Predictive communication network management based on real-time weather patterns

This technology is a machine learning-based algorithm that predicts weather-induced signal degradation in wireless communication networks and proactively manages network resources to maintain quality of service.

Unmet Need: Automated network management for weather-related signal attenuation in wireless networks.

Current 5G and next-generation wireless networks increasingly rely on millimeter-wave (mmWave) frequencies for high-bandwidth connectivity and cellular infrastructure. However, current millimeter-wave (mmWave) and wireless networks are heavily influenced by weather, specifically rain, which can severely which can degrade or completely disrupt network links. Existing network management systems react to disruptions only after they occur, leading to service interruptions and poor Quality of Service (QoS). There is no solution on the market that can predict weather-induced network degradation and proactively adapt network routing, power, and modulation before disruptions occur.

The Technology: Predictive network management using in-network real-time weather sensing

This technology enables the predictive management of wireless networks by leveraging a network's own commercial microwave links (CMLs) to detect and preempt weather-related signal disturbances in real time. A deep learning model analyzes signal attenuation patterns across the network to forecast conditions at each link. Based on these predictions, the system proactively manages network resources through dynamic rerouting, adaptive transmission power control (ATPC), adaptive modulation and coding (AMC), and channel switching, before weather-related network degradation occurs. This approach is self-contained and does not require external weather data sources such as radar.

The technology has been validated using real-world data from a smart-city mmWave network deployment.

Applications:

• Network resource allocation during weather events
• 5G and next-generation cellular network management
• Smart-city wireless infrastructure optimization and resilience
• Critical communication systems (autonomous vehicles, emergency services, public safety)
• Military and government communication networks
• Real-time weather and rainfall monitoring

Advantages:

• Real-time weather prediction
• Proactive and preventative network management
• Maintains high network uptime and QoS through adverse weather
• Self-contained (uses the network’s own CMLs)

Lead Inventor:

Gil Zussman, Ph.D.

Patent Information:

Patent Issued (US 12,556,937)

Related Publications:

• Jacoby D, Ostrometzky J, Messer H. "Short-Term Prediction of the Attenuation in a Commercial Microwave Link Using LSTM-based RNN." 28th European Signal Processing Conference (EUSIPCO), 2021, pp. 1628-1632

• Ostrometzky J, Raich R, Bao L, Hansryd J, Messer H. "The Wet-Antenna Effect—A Factor to be Considered in Future Communication Networks." IEEE Transactions on Antennas and Propagation, 2018 Jan 66(1), pp. 315-322

• J. Ostrometzky and H. Messer, "Dynamic Determination of the Baseline Level in Microwave Links for Rain Monitoring From Minimum Attenuation Values." IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing. 2018 Jan. 11(1) pp. 24-33

Tech Ventures Reference:

• IR CU19208

• Licensing Contact: Greg Maskel