity of bit quantity used for data and a reference quantity in which a bit quantity for sub information is taken into consideration.

35. An efficient encoding apparatus as set forth in claim 25, wherein the fixed reference quantity determined in advance is a reference quantity in which a bit quantity for sub information is taken into consideration, the apparatus including means for specifying a channel in which a bit quantity greater than the reference quantity is allocated.

36. An efficient encoding apparatus as set forth in claim 25, wherein the allocation of bits is based on independent allocation in which bits are independently allocated every respective channels and dependent allocation in which bits are correlatively allocated between the respective channels.

37. An efficient encoding apparatus as set forth in claim 36,

wherein the first bit quantity is based on bit allocation by the independent allocation, and

wherein the second bit quantity is based on bit allocation by the dependent allocation.

38. An efficient encoding apparatus as set forth in claim 35, wherein alteration of block size is independently carried out every output bands of analysis of at least two non-blocking frequency characteristics.

39. An efficient encoding apparatus as set forth in claim 25, wherein sum of the first bit allocation portion and the second bit allocation portion of the respective channels change by maximum value of scale factors or sample data of the respective channels.

40. An efficient encoding apparatus as set forth in claim 36, wherein dependent allocation is changed by changes in point of time of amplitude information of energy values, peak values or mean values of signals of the respective channels.

41. An efficient encoding apparatus as set forth in claim 36, wherein dependent allocation is changed by changes in point of time of scale factors of the respective channels.

42. A recording medium adapted so that encoded signals formed by the efficient encoding apparatus as set forth in any one of claims 25 to 39 are recorded thereonto or therein.

43. A recording medium adapted so that encoded signals as set forth in claim 42 are recorded thereon or therein, wherein the recording medium is cinema film.

44. An efficient encoding/decoding system for re-quantizing sample data of respective digital signals of a plurality of channels by a predetermined bit quantity to encode re-quantized data to decode encoded data,

the efficient encoding/decoding system comprising:

means for specifying a channel in which a bit quantity greater than a fixed reference quantity determined in advance is allocated;

means for allocating a bit quantity allocated to the channel to a first bit quantity which is not above the fixed reference quantity at most and the remaining second bit quantity;

means for re-quantizing a portion of the sample data by using bits by the first bit quantity;

means for re-quantizing at least the other portion of the sample data by using bits by the second bit quantity;

means for synthesizing a portion of the re-quantized sample data and the other portion of the re-quantized sample data;

means for recording the synthesized sample data onto or into a recording medium or transmitting it to a transmission path; and

means for decoding sample data obtained by re-quantizing a portion of sample data by using the sample data by at least the first bit quantity of the recorded or transmitted sample data to obtain a digital of at least one of the plurality of channels.

45. An efficient encoding/decoding system for re-quantizing sample data of respective digital signals of a plurality of channels by a predetermined bit quantity to encode re-quantized data to decode encoded data,

the efficient encoding/decoding system comprising:

means for specifying a channel in which a bit quantity greater than a fixed reference quantity determined in advance is allocated;

means for allocating a bit quantity allocated to the channel to a first bit quantity which is not above the fixed reference quantity at most and the remaining second bit quantity;

means for re-quantizing a portion of the sample data by using bits by the first bit quantity;

means for re-quantizing at least the other portion of the sample data by using bits by the second bit quantity;

means for synthesizing a portion of the re-quantized sample data and the other portion of the re-quantized sample data;

means for recording the synthesized sample data onto or into a recording medium, or transmitting it to a transmission path; and

means for decoding, as sample data of the corresponding channel, only sample data by the first bit quantity of the recorded or transmitted sample data to obtain a digital signal of at least one of the plurality of channels.