We propose a discrete and average approach, focusing on the code set rather than individual codes, for the PON monitoring system based on the optical frequency hopping/periodic code scheme. The discrete and averaged codes are designed to support large-capacity network monitoring, particular for next-generation PONs, while preserving the original encoder structure of the PON monitoring system. The proposed approach discretizes and averages optical frequency hopping/periodic codes with small and large intervals, employing a modified greedy algorithm that mitigates falling into local optimality and an efficient backtracking algorithm, respectively. Our proposed algorithms demonstrate exceptional efficiency and performance, as exemplified by the modified greedy algorithm, which surpasses the exhaustive method by over 2 millionfold in terms of speed. This approach significantly reduces the more rigorously defined probability of multiple-customer interference and lowers the final recognition difficulty, enhancing the practicality of the PON monitoring system. Additionally, we develop a cost-effective model to analyze optimal system costs for discrete-averaged codes in the cost-sensitive PON market, facilitating the practical application of the PON monitoring system in engineering practice.

中文翻译：

---

基于OFH/PC方案的PON监控系统增强实用性的优化研究

我们为基于光跳频/周期码方案的 PON 监控系统提出了一种离散平均方法，重点关注代码集而不是单个代码。离散和平均编码旨在支持大容量网络监控，特别是下一代PON，同时保留PON监控系统的原始编码器结构。所提出的方法分别采用改进的贪婪算法和有效的回溯算法，以小间隔和大间隔对光学跳频/周期码进行离散化和平均化，以减轻陷入局部最优的情况。我们提出的算法展示了卓越的效率和性能，例如修改后的贪婪算法，其速度比穷举方法高出 200 万倍以上。这种方法显着降低了更严格定义的多客户干扰概率，降低了最终识别难度，增强了PON监控系统的实用性。此外，我们开发了一个成本效益模型来分析成本敏感的PON市场中离散平均码的最佳系统成本，促进PON监控系统在工程实践中的实际应用。