Abstract
The rapid uptake of mobile devices and the rising popularity of mobileapplications and services pose unprecedented demands on mobile and wirelessnetworking infrastructure. Upcoming 5G systems are evolving to supportexploding mobile traffic volumes, agile management of network resource tomaximize user experience, and extraction of fine-grained real-time analytics.Fulfilling these tasks is challenging, as mobile environments are increasinglycomplex, heterogeneous, and evolving. One potential solution is to resort toadvanced machine learning techniques to help managing the rise in data volumesand algorithm-driven applications. The recent success of deep learningunderpins new and powerful tools that tackle problems in this space. In this paper we bridge the gap between deep learning and mobile and wirelessnetworking research, by presenting a comprehensive survey of the crossoversbetween the two areas. We first briefly introduce essential background andstate-of-the-art in deep learning techniques with potential applications tonetworking. We then discuss several techniques and platforms that facilitatethe efficient deployment of deep learning onto mobile systems. Subsequently, weprovide an encyclopedic review of mobile and wireless networking research basedon deep learning, which we categorize by different domains. Drawing from ourexperience, we discuss how to tailor deep learning to mobile environments. Wecomplete this survey by pinpointing current challenges and open futuredirections for research.