Compressor/decompressor selecting apparatus and method of the same

A Codec selecting apparatus, for selecting a Codec according to an internet communication system, the Codec selecting apparatus comprising:

• a performance analyzer, for analyzing the performance of the current internet communication system, and accordingly outputting a plurality of environmental parameters;
• a training server, for receiving a plurality of training samples and accordingly outputting a plurality of learning data through a SVM calculation, wherein each of the learning data is corresponding to a hyperplane in a high-dimensional feature space mapped by the training samples, and the hyperplanes divide the high-dimensional feature space into a plurality of Codec regions;
• a storage unit, for storing the learning data;
• and
• a selecting unit, for performing the SVM calculation to select the Codec suitable for the current internet communication system by determining the Codec region comprising a function point in the high-dimensional space mapped by the environmental parameters.

A Codec selecting method, for selecting a Codec according to an internet communication system, the selecting method comprising:

• analyzing the performance of the current internet communication system to obtain a plurality of environmental parameters;
• obtaining a plurality of learning data through a SVM calculation, wherein each of the learning data forms a hyperplane in a high-dimensional feature space and the hyperplanes divide the high-dimensional feature space into a plurality of Codec regions;
• and
• performing the SVM calculation to determine the Codec region comprising a function point in the high-dimensional feature space mapped by the environmental parameters and accordingly select the Codec suitable for the current internet communication system.

A Codec selecting method, for selecting a Codec according to an internet communication system, the selecting method comprising:

• analyzing the performance of the current internet communication system to obtain a plurality of environmental parameters;
• having a training according to a plurality of training samples to obtain a plurality of learning data through a SVM calculation, wherein each of the learning data is corresponding to a hyperplane in a high-dimensional feature space mapped by the training samples, and the hyperplanes divide the high-dimensional feature space into a plurality of Codec regions;
• and
• determining the Codec region comprising a function point in the high-dimensional feature space mapped by the environmental parameters to select the Codec suitable for the current internet communication system.