The prevalence of fiber optic cable failures has been identified as a key contributor to failures across multiple network systems in the realm of network operations and maintenance. Meanwhile, with the continued expansion of fiber optic cable adoption and its widening coverage, the task of maintenance has grown increasingly complex for network operators. In the context of today’s highly efficient, expansive, and intricate networks, the limitations of conventional methods for operating and maintaining fiber optic cables have become unmistakably evident. To meet the pressing need for cost reduction, this paper introduces an innovative optical cable fault location method, leveraging automation and artificial intelligence technology. Tailored for practical network applications, this method conducts a comprehensive analysis of concurrent faults across multiple systems, pinpointing their origin in optical cable faults. Through the implementation of diverse navigation strategies, it achieves automatic optical cable fault location. The adoption of this method delivers a substantial reduction in fault location time, diminishing it from minutes to mere milliseconds. This dramatic increase in location speed leads to significant labor cost reductions, thereby greatly enhancing the efficiency of network maintenance.

Satellite and ground systems have traditionally operated independently, each providing unique services. However, the integration of satellite-ground networks necessitates collaboration to ensure consistent Quality of Service (QoS) for users in the Satellite-Ground Integrated Network (SGIN). This paper focuses on achieving highly reliable shared QoS orchestration for SGINs. We analyze eight typical SGIN services, including ultra-long-distance real-time, densely connected, high-speed mobile broadband, real-time large-capacity transmission, non-real-time large-capacity transmission, highly secure, deterministic guarantee transmission, and high-reliability guaranteed transmission services. The constantly changing SGIN topology, due to rapid inter-satellite link switching, prompts us to explore data-driven multi-service identification using Artificial Intelligence (AI) techniques. For accurate identification, two approaches are proposed in this paper: a Decision Tree (DT)-based scheme and an ensemble learning-based scheme. Additionally, we present the construction of a shared QoS system in SGINs, along with the evaluation and assurance of AI services using the concept of QoS of the SGIN AI service (QoSGINAI). Simulation results demonstrate the effectiveness of the proposed AI-based multi-service identification schemes.

It is difficult to find the root cause alarms by the traditional operation and maintenance methods quickly and accurately, due to the complex architecture of the 5G core network. At present, the library of cross-layer alarm association rules for the 5G core network is not complete. Therefore, the cross-layer alarm association rules mining algorithm of 5G core network is proposed in this paper, which is based on OPTICS algorithm and the weighted FP-growth algorithm. Initially, a time-correlated alarm transaction set is generated by OPTICS algorithm, and then a spatially correlated alarm transaction set is further extracted according to the three-layer topology. Finally, the weighted FP-growth algorithm is used to generate cross-layer alarm association rules. According to the experimental results, it suggests that the alarm transaction set is effectively compressed, and the accuracy of alarm rules mining is improved greatly by this algorithm.

Future networks are towards autonomous, with a high level of automatic and intelligent abilities. There are several domains and layers in operator networks, malfunctions can be transmitted from lower layers to upper layers, and from one domain to another domain. At present, cross-domain network malfunctions are mainly relied on the operation and maintenance staff of each professional network to analyze and dispatch orders, resulting in repeated orders and increased human cost. The first and important step of malfunction diagnosis is the construction of network topology. However, cross-domain network topology cannot be associated automatically at present. Based on the performance data, a new method using AI technologies is proposed in this paper, which can associate the connecting cross-domain network ports automatically. The principle is that a same time sequence similarity is shared by the connected ports. Taking the data from real networks and comparing with the existing topology, the connecting relations can be 100% correctly recognized. This method can be widely used to any cross-domain networks, without changing current network equipment.

With the trend of 5G-A (5G-Advanced) harmonized network communications and sensing harmonized communication and sensing, network capability exposure technology will help operators, business providers and 3rd business parties to realize harmonized network communications and sensing harmonized communication and sensing business and applications. This paper will focus on capability exposure technology based on 5G-A harmonized communication and sensing. Initially, this paper discusses the capability exposure hierarchical architecture based on harmonized communication and sensing, secondly proposes the harmonized communication and sensing network architecture and basic network signaling process combined with capability exposure technology. Then, this paper discusses capability exposure application scenarios based on communication sensing. This paper can provide relevant reference for the technological evolution, network deployment and application discussion of the harmonized communication and sensing capability in the operator network.