The problem of designing and simulating optimal com-munication protocols for energy harvesting wireless net-works has recently received considerable attention, thus re-quiring an accurate modeling of the energy harvesting pro-cess and a consequent redesign of the simulation framework to include this. While the current ns-3 energy framework al-lows the definition of new energy sources that incorporate the contribution of an energy harvester, integrating an en-ergy harvester component into an existing energy source is not straightforward using the existing energy framework. In this paper, we propose an extension of the ns-3.20 energy framework in order to explicitly introduce the concept of an energy harvester. Starting from the definition of a general en-ergy harvester, we provide the implementation of two simple models for the energy harvester. In addition, we introduce the concept of an energy predictor, that gathers information from the energy source and harvester and uses this informa-tion to predict the amount of energy that will be available in the future. Finally, we extend the current energy frame-work to include a model for a supercapacitor energy source and a device energy model for the energy consumption of a sensor. Example simulation results show the benefit of our contributions to the ns-3 energy framework.

With the promise of delivering immortality, energy har-vesting and wireless energy transfer have become the next research frontier for pervasive computing and networks. The challenge still is to adapt operations along all aspects of a sensing system (sensing, computation, communication), to deal with the varied, unpredictable or possibly intermittent supply of such energy. To understand these challenges, we currently lack sufficient tools to evaluate the impact of dif-ferent energy harvesting and transfer models- which is fur-ther exacerbated due to the high cost of deploying such sys-tems at a large scale. We present a generic, TOSSIM-based simulation framework to model energy harvesting and en-ergy transfer, enabling rapid development of harvesting- and transfer-aware applications, protocols, and system software. Our evaluation shows that even an abstract simulation model can provide useful insights, such as frequent power outages and node reboots due to intermittent energy supply. Based on these insights, we further establish the utility of this frame-work by demonstrating how high level simulations can lead to a better choice of energy scheduling algorithms.