SYSTEM AND METHOD FOR ADAPTIVE CELL SEARCH BASED ON ARTIFICIAL INTELLIGENCE MODEL OF AN ELECTRONIC DEVICE

Smart overview of the Invention

The patent application introduces a system for adaptive cell search in electronic devices using an artificial intelligence (AI) model. The process involves selecting a target cell by user equipment (UE) based on movement detection and changes in signal power. Personalized user data is determined from various contextual parameters, aiding in predicting a user’s destination and potential target cells along the path. The AI model updates weights of network quality parameters to optimize cell selection.

Background

Traditional cell reselection involves switching cells when a UE is in idle mode, selecting the best available signal. In 5G networks, this process relies heavily on Reference Signal Received Power (RSRP) measurements. However, current methods do not account for factors like cell size and network load, leading to issues such as high battery drain and low throughput. The need for dynamic adjustment of network parameter weights is emphasized to improve user experience and resource optimization.

Challenges

• Fixed weights for network parameters fail to consider user behavior, degrading throughput.
• Call drops and interference increase when RF link imbalance is undervalued.
• Battery life suffers due to static weight assignment regardless of battery strength.
• Network resource utilization remains suboptimal without dynamic adjustments.

Proposed Solution

The proposed method detects UE movement through location and signal power changes, using AI to adaptively select target cells. Personalized data helps predict destinations, with the AI model dynamically updating network parameter weights based on user behavior and path predictions. This approach aims to enhance throughput, reduce call drops, and conserve battery life by optimizing network resource use.

Implementation

The system employs processors within the UE to execute the AI-driven selection process. By integrating personalized data and dynamic weight adjustments, the system anticipates user movements and selects optimal target cells. This methodology promises improved connectivity and efficiency by learning from user patterns and adapting to real-time conditions.