Method and apparatus for voice latency reduction in a voice-over-data wireless communication system

A method for reducing voice latency in a voice-over-data wireless communication system, comprising the steps of:

• generating a plurality of data frames;
• dropping an entire one or more of said plurality of data frames to keep a plurality of remaining data frames, wherein said step of dropping further comprises:
• determining a communication channel latency;
• and
• dropping entire ones of each of said plurality of data frames having an encoded rate equal to a first encoding rate out of a number of possible encoder rates if said communication channel latency exceeds a predetermined threshold;
• and
• storing said plurality of remaining data frames in a queue.

An apparatus for reducing voice latency in a voice-over-data wireless communication system, comprising:

• means for generating data flames;
• a processor connected to said data frame generating means for determining a communication channel latency and for dropping an entire one or more of said data frames to keep remaining data frames;
• and
• a queue for storing said remaining data frames, wherein:
• entire ones of said data frames having an encoded rate equal to a first encoding rate out of a number of possible encoder rates are dropped if said communication channel latency exceeds a predetermined threshold.

An apparatus for reducing voice latency in a voice-over-data wireless communication system, comprising:

• a receiver for receiving a wireless communication signal;
• a demodulator for demodulating said wireless communication signal and for producing data frames;
• means for determining a voice frame integrity;
• a processor connected to said demodulator for dropping one or more of said data frames to keep remaining data frames, said processor further for comparing said voice frame integrity with a predetermined value, said predetermined value representing a minimum desired voice quality, for increasing a variable queue threshold if said voice frame integrity is less than said predetermined value, for decreasing said variable queue threshold if said voice frame integrity is greater than said predetermined value, for dropping frames at a first rate if a length of said queue is less than said variable queue threshold, and for dropping frames at a second rate if said length is greater than said variable queue threshold;
• and
• a queue for storing said remaining data frames.