Deep Reinforcement Learning for Interference Suppression in RIS-Aided Space–Air–Ground Integrated Networks

1 Pujitha Mamillapalli, 2,3 Shikhar Verma, 3 Tiago Koketsu Rodrigues and 1 Abhinav Kumar · 2026

Read Deep Reinforcement Learning for Interference Suppression in RIS-Aided Space–Air–Ground Integrated Networks by 1 Pujitha Mamillapalli, 2,3 Shikhar Verma, 3 Tiago Koketsu Rodrigues and 1 Abhinav Kumar (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2602-06982.txt.

Description

Pujitha Mamillapalli and three co-authors tame interference in space-air-ground networks for future 6G. High-altitude platforms and satellites extend ground networks for wide, low-lag coverage. Yet rapid growth in ground devices forces spectrum sharing between ground and non-ground parts, causing severe cross-tier interference. Sharing between the platform-to-satellite uplink and the platform-to-ground downlink boosts spectrum use, but the platform antenna's back-lobe leaks in.

Old zero-forcing codebooks falter when channels shift fast. The team instead uses a reconfigurable intelligent surface with a deep deterministic policy gradient algorithm. It tunes beamforming weights to carve spatial nulls toward interference sources while keeping wanted links strong. Simulations show it beats standard zero-forcing across surface setups, gaining up to 11.3 percent in throughput for a four-by-four surface.

Catalog record

CREATOR1 Pujitha Mamillapalli, 2,3 Shikhar Verma, 3 Tiago Koketsu Rodrigues and 1 Abhinav Kumar
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE30 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2602-06982

Browse the full library · About the Tower of Babel