Department of Information Strategy, Electronics and Telecommunications Research Institute, 218 Gajeong-ro, Yuseong-gu, Daejeon 34129, Korea.
Department of Computer Engineering, Chosun University, 309 Pilmun-daero, Dong-gu, Gwangju 61452, Korea.
Sensors (Basel). 2020 Feb 19;20(4):1128. doi: 10.3390/s20041128.
In the last few years, multicast device-to-device (D2D) cellular networks has become a highly attractive area of research. However, a particularly challenging class of issues in this area is data traffic, which increases due to increase in video and audio streaming applications. Therefore, there is need for smart spectrum management policies. In this paper, we consider a fractional frequency reuse (FFR) technique which divides the whole spectrum into multiple sections and allows reusing of spectrum resources between the conventional cellular users and multicast D2D users in a non-orthogonal scenario. Since conventional cellular users and multicast D2D users shared same resources simultaneously, they generate severe data traffic and high communication overhead. To overcome these issues, in this paper we propose Lagrange relaxation technique to solve the non-convex problem and combinatorial auction-based matching algorithm to select the most desirable resource reuse partners by fulfilling the quality of service (QoS) requirements for both the conventional cellular users and multicast D2D users. Then, we formulate an optimization problem to maximize the overall system performance with least computational complexity. We demonstrate that our method can exploit a higher data rate, spectrum efficiency, traffic offload rate, coverage probability, and lower computational complexity.
在过去的几年中，多播设备到设备（D2D）蜂窝网络已成为一个极具吸引力的研究领域。然而，在这个领域中一个特别具有挑战性的问题是数据流量，由于视频和音频流应用程序的增加而增加。因此，需要智能的频谱管理策略。在本文中，我们考虑一种分数频率复用（FFR）技术，该技术将整个频谱划分为多个部分，并允许在非正交场景中在传统蜂窝用户和多播 D2D 用户之间复用频谱资源。由于传统蜂窝用户和多播 D2D 用户同时共享相同的资源，因此它们会产生严重的数据流量和高通信开销。为了克服这些问题，在本文中，我们提出了拉格朗日松弛技术来解决非凸问题，并提出了基于组合拍卖的匹配算法，通过满足传统蜂窝用户和多播 D2D 用户的服务质量（QoS）要求，选择最合适的资源重用伙伴。然后，我们制定了一个优化问题，以最小的计算复杂度最大限度地提高整体系统性能。我们证明了我们的方法可以利用更高的数据速率、频谱效率、流量卸载率、覆盖概率和更低的计算复杂度。
