Area minimization algorithm for parallel prefix adders under bitwise delay constraints

This paper addresses parallel prefix adder synthesis which targets area minimization under given timing constraints. This problem is treated as synthesis of prefix graphs which represent global structures of parallel prefix adders, and a two-folded robust heuristic is proposed. The first process is dynamic programming based area minimization (DPAM), where the search space is limited to a specific subset of the whole set of prefix graphs by imposing some restrictions on structure of prefix graphs, and an exact minimum prefix graph for the limited space can be found efficiently by dynamic programming. The second process is area reduction with re-structuring (ARRS),which removes imposed restrictions on structure, and restructures the result of DPAM for further area reduction. Experimental results show that the size of prefix graph can be reduced by about 10% compared to an existing approach, and area at gate level can also be reduced by more than 30% compared to a commercial tool in some case.