Method for decoding MPEG audio data
DCFirst Claim
Patent Images
1. A method for decoding an MPEG audio data frame, comprising the steps of:
- decoding subband data from an audio data frame to generate a first vector, the first vector having components which represent frequency-domain components of a sound sample;
combining two or more components of the first vector, using a butterfly unit;
determining a product of the combination and a matrixing coefficient, using a multiplier;
accumulating the product into a memory location; and
repeating the combining, determining, and accumulating steps one or more times to determine a component of a second vector, the second vector having components which represent frequency-domain components of a sound sample.
4 Assignments
Litigations
0 Petitions
Accused Products
Abstract
An MPEG audio/video decoder has memories, a signal processing unit (SPU) including a multiplier and a butterfly unit, a main CPU, and a memory controller which are time division multiplexed between decoding video and audio data. For audio decoding, the butterfly unit determines combinations of components of a frequency-domain vector to reduce the number of multiplies required to transform to the time domain (matrixing). Matrixing is interwoven with MPEG filtering to increase throughput of the decoder by increasing parallel use of the multiplier, the butterfly unit, and a memory controller.
9 Citations
15 Claims
-
1. A method for decoding an MPEG audio data frame, comprising the steps of:
-
decoding subband data from an audio data frame to generate a first vector, the first vector having components which represent frequency-domain components of a sound sample; combining two or more components of the first vector, using a butterfly unit; determining a product of the combination and a matrixing coefficient, using a multiplier; accumulating the product into a memory location; and repeating the combining, determining, and accumulating steps one or more times to determine a component of a second vector, the second vector having components which represent frequency-domain components of a sound sample. - View Dependent Claims (2, 3, 4, 5, 6)
-
-
7. A method for decoding an MPEG audio data frame, comprising:
-
decoding subband data from an audio data frame to generate a sample vector S having components Si which represent frequency-domain components of a sound sample, wherein index i ranges from 0 to M; combining two or more of said components Si to produce a linear, unscaled combination Tk of said two or more components Si; repeating said combining M times to obtain a plurality of said combinations Tk; multiplying selected ones of said combinations Tk with selected ones of coefficients from a transformation matrix N to form a plurality of products; adding selected products from said multiplying to obtain a vector component Vj of a vector V; and determining a time-domain sound amplitude from said vector component Vj. - View Dependent Claims (8, 9, 10, 11, 12, 13, 14, 15)
-
Specification