Hao Shi

Mr. Hao Shi

orcid

Research Focus & Metrics

Academic Career Length

2Years (2025-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
21 June 2025 in Technologies
Verified
Journal Article
Open access
PDF file
HSHao Shi
Hao Shi
WGWenpu Guo
Wenpu Guo
SGShaoyuan Gao
Shaoyuan Gao
For the process of covert transmission of text information, in addition to the need to ensure the quality of the text at the same time, it is also necessary to make the text content match the current context. However, the existing text steganography methods excessively pursue the quality of the text, and lack constraints on the content and emotional expression of the generated steganographic text (stegotext). In order to solve this problem, this paper proposes an emotionally controllable text steganography based on large language model and named entity. The large language model is used for text generation to improve the quality of the generated stegotext. The named entity recognition is used to construct an entity extraction module to obtain the current context-centered text and constrain the text generation content. The sentiment analysis method is used to mine the sentiment tendency so that the stegotext contains rich sentiment information and improves its concealment. Through experimental validation on the generic domain movie reviews dataset IMDB, the results prove that the proposed method has significantly improved hiding capacity, perplexity, and security compared with the existing mainstream methods, and the stegotext has a strong connection with the current context.

13 January 2026 in Journal of King Saud University Computer and Information Sciences
Verified
Journal Article
Open access
PDF file
SGShaoyuan Gao
Shaoyuan Gao
WGWenpu Guo
Wenpu Guo
HSHao Shi
Hao Shi
RPRuiyan Peng
Ruiyan Peng
Single-channel blind source separation (SCBSS) remains a formidable challenge in modern wireless communication systems, especially when applied to long-duration, spectrally overlapped complex baseband signals. Existing deep learning approaches are hindered by limited receptive fields, vanishing gradients, or quadratic computational complexity, making them impractical for real-time separation of extended signal frames. To overcome these limitations, we propose IQUMamba-1D, a one-dimensional U-Net architecture enhanced with selective structured state-space model (SSM). The framework synergistically integrates long-range temporal modeling through Mamba blocks while preserving multi-scale feature extraction via U-Net. Three communication-specific innovations are introduced: adaptive dual tokenization mitigates quadrature imbalance by stage-dependent tokenization mode selection between temporal and channel token modes; complex-valued selective state space modeling enforces geometric consistency in the complex plane by jointly processing in-phase and quadrature (I/Q) components; and an adaptive skip connection processor alleviates semantic gaps induced by aggressive downsampling, thereby preserving phase coherence vital for signal reconstruction. Evaluated across synthetic and public datasets with signal lengths up to 32 768 samples, IQUMamba-1D achieves state-of-the-art separation performance with linear computational complexity. The proposed method provides a theoretically grounded and computationally efficient solution for SCBSS of spectrally overlapped communication signals under single-channel observation.