Impact of Code Compression on the Power Consumption in Embedded Systems

Future embedded systems (ES) will offer higher computation capacity than existing embedded systems. New applications of these ES require more and more memory resources and more and more energetic autonomy. For this reason, limited memory capacity and limited power consumption are two of the most important constraints in these systems. Since many ES are battery powered, reduced power consumption can be directly translated into extended battery life. Several studies have been consecrated to the reduction of power consumption, and some very elaborate techniques have been proposed to attain this goal. Compression technique is one of them. In this paper, we describe how to evaluate the effect of code compression on power consumption. For this purpose, a benchmark dedicated to embedded applications (Mibench) has been used to compare several microarchitecture configurations