No abstract

This paper presents main concepts and issues concerned with the simulation of Internet of Things (IoT). The heterogeneity of possible scenarios, arising from the massive deployment of an enormous amount of sensors and devices, imposes the use of sophisticated modeling and simulation techniques. In fact, the simulation of IoT introduces several issues from both quantitative and qualitative aspects. We discuss novel simulation techniques to enhance scalability and to permit the real-time execution of massively populated IoT environments (e.g., large-scale smart cities). In particular, we claim that agent-based, adaptive Parallel and Distributed Simulation (PADS) approaches are needed, together with multi-level simulation, which provide means to perform highly detailed simulations, on demand. We present a use case concerned with the simulation of smart territories.Comment: Proceedings of the IEEE 2016 International Conference on High Performance Computing and Simulation (HPCS 2016

This work presents a comprehensive and structured taxonomy of available techniques for managing the handover process in mobility architectures. Representative works from the existing literature have been divided into appropriate categories, based on their ability to support horizontal handovers, vertical handovers and multihoming. We describe approaches designed to work on the current Internet (i.e. IPv4-based networks), as well as those that have been devised for the "future" Internet (e.g. IPv6-based networks and extensions). Quantitative measures and qualitative indicators are also presented and used to evaluate and compare the examined approaches. This critical review provides some valuable guidelines and suggestions for designing and developing mobility architectures, including some practical expedients (e.g. those required in the current Internet environment), aimed to cope with the presence of NAT/firewalls and to provide support to legacy systems and several communication protocols working at the application layer. application level. While throughput remains a major goal of system design, the main concern of mobility architectures is how to best manage situations where a MN changes network. This event is currently referred to as handover (or handoff ).By default, current operating systems installed on smartphones adopt the following strategy for data transmission: one Network Interface Card (NIC) at a time is configured and employed to send data. If a WiFi network is available, the terminal switches to WiFi; otherwise a cellular network is utilized, if the latter is available too. During the handover, communications are interrupted. While the widespread use of current smartphones confirms that in general such a simple approach may be a viable solution, in some cases this strategy has some severe limitations. Just as an example (which is actually a true story), let us consider the case of an employee working in an institution/company composed of several buildings, all covered by a WiFi network (e.g. a researcher in a university campus). Suppose that the researcher is a commuter and, just before leaving for going home, he/she receives an important Voice over IP (VoIP) phone call. Since he needs to leave to take the last train home, he decides to answer the call using his/her mobile phone; today, there are plenty of smartphone apps that offer very efficient VoIP services. At that moment, the device is connected through WiFi, but when he gets out of the building, the WiFi signal is lost and the smartphone automatically switches to 4G without any handover management at the application level, thus experiencing a first communication interruption. While moving, he passes through other buildings (hence, within their WiFi coverage); as a consequence, the smartphone switches back to WiFi (i.e. a second communication interruption occurs), and then back to 4G (i.e. yet another communication interruption) and so on. One might suggest that the employee should turn off the WiFi NIC before leaving, thus using the cellular...

In this paper, a methodology is presented and employed for simulating the Internet of Things (IoT). The requirement for scalability, due to the possibly huge amount of involved sensors and devices, and the heterogeneous scenarios that might occur, impose resorting to sophisticated modeling and simulation techniques. In particular, multi-level simulation is regarded as a main framework that allows simulating large-scale IoT environments while keeping high levels of detail, when it is needed. We consider a use case based on the deployment of smart services in decentralized territories. A two level simulator is employed, which is based on a coarse agent-based, adaptive parallel and distributed simulation approach to model the general life of simulated entities. However, when needed a finer grained simulator (based on OMNeT++) is triggered on a restricted portion of the simulated area, which allows considering all issues concerned with wireless communications. Based on this use case, it is confirmed that the ad-hoc wireless networking technologies do represent a principle tool to deploy smart services over decentralized countrysides. Moreover, the performance evaluation confirms the viability of utilizing multi-level simulation for simulating large scale IoT environments.Keywords: Simulation, Smart cities, Internet of Things, Multi-level Simulation, Parallel And Distributed Simulation (PADS) 0 Some parts of the research work described in this paper previously appeared in [1,2,3]. This paper is an extensively revised and extended version of the previous work in which more than 30% is new material. Please note that, some parts of the performance evaluation appeared in the extended version of [2] that is available in https://arxiv.org/abs/1604.07076 but not included in the official conference proceedings.