D2D communications are considered as effective means to achieve vehicle-to-vehicle (V2V) communications, which can be used to improve the performance in vehicular ad hoc networks (VANETs). In this paper, we investigate the resource allocation and interference management based on clustering mechanism in the D2D communications under-laying VANETs aiming to optimize resource utilization efficiency. Firstly, given the transmission resources of intra-cluster for D2D communications, we design a vehicle clustering algorithm to improve intra-cluster resource efficiency and nullify the intra-cluster interference. Secondly, by analyzing the wireless service in VANETs, a cooperative communication scheme with optimized resource utilization is proposed to maximize intra-cluster spectral efficiency. Thirdly, when uplink spectrums are shared, an interference neutralization (IN) model is developed to manage inter-cluster interference, while each vehicle is equipped with one antenna and there is no extra antenna for interference cancellation. In accordance with the IN model, a cross-layer optimization frame for multi-hop VANETs is proposed to maximize the end-to-end throughput of multiple coexisting communication sessions. To evaluate the performance, we select the method without cooperative communication and IN as a benchmark and compare it against the method with cooperative communication only and the method with both cooperative communication and IN. The simulation results show that the method that uses cooperative communication and IN can dramatically increase the throughput. Furthermore, the throughput gains increase along with vehicle density growing in VANETs.

中文翻译：

---

VANET 中多跳 D2D 通信的频谱效率优化和干扰管理


D2D 通信被认为是实现车对车（V2V）通信的有效手段，可用于提高车载自组织网络（VANET）的性能。在本文中，我们研究了D2D通信底层VANET中基于集群机制的资源分配和干扰管理，旨在优化资源利用效率。首先，考虑到D2D通信的簇内传输资源，我们设计了一种车辆聚类算法，以提高簇内资源效率并消除簇内干扰。其次，通过分析车载自组网中的无线服务，提出了一种优化资源利用的协作通信方案，以最大化簇内频谱效率。第三，当共享上行链路频谱时，开发了干扰中和（IN）模型来管理簇间干扰，而每辆车配备一根天线，并且没有额外的天线用于干扰消除。根据IN模型，提出了一种多跳VANET的跨层优化框架，以最大化多个共存通信会话的端到端吞吐量。为了评估性能，我们选择没有协作通信和智能的方法作为基准，并将其与仅使用协作通信的方法和同时使用协作通信和智能的方法进行比较。仿真结果表明，采用协作通信和智能网的方法可以显着提高吞吐量。此外，吞吐量增益随着 VANET 中车辆密度的增加而增加。