一种可伸缩的支持向量机的硬件实现方法

提出了支持向量机（Support Vector Machine）学习过程硬件实现的一种新的数字电路结构，它可以在保证向量机学习速度的同时，提高支持向量机的硬件资源利用效率；此外基于此结构的支持向量学习机还可以适用于低于设计数的样本集．由于该设计的灵活性以及其对硬件要求的减小，使得支持向量学习机可以更好、更方便地应用于嵌入式系统中．

A scalable hardware implementation of the support vector machine.

This paper presented a new digital structure for support vector machine(SVM) learning,which can get a good performance with less hardware resource while the process speed is not reduced.Also the SVM learning based on the new structure can make itself to suit the number of the sample which is less than the expected.Because of this adaptions and the reduction of the demand for hardware,SVM learning can be applied more widely and conveniently in the embedded system.