Design And Implementation Of A Multichannel Real-time Audio Coding Based On DSP

Today,digital broadcasting,consumer electronics and communications audio equipments developed rapidly, economic and market prospects are very impressive. More and more attention has been paid on digital audio encoding and decoding products after decades of development, new audio coding class technologies continue to emerge, the corresponding audio coding standards have also been put forward. With the advent of higher performance DSP chips at the same time, the digital broadcast of audio coding equipment are packed with options. Because of the DSP chips and code charges provided by TI and some audio coding accuracy and coding quality problems,the existing MPEG1-Layer2 encoding for the fixed-point and optimization as well as AAC code optimization do have very high research values.MPEG1-Layer2 has been widely applied in digital audio and video broadcasting. Based on TMS320C6455 chip of TI company, this paper designs a multi-channel real-time audio encoding system,and achieves 4 channels real-time audio coding on the device uses TMS320C6455 as the main chip. This design uses C language to the floating-point MPEG1-Layer2 audio coding modules for fixed-point and optimization design, then improves some optimization methods after several revisions and comparisons, finally obtains a high accuracy real-time MP2 audio coding. Meanwhile this paper uses some corresponding optimized design to HE-AAC V2 Parametric Stereo technologies (PS) and QMF filter,together with AAC-LC audio encoding, the three coding modes comprise the 4-channel encoding, and the use artificial selection to encode the input signals. At the same time, this design uses many selections of bit rate and channel, each encoding mode has multiple bit rates and channels to choose accordingly. Finally,this paper uses the DSP/BIOS technology, such as using software interrupts and task scheduling. so as to achieve the real-time performance of each encoding mode.