Title :
A Common Subexpression Elimination Tree Algorithm
Author :
Al-Hasani, Firas ; Hayes, Michael P. ; Bainbridge-Smith, Andrew
Author_Institution :
Dept. of Electr. & Comput. Eng., Univ. of Canterbury, Christchurch, New Zealand
Abstract :
A common subexpression elimination algorithm is proposed to minimize the complexity of the multiple constant multiplication operation. The coefficients (constants) of the multiple constant multiplication are represented using the binary signed digit number system. The binary signed digit representations of each coefficient are enumerated using the representation tree. The algorithm traverses the tree to calculate the possible subexpressions at each node. Each subexpression is used to find a possible decomposition for the coefficient to be encoded. A complexity formula is proposed to compare the decompositions. The algorithm is designed to prune the tree when it finds a decomposition with minimum complexity. This reduces the search space while minimizing the hardware complexity. Results show that the algorithm has better performance than other published algorithms including linear programming optimization methods. The algorithm outperforms the subexpression sharing method in that uses only the canonical signed-digit representations.
Keywords :
computational complexity; linear programming; trees (mathematics); binary signed digit number system; hardware complexity; linear programming optimization methods; minimum complexity; multiple constant multiplication operation; representation tree; subexpression elimination tree algorithm; Binary signed digit (BSD); common subexpression elimination (CSE); multiple constant multiplication (MCM); subexpression space;
Journal_Title :
Circuits and Systems I: Regular Papers, IEEE Transactions on
DOI :
10.1109/TCSI.2013.2244328