Radio Resource Allocation for Device-to-Device Communications Underlaying Cellular Networks
device-to-device communication , resource allocation , Cellular network , Communication network
Inexpensive connectivity and computing power have caused the number of communicating devices to explode in the last decade. New applications are emerging every day to take advantage of the proximity and abundance of these devices. Device-to-Device (D2D) communication using cellular spectrum to increase spectral efficiency of the network is a technology component of Long Term Evolution-Advanced (LTE-A). In D2D communication underlaying cellular networks, devices communicate with each other using a direct link using cellular resources without going through the evolved Node B (eNB) but remaining under the control of it. D2D communication is expected to be one of the prominent features supported by future cellular networks because of reusing the cellular spectrum to increase the system performance of cellular networks. However, due to the limitations of a licensed spectrum when these devices share a cellular spectrum to communicate directly among themselves, the same resource may need to be shared among cellular and D2D communicating pairs. This resource sharing introduces a new interference scenario, which needs to be coordinated through a new resource allocation scheme. We investigate this problem of interference coordination and explore three different perspectives from which this problem can be viewed, namely a) interference minimization; b) fair allocation while minimizing interference; c) Quality of Service (QoS) satisfaction while maximizing total system sum rate of the cellular network. We show that the proposed schemes are suitable for the short LTE scheduling period of 1 ms as they are computationally tractable. Our schemes can allocate radio resources to D2D pairs underlaying a cellular network in a short time, ensuring fairness in allocation while minimizing interference, and increasing the total system sum rate of the network while maintaining a QoS target.