In this paper, a new algorithm is introduced for power efficient implementation of the folding algorithm transformation for the adaptive FIR filters. The algorithm is based on the known minimum spanning tree problem which is a modification of the traveling sales person NP-hard problem. Sub-optimal solution can be obtained leading to dramatic reduction in the folded architecture switching activities. The optimum ordering within the folding sets is done in the steady state condition of the filtering operation. For a typical word length, and for large enough folding factor, the projected reduction in switching activity can be as large as 60%. HEAT tool was used to simulate the effect of this reduction on a typical Baugh-Wooley multiplier and reduction in power consumption for a typical wireline equalization implementation was found to be between 25% and 60%. The new algorithm is based on a simple breadth first search approach and can be easily performed for one time only when the filter is geared to the steady state mode. Dramatic improvement in power consumption can be achieved with minimal overhead.
|Original language||English (US)|
|Number of pages||5|
|Journal||Conference Record of the Asilomar Conference on Signals, Systems and Computers|
|State||Published - Dec 1 1998|