G.726 Voice Encoding And Embedded Adhoc Network Realization

In highly development of informational technology today, multimedia and network transmission gradually become the hot fields of the embedded application. Then the speech coding technology as an important branch of the voice processing technology whose application has become an important part of embedded development. Thus, according to the different hardware platforms and the application backgrounds, different coding algorithm technology for different occasions. Among them, the ITU G.726 ADPCM speech coding standard has been concerned for its high fidelity, small delay and low algorithm complexity. And the voice wireless networks transmission as a direction for application of embedded system that has been more extensive used in the consumer electronic area. The embedded technology can easy achieve the voice compression that will get the expansion and upgrading of transmission system, and make the more perfect application system.Using a large number of associated documents that be referred to that we can find the most literature placed on the implementation of the ADPCM speech codec, and mainly through the hardware MCU, FPGA and DSP to design its codec modules, or the simulation of pure software. But in the majority of building the voice system only adapt to the local voice recording, but not relate to standard voice algorithm with the embedded network transmission.Now, in the thesis, my starting point is that comprehensively study the basis for speech coding theory at first, with emphasis on G.726 ADPCM standard algorithm analysis and research. Then, according to the hardware platform, we can simplify and improve the algorithm. Taking into account the embedded application environment, with the improved speech coding algorithms, we can use the MATLAB or such as VC++ software to make simulation for it.The second, we build the embedded the Linux cross-build environment according to the hardware platform, and to achieve transplantation of AODV routing protocol, so it can be build an Ad Hoc wireless network environment.At last, under the premise of ensuring voice quality, we can make the optimizational algorithm transplantate to the MagicS3C2410 platoform constitutes a G.726 voice codec system. After the study the implementation mechanism under the multi-task in Linux, with using the multi-threading technology in the wireless network of voice full duplex communication.