COMMUNICATIONS DEVICE, INFRASTRUCTURE EQUIPMENT AND METHODS FOR PERFORMING HANDOVER USING A MODEL BASED ON MACHINE LEARNING

Smart overview of the Invention

The patent application describes a method for enhancing handover processes in wireless communications networks using machine learning models. This involves establishing a connection in a first cell, evaluating certain input parameters, and using these parameters as inputs to a pre-trained machine learning model. Based on the model's output, a decision is made whether to perform a handover to a second cell, and if so, a handover message is transmitted to initiate this process.

Background

Wireless communication systems, such as those based on 3GPP's UMTS and LTE architectures, have evolved to support high data rate applications like video streaming and conferencing. Future networks, including 5G and New Radio (NR) systems, are expected to efficiently support a broad spectrum of devices and data profiles, ranging from IoT devices to high-resolution video displays. The increasing variety of devices and traffic profiles presents challenges in maintaining efficient communications, particularly in handling handovers.

Summary

The application proposes using machine learning techniques to enhance handover decisions in wireless networks. By evaluating input parameters through a trained model, the system dynamically determines whether a handover is necessary based on current radio conditions. This approach aims to improve the continuity and quality of data transmission and reception as devices move across different network cells.

Detailed Description

The document provides a detailed explanation of the method's implementation within LTE Advanced and 5G network architectures. In LTE, the network consists of base stations connected to a core network, providing coverage areas for communication devices. In 5G, the architecture involves controlling nodes and distributed units, each responsible for providing radio access within specific coverage areas. The machine learning-enhanced handover method can be applied across these different generations and architectures.

Applications and Benefits

This technique addresses the challenges of supporting diverse device types and traffic profiles in future wireless networks. It is particularly beneficial for applications requiring Ultra Reliable Low Latency Communications (URLLC), where high reliability and low delay are critical. By leveraging machine learning for handover decisions, the method aims to maintain optimal connectivity and service quality across varying network conditions.