There is no the method and system of the frequency domain active matrix decoding of feedback

There is no the method and system of the frequency domain active matrix decoding of feedback

  • CN 102,282,868 B
  • Filed: 01/12/2010
  • Issued: 11/02/2016
  • Est. Priority Date: 01/14/2009
  • Status: Active Grant
First Claim
Patent Images

1. one kind is used for decoding to determine M audio output signal N number of audio input signalMatrix coding/decoding method, wherein, M and N is integer, and M is bigger than N, and N=2,Described method comprises the following steps:

  • N number of audio input signal is converted into frequency domain (10,11) to produce input frequency from time domainN number of stream of rate composition;

    Power ratio (17,30,31,32,33) is determined from the stream of incoming frequency composition, describedPower ratio comprises at least one the power ratio about each critical band in one group of critical band,Wherein, described one group of critical band is that the consideration excited according to consciousness determines;

    About each critical band in critical band from power ratio determine gain control value (17,38), comprise by a non-linear fashion by power than shaping, wherein, described with non-linearMode power is comprised the exponent pair with substantially equal at least 3 from power ratio than shapingThe step of at least one at least one value exponentiation determined, wherein, for each critical bandGain control value can be independently adjusted;

    Operation active matrix subsystem (16) is with the miscarriage responding incoming frequency composition raw output frequencyM stream of rate composition, wherein, uses gain control value to control active matrix subsystem (16),And the stream of incoming frequency composition is applied and is organized matrix coefficient more by active matrix subsystem (16),Often group matrix coefficient is used for different in critical band one;

    AndThe stream of output frequency composition is converted into time domain (20) from frequency domain, thus produces M soundFrequently output signal.

View all claims

    Thank you for your feedback