Decoder for compressed digital signals

A decoder for decoding compression data attained by conducting a compression encoding operation for digital data through a high efficiency encoding in which the data compression is conducted such that one sample is subdivided according to a frequency of a signal represented by the digital data into a set of subband samples representing frequency components of a plurality of subbands and there are extracted from each of the subband samples first information data having a variable bit length and indicating a level of the subband sample logarithmically compressed, second information data representing a factor of the logarithmic compression, third information data denoting the number of the second information items associated with a predetermined data length, and fourth information data representing at least a bit length of the first information data, said decoder comprising:

• first storage means for provisionally storing therein at least the second information data and the fourth information data;
• first write means for writing at least the second information data and the fourth information data in a predetermined storage area of the first storage means according to the third information data;
• first read means for reading at least the second information data and the fourth information data from the predetermined storage area of the first storage means;
• second storage means for temporarily storing therein the first information data;
• second write means for writing the first information data in the second storage means according to the fourth information data;
• second read means for reading the first information data from the second storage means;
• operating means for conducting an inverse quantization process obtaining a bit length of each sample data from the first information data according to the fourth information data, executing an inverse scaling process attaining subband sample data indicating a level of the subband sample from each sample data through an exponent operation specified by the second information data, carrying out a complex subband filtering process combining the subband sample data related to the set of subband samples into a sequence of numeric values associated with the original digital data, and thereby obtaining the original digital data from the sequence of numeric values;
• said operating means further producing an operation formula by subdividing an exponent of the exponent operation into an integer part and a fraction part and conducting through a multiplication an operation related to the fraction part according to the operation formula, conducting through a shift operation an operation related to the integer part, thereby attaining the subband sample data;
• said operating means calculating the sequence of numeric values on the basis of the subband sample data associated with the set of subband samples;
• third storage means for temporarily storing therein the sequence of numeric values calculated by the operating means;
• third write means for writing the sequence of numeric values in a predetermined area of the third storage means;
• third read means for reading the sequence of numeric values from the third storage means;
• and
• output means for outputting therefrom digital data obtained by the operation means;
• and wherein:
• the second write means writes the first information in the second storage means according to the fourth information in a state in which the write operation is not necessarily aligned with address boundaries of the second storage means;
• and
• the second read means reads the first information from the second storage means in an information-by-information manner, the first information being written in the second storage means according to the fourth information in a state in which the write operation is not necessarily aligned with address boundaries of the second storage means.

A decoder for decoding compression data attained by conducting a compression encoding operation for digital data through a high efficiency encoding in which the data compression is conducted such that one sample is subdivided according to a frequency of a signal represented by the digital data into a set of subband samples representing frequency components of a plurality of subbands and there are extracted from each of the subband samples first information data having a variable bit length and indicating a level of the subband sample logarithmically compressed, second information data representing a factor of the logarithmic compression, third information data denoting the number of the second information items associated with a predetermined data length, and fourth information data representing at least a bit length of the first information data, said decoder comprising:

• first storage means for provisionally storing therein at least the second information data and the fourth information data;
• first write means for writing at least the second information data and the fourth information data in a predetermined storage area of the first storage means according to the third information data;
• first read means for reading at least the second information data and the fourth information data from the predetermined storage area of the first storage means;
• second storage means for temporarily storing therein the first information data;
• second write means for writing the first information data in the second storage means according to the fourth information data;
• second read means for reading the first information data from the second storage means;
• operating means for conducting an inverse quantization process obtaining a bit length of each sample data from the first information data according to the fourth information data, executing an inverse scaling process attaining subband sample data indicating a level of the subband sample from each sample data through an exponent operation specified by the second information data, carrying out a complex subband filtering process combining the subband sample data related to the set of subband samples into a sequence of numeric values associated with the original digital data, and thereby obtaining the original digital data from the sequence of numeric values;
• said operating means further producing an operation formula by subdividing an exponent of the exponent operation into an integer part and a fraction part and conducting through a multiplication an operation related to the fraction part according to the operation formula, conducting through a shift operation an operation related to the integer part, thereby attaining the subband sample data;
• said operating means calculating the sequence of numeric values on the basis of the subband sample data associated with the set of subband samples;
• third storage means for temporarily storing therein the sequence of numeric values calculated by the operating means;
• third write means for writing the sequence of numeric values in a predetermined area of the third storage means;
• third read means for reading the sequence of numeric values from the third storage means;
• and
• output means for outputting therefrom digital data obtained by the operation means;
• and wherein:
• the operating means calculates only a first portion of the sequence of numeric values according to a correlation in a plurality of numeric values included in the sequence of numeric values;
• and
• the operating means obtains the original digital data according to, in place of a second portion of the sequence of numeric values, the first portion of the sequence of numeric values having a correlation with the second portion of the sequence of numeric values.

