Low power high level synthesis by increasing data correlation

With the increasing performance and density of VLSI circuits as well as the popularity of portable devices such as personal digital assistance, power consumption has emerged as an important issue in the design of electronic systems. Low power design techniques have been pursued at all design levels. However, it is more effective to attempt to reduce power dissipation at higher levels of abstraction which allows a wider view. In this paper, we propose a simultaneous scheduling and binding scheme which increases the correlation between consecutive inputs to an execution unit so that the switched capacitance of the execution unit is reduced. The proposed method is implemented and integrated into the scheduling and assignment part of the HYPER synthesis environment. Compared with the original HYPER synthesis system, average power saving of 23.0% in execution units and 14.2% in the whole circuit, is obtained for a set of benchmark examples.