Crowdsourcing for Mobile Networks and IoT
Wireless Communications and Mobile Computing
As the deep integration of ubiquitous sensors, intelligent devices, and social networks, mobile networks and IoT are formed by the opportunity of virtual mobile communication networks and social communities between mobile carriers. People involved in a mobile network can easily interact and share information with each other anytime and anywhere through the popular use of intelligent devices. As a result, there is a remarkable trend to enable crowdsourcing for mobile networks and IoT to address
... arious problems that involve real-time collection, processing, and collaborations among participants in highly mobile environments. Thus, crowdsourcing could be an efficient strategy to improve quality and user experiences of applications in mobile networks and IoT, which not only potentially brings enormous benefits for economics but also leads to revolution for our daily life. The embedded sensors including accelerometer, compass, gyroscope, GPS, microphone, and camera in mobile phones are leveraged to gather the required information to support location-based services, for example, environmental measurements, personal activity sharing, and online recommendation. In this special issue on crowdsourcing for mobile networks and IoT, we have invited some papers that address such issues. L. Nie et al. introduce a network traffic prediction method in wireless mesh backbone networks based on deep learning and spatiotemporal compressive architecture. This method applies discrete wavelet transform to extract the low-pass component of network traffic. The performance of this method is verified by comparing with three widely used traffic prediction methods. The paper entitled "CPSFS: A Credible Personalized Spam Filtering Scheme by Crowdsourcing" proposes a credible personalized spam filtering scheme and classifies spam into two categories, that is, complete-spam and semispam, before filtering them. According to the social trust and interest similarity, complete-spam can be filtered by the Bayesian filtering, and semispam can be estimated by crowdsourcing at the client side. In order to maximize network utility, H. Meng et al. demonstrate an optimal real-time pricing strategy for computing resource management in mobile crowdsourcing. Furthermore, the existence of real-time prices is proved, which can align individual optimality with systematic optimality. H. Zhu et al. allocate the sharing resource to users across the network edge. A novel architecture is proposed to share resource of physical customer-premised equipment nodes across the network edge and assign virtual customerpremised equipment instances to a cost-efficient node. The paper entitled "An SAT-Based Method to Multithreaded Program Verification for Mobile Crowdsourcing Networks" presents a novel IC3-based algorithm on the safety verification of the multithreaded programs for mobile crowdsourcing networks. The performance of the proposed algorithm is evaluated by the SAT-based model checking algorithms, focusing on memory consumption. Y. Ye et al. introduce a color distribution pattern metric method, concentrating on reidentification in video searching for surveillance and forensic fields in crowdsourcing IoTs. Performance evaluations show that the presented method on different datasets can obtain higher network accuracy.