Device-to-device communications are considered as a key feature to enhance the performance of the fifth generation (5G) wireless networks. Several radio access technologies such as LTE Direct, Bluetooth, Wi-Fi, and ZigBee are expected to provide the opportunity of D2D communications. Therefore, it is possible to choose any of them autonomously to establish a D2D link. The primary focus of this work is to investigate the radio interface selection, where end users select an interface opportunistically among different available radio interfaces to establish outband D2D connectivity against interference. We model a non-cooperative game to select a radio interface for D2D users to minimize their communication cost. We have investigated Nash equilibrium in the game and argue that without any cooperation users can achieve a balanced strategy. In our model, each pair selects a radio interface based on a utility function that associates communication performance and cost. Finally, we propose three heuristic algorithms: Social, Greedy, and Local, that achieve Nash equilibrium with different information. Event-driven simulation experiments are then conducted to evaluate the utility and cost of the equilibrium strategy. Our results confirm that the proposed schemes can increase the utility, lower the cost, and lead to higher efficiency in terms of achievable throughput per consumed energy. INDEX TERMS D2D communications, game theory, multiple radio interfaces, Nash equilibrium. I. INTRODUCTION Wireless communications have boosted the opportunity for smart devices with a number of standards and technologies. Smart devices are now the most important computing and communication platform. In previous years wireless connectivity was only possible with a single operator/ technology. However, these days smart devices are capable of multiple wireless opportunities. These end users are often equipped with multiple radio interfaces (i.e., 3G/LTE, Bluetooth, Zigbee, and Wi-Fi), which complements their cellular communication capabilities. According to a recent market research report, 70% of the mobile phones have Bluetooth interface, while 80% are enabled WiFi [1]. The proliferation of smart devices and exponential demand of bandwidth have created spacious performance requirement on the future wireless networks [2]. Device-to-device communication (D2D) [3] is considered one of the major technology to enhance the boosting demand [4] of users. Motivated by the performance gain, many telecommunication The associate editor coordinating the review of this manuscript and approving it for publication was Fang Yang.