The proliferation of radio communication systems and the significant advances in enabling device technologies are paving the way towards the Internet-of-Things (IoT) and opening new perspectives for Smart City services and applications. Such evolution becomes essential in order to enhance quality of urban services, to reduce costs, and to engage citizens more effectively. In this context, novel simulation tools are required to prepare the future deployments of large-scale IoT infrastructure for Smart cities in the best conditions in terms of reliability, energy consumption, and cost. This keynote presents the CupCarbon framework: a platform for designing smart-city & IoT wireless sensor networks (SCI-WSN). CupCarbon is developed as a part of the French ANR project PERSEPTEUR to meet the following three requirements: (1) to be very accurate, to take into account all details related to the deployment environment and to quantify the simulation uncertainty, (2) to allow the representation of mobile nodes and dynamic environments, (3) to allow the study of the behaviour of a whole network (or multiple networks) containing a large number of nodes in realistic environments (city, mountain, etc.).