Predicting Buffer Status Report (BSR) for 6G Scheduling using Machine Learning Models

Abstract

In 6G communication, many state-of-the-art ma-chine learning algorithms are going to be implemented to enhance the performances, including the latency property. In this paper, we apply Buffer Status Report (BSR) prediction to the uplink scheduling process, in order to improve the resource allocation solution currently used at the base station. According to the current solution, the base station allocates the resources based on the BSRs. However, since the BSRs do not include information for data arriving after their transmissions, the base station allocates the resources without taking into consideration the new arrival data, which may lead to the increased latency. To solve this problem, we decide to make BSR predictions at the base station side and allocate more resources than BSRs indicate. Making an accurate BSR prediction is a challenging task since there are numerous features that may influence the BSRs. Another challenge in this task is that the time intervals are tremendously short (in the order of milliseconds). After cleaning the data collected from real networks, we convert the time series forecasting problem into a supervised learning problem. State-of-the-art algorithms such as Random Forest (RF), XGboost, and Long Short Term Memory (LSTM) are leveraged to predict the data arrival rate, and one K-Fold Cross-Validation is followed to validate the models. The results show that even the time intervals are small, the data arrival rate can be predicted and the downlink data, downlink quality indicator and rank indicator can boost the forecasting performance.