[Paper Review] A new communication paradigm: from bit accuracy to semantic fidelity

Guangming Shi, Dahua Gao+6 more|arXiv (Cornell University)|Jan 29, 2021

Wireless Signal Modulation ClassificationComputer Science11 references47 citations

TL;DR

The paper proposes a semantic fidelity-based communication framework (CTSf) that transmits semantic symbols rather than exact bit-level messages, enabling large bandwidth savings while maintaining semantic integrity. It introduces semantic transformation, semantic level confirmation, and semantic inverse processing, with a case study on audio transmission.

ABSTRACT

Wireless communication has achieved great success in the past several decades. The challenge is of improving bandwidth with limited spectrum and power consumption, which however has gradually become a bottleneck with evolution going on. The intrinsic problem is that communication is modeled as a message transportation from sender to receiver and pursues for an exact message replication in Shannon's information theory, which certainly leads to large bandwidth and power requirements with data explosion. However, the goal for communication among intelligent agents, entities with intelligence including humans, is to understand the meaning or semantics underlying data, not an accurate recovery of the transmitted messages. The separate first transmission and then understanding is a waste on bandwidth. In this article, we deploy semantics to solve the spectrum and power bottleneck and propose a first understanding and then transmission framework with high semantic fidelity. We first give a brief introduction of semantics covering the definition and properties to show the insights and scope of this paper. Then the proposed communication towards semantic fidelity framework is introduced, which takes the above mentioned properties into account to further improve efficiency. Specially, a semantic transformation is introduced to transform the input into semantic symbols. Different from the conventional transformations in signal processing area, for example discrete cosine transform, the transformation is with data loss, which is also the reason that the proposed framework can achieve large bandwidth saving with high semantic fidelity. Besides, we also discuss semantic noise and performance measurement. To evaluate the effectiveness, a case study of audio transmission is carried out. Finally, we discuss the typical applications and open challenges.

Motivation & Objective

• Motivate a shift from bit-level fidelity to semantic fidelity in wireless communications.
• Define semantics in the context of communication and establish a semantic transmission framework (CTSF).
• Propose a semantic transformation to convert signals into semantic symbols and an abstraction mechanism via a semantic library.
• Introduce semantic level confirmation to reconcile sender/receiver knowledge and ensure understandability.
• Demonstrate the approach with a case study on audio transmission to show bandwidth savings and semantic fidelity benefits.

Proposed method

• Define semantics and semantic symbols as data subsets mapped through a many-to-one transformation to enable data loss yet high semantic fidelity.
• Introduce a semantic transformation pipeline comprising semantic representation (SR) and semantic symbol abstraction (SSA) and organize semantics in a hierarchical semantic library (SL).
• Propose semantic level confirmation to adapt to differences in sender/receiverSL, using feedback to choose the appropriate abstraction level.
• Describe semantic inverse processing including semantic symbol recognition (SSR) and semantic symbol inverse representation (SSIR) to recover signals or generate semantically equivalent content.
• Discuss semantic noise sources (transformation, channel, SL mismatch, ambiguity) and define semantic symbol error rate as a fidelity metric.
• Present a case study on audio transmission comparing conventional CCS with bit-fidelity against CTSF, using ASR-based SR and ZIP-based encoding for semantics.

Experimental results

Research questions

• RQ1 How can semantic symbols and semantic transformation reduce bandwidth while preserving semantic fidelity in wireless communication?
• RQ2 What mechanisms (semantic libraries, level confirmation) are needed to align sender and receiver understanding of semantics?
• RQ3 How can semantic noise be modeled and measured, and how does semantic fidelity compare to traditional signal fidelity?
• RQ4 What is the trade-off between data-loss in semantic transformation and achieved semantic fidelity in practical cases such as audio transmission?

Key findings

• CTSF dramatically reduces required bandwidth while maintaining high semantic fidelity in the case study (audio transmission).
• Semantic fidelity is measured via semantic symbol accuracy rather than traditional bit error rate, enabling meaningful comparisons with reduced data rates.
• A semantic transformation pipeline (SR and SSA) with a hierarchical semantic library enables efficient abstraction and cross-modality reconstruction.
• Semantic level confirmation and SSR/SSIR enable robust transmission even with SL mismatches and channel noise.
• The approach demonstrates significant bandwidth savings (e.g., up to two orders of magnitude in the example) compared to WAV/AAC baselines when aiming for high semantic fidelity.