ENHANCING 5G NON-TERRESTRIAL NETWORK (NTN) COMMUNICATIONS WITH MACHINE LEARNING-DRIVEN NON-LINE-OF-SIGHT (NLOS) PREDICTION AND MITIGATION

Aug 5, 2026

This innovation leverages machine learning to predict Non-Line-of-Sight (NLOS) conditions in 5G Non-Terrestrial Networks (NTN), improving communication reliability for satellite-based connectivity. Predicted NLOS behavior is dynamically encoded into user equipment (UE) Route Selection Policies (URSPs) to optimize data transfer timing and route selection. The approach focuses on efficient satellite coverage usage while maintaining minimal changes to the existing 5G core infrastructure, enabling more robust performance in discontinuous signal environments.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages14

Reading Time35 min

AccessFree to download

Overview

Motivation: LOS dependence and NLOS impact

Proposed approach and system goal

Detailed Operation

Initial connection and packet exchange

Latency calculation

Machine learning implementation and prediction usage

Technical Disclosure Commons Defensive Publications Series 22 Jul 2025 ENHANCING 5G NON-TERRESTRIAL NETWORK (NTN) COMMUNICATIONS WITH MACHINE LEARNING-DRIVEN NON-LINE-OF-SIGHT (NLOS) PREDICTION AND MITIGATION Alwin Xavier Wojciech Grobel Snezana Mitrovic Vinay Saini Follow this and additional works at: [https://www.tdcommons.org/dpubs_series](https://www.tdcommons.org/dpubs_series) Recommended Citation Xavier, Alwin; Grobel, Wojciech; Mitrovic, Snezana; and Saini, Vinay, "ENHANCING 5G NON-TERRESTRIAL NETWORK (NTN) COMMUNICATIONS WITH MACHINE LEARNING-DRIVEN NON-LINE-OF-SIGHT (NLOS) PREDICTION AND MITIGATION", Technical Disclosure Commons,(July 22, 2025) [https://www.tdcommons.org/dpubs_series/8382](https://www.tdcommons.org/dpubs_series/8382) This work is licensed under a Creative Commons Attribution 4.0 License. This Article is brought to you for free and open access by Technical Disclosure Commons. It has been accepted for inclusion in Defensive Publications Series by an authorized administrator of Technical Disclosure Commons. ENHANCING 5G NON-TERRESTRIAL NETWORK (NTN) COMMUNICATIONS WITH MACHINE LEARNING-DRIVEN NON-LINE-OF-SIGHT (NLOS) PREDICTION AND MITIGATION AUTHORS: Alwin Xavier Wojciech Grobel Snezana Mitrovic Vinay Saini ABSTRACT The innovation proposed herein leverages machine learning to predict Non-Lineof-Sight (NLOS) conditions in a 5G Non-Terrestrial Network (NTN), significantly enhancing communication reliability in such networks. By dynamically encoding these predictions into user equipment (UE) Route Selection Policies (URSPs), the innovation proposed herein optimizes data transfers, ensuring efficient use of satellite coverage and improving overall network performance with minimal changes to the existing 5G core infrastructure. DETAILED DESCRIPTION There are numerous challenges in Third Generation Partnership Project (3GPP) Fifth Generation (5G) Non-Terrestrial Network (NTN) communications that need to be tackled to ensure the success of 5G NTN deployments. During the International Forum on Connecting the World from the Skies, which took place from November 8th to 10th, 2022, in Riyadh, an intriguing study highlighted the critical importance of the line-of-sight (LOS) requirement in satellite communications. In satellite communications, the received signal power is significantly diminished under Non-Line-of-Sight (NLOS) conditions. A comparison of received signal strength under NLOS and LOS for Low Earth Orbit (LEO) connected User Equipment (UE), and terrestrial connected devices illustrates the importance of LOS. For example, in NLOS situations, the received signal strength fades dramatically as compared to LOS situations. NLOS refers to scenarios where the communication signal between a satellite anda receiver is obstructed by physical objects such as buildings, trees, mountains, or even the 1 8012 Published by Technical Disclosure Commons, 2025 2 curvature of the Earth. Such NLOS scenarios are common in satellite communications. These random signal strength variations make 5G NTN unreliable for many applications. An innovative solution is proposed herein that leverages 5G and machine learning (ML) to mitigate these limitations and ensure improved and more efficient data transfers in 5G NTN environments. From the perspective of user equipment (UE) and/or Internet of Things (IoT) clientsand servers, a primary challenge in satellite communication is the movement of satellites relative to a UE/IoT client. This movement can cause NLOS conditions due to the curvature of the Earth or other physical obstructions. The NLOS behavior can vary among clients based on the geographical location of each client. Even if a specific area is covered by multiple satellites over time, there is still a chance of experiencing NLOS communication due to the distances between the satellites. Additionally, physical objects near a UE/IoT client can intermittently a

FAQs

What problem does the submission address in 5G NTN communications?

It addresses unreliable communication caused by Non-Line-of-Sight (NLOS) signal conditions, which significantly reduce received signal power and vary with satellite movement, geography, and nearby obstructions.

How does the innovation improve reliability under NLOS conditions?

It uses machine learning to predict when NLOS is likely, then encodes those predictions into UE Route Selection Policies (URSPs) to optimize data transfers and routing toward LOS periods.

What is the role of the AF and the initial packet exchange?

The Application Function (AF) generates a UDP packet containing the current time and a sequence number, sends it to the client, and receives the client’s response to enable round-trip and uplink latency calculation before ML prediction.

Aug 5, 2026

Upload by Seraphina