AI Wide Beam Optimization for 5G Coverage and Data Rate
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing wireless communication networks face challenges in efficiently managing radio resource management and beam optimization for various use cases in 5G systems, particularly in high-frequency bands, due to varying coverage requirements and propagation losses.
Innovation Solution
Implementing machine learning-based wide beam optimization in cellular networks, where a central controller selects and optimizes beams using a candidate beam pool, calculates beam scores based on measurement results, and determines the highest-scoring beam for improved network performance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If beamforming and narrow beams are used in 5G high-frequency bands, then data rates and transmission distance are improved, but coverage area is reduced and propagation loss increases
Solution Approach 1:
The patent implements dynamic beam width adjustment where the beam configuration changes based on real-time measurement reports from user equipment. The system transitions between narrow beams (for high data rates) and wide beams (for extended coverage) dynamically, allowing the beam width to adapt to different operational requirements and propagation conditions.
Solution Approach 2:
The system changes the beam parameter (beam width) based on measured propagation characteristics and coverage requirements. By modifying the beam width parameter dynamically, the system optimizes the trade-off between data rate and coverage area, selecting appropriate beam configurations for different operational scenarios.
2Adaptability or versatility
If multiple beams are managed for different use cases, then network adaptability is improved, but radio resource management complexity increases
Solution Approach 1:
The system implements self-service beam management where user equipment autonomously measures synchronization signals and reference signals from multiple beams, generates measurement reports, and provides feedback for beam selection. This autonomous operation reduces the complexity of centralized radio resource management while maintaining high network adaptability.
Solution Approach 2:
The patent establishes a feedback mechanism where user equipment reports measurement results of synchronization signals and reference signals to the network. Based on this feedback, the system dynamically adjusts beam configurations, enabling adaptive resource management that reduces complexity through automated decision-making.
3Reliability
If beam selection is optimized manually, then beam performance is improved, but system response time increases
Solution Approach 1:
The system performs automated beam selection based on measurement reports generated by user equipment. The automated process eliminates manual intervention and reduces system response time while maintaining reliable beam performance through objective measurement-based decision-making.
Solution Approach 2:
The patent implements preliminary beam configuration where a set of candidate beams is pre-defined and measured in advance. The system uses pre-measured synchronization signals and reference signals to quickly select optimal beams, reducing real-time processing delays and improving system response time.
Data Source
Figure 1
Figure 2A
Figure 2B
AI summary
The present disclosure relates to a pre-5th-Generation (5G) or 5G communication system to be provided for supporting higher data rates Beyond 4th-Generation (4G) communication system such as Long Term Evolution (LTE). The present disclosure relates an artificial intelligence (AI) system and its application that sumltate functions such as recognition and judgment of a human brain using a machine learning algorithm such as deep learning. An apparatus and method for controlling and optimizing the broadcast beam for base stations (BS) using user equipment (UE) measurements with machine learning is provided. The apparatus and method is configured to select a first beam for each BS, send selected beams for each BS, receive measurement information of a first beam from UEs via BSs, preprocess the measurement results, use a neural network or a table for each BS to give a score for each broadcast beam in the beam pool, select a second beam with the highest score for each BS either from a neural network or a table, train the neural network for broadcast beam optimization offline based on a UE distribution pattern and ray-tracing data, identify typical UE distribution patterns based on AI classification algorithms and UE history measurement and location infomraiton, and create scenario-specific ray-tracing data based on typical UE distribution patterns.