In the near future, portable embedded devices must run multimedia applications with enormous computational requirements at low energy consumption. These applications demand extensive memory footprint and must rely on dynamic memory due to the unpredictability of input data (e.g. 3D streams features) and system behaviour (e.g. variable number of applications running concurrently). Within this context, the dynamic memory subsystem is one of the main sources of power consumption and embedded systems have very limited batteries to provide efficient general-purpose dynamic memory management. As a result, consistent design methodologies that can efficiently tackle the complex dynamic memory behaviour of these new applications for low power embedded systems are in great need. In this paper we propose a step-wise system-level approach that allows the design of platform-specific dynamic memory management mechanisms with low power consumption for such kind of dynamic applications. The experimental results in reallife case studies show that our approach improves power consumption up to 89% over current state-of-the-art dynamic memory managers for complex applications.