CAJ | 학술논문

提出了一种新的最优小波包基搜寻算法,在BBS（best basis selection）基础上,选择对数能量熵作为信息花费代价函数,自上而下搜索,将父节点对数能量熵与子节点和的对数能量熵进行比较,标记最小熵节点,直至标记到根节点,按所标记的节点构成小波包树,则此结构即为最优小波包基,此算法打破了一直以来以香农熵为代价函数的惯例。
제출료일충신적최우소파포기수심산법,재BBS（best basis selection）기출상,선택대수능량적작위신식화비대개함수,자상이하수색,장부절점대수능량적여자절점화적대수능량적진행비교,표기최소적절점,직지표기도근절점,안소표기적절점구성소파포수,칙차결구즉위최우소파포기,차산법타파료일직이래이향농적위대개함수적관례。
This paper proposed a new kind of optimum wavelet packet basis search algorithm.Based on the BBS（Best Basis Selection）,the paper chose long energy entropy as the information cost function to search from bottom to top,and compared father node log energy entropy with son node,and marked min-entropy node until root node.According to mark-node,it constituted the wavelet packet tree,and then the structure would be the optimal wavelet packet basis.This algorithm has broken that Shannon as information cost function.This paper also simply introduced application field of this algorithm in the last.