A new review examines how insertion and deletion (indel) errors disrupt data synchronization in modern communication systems. By surveying both traditional and Deep Learning-driven approaches, the study provides a comprehensive overview of existing solutions and outlines future directions for more reliable communication in 6G, satellite networks, and beyond.
Insertion and deletion (indel) errors present a critical challenge in modern communication systems, especially as technologies evolve toward 5G/6G networks, satellite communications, and large-scale Internet of Things (IoT) deployments. Unlike conventional bit errors, indel errors disrupt sequence alignment, making accurate data recovery significantly more difficult. Addressing this issue, Peng Zhu, Hui Yang, Chao Yu, and Wenwu Xie from Hunan Institute of Science and Technology, and Ji Wang from Central China Normal University, have published a comprehensive review in Advanced Information and Communication , systematically examining existing indel error detection and correction techniques and outlining future research directions.
“Indel errors are not just about incorrect bits. They break the structure of the data itself,” explains Professor Peng Zhu.
“As a result, synchronization recovery becomes a key challenge for reliable communication, particularly in complex and dynamic environments,” note Professors Chao Yu and Wenwu Xie.
The review organizes existing solutions into two major paradigms: traditional model-driven approaches and emerging data-driven methods. Traditional techniques—including synchronization marker-based methods, edit-distance coding, sequence alignment, and probabilistic models—have shown solid performance in relatively stable conditions. However, their effectiveness often degrades in scenarios involving high error rates, high mobility, and complex channel environments.
To address these limitations, recent research has explored the integration of deep learning models into synchronization recovery and decoding processes. Architectures such as convolutional neural networks (CNNs), recurrent neural networks (RNNs), Transformers, and deep neural networks (DNNs) are increasingly used to assist tasks such as synchronization-state estimation, sequence alignment, and error detection in complex and time-varying communication environments.
The review also highlights the growing interest in semantic communication. Instead of focusing solely on bit-level accuracy, this paradigm considers whether useful information can still be recovered when data is partially lost or misaligned. By combining ideas such as joint source-channel coding and data-driven modeling, semantic communication aims to recover meaningful information and improve communication reliability under severe distortion conditions.
At the same time, the study carefully examines current challenges. Deep learning-based methods often require large training datasets and significant computational resources, while their performance under varying channel conditions remains a challenge. Semantic communication-based approaches, although promising, still face issues related to domain adaptation and compatibility with existing communication system architectures.
A key contribution of the review lies in its systematic discussion of future research directions. The authors identify several important areas for further study, including synchronization recovery in highly dynamic and complex channels, lightweight deep learning model design, integration of mathematical optimization with learning-based methods, semantic communication-driven error correction, unified end-to-end evaluation frameworks, and customized solutions for emerging applications such as DNA storage, quantum communication, and wearable systems.
"These directions highlight the need for more adaptive and efficient frameworks that can operate across diverse communication scenarios," Hui Yang and Professor Ji Wang note.
Overall, the study provides a comprehensive overview of indel error detection and correction techniques, bridging traditional and emerging approaches while identifying key challenges and opportunities for future research.
This paper, "A survey on insertion/deletion error detection and correction: progress and future directions," was published in Advanced Information and Communication .
Zhu P, Yang H, Yu C, Xie W, Wang J. A survey on insertion/deletion error detection and correction: progress and future directions. Adv. Inf. Commun . 2026(1):0003, https://doi.org/10.55092/aic20260003.
Literature review
Not applicable
A survey on insertion/deletion error detection and correction: progress and future directions
31-Mar-2026