Core Concepts
Revolutionizing error detection in image transmission using Topological Data Analysis.
Abstract
The article introduces a novel approach, SC-TDA-HARQ, combining swin transformer-based JSCC and IK-HARQ for semantic image transmission. It proposes a TDA-based error detection method to capture semantic information effectively. The paper highlights the limitations of traditional methods like CRC and Sim32 in handling semantic ambiguity in image transmission. By leveraging TDA, the proposed framework shows superior performance under limited bandwidth conditions.
Stats
Method Similarity Mean Variance Sim32 0.5416 ∼ 0.5423 0.5421 3.1151 × 10−8
TDA-based decision network 0.5284 ∼ 0.5311 0.5297 4.021 × 10−7
Quotes
"There emerges a strong incentive to revolutionize the CRC mechanism."
"Extensive numerical results validate the effectiveness and efficiency of the proposed SC-TDA-HARQ framework."