======= Review 1 ======= *** What is the contribution of this paper? *** This paper considers WSN using energy harvesting with PV. The authors present a harvesting analytical model for a single node, linking three components: the environment, the battery, and the application. The model is employed in several use cases with various indoor and outdoor locations, battery types, and application requirements. No real deployments are considered. *** What are the main reasons to accept this paper? *** The topic fits well with the conference cfp. *** What are the main reasons to reject this paper? *** The paper does not go beyond the state of the art of the subject. There are numerous papers in the area and it is unclear, how this paper goes beyond the state of the art. Many papers consider more elaborate models for storage elements (battery aging and leakage flows are not considered) and also a more detailed model for the energy income. Consider for example these papers: Christian Renner, Stefan Unterschütz, Volker Turau und Kay Römer. Perpetual Data Collection with Energy-Harvesting Sensor Networks. ACM Transactions on Sensor Networks, 11(12):1–12, November 2014. Christian Renner und Volker Turau. Adaptive Energy-Harvest Profiling to Enhance Depletion-Safe Operation and Efficient Task Scheduling. Sustainable Computing: Informatics and Systems, 2(1):43–56, März 2012. The model allows to predict which type of battery is most suitable. The results are obvious and allow no further insight. *** Reasons for your recommendation and suggestions *** The paper does not go beyond the state of the art. The model is not verified with a real deployment. After more then 10 years of research in this field, the contribution is too low for AdhocNow. *** If accepted, which changes would you recommend? *** *** Sub-reviewer Argumentation *** The paper refers to the concept of energy harvesting in WSNs described as a remedy to the problem of finite capacity of WSN batteries. The authors have proposed a harvesting analytical model for a WSN node that seems adequate. Results provided in the second part of the paper also seem to be valuable for the conference audience. The only remarks refer to: - the Energy Collection model from Section 3.1 which is introduced by the authors without any discussion / comparison with other models from the literature - the battery model without similar discussion (as for the issue given above) provided in Section 4.1 with respect to the introduced formulas (necessary to explain why the formulas from page 4 are defined as they are on page 4) These two isses given above lower the overall score of the paper. ======= Review 2 ======= *** What is the contribution of this paper? *** The authors present a model to predict energy expenditure considering the battery parameters, the duty cycle and the energy source (in this case only photovoltaic energy is considered). Some recommendations are then proposed by analysing the operational times and maximum duty cycles obtained for a number of use cases. *** What are the main reasons to accept this paper? *** I think the conclusions obtained are highly relevant to WSN practitioners as it allows them to easily select the most appropriate battery to use based on the scenario and foreseen amount of data to be delivered. *** What are the main reasons to reject this paper? *** The model is not validated and it is fairly simplistic. *** Reasons for your recommendation and suggestions *** I consider that the analytical model presented should be validated, if possible using experimental data. Note that small errors in the energy consumption computation can result in large errors in operational times, which may change the recommendations quite drastically. Note that one would expect the same qualitative results even if the model is slightly inaccurate but that quantitative results may change considerably. The model does not consider interactions among other nodes in the network, network dynamics nor traffic variations. The authors mention some of these aspects as future work. *** If accepted, which changes would you recommend? *** I suggest the authors to either include some experiments to validate the model or, from parts taken from other sources, to explicitly discuss their accuracy. The scope of the model should be better clarified early in the paper. I recommend the authors to describe that they do not consider network interactions, dynamics and traffic variations in the Introduction or at the beginning of Section 2. It appears to me that the contribution of the authors are small variations/extensions as well as linking together models taken from the literature. I believe the authors should clearly state which are their contributions for Sections 3, 4 and 5. Minor changes: - function of time -> as a function of time (it appears twice) - figures are hard to read, I suggest increasing the size of bars and markers and changing the patterns so that they are more easily distinguishable, especially when printed in black and white. ======= Review 3 ======= *** What is the contribution of this paper? *** A model for determining trade-offs between the environment (indoor/outdoor), the application type (tolerates outages or not) and the duty cycle i WSNs are analyzed with the help of a corresponding model. *** What are the main reasons to accept this paper? *** The paper is well written, I found it interesting and informative. *** What are the main reasons to reject this paper? *** I was unable to assess the novelty and the limitations of the approach, because they are not covered in sufficient depth in the paper. *** Reasons for your recommendation and suggestions *** Not being a specialist in the area, expect from the paper that it convinces me in its novelty and usefulness, as I cannot do it myself. Unfortunately, this is not the case with this paper. *** If accepted, which changes would you recommend? *** State previous work on modeling, your contribution and limitations very clearly. What is the novelty in your approach, i comparison to other works of this type?