MPEG Audio Decoder
by Atlanta Signal Processors, Inc.

Software Overview

MPEG Audio Decoder is an audio- decompression algorithm that decompresses 16-bit sampled data (32 
kHz, 44.1 kHz, or 48 kHz). The compressed data rates range from 32 kbps to 384 kbps, for 
compression ratios from 24:1 to 4:1. The algorithm includes options for various sample rates and 
compression ratios in mono, stereo, joint stereo, or dual-channel mode.
The decoder uses the configuration information in each frame of compressed audio to recover the 
algorithm parameters used by the encoder. These parameters may change at any time during the 
compressed sequence without disrupting the decoder operation. In particular this allows segments of 
arbitrary MPEG audio data to be edited and inter-mixed at any frame boundary irrespective of the 
coding parameters used in the individual segments.
In real environments, errors in the compressed-audio data are unavoidable, even if only in the form 
of occasional single bit errors in storage and transmission media. The decoder detects and attempts 
to mask many errors, and always recovers normal operation immediately. During the recovery period, 
the decoder mutes the audio signal to minimize the impact of the errors.

Features and Benefits

•	Runs real-time on a single TMS320C3x DSP
•	32-kHz, 44.1-kHz, 48-kHz sample rates 
•	Transparent quality at 6:1 using layer 2 coding and 4:1 using layer 1
•	Uses `C3x floating-point internally for high precision
•	Mono, stereo, joint-stereo, and dual-channel modes

Processor and System Specifications

•	CPU loading (static memory):

Sample Rate (Hz)
Max %
60-ns
Zero-Wait State
Max %
60-ns
One-Wait State
Max %
74-ns
Zero-Wait State
Max %
74-ns
One-Wait State
32,000
40
50
45
60
44,100
50
65
65
80
48,000
55
70
70
90
These figures were measured using the recommended system setup which places the decoder's temporary 
work space in on-chip `C3x RAM and all else in external memory.
•	Minimum analog requirement: 16-bit A/D, D/A
•	Ready to run on the Elf DSP Platform from Atlanta Signal Processors
•	Portable to any TMS320C3x target platform

Usage Limitations or Performance Considerations

•	The free-format bit-rate mode of operation of the MPEG Audio layer 1 and layer 2 specification is 
not implemented at this time.

Algorithm Verification

•	Fully compliant with the ISO MPEG audio standard layers 1 and 2. Extensively test by several 
major multinational consumer electronic manufacturers.

Availability

•	Written by Texas Instruments and available now through Atlanta Signal Processors.

Company Background and Contact Information

Atlanta Signal Processors was founded in 1981 by Ronald W. Schafer, Thomas P. Barnwell III, and 
Russell M. Mersereau, three Professors of Electrical Engineering at the Georgia Institute of 
Technology with over 80 years of experience in the field. The company has focused its efforts on 
developing tools for professional signal-processing design engineers and now offers digital 
signal-processing hardware and software products to DSP designers worldwide.
Contact:	Atlanta Signal Processors, Inc.
		1375 Peachtree St. NE, Suite 690
		Atlanta, Georgia  30309-3115
		(404) 892-7265
		Fax: (404) 892-2512
		e-mail: aspi@aspi.com