This paper proposed two neighbor discovery schemes in the wireless D2D networks using directional antennas. The neighbor discovery process is modeled as a deterministic estimator learning automaton. Pursuit algorithm and Generalized Pursuit algorithm are used to further improve the efficiency of neighbor discovery. The nodes in the network adjust the direction of the directional antennas through learning the feedback given by the environment in the history discovery process. Finally, OPNET is used to simulate the proposed schemes, and the simulation results show that the proposed two schemes in this paper can effectively improve the efficiency of neighbor discovery.

中文翻译：

---

具有定向天线的D2D网络基于确定性估计器学习的基于自动机的邻居发现方案

本文提出了使用定向天线的无线D2D网络中的两种邻居发现方案。邻居发现过程被建模为确定性估计器学习自动机。追踪算法和广义追踪算法被用来进一步提高邻居发现的效率。网络中的节点通过学习历史发现过程中环境给出的反馈来调整定向天线的方向。最后，利用OPNET对提出的方案进行了仿真，仿真结果表明，本文提出的两种方案可以有效地提高邻居发现的效率。