[Mp4-tech] Re: [audio] AAC bitstream - headers

[Ralph Sperschneider]

Anubrata Chakarabarti wrote:
> Ralph Sperschneider wrote:
> 
> 
>>Levels specify the complexity of the decoding process. This might be due to the
>>number of channels, due to the sampling frequency, due to the number of objects,
>>due to the permitted amount of FEC etc.
> 
> 
> That means if I'm analyzing the input (but not decoding) then I need not worry about levels, right?

That should be right.
>>>2. If CRC check is present in ADTS frame then what measures will have to be taken at the decoder side?
>>
>>You don´t need to do anything but just skipt the crc word. If you wonna make
>>sure that the received data is integer, you can derive the crc on decoder side
>>and compare it with the transmitted crc word.
> 
> 
> I believe for derivation of CRC i''ll need the CRC polynomial. Where do I get it?

It is defined in the standard.
>>>If yes then what should be the treatment for Data Stram Elemnts and/or Fill elements (from the perspective of an analyzer - do
>>>they have any significance?) My question is based on the basic AU definition that all values in an AU are associated with time data.
>>
>>Data transmitted in the extension payload (I guess you talk about those data)
>>might be of different kind. It might be directly related to the transmitted
>>audio data (e.g. SBR, DRC) or just cover any kind of not specifically defined
>>ancillary data.
>>
> 
> 
> Can they be considered as part of AU? I mean should I ignore them while decoding?

The extension payload is part of the AU. Some kinds of extension payload are 
optional, while others are mandatory to be evaluate by a decoder.
Best regards,
Ralph
-- 
Dipl.-Ing. Ralph Sperschneider  | Phone: +49 9131 776 344
Fraunhofer IIS                  | Fax:   +49 9131 776 398
Am Wolfsmantel 33               | mailto:ralph.sperschneider iis.fraunhofer.de
D 91058 Erlangen                | http://www.iis.fraunhofer.de/amm/