In many Wireless Sensor Network (WSN) applications, it is important to optimize the global energy efficiency in order to enhance node autonomy. Several factors impact the energy consumption in WSN, such as the purpose of the application, the network architecture, the hardware and software of the nodes. With all these factors in view, the definition of a model can provide a useful way to estimate this energy consumption. In this context, we propose a model that is energy-aware, phase-based and protocol-independent. Based on the energy consumption observation of typical WSN applications, our model introduces the pattern concept. It consists of different phases that take place periodically following a frequency Fp. The proposed model is first validated against a well-known simulator (NS2) using different scenarios and considering two wireless link protocols: 802.11a and 802.15.4. Our idea is to use this kind of simulator to monitor the energy consumed in WSN from different points of view. On one hand, the cross-level concept would be applied to minimize the computation resources needed to study the energy consumption of the entire system. On the other hand, at the same time, it offers the possibility to point out for details at different levels of the network. For example, with this simulator, both the consumed energy at the system level and the instantaneous power at the node level could be obtained and analyzed at the same time.< Leer menos