Decomposition Based Interference Management Framework for Local 6G Networks

TL;DR

This paper presents a novel interference management framework for local 6G networks that predicts interference using advanced signal processing and transformer models to improve resource allocation and reliability for URLLC applications.

Contribution

It introduces an innovative interference prediction and cancellation framework using empirical mode decomposition and sequence-to-one transformer models for 6G networks.

Findings

Root mean squared error reduced by up to 55% compared to baseline algorithms.

Effective interference prediction enhances resource allocation for URLLC.

The transformer-based approach demonstrates robustness in interference forecasting.

Abstract

Managing inter-cell interference is among the major challenges in a wireless network, more so when strict quality of service needs to be guaranteed such as in ultra-reliable low latency communications (URLLC) applications. This study introduces a novel intelligent interference management framework for a local 6G network that allocates resources based on interference prediction. The proposed algorithm involves an advanced signal pre-processing technique known as empirical mode decomposition followed by prediction of each decomposed component using the sequence-to-one transformer algorithm. The predicted interference power is then used to estimate future signal-to-interference plus noise ratio, and subsequently allocate resources to guarantee the high reliability required by URLLC applications. Finally, an interference cancellation scheme is explored based on the predicted interference signal with the transformer model. The proposed sequence-to-one transformer model exhibits its robustness for interference prediction. The proposed scheme is numerically evaluated against two baseline algorithms, and is found that the root mean squared error is reduced by up to 55% over a baseline scheme.