Analysis and Modelling of Group-based Communication in Internet Applications for improved Edge Caching
Final Report Abstract
Group-based communication, in which a group of users communicates with each other, is a communication paradigm that is becoming increasingly popular on the Internet. Internet applications, e.g., mobile instant messaging (MIM) applications such as WhatsApp, extend the classic Internet communication models (one-to-one, one-to-many) through group communication, which makes for a large user base. Communication behavior changes with group-based communication; for example, communication speeds up due to the large number of participants, and sent data is duplicated for all recipients. In this project, group-based communication and its impact on communication networks were analyzed and modeled. As large data volumes in group-based communications create a lot of load, especially in mobile networks, network operators can use various mechanisms to optimize data traffic in the network. One such mechanism is edge computing, i.e., processing traffic at the edge of the network where it originates. This shortens access times for users and reduces the load on the core network. This is done, for example, by caching, i.e., by temporarily storing popular data at suitable locations. With edge caching, such caches are operated directly at the edge of the network, i.e., at the network operators' resources close to the users, e.g., at base stations of mobile cells in 5G networks. This allows data to be delivered directly from the cache to all users in the same mobile cell without traversing the entire network. This speeds up access for users and further reduces the amount of data on the network. In addition, direct device-to-device (D2D) transmission, which takes place directly between terminals, can be used when two users are close to each other. The project analyzed the communication structure in WhatsApp group chats, measured and modeled MIM application traffic, and implemented a framework for detailed simulation of MIM (group) communication. The resulting models for group communication (communication frequency, message types, group structure, contact network, and network traffic) were implemented in simulations. The framework allows realistic movement of communicating people, considering real traffic routes, and it allows the implementation of edge caching and D2D with realistic network infrastructure. The simulation framework can be used to evaluate the performance of current and future edge caching mechanisms. The results have shown that in a densely populated and therefore well-suited environment for group communication savings, there is a potential saving of over 10% of message volume by edge caching and about 1% by D2D. In addition, subjective studies have investigated the impact of groups on the subjective perception (Quality of Experience, QoE) of group videoconferencing.
Publications
-
Pandemic in the digital age: analyzing WhatsApp communication behavior before, during, and after the COVID-19 lockdown. Humanities and Social Sciences Communications, 9(1).
Seufert, Anika; Poignée, Fabian; Hoßfeld, Tobias & Seufert, Michael
-
Share and Multiply: Modeling Communication and Generated Traffic in Private WhatsApp Groups. IEEE Access, 11, 25401-25414.
Seufert, Anika; Poignée, Fabian; Seufert, Michael & Hoßfeld, Tobias
-
“WhatsApp Data Set.” Figshare dataset
Anika Seufert, Fabian Poignée, Tobias Hoßfeld, Michael Seufert
