Chen Yi
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Research Interests
Chen Yi has a total of 61 co-authors. Here are the top co-authors by number of shared publications:
- Yong Li
College of Computer Science, Chongqing University, Chongqing, China
9 shared publications - Zeliang An
School of Computer and Information, Anhui Polytechnic University, Wuhu, China
5 shared publications Baoze Ma
4 shared publications- Jihua Zhou
Southwest University
4 shared publications - Baoze Ma
School of Communication and Information Engineering, Chongqing University of Posts and Telecommunications, Chongqing, China
3 shared publications - Francis Chung Ming Lau
The Hong Kong Polytechnic University
3 shared publications
Latest Publications
- 12 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Communications Letters
- Vol. 30 (10897798) , 887-891
Blind Recognition of Polar Code Information Bits Based on Multi-Threshold Voting and Partial Orders
Blind recognition of polar codes in non-cooperative scenarios remains insufficiently addressed. Although existing methods have achieved reliable performance in code length recognition, reliably identifying information bits remains challenging under moderate-to-high bit error rate (BER). In this letter, assuming the code length is known, we propose a two-stage blind information bits recognition scheme. In the first stage, multi-threshold voting judgment is performed to obtain the initial frozen set. In the second stage, we perform partial orders (POs) correction to enforce structural consistency of the frozen set. When BER ranges from 0 to 0.2, simulations show that the proposed method, at the cost of modest computational complexity, significantly improves the information bits identification accuracy compared with existing estimation and derivation-based approaches, achieving relative gains of approximately 40.0%, 20.7% and 13.8% for polar codes (32,15), (64,30), and (128,64), respectively.
- 12 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Communications Letters
- Vol. 30 (10897798) , 887-891
Blind Recognition of Polar Code Information Bits Based on Multi-Threshold Voting and Partial Orders
Blind recognition of polar codes in non-cooperative scenarios remains insufficiently addressed. Although existing methods have achieved reliable performance in code length recognition, reliably identifying information bits remains challenging under moderate-to-high bit error rate (BER). In this letter, assuming the code length is known, we propose a two-stage blind information bits recognition scheme. In the first stage, multi-threshold voting judgment is performed to obtain the initial frozen set. In the second stage, we perform partial orders (POs) correction to enforce structural consistency of the frozen set. When BER ranges from 0 to 0.2, simulations show that the proposed method, at the cost of modest computational complexity, significantly improves the information bits identification accuracy compared with existing estimation and derivation-based approaches, achieving relative gains of approximately 40.0%, 20.7% and 13.8% for polar codes (32,15), (64,30), and (128,64), respectively.
- 1 October 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Communications
- Vol. 73 (12) , 13163-13172
Correcting Burst Deletion/Insertion and Random Substitution Errors With Stacking Codes in Non-Binary Segmented Burst Deletion/Insertion Channels
In this paper, we propose stacking codes and decoding algorithms to correct burst deletion/insertion and random substitution errors in a non-binary segment-L maxburst-D deletion and maxburst-S insertion (NB-SBDI(L,D,S)) channel without knowing the codeword boundaries. Specifically, in the NB-SBDI(L,D,S) channel, at most a single non-binary burst (a block of consecutive bits/symbols) of deletions or insertions of length up to D or S, respectively, exists in a block of L consecutive non-binary symbols. The proposed stacking-codes are constructed by stacking multiple Reed-Muller(RM)/modified BCH/modified binary representation of Reed-Solomon (BRRS) codes, respectively, and no marker codes are involved. We show that, in the NB-SBDI(L,D,S) channel, the proposed approach can not only correct both burst deletions/insertions and random substitution errors, but also achieve error-free decoding for stacking-RM/modified stacking-BCH/modified stacking-BRRS codes for burst deletions/insertions as long as no substitution errors exist. Also, the modified stacking-BRRS code is more rate efficient than the existing Song code [1], the stacking-RM/modified stacking-BCH codes are more rate efficient than the Song code if max(D,S) and code length are small.
