In this paper, a new mode selection (MS) scheme for device-to-device (D2D) enabled cellular networks is proposed. The MS is carried out using a predefined signal-to-interference plus noise ratio threshold and it aims to guarantee a minimum level of quality of service.With the new scheme, interfering effects are taken into consideration, while realistic assumptions for the acquisition of the channel information are made. Assuming a single-user case, an analytical framework is developed, which is based on the Markov-chain theory, and is used to study the system's performance. The various performance results that have been obtained reveal that with the proposed MS, the probability of mode switching can be reduced (as compared to other systems) and thus, an increase on the overhead and signal processing is avoided. Moreover, the new approach is applied in a multi-user communication scenario, where a joint MS, resource allocation (RA), and scheduling optimization problem is formulated. The solution to this problem is based on a greedy heuristic algorithm that properly assigns the available resources among cellular users and D2D pairs. The joint MS, RA, and scheduling algorithm considers limited SINR awareness and it provides significant performance improvement in terms of sum-rate.
- Device-to-device communication
- Markov chain
- Performance analysis
- Resource allocation
- SINR-aware mode selection