Special issue introduction

This special issues highlights research advances that were presented at the Thirteenth International Conference on Pervasive Computing and Communications (PerCom 2015). The PerCom 2015 conferencewas held in St. Louis March 23–27, 2015, and provided an opportunity for researchers to present their original ideas and results, to relate practical development experiences, and to spark ideas for new research directions and collaborations. Selection of papers for this special issue was very competitive. A total of 243 papers were submitted to the PerCom conference and were reviewed by our 99 technical committee members. The acceptance rate for the conference was 12%. The papers presented state-of-the-art ideas in activity sensing, sensor data mining, energy analytics, mobile assistance, pervasive health, pervasive transportation, sensing applications, and security. In the case of the eleven full-length papers that were accepted for the conference, we invited the authors to submit an extended version of their more recent work for possible publication in this special issue. We received nine submissions and selected six of these papers for inclusion in this special issue. The first paper of this special issue introduces a pervasive computing-based approach to enhancing exercising through video games. In the paper ‘‘User-optimized activity recognition for exergaming’’, BobakMortazavi, Mohammad Pourhomayoun, Sunghoon Lee, Suneil Nyamathi, Brandon Wu, and Majid Sarrafzadeh use wearable sensors to monitor movements in their augmented reality soccer exergame called SoccAR. The authors found that not only can multiple classifiers be combined to improve motion recognition performance based on leave-one-subject-out analysis, but user-specific optimization can be integrated to further improve recognition by adapting to the characteristics of each user. Such activity recognition can heighten the success of a mobile game and thus serve as a mechanism for improving fitness through virtual gaming. The next four papers focus on pervasive computing in the context of smart phones. The second paper in this special issue, ‘‘Towards attention-aware adaptive notification in smart phones’’ by Tadashi Okoshi, Hiroki Nozaki, Jin Nakazawa, and Hideyuki Tokuda, analyzes the timing of smart phone notifications. These authors recognize that interrupting users who are in the middle of a task is likely to reduce the effectiveness of the notification as well as introduce task errors and increase the user’s cognitive load. To ameliorate the impact of smart phone interruptions, the authors introduce a middleware called Attelia that detects breakpoints in a user’s interactions. Attelia relies on supervised learning of breakpoint situations based on data that is collected by the phone itself. Evaluation of Attelia for 30 users indicates that not only can the middleware detect breakpoints but information delivery that occurs during these breakpoints reduces frustration over random timings. In the third paper titled ‘‘Robust and ubiquitous smartphone-based lane detection’’, Heba Aly, Anas Basalamah, Moustafa Youssef utilize smart phones to provide intelligent transportation assistance. These authors introduce the LaneQuest system that detects the position of a vehicle with respect to the lanes on a road. This technology can provide valuable assistance to manual and automated drivers. LaneQuest utilizesmultiple sources of information inside and outside the vehicle to enhance the robustness of its position estimation. The authors report that not only does LaneQuest detect the correct car lane in 92% of the evaluation cases but it does so with a very small energy footprint, making it easily usable for Android devices. Energy usage of mobile devices was a common theme for this year’s PerCom conference. While the previous paper designed a system to be energy efficient, the fourth paper in this special issue analyzes existing interaction modalities to explicitly determine their impact on energy consumption. Fangzhou Jiang, Eisa Zarepour, Mahbub Hassan, Aruna Seneviratne, and Prasant Mohapatra consider the text input modalities of soft keyboard input, speech to text, and swyping in their paper ‘‘Type, talk, or swype: Characterizing and comparing energy consumption of mobile input modalities.’’ The authors find that not only do these modalities consume different amounts of battery energy but their consumption is dependent upon factors such as the length of the user input. In addition, they estimate the percentage of battery usage that can be expected for typical scenarios when one of these modalities is used constantly. The fifth paper in this special issue builds a more general model of smart phone battery consumption. In the paper ‘‘Constella: Crowdsourced system setting recommendations for mobile devices’’, Ella Peltonen, Eemil Lagerspetz, Petteri