Section: New Results
Data offloading decision via mobile crowdsensing
Participants : Emanuel Lima, Aline Carneiro Viana, Ana Aguiar [FEUP (Portugal) - Dept. of Electrical and Computer Engineering] , Paulo Carvalho [FEUP (Portugal) - Dept. of Electrical and Computer Engineering] .
According to Cisco forecasts (https://www.cisco.com/c/en/us/solutions/collateral/service-provider/visual-networking-index-vni/mobile-white-paper-c11-520862.html), mobile data traffic will grow at a compound annual growth rate of 47 % from 2016 to 2021 with smartphones surpassing four-fifths of mobile data traffic. It is known that mobile network operators are struggling to keep up with such traffic demand, and part of the solution is to offload communications to WiFi networks. Mobile data offloading systems can assist mobile devices in the decision making of when and what to offload to WiFi networks. However, due to the limited coverage of a WiFi AP, the expected offloading performance of such a system is linked with the users mobility. Unveiling and understanding human mobility patterns is a crucial issue in supporting decisions and prediction activities for mobile data offloading.
Several studies on the analysis of human mobility patterns have been carried out focusing on the identification and characterization of important locations in users' life in general. We intend to extend these works by studying human mobility from the perspective of mobile data offloading. This brings two major differences compared to the related work. First, high temporal resolution of positioning datasets is needed. In the majority of the related work, important locations have a temporal dimension representing the time spent by a user in that location, which confers its degree of importance. This time is usually in the order of several minutes which is suitable for the case of detecting important locations but not for a mobile data offloading scenario. Here, according to the amount of data traffic that needs to be offloaded, locations with a visiting temporal resolution of few seconds may be enough for data offloading. Thus, we expect to discover additional offloading opportunities, which were not visible with a coarser temporal resolution. Second, while important locations are usually limited in size, offloading locations can have any arbitrary shape and size.
In this work, offloading regions are defined as spatially aggregated locations where users have mobility suitable to offload. The main contribution of this work are: (a) the identification of offloading regions on an individual basis through unsupervised learning; (b) the characterization of these regions in terms of availability, sojourn, and transition time based on their relevance; (c) the study of the impact of the users mobility on the design of mobile offloading systems. This work was published at ACM CHANTS 2018.
We now working on the extension of this work, which will incorporate the mobility prediction of the users. Such prediction is essential to the design of the decision offloading strategy. Such strategy will be used to allow a mobile phone of a user deciding if offload or not her traffic, i.e., when, where (in which offloading region) and how (if the traffic will be offloadied to one or more Access Points). This is an on-going work with the the PhD Emanuel Lima, who spent 4 months as an intern in our team, and his advisors.