Flexible synthetic mobility modeling to discover trajectories for complex areas of mobile wireless networks

Abstract

Mobility modeling is one of the most influential pillars in improving the performance of wireless networks. Understanding mobility features is relevant for the design and analysis of proper motion schemes for any network. Up to now, a variety of entity mobility models have been suggested which do not take into account all the characteristics of real-life movements (time, space, environment). In order to obtain an improved model that overcomes such limitations, in this paper a new hybrid synthetic entity mobility model is proposed, called the Maze mobility model (MzMM). It takes into consideration the timeline history, spatial dependencies, and geographical restrictions as well. At the same time, it respects the laws of motion to reflect actual scenarios using a flexible discovery mechanism, where it permits nodes arriving at predefined destinations following the most appropriate trajectories in the presence of several obstacles. This approach allows mobile nodes to move correctly even in the presence of múltiple mobility constraints. The significance of this research is that the new approach considers a realistic combination of parameters to achieve a flexible and robust mobility model that can be applied for autonomous or human mobility, even in complex environments to provide optimized performances for networks, as demonstrated by its high performance results.

DMLAR: Distributed Machine Learning-Based Anti-Collision Algorithm for RFID Readers in the Internet of Things

“…Considering a healthcare infrastructure as a base for our deployment simulation, we considered that our RFID network included sensor tags carried by patients and RFID readers by medical staff. We conducted MATLAB simulations for the reader mobility models (Figure 7) [51]. After the collection and learning phase, reader R1 initiates the first test at period P1 by selecting data channel F1 and time slot TS1, and then calculates the number of readers that can be in RRI and RTI (Figure 6).…”

“…Considering a healthcare infrastructure as a base for our deployment simulation, we considered that our RFID network included sensor tags carried by patients and RFID readers by medical staff. We conducted MATLAB simulations for the reader mobility models (Figure 7) [51]. The goal was to evaluate our new anti-collision mechanism according to different criteria of mobility, resources and density.…”