Recent developments of self-interference suppression techniques have enabled practical implementations of full-duplex (FD) cognitive-radio (CR) communication systems. Such systems can significantly enhance spectrum utilization by allowing a CR user to simultaneously transmit and receive over the same frequency channel. However, the CR FD-capabilities challenge the effectiveness of existing CR-based routing protocols as joint FD-aware channel-assignment and routing are essential to enhance network performance. In this letter, we propose a joint FD-aware channel-assignment and route selection protocol in FD-based CR networks (CRNs) under time-varying channel conditions and transmission rates. Specifically, for a given set of paths between a CR source-destination pair, our protocol computes the channel-assignment over each path that maximizes the end-to-end network throughput subject to interference constraints. This assignment problem is shown to be an NP-hard binary linear programming (BLP) that can be sub-optimally solved in polynomial-time using the sequential-fixing procedure. Then, our protocol determines the path with the highest end-to-end network throughput. Simulation results reveal that our proposed routing protocol significantly improves the end-to-end throughput compared to previous FD-aware routing protocols.

中文翻译：

---

不同信道条件下全双工认知网络中的智能频谱分配路由

自干扰抑制技术的最新发展使全双工 (FD) 认知无线电 (CR) 通信系统的实际实现成为可能。通过允许 CR 用户在同一频率信道上同时发送和接收，此类系统可以显着提高频谱利用率。然而，CR FD 功能对现有基于 CR 的路由协议的有效性提出了挑战，因为联合 FD 感知信道分配和路由对于提高网络性能至关重要。在这封信中，我们在时变信道条件和传输速率下提出了基于 FD 的 CR 网络 (CRN) 中的联合 FD 感知信道分配和路由选择协议。具体来说，对于 CR 源-目标对之间的一组给定路径，我们的协议计算每条路径上的信道分配，以最大化受干扰约束的端到端网络吞吐量。这个分配问题被证明是一个 NP 难二元线性规划 (BLP)，可以使用顺序固定程序在多项式时间内次优地解决。然后，我们的协议确定具有最高端到端网络吞吐量的路径。仿真结果表明，与以前的 FD 感知路由协议相比，我们提出的路由协议显着提高了端到端吞吐量。我们的协议确定具有最高端到端网络吞吐量的路径。仿真结果表明，与以前的 FD 感知路由协议相比，我们提出的路由协议显着提高了端到端吞吐量。我们的协议确定具有最高端到端网络吞吐量的路径。仿真结果表明，与以前的 FD 感知路由协议相比，我们提出的路由协议显着提高了端到端吞吐量。