Burst Aware Forecasting of User Traffic Demand in LEO Satellite Networks

TL;DR

Addresses forecasting bursty downlink demand in LEO beam hopping to prevent buffer overflow. The method introduces three enhancements to a transformer: a burst distance embedding, two decoder heads for burst presence and adjustment, and an asymmetric loss, evaluated on synthetic data modeled by a second-order self-similar process with $H = (3 - a_{min})/2$ and fractional Brownian motion $B_H(t)$. Results show up to $94\%$ reduction in burst MSE at one-step horizon and sustained gains across longer horizons, outperforming ARIMA, FARIMA and standard transformer baselines. The approach enables more reliable BH planning and could generalize to other bursty networks.

Abstract

In Low Earth Orbit (LEO) satellite networks, Beam Hopping (BH) technology enables the efficient utilization of limited radio resources by adapting to varying user demands and link conditions. Effective BH planning requires prior knowledge of upcoming traffic at the time of scheduling, making forecasting an important sub-task. Forecasting becomes particularly critical under heavy load conditions where an unexpected demand burst combined with link degradation may cause buffer overflows and packet loss. To address this challenge, we propose a burst aware forecasting solution. This challenge may arise in a wide range of wireless networks; therefore, the proposed solution is broadly applicable to settings characterized by bursty traffic patterns where accurate demand forecasting is essential. Our approach introduces three key enhancements to a transformer architecture: (i) a distance from the last burst embedding to capture burst proximity, (ii) two additional linear layers in the decoder to forecast both upcoming bursts and their relative impact, and (iii) use of an asymmetric cost function during model training to better capture burst dynamics. Empirical evaluations in an Earth-fixed cell under high-traffic demand scenario demonstrate that the proposed model reduces prediction error by up to 94% at a one-step horizon and maintains the ability to accurately capture bursts even near the end of longer prediction horizons following Mean Square Error (MSE) metric.

Figures (4)

• Figure 1: SAT1 serves two user cells experiencing high downlink demand. As SAT1 maintains a stable backhaul connection, packets continue to arrive steadily. Downlink user cells may experience sudden traffic bursts combined with channel quality degradation, which can lead to buffer congestion and even packet loss.
• Figure 2: Snapshot of normalized data with one-step-ahead forecasts from the Proposed Solution, Informer, and FARIMA, with a zoomed view around an observed burst.
• Figure 3: Proposed system architecture, with blue boxes marking enhancements to the forecasting transformer.
• Figure 4: Heatmap showing MSE for burst samples, measured at forecast indices where bursts occur, across different prediction lengths.