A radio signal with layered modulation resembles the cultural layer of an ancient city: each level carries its own information, but over time they get jumbled by interference. Conventional algorithms get lost in this mess. The new TSDN method acts as a neural network archaeologist. Its secret lies in logarithmic space, where multiplying layers turns into addition: if signal S is the product of layers A and B, then log(S) = log(A) + log(B). Then a trainable geometric transformer arranges the layers in order, as if reconstructing the chronology of an excavation.
Accuracy exceeds 93%, even when the signal is weaker than noise. The network recognizes modulation combinations it has never encountered before — as if an archaeologist could deduce the culture of a vanished civilization from just a few pottery shards.
The word 'modulation' comes from the Latin modulatio, meaning measured rhythm. In radio engineering, it's a way of packing information onto a carrier signal — like a voice onto a radio wave.
Technologies like TSDN may one day help decode alien messages, as in the film 'Contact'.