Development of MPEG-1 (Layer 1 & 2) Audio Codecs

• Encoder

  MPEG 1 audio encoder was developed using a 16 bit fixed point DSP. All sampling rates and bit rates were supported. That means user could program these parameters accordingly. Single and Stereo modes for Layer I and II were supported.

  The DSP processor is a 16 bit fixed point processor which runs at 33 MHz. The performance parameters were as follows:

  	>Layer I	>Layer II
  Processor loading	64% (21 MIPs)	50% (16 MIPs)
  Program Memory	1214 Words	1500 Words
  Data Memory
  Internal	096 Words	4096 Words
  External	12256 Words	20000 Words

  The encoder was tested against a standard reference decoder. The decoded output was tested for subjective quality by listening tests. It is also tested with tones with different power levels and output compared with input spectrum (Spectrum Matching).

• Decoder

  Audio decoder was developed in a 16 bit fixed point DSP running at 33 MHz. Both 'C' language fixed point code as well as in assembly language code was developed and delivered. The performance parameters were as follows:

  	Layer I	Layer II
  Processor loading	20% ( 6.6 MIPs )	19% ( 6.7 MIPs )
  Program Memory	450 Words	750 Words
  Data Memory
  Internal	3330 Words	3900 Words
  External	3072 Words	9200 Words

  The decoder was tested using the output from a standard or reference encoder. Subjective testing by listening methods was carried out. For objective analysis, tone testing and spectrum matching method were followed.