China
Africa
Explore chapters and articles related to this topic
Testing and Debugging Sensor Network Applications
Published in S. Sitharama Iyengar, Richard R. Brooks, Distributed Sensor Networks, 2016
Sally K. Wahba, Jason O. Hallstrom, Nigamanth Sridhar
Network simulation has been used as a testing tool in networked and distributed systems for many years. The use of network simulators in testing sensor network applications is only natural. The primary goal of simulation is to provide developers with an inexpensive means of rapidly testing their applications before deployment. Simulators provide a controlled environment that allows developers to repeatedly evaluate their applications. Simulators also scale; they are not limited by physical deployment constraints. Users can test their applications with many sensor nodes.
Vehicular Network Simulation via ns-3 with Software-Defined Networking Paradigm
Published in Fei Hu, Vehicle-to-Vehicle and Vehicle-to-Infrastructure Communications A Technical Approach, 2018
The communication networks can be huge and complex; traditional analytical methods are not enough to estimate the performance and the behavior accurately. Thus, network simulators have been introduced to improve the network evaluation. Normally, a network simulator contains various models to achieve the functionality of computer networks, such as a channel model, a mobility model, a routing model, and an application model. Since there are a huge number of functionalities in different layers (e.g., physical layer, data link layer, routing layer, application layer) of a computer network, a network simulator could be a large and complex software system with the coverage of all these functionalities. Besides various network functionalities, a network simulator also needs to logically integrate the models and make them work together as one whole system. This procedure involves some critical aspects, such as network synchronization, parameter statistics, thread management, model integration, and internal simulator clock. In a network simulator, all the APIs of the models need to be well designed to cooperate with other relative models. Simulating a new protocol in a simulator would incur lots of work, since a huge number of models must be modified to adapt the new model. Based on this consideration, it is preferred to select a professional network simulator that covers as many of the functionalities we need as possible. Nevertheless, the SDN paradigm hasn’t been well built in most network simulators. Especially, the current research of SDN is focusing more on wired networks and none of the simulation tools have combined the SDN framework with a wireless network environment. On the contrary, there is also a specialized network simulator for the SDN structure, called Mininet. Mininet is mainly focused on wired network structures. It is short of the functionalities of WMN, in which the ad hoc protocol and mobility functions cannot be achieved. Furthermore, as a lightweight SDN simulator, Mininet is not able to simulate any traditional distributed networking structures either. Users cannot make comparisons between an SDN structure and a traditional distributed network structure by Mininet.
Optimal travel information provision strategies: an agent-based approach under uncertainty
Published in Transportmetrica B: Transport Dynamics, 2018
Chenfeng Xiong, Zheng Zhu, Xiqun Chen, Lei Zhang
Secondly, strategies such as ATIS/ATMS market penetration need a reexamination. In a number of studies, results show that transportation, social and economic benefit to informed travelers is significant but may decrease with increasing market penetration under certain circumstances (Yang 1998; Yin and Yang 2003; Nakamura and Keitoku 2008). Optimal strategies in providing information are worth exploitation and may be analyzed using a simulation-based approach. Compared to bi-level optimization that is widely applied in network-level optimizations (e.g. Yang 1999; Yang and Huang 2004; Shimamoto et al. 2010; Qiao et al. 2014), simulation-based optimization (SBO) adopts response surface and meta models to approximate optima, which reduces the computational load for expensive-to-evaluate studies (interested readers can get more information regarding this approach in Chen et al. 2014; Osorio and Bierlaire 2013). It is especially promising in this research since sophisticated multidimensional agent-based simulation is integrated to large-scale network simulation. Furthermore, travelers benefit from information provision not only from the ability to save time, but also from the value of certainty. Thus, various measures of network, path, and origin–destination (O–D)-based reliability need to be considered in this research. However, typical utility-maximizing behavior and equilibrium-based models are computationally expensive-to-evaluate those measures since the applications in a relatively large network can involve millions of agents. This is especially true when the evaluation involves multiple scenarios using simulation (e.g. in the case of information provision, different combinations of pre-trip and en-route information percentages available to agents).
Simulation tools, environments and frameworks for UAVs and multi-UAV-based systems performance analysis (version 2.0)
Published in International Journal of Modelling and Simulation, 2023
Aicha Idriss Hentati, Lamia Chaari Fourati, Essia Elgharbi, Sana Tayeb
The network simulator [version 2), known as NS-2, [22], is an event-driven simulation tool, it has proven to be very useful for studying the dynamic characteristics of communication networks. NS-2 can simulate wired and wireless network functions and protocols [such as routing algorithms, TCP, UDP …]. NS-2 provides users with a way to specify and simulate such network protocols and their corresponding behavior.