A single pretrained wireless time-series model, using univariate decomposition and a causal Transformer, predicts channel, angle, and traffic data, and zero-shot predicts a new delay task better than traditional full-shot baselines.
Addressing the curse of mo- bility in massive mimo with prony-based angular-delay domain channel predictions,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.AI 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
A Wireless Foundation Model for Multi-Task Prediction
A single pretrained wireless time-series model, using univariate decomposition and a causal Transformer, predicts channel, angle, and traffic data, and zero-shot predicts a new delay task better than traditional full-shot baselines.