Realization of Embedded Automotive System based on Dual Core Processor

Abstract

The power Consumption of modern high-performance
processors is becoming a major concern because it leads to increased heat dissipation and decreased reliability.A heterogeneous multiprocessor (HeMP) system consists of several heterogeneous processors each would have the capability of producing better energy saving performance. The problem of minimizing the energy consumption without missing deadlines has become an important issue in constructing low-power real-time HeMP systems. Each
processor may have its own instruction-set architecture, specially designed to provide the best performance for a particular category of applications. Here pruning-based algorithm is used to obtain the optimal solution and a heuristic algorithm is also proposed to derive
an approximate solution. Each processor’s speed is determined by its final workload. In simulations, we model more than a couple dozens of off-the-shelf embedded processors including ARM processor and TI DSP. The results show that the pruning-based algorithm reduces
the time needed to derive the optimal solution by at least 98%, compared with the exhaustive search.