MOCHA: A Tool for Mobility Characterization

Author(s)

Souza, Fabricio R. ; Domingues, Augusto C. S. A. ; Vaz de Melo, Pedro O. S. ; Loureiro, Antonio A. F. ; ASSOC COMP MACHINERY

• Document type: Journal article
• Source: MSWIM'18: PROCEEDINGS OF THE 21ST ACM INTERNATIONAL CONFERENCE ON MODELING, ANALYSIS AND SIMULATION OF WIRELESS AND MOBILE SYSTEMS; v. N/A, p. 8-pg., 2018-01-01.

Abstract

There are many mobility models in the literature with diverse formats and origins. Besides the existence of studies that analyze and characterize these models, there is a need for a framework that can compare them in an easy way. MOCHA (Mobility framework for CHaracteristics Analysis) is a tool that characterizes and makes possible the comparison of mobility models without any hard work. We implemented 9 social, spatial and temporal characteristics, which were extracted from various (real and synthetic) distinct mobility traces. MOCHA has a classifying module that attributes each characteristic the statistic distribution that better describes it. As a validation process, all the traces were compared using the T-SNE method for data visualization, resulting in the approximation of similar traces. One of the advantages of using MOCHA is its ease of use, being able to read diverse traces formats and converting them to its standard format, allowing that different types of traces, such as check-in, GPS, contacts, and so on, to be compared. The metrics used in the tool can become a standard for trace analysis and comparison in the literature, allowing a better vision of where one trace belongs related to others. MOCHA is available for download at https://github.com/wisemap-ufmg/MOCHA (AU)

• FAPESP's process: 15/24494-8 - Communications and processing of big data in cloud and fog computing
• Grantee: Nelson Luis Saldanha da Fonseca
• Support Opportunities: Research Projects - Thematic Grants