A decoder for decoding compression data attained by conducting a compression encoding operation for digital data through a high efficiency encoding in which the data compression is conducted such that one sample is subdivided according to a frequency of a signal represented by the digital data into a set of subband samples representing frequency components of a plurality of subbands and there are extracted from each of the subband samples first information data having a variable bit length and indicating a level of the subband sample logarithmically compressed, second information data representing a factor of the logarithmic compression, third information data denoting the number of the second information items associated with a predetermined data length, and fourth information data representing at least a bit length of the first information data, said decoder comprising:

• first storage means for provisionally storing therein at least the second information data and the fourth information data;
• first write means for writing at least the second information data and the fourth information data in a predetermined storage area of the first storage means according to the third information data;
• first read means for reading at least the second information data and the fourth information data from the predetermined storage area of the first storage means;
• second storage means for temporarily storing therein the first information data;
• second write means for writing the first information data in the second storage means according to the fourth information data;
• second read means for reading the first information data from the second storage means;
• operating means for conducting an inverse quantization process obtaining a bit length of each sample data from the first information data according to the fourth information data, executing an inverse scaling process attaining subband sample data indicating a level of the subband sample from each sample data through an exponent operation specified by the second information data, carrying out a complex subband filtering process combining the subband sample data related to the set of subband samples into a sequence of numeric values associated with the original digital data, and thereby obtaining the original digital data from the sequence of numeric values;
• said operating means further producing an operation formula by subdividing an exponent of the exponent operation into an integer part and a fraction part and conducting through a multiplication an operation related to the fraction part according to the operation formula, conducting through a shift operation an operation related to the integer part, thereby attaining the subband sample data;
• said operating means calculating the sequence of numeric values on the basis of the subband sample data associated with the set of subband samples;
• third storage means for temporarily storing therein the sequence of numeric values calculated by the operating means;
• third write means for writing the sequence of numeric values in a predetermined area of the third storage means;
• third read means for reading the sequence of numeric values from the third storage means;
• and
• output means for outputting therefrom digital data obtained by the operation means;
• and
• wherein a sequence of processes of attaining the original data from the compression data and outputting the data to an external device are conducted in a pipeline process.

A decoder for decoding compression data attained by conducting a compression encoding operation for digital data by using a high efficiency encoding in which the data compression is conducted such that one sample is subdivided according to a frequency of a signal represented by the digital data into a set of subband samples representing frequency components of a plurality of subbands and there are extracted from each of the subband samples first information data having a variable bit length and indicating a level of the subband sample logarithmically compressed, second information data representing a factor of the logarithmic compression, third information data denoting the number of the second information items associated with a predetermined data length, and fourth information data representing at least a bit length of the first information data, said decoder comprising:

• first operating means for calculating a sequence of numeric values related to the original data according to the first to fourth information data items;
• storage means for storing therein the sequence of numeric values calculated by the first operating means;
• write means for writing the sequence of numeric values in a predetermined storage area of the storage means;
• read means for reading the sequence of numeric values from the storage means;
• read control means for controlling the read means to read, when a numeric value calculated by the first operating means is beforehand stored in the storage means, a portion of the sequence of numeric values to which the value belongs from the storage means;
• second operating means for achieving multiplications and additions according to at least the portion of the sequence of numeric values read from the storage means under control of the read control means and a predetermined coefficient and thereby producing digital data items in an item-by-item fashion;
• and
• output means for outputting, each time the original digital data is reproduced in the second operation, the digital data to an external device.

A decoder for decoding compression data which is attained by conducting a compression encoding operation for digital data through a high efficiency encoding and which includes sample data having a variable bit length;

• the decoder comprising:
• storage means for storing therein the sample data;
• write means for writing the sample data in the storage means in a state in which the write operation is not necessarily aligned with address boundaries of the storage means;
• and
• read means for reading the sample data items from the storage means in an item-by-item manner, the sample data being written in the storage means in a state in which the write operation is not necessarily aligned with address boundaries of the storage means;
• and wherein
• said compression data is audio compressed data conforming to moving picture experts group (MPEG) standards.

A decoder for decoding compression data which is attained by conducting a compression encoding operation for digital data through a high efficiency encoding and which includes sample data having a variable bit length;

• the decoder comprising:
• storage means for storing therein the sample data;
• write means for writing the sample data in the storage means in a state in which the write operation is not necessarily aligned with address boundaries of the storage means;
• and
• read means for reading the sample data items from the storage means in an item-by-item manner, the sample data being written in the storage means in a state in which the write operation is not necessarily aligned with address boundaries of the storage means;
• and
• wherein the read means includes:
• sample specifying means for specifying a position of one sample data item to be read from the storage means:
• first latch means for sequentially keeping therein data from the storage means at an address associated with the position specified by the sample specifying means;
• second latch means for sequentially keeping therein data outputted from the first latch means;
• decision means for deciding that the sample data item to be read from the storage means is stored in an area thereof indicated by one address or in an area indicated by two addresses;
• first acquiring means for acquiring, when the decision means decides that the sample data item is in an area indicated by one address, the sample data item as the read object from data kept in the second latch means;
• and
• second acquiring means for acquiring, when the decision means decides that the sample data item is in an area indicated by two addresses, the sample data item as the read object from data items respectively kept in the first and second latch means.

A decoder for decoding compression data attained by conducting a compression encoding operation for digital data by using a high efficiency encoding, the decoder comprising:

• operating means for calculating a portion of a sequence of numeric values related to the original digital data according to the compression data, the portion of the sequence of numeric values having a correlation with another portion of the sequence of numeric values;
• storage means for storing therein at least the portion of the sequence of numeric values calculated by the operating means;
• write means for writing at least the portion of the sequence of numeric values in a predetermined area of the storage means;
• read means for reading at least the portion of the sequence of numeric values from the storage means;
• and
• read control means for controlling the read means to read, in response to a use request of a numeric value, the numeric value from the storage means, and controlling the read means to read, when the requested numeric value is missing in the storage, a numeric value having a correlation with the requested numeric value from the storage means.