Decomposition Based Interference Management Framework for Local 6G Networks

Samitha Gunarathne,Thushan Sivalingam,Nurul Huda Mahmood,Nandana Rajatheva,Matti Latva-Aho

2023-10-09

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.

Signal Processing

What problem does this paper attempt to address?

This paper aims to address the issue of inter-cell interference management in wireless networks, particularly in Ultra-Reliable Low-Latency Communication (URLLC) applications. Specifically, the paper proposes an intelligent interference management framework based on interference prediction for local 6G networks. This framework utilizes Empirical Mode Decomposition (EMD) to preprocess composite interference signals and employs a Sequence-to-One Transformer algorithm to predict each decomposed Intrinsic Mode Function (IMF). The predicted interference power is used to estimate the future Signal-to-Interference-plus-Noise Ratio (SINR), thereby allocating resources to meet the high reliability requirements of URLLC applications. Additionally, an interference cancellation scheme based on predicted interference signals is explored. Experimental results show that the proposed Sequence-to-One Transformer model performs excellently in interference prediction, reducing the Root Mean Square Error (RMSE) by up to 55% compared to baseline methods.