Abstract
The Internet of Things is not just a theoretical topic anymore, but it has become a reality that affects everyday life in many aspects. Users and devices can now be connected to the Internet whenever, wherever and interact with each other in revolutionary ways. Mobile Crowdsensing (MCS) is a collaborative demonstration of this interaction. However, MCS introduced a series of challenges surrounding the areas of power management, privacy preservation, and data quality. These may result in compromised user experience, limited user acceptance, and minimise the potential this area has to offer. This study addresses the discussed problems by examining the impact Internet of Things (IoT) communication protocols have on modern mobile crowdsensing systems. To do so, an end-to-end crowdsensing system is designed to evaluate the most popular IoT protocols. Simulation and off-the- shelf-device-based experiment runs provided an insight on the performance of the said protocols. Based on the findings, a new sensing approach is introduced aiming to improve system’s robustness and minimise energy requirements. Finally, a user-preference efficient crowdsensing algorithm that determines the most efficient communication protocol based on user input and experiment parameters is proposed.