Abstract
Mobile data offloading can greatly decrease the load on and usage of current and future cellular data networks by exploiting opportunistic and frequent access to Wi-Fi connectivity. Unfortunately, Wi-Fi access from mobile devices can be difficult during typical work commutes, e.g., via trains or cars on highways. In this paper, we propose a new approach: to preload the mobile device with content that a user might be interested in, thereby avoiding the need for cellular data access. We demonstrate the feasibility of this approach by developing a supervised machine learning model that learns from user preferences for different types of content, and propensity to be guided by the user interface of the player, and predictively preload entire TV shows. Testing on a data set of nearly 3.9 million sessions from all over the U.K. to BBC TV shows, we find that predictive preloading can save over 71% of the mobile data for an average user.
Highlights
I NTERNET video services are increasingly going mobile
A recent study of 48 metro systems from 28 countries suggests that the lack of good Internet connectivity underground is a common problem for many developed cities across the globe [31]. To address these difficulties in finding opportunistic Wi-Fi during commutes, we propose predictive preloading, a new approach to mobile data offloading for current and future generation 5G networks: In contrast to predicting mobility patterns, we propose to predict the content that a user is likely to watch during the commute and preload that content on her mobile device in advance, when she might have access to reliable Wi-Fi connectivity, with sufficient spare bandwidth, 1Throughout this paper, we use Wi-Fi to denote access through a fixed-line broadband connection, potentially via a Wi-Fi access point
We find that a vast majority of users are influenced by the User Interface of the video player, and tend to access items which are featured by the BBC content editors on the iPlayer homepage, or access items on “most popular” lists (25% of accesses by average users and more than 80% of accesses by the top 10% of users are for such content items, which appear prominently on the user interface)
Summary
I NTERNET video services are increasingly going mobile. Conveniences offered by high bandwidth mobile networks and the availability of dedicated mobile video apps have raised the volume of per-user mobile video traffic by an incredible 262% in recent years [7]. A recent study of 48 metro systems from 28 countries suggests that the lack of good Internet connectivity underground is a common problem for many developed cities across the globe [31] To address these difficulties in finding opportunistic Wi-Fi during commutes, we propose predictive preloading, a new approach to mobile data offloading for current and future generation 5G networks: In contrast to predicting mobility patterns, we propose to predict the content that a user is likely to watch during the commute and preload that content on her mobile device in advance, when she might have access to reliable Wi-Fi connectivity, with sufficient spare bandwidth, 1Throughout this paper, we use Wi-Fi to denote access through a fixed-line broadband connection, potentially via a Wi-Fi access point. Our results suggest that predictive preloading allows to offload up to 71% of mobile data usage for an average user (over 95% for top 10% of users) and significantly outperforms naïve greedy techniques (which can only save ≈ 22% of per-user mobile data on average)
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