- 1 October 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Communications
- Vol. 73 (12) , 13163-13172
Correcting Burst Deletion/Insertion and Random Substitution Errors With Stacking Codes in Non-Binary Segmented Burst Deletion/Insertion Channels
In this paper, we propose stacking codes and decoding algorithms to correct burst deletion/insertion and random substitution errors in a non-binary segment-L maxburst-D deletion and maxburst-S insertion (NB-SBDI(L,D,S)) channel without knowing the codeword boundaries. Specifically, in the NB-SBDI(L,D,S) channel, at most a single non-binary burst (a block of consecutive bits/symbols) of deletions or insertions of length up to D or S, respectively, exists in a block of L consecutive non-binary symbols. The proposed stacking-codes are constructed by stacking multiple Reed-Muller(RM)/modified BCH/modified binary representation of Reed-Solomon (BRRS) codes, respectively, and no marker codes are involved. We show that, in the NB-SBDI(L,D,S) channel, the proposed approach can not only correct both burst deletions/insertions and random substitution errors, but also achieve error-free decoding for stacking-RM/modified stacking-BCH/modified stacking-BRRS codes for burst deletions/insertions as long as no substitution errors exist. Also, the modified stacking-BRRS code is more rate efficient than the existing Song code [1], the stacking-RM/modified stacking-BCH codes are more rate efficient than the Song code if max(D,S) and code length are small.
- 20 August 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Communications
- Vol. 73 (12) , 12993-13004
Identification of the Whole Interleaving Pattern From Randomly Interleaved Linear Block Codes With Known Minimum Distances
In this paper, an effective approach is proposed to recognize the order among codewords from the randomly binary and non-binary interleaved linear block coded streams if the code minimum distance is known, where the interleaver length Nn + d is not the multiple of the linear block code length n. Then, an efficient approach is developed without knowing linear block code parameters such as code rate and the code generator matrix (except the code length) to recognize the whole random interleaving pattern if the greatest common divisor (GCD) of n and d is GCD(n, d) = 1 for both binary and non-binary interleaved linear block coded symbols. Further, for the case GCD(n, d) = q ̸= 1, the whole random interleaving pattern can be recognized with the probability of no more than (q!)−(Nn+d)/q for the following cases: (1) the non-binary interleaved coded symbols without knowing the non-binary code generator matrix; (2) the binary interleaved coded symbols.
- 20 August 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Communications
- Vol. 73 (12) , 12993-13004
Identification of the Whole Interleaving Pattern From Randomly Interleaved Linear Block Codes With Known Minimum Distances
In this paper, an effective approach is proposed to recognize the order among codewords from the randomly binary and non-binary interleaved linear block coded streams if the code minimum distance is known, where the interleaver length Nn + d is not the multiple of the linear block code length n. Then, an efficient approach is developed without knowing linear block code parameters such as code rate and the code generator matrix (except the code length) to recognize the whole random interleaving pattern if the greatest common divisor (GCD) of n and d is GCD(n, d) = 1 for both binary and non-binary interleaved linear block coded symbols. Further, for the case GCD(n, d) = q ̸= 1, the whole random interleaving pattern can be recognized with the probability of no more than (q!)−(Nn+d)/q for the following cases: (1) the non-binary interleaved coded symbols without knowing the non-binary code generator matrix; (2) the binary interleaved coded symbols.
- 30 July 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Communications Letters
- Vol. 29 (10) , 2321-2325
Secure Energy Efficiency for RIS-Aided Cognitive Radio Networks With Hardware Impairments
In practical applications, transceiver hardware impairments (HWIs) and phase errors (PEs) in reconfigurable intelligent surfaces (RIS) can distort signals in cognitive radio (CR) networks, reducing the system’s anti-interference and security performance. This paper investigates secure communication design for RIS-assisted CR systems affected by HWIs and RIS PEs. To maximize the security energy efficiency (SEE) of cognitive users (CUs), a joint optimization method for base station (BS) beamforming, artificial noise (AN), and RIS reflection phase is proposed. An alternating iterative approach addresses this optimization problem by first optimizing beamforming and AN through variable substitution and successive convex approximation (SCA), then optimizing the passive beamforming using Semidefinite Relaxation (SDR) and singular value decomposition (SVD), assuming PEs follow a uniform or von mises distribution. Numerical results show the proposed method surpasses benchmark schemes in security and robustness.
