Performance detection of an embedded system using Boosting Algorithm
Wenting Li, Yan Lin · 2009
The boosting algorithm which we introduce is a representational ensemble classification methodology. It is well known that the boosting algorithm can improve the accuracy of any given learning algorithm and train the strong classifiers efficiently. A specific embedded hardware is designed for target identification task. This paper also evaluates the performance and implementation issues for the boosting classification on the embedded hardware. Considering the limited source and the characteristics of the embedded system, this paper proposed an optimal memory allocation method for system optimization which is combining the general software optimization methods. And the method we proposed can also be used for the other embedded system which is support the cache configuration. Some testing samples show the effectiveness of the proposed technique.