Edge computing offers a promising paradigm for implementing the industrial Internet of things (IIoT) by offloading intensive computing tasks from resource constrained machine type devices to powerful edge servers. However, efficient spectrum resource management is required to meet the quality of service requirements of various applications, taking into account the limited spectrum resources, batteries, and the characteristics of available spectrum fluctuations. Therefore, this study proposes intelligent dynamic spectrum resource management consisting of learning engines that select optimal backup channels based on history data, reasoning engines that infer idle channels based on backup channel lists, and transmission parameter optimization engines based genetic algorithm using interference analysis in time, space and frequency domains. The performance of the proposed intelligent dynamic spectrum resource management was evaluated in terms of the spectrum efficiency, number of spectrum handoff, latency, energy consumption, and link maintenance probability according to the backup channel selection technique and the number of IoT devices and the use of transmission parameters optimized for each traffic environment. The results demonstrate that the proposed method is superior to existing spectrum resource management functions.

中文翻译：

---

基于时空频域感知数据的智能动态频谱资源管理

边缘计算通过将密集计算任务从资源受限的机器类型设备卸载到强大的边缘服务器，为实现工业物联网 (IIoT) 提供了一种有前景的范例。然而，需要有效的频谱资源管理来满足各种应用的服务质量要求，同时考虑到有限的频谱资源、电池以及可用频谱波动的特点。因此，本研究提出了智能动态频谱资源管理，包括根据历史数据选择最佳备份信道的学习引擎，根据备份信道列表推断空闲信道的推理引擎，以及基于及时干扰分析的遗传算法的传输参数优化引擎，空间域和频域。根据备用信道选择技术和物联网设备的数量以及使用针对每个交通环境优化的传输参数。结果表明，所提出的方法优于现有的频谱资源管理功能。根据备用信道选择技术和物联网设备的数量以及针对每个流量环境优化的传输参数的使用，以及链路维护概率。结果表明，所提出的方法优于现有的频谱资源管理功能。根据备用信道选择技术和物联网设备的数量以及针对每个流量环境优化的传输参数的使用，以及链路维护概率。结果表明，所提出的方法优于现有的频谱资源管理功能。