- 30 July 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Communications Letters
- Vol. 29 (10) , 2321-2325
Secure Energy Efficiency for RIS-Aided Cognitive Radio Networks With Hardware Impairments
In practical applications, transceiver hardware impairments (HWIs) and phase errors (PEs) in reconfigurable intelligent surfaces (RIS) can distort signals in cognitive radio (CR) networks, reducing the system’s anti-interference and security performance. This paper investigates secure communication design for RIS-assisted CR systems affected by HWIs and RIS PEs. To maximize the security energy efficiency (SEE) of cognitive users (CUs), a joint optimization method for base station (BS) beamforming, artificial noise (AN), and RIS reflection phase is proposed. An alternating iterative approach addresses this optimization problem by first optimizing beamforming and AN through variable substitution and successive convex approximation (SCA), then optimizing the passive beamforming using Semidefinite Relaxation (SDR) and singular value decomposition (SVD), assuming PEs follow a uniform or von mises distribution. Numerical results show the proposed method surpasses benchmark schemes in security and robustness.
- 11 April 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Signal Processing Letters
- Vol. 32 (10709908) , 2015-2019
Speech Conv-Mamba: Selective Structured State Space Model With Temporal Dilated Convolution for Efficient Speech Separation
As a selective state space model, Mamba exhibits outstanding performance and efficiency in sequence modeling tasks. Therefore, in this paper, we use Mamba as the fundamental network component to construct a novel speech separation model, Speech Conv-Mamba. Specifically, this model embeds Mamba within a U-shaped convolutional network to build the encoder and decoder network for high-dimensional representation and waveform reconstruction of speech signals. Additionally, we stack multiple temporal dilated convolutions and Mamba to create the separation network for separation task. Our comparative experiments on the GRID2Mix and Libri2Mix datasets demonstrate that the proposed model Speech Conv-Mamba, which achieves 98% and 89% of SepFormer's separation accuracy on two datasets using only 9% (2.4 M) of its model size, provides much less computational complexity and training cost.
- 11 April 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Signal Processing Letters
- Vol. 32 (10709908) , 2015-2019
Speech Conv-Mamba: Selective Structured State Space Model With Temporal Dilated Convolution for Efficient Speech Separation
As a selective state space model, Mamba exhibits outstanding performance and efficiency in sequence modeling tasks. Therefore, in this paper, we use Mamba as the fundamental network component to construct a novel speech separation model, Speech Conv-Mamba. Specifically, this model embeds Mamba within a U-shaped convolutional network to build the encoder and decoder network for high-dimensional representation and waveform reconstruction of speech signals. Additionally, we stack multiple temporal dilated convolutions and Mamba to create the separation network for separation task. Our comparative experiments on the GRID2Mix and Libri2Mix datasets demonstrate that the proposed model Speech Conv-Mamba, which achieves 98% and 89% of SepFormer's separation accuracy on two datasets using only 9% (2.4 M) of its model size, provides much less computational complexity and training cost.
- 18 September 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE/ACM Transactions on Audio, Speech, and Language Processing
- Vol. 32 (23299290) , 4647-4660
Audio-Visual Fusion With Temporal Convolutional Attention Network for Speech Separation
Currently, audio-visual speech separation methods utilize the speaker's audio and visual correlation information to help separate the speech of the target speaker. However, these methods commonly use the approach of feature concatenation with linear mapping to obtain the fused audio-visual features, which prompts us to conduct a deeper exploration for audio-visual fusion. Therefore, in this paper, according to the speaker's mouth landmark movements during speech, we propose a novel time-domain single-channel audio-visual speech separation method: audio-visual fusion with temporal convolution attention network for speech separation model (AVTCA). In this method, we design temporal convolution attention network (TCANet) based on the attention mechanism to model the contextual relationships between audio and visual sequences, and use TCANet as the basic unit to construct sequence learning and fusion network. In the whole deep separation framework, we first use cross attention to focus on the cross-correlation information of the audio and visual sequences, and then we use the TCANet to fuse the audio-visual feature sequences with temporal dependencies and cross-correlations. Afterwards, the fused audio-visual features sequences will be used as input to the separation network to predict mask and separate the source of each speaker. Finally, this paper conducts comparative experiments on Vox2, GRID, LRS2 and TCD-TIMIT datasets, indicating that AVTCA outperforms other state-of-the-art (SOTA) separation methods. Furthermore, it exhibits greater efficiency in computational performance and model size.
- 18 September 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE/ACM Transactions on Audio, Speech, and Language Processing
- Vol. 32 (23299290) , 4647-4660
Audio-Visual Fusion With Temporal Convolutional Attention Network for Speech Separation
Currently, audio-visual speech separation methods utilize the speaker's audio and visual correlation information to help separate the speech of the target speaker. However, these methods commonly use the approach of feature concatenation with linear mapping to obtain the fused audio-visual features, which prompts us to conduct a deeper exploration for audio-visual fusion. Therefore, in this paper, according to the speaker's mouth landmark movements during speech, we propose a novel time-domain single-channel audio-visual speech separation method: audio-visual fusion with temporal convolution attention network for speech separation model (AVTCA). In this method, we design temporal convolution attention network (TCANet) based on the attention mechanism to model the contextual relationships between audio and visual sequences, and use TCANet as the basic unit to construct sequence learning and fusion network. In the whole deep separation framework, we first use cross attention to focus on the cross-correlation information of the audio and visual sequences, and then we use the TCANet to fuse the audio-visual feature sequences with temporal dependencies and cross-correlations. Afterwards, the fused audio-visual features sequences will be used as input to the separation network to predict mask and separate the source of each speaker. Finally, this paper conducts comparative experiments on Vox2, GRID, LRS2 and TCD-TIMIT datasets, indicating that AVTCA outperforms other state-of-the-art (SOTA) separation methods. Furthermore, it exhibits greater efficiency in computational performance and model size.
- 12 June 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Signal Processing Letters
- Vol. 31 (10709908) , 2445-2449
A Tensor Based Unified Framework for Streaming Signal Online Analysis
To explore the online tensor analysis technique in single-aspect streaming model, a tensor based unified framework encompassed more tensor formats than one is proposed according to the structure of the last tensor factor. We found that the temporal slice only changes the size of factor matrix corresponding to the temporal dimension in unified framework. Consequently, the block-coordinate descent strategy is utilized to calculate the temporal tensor factor and non-temporal decomposition components respectively based on exponentially weighted cost function. The simulation results indicate that the presented unified framework provides a superior reconstruction performance compared to the existing tensor tracking methods in real video scenarios.
- 12 June 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Signal Processing Letters
- Vol. 31 (10709908) , 2445-2449
A Tensor Based Unified Framework for Streaming Signal Online Analysis
To explore the online tensor analysis technique in single-aspect streaming model, a tensor based unified framework encompassed more tensor formats than one is proposed according to the structure of the last tensor factor. We found that the temporal slice only changes the size of factor matrix corresponding to the temporal dimension in unified framework. Consequently, the block-coordinate descent strategy is utilized to calculate the temporal tensor factor and non-temporal decomposition components respectively based on exponentially weighted cost function. The simulation results indicate that the presented unified framework provides a superior reconstruction performance compared to the existing tensor tracking methods in real video scenarios.
- 14 May 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Communications
- Vol. 72 (10) , 6001-6013
Error-Correcting Codes With Large Field Size Under Non-Binary Segmented Burst Deletion/Insertion Channels and Unknown Codeword Boundaries
In this paper, we construct non-binary codes of length N which correct errors under a non-binary segment- N maxburst- D deletion and maxburst- S insertion (NB-SBDI( N, D, S )) channel without knowing the codeword boundaries. In this NB-SBDI( N, D, S ) channel, at most a single non-binary burst (a block of consecutive bits/symbols) of deletions or insertions of length up to D or S , respectively, exists in a block of N consecutive non-binary symbols. One code named as BM-DB-MDS consists of a maximum distance separable (MDS) code, a block of periodic de Bruijn (DB) symbols, and a block of proposed periodic binary marker (BM) patterns with a period of S + D + 2. The other code called BM-MDS code consists of a BM code and an MDS code. We show that the rates of BM-DB-MDS and BM-MDS codes achieve λ/λ+1 (1 - 1/2 t ), λ ∈ N + and 1 - 1/ t , respectively, when N → +∞, where λ represent the MDS code shortening factor, and 1/ t is the rough proportion of the maximum length of burst deletions or insertions allowed in a code.
- 14 May 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Communications
- Vol. 72 (10) , 6001-6013
Error-Correcting Codes With Large Field Size Under Non-Binary Segmented Burst Deletion/Insertion Channels and Unknown Codeword Boundaries
In this paper, we construct non-binary codes of length N which correct errors under a non-binary segment- N maxburst- D deletion and maxburst- S insertion (NB-SBDI( N, D, S )) channel without knowing the codeword boundaries. In this NB-SBDI( N, D, S ) channel, at most a single non-binary burst (a block of consecutive bits/symbols) of deletions or insertions of length up to D or S , respectively, exists in a block of N consecutive non-binary symbols. One code named as BM-DB-MDS consists of a maximum distance separable (MDS) code, a block of periodic de Bruijn (DB) symbols, and a block of proposed periodic binary marker (BM) patterns with a period of S + D + 2. The other code called BM-MDS code consists of a BM code and an MDS code. We show that the rates of BM-DB-MDS and BM-MDS codes achieve λ/λ+1 (1 - 1/2 t ), λ ∈ N + and 1 - 1/ t , respectively, when N → +∞, where λ represent the MDS code shortening factor, and 1/ t is the rough proportion of the maximum length of burst deletions or insertions allowed in a code.
- 22 April 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Communications Letters
- Vol. 28 (6) , 1253-1257
Rate Efficient Codes Correcting a Burst of Deletions or Insertions
In this paper, we construct two rate efficient codes of length n named as marker-MDS and marker-SVT codes which correct a burst of deletions/insertions of length b (error-free decoding), where b is not necessarily fixed as a constant but is proportional to n , i.e., b = tn , 0 < t < 1. Both of these two codes consist of binary marker codes which are employed to locate the burst of deletions/insertions. Also, the marker-MDS and marker-SVT codes consist of the maximum distance separable (MDS) codes and shifted Varshamov-Tenengol’ts (SVT) codes, respectively, which are responsible for correcting erasures caused in the synchronization stage. Both the theoretical and simulation results verify that the constructed marker-MDS and marker-SVT codes provide the higher code rate than the existing run-length limited Varshamov-Tenengol’ts shifted Varshamov-Tenengol’ts (RLLVT-SVT) codes if n is not smaller than a lower bound f ( t ) which is determined by t .
- 22 April 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Communications Letters
- Vol. 28 (6) , 1253-1257
Rate Efficient Codes Correcting a Burst of Deletions or Insertions
In this paper, we construct two rate efficient codes of length n named as marker-MDS and marker-SVT codes which correct a burst of deletions/insertions of length b (error-free decoding), where b is not necessarily fixed as a constant but is proportional to n , i.e., b = tn , 0 < t < 1. Both of these two codes consist of binary marker codes which are employed to locate the burst of deletions/insertions. Also, the marker-MDS and marker-SVT codes consist of the maximum distance separable (MDS) codes and shifted Varshamov-Tenengol’ts (SVT) codes, respectively, which are responsible for correcting erasures caused in the synchronization stage. Both the theoretical and simulation results verify that the constructed marker-MDS and marker-SVT codes provide the higher code rate than the existing run-length limited Varshamov-Tenengol’ts shifted Varshamov-Tenengol’ts (RLLVT-SVT) codes if n is not smaller than a lower bound f ( t ) which is determined by t .
- 22 April 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Vehicular Technology
- Vol. 73 (9) , 13997-14002
Accelerating Neural BP-Based Decoder Using Coded Distributed Computing
While neural BP-based (NBP) decoders exhibit superior error correction performance compared to belief- propagation (BP) decoders, the NBP decoder's high computational and memory requirements impede its practical deployment in communication systems. To overcome this challenge, we propose a Coded Neural BP (CNBP) scheme to accelerate the NBP decoder in distributed environments, while considering storage constraints and providing resilience to stragglers. The key idea is to reformulate the primary operations of the NBP decoder as matrix- vector multiplications by introducing weight matrices and transformations. Based on this, the acceleration of the NBP decoder is achieved by speeding up matrix- vector multiplications using coded distributed computing. Extensive experiments conducted on Amazon EC2 cluster demonstrate that CNBP achieves notable acceleration and scalability performance without any loss in error correction performance.
- 22 April 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Vehicular Technology
- Vol. 73 (9) , 13997-14002
Accelerating Neural BP-Based Decoder Using Coded Distributed Computing
While neural BP-based (NBP) decoders exhibit superior error correction performance compared to belief- propagation (BP) decoders, the NBP decoder's high computational and memory requirements impede its practical deployment in communication systems. To overcome this challