Emergency communication network self-repairing method and system based on federated learning

By employing a self-healing method for distributed emergency communication networks based on federated learning, combined with multimodal data fusion and self-healing strategies, the problem of rapid response and autonomous repair of emergency communication systems under natural disasters is solved, improving the network's resilience and endurance while ensuring data privacy and security.

CN122120113APending Publication Date: 2026-05-29HUTCHISON CAPITAL TECHNOLOGY (SHENZHEN) CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
HUTCHISON CAPITAL TECHNOLOGY (SHENZHEN) CO LTD
Filing Date
2026-02-08
Publication Date
2026-05-29

AI Technical Summary

Technical Problem

Existing emergency communication systems face challenges such as limited data sources, weak endurance, reliance on manual troubleshooting for fault diagnosis, lack of intelligent network topology reconstruction, and contradictions between data privacy protection and information sharing when confronted with the combined challenges of "road outages, power outages, and network outages" caused by natural disasters. These issues make it difficult to achieve rapid response and autonomous repair.

Method used

A self-healing method for distributed emergency communication networks based on federated learning is adopted. A cluster is formed by portable terminals, UAV relays, vehicle-mounted nodes and sensor nodes to perform multimodal data fusion and spatiotemporal feature extraction, thereby realizing fault diagnosis and predictive maintenance. Combined with self-healing strategies such as communication protocol switching, power consumption mode adjustment and network topology reconstruction, self-healing strategies are dynamically generated and model parameters are optimized.

Benefits of technology

It enables rapid response and autonomous repair of faults, enhances network resilience and endurance, ensures data privacy and security, adapts to complex terrain and extreme environments, and improves the reliability and efficiency of emergency communications.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122120113A_ABST
    Figure CN122120113A_ABST
Patent Text Reader

Abstract

The application relates to the technical field of emergency communication, in particular to an emergency communication network self-repairing method and system based on federal learning, which solves the communication guarantee in the three-break scenarios of circuit break, power failure and network break. The method comprises the following steps: deploying a distributed emergency communication node cluster; training a fault diagnosis model independently based on a fault feature library, and generating local model parameters; calculating aggregation weights according to the historical diagnosis accuracy, data quality and residual power of the nodes, generating global model parameters and issuing them; diagnosing faults based on the updated model, identifying network break, power failure, circuit break or composite faults; executing self-repairing strategies such as communication protocol switching, power consumption mode adjustment, node role switching and network topology reconstruction according to the fault type; and taking the repair effect data as new training samples to iteratively optimize the model. The application improves the invulnerability and self-adaptive ability of the emergency communication network in extreme environments.
Need to check novelty before this filing date? Find Prior Art