A multi-modal affective semantic perception computing and transmission method

By using feature mapping of multimodal data and a continuous-time state-space model, the problems of geometric distortion and temporal asynchrony in multimodal semantic communication are solved, achieving robust and accurate transmission under complex channel conditions and improving the reliability of sentiment analysis.

CN122294147APending Publication Date: 2026-06-26CHONGQING UNIV OF POSTS & TELECOMM
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
CHONGQING UNIV OF POSTS & TELECOMM
Filing Date
2026-04-14
Publication Date
2026-06-26

AI Technical Summary

Technical Problem

Existing technologies in multimodal semantic communication suffer from problems such as geometric distortion, temporal asynchrony, and insufficient robustness to channel fading, leading to unstable feature extraction, information loss, and semantic information distortion.

Method used

By using feature mapping, geometric consistency correction, continuous-time state-space model, and structural consistency mapping of multimodal data, we achieve accurate alignment and robust transmission of multimodal features, including feature extraction, fusion, and channel symbol mapping of text, visual, and audio modalities.

Benefits of technology

It achieves robustness and transmission efficiency of multimodal semantic communication in wireless scenarios with limited bandwidth and changing environments, ensuring the accuracy and integrity of sentiment analysis.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122294147A_ABST
    Figure CN122294147A_ABST
Patent Text Reader

Abstract

This application relates to the field of semantic communication and information processing technology, and in particular to a method for multimodal sentiment semantic perception calculation and transmission. It is implemented based on a signal transmitter and a signal receiver. The transmitter acquires multimodal data and forwards it to the receiver via a wireless channel. The receiver performs sentiment analysis and outputs the predicted sentiment analysis results. This application effectively solves the robustness challenges of multimodal semantic communication under geometric distortion, temporal asynchrony, and channel fading by jointly optimizing geometrically consistent feature alignment, continuous-time dynamic fusion, and semantic mapping that preserves topological structure.
Need to check novelty before this filing date? Find Prior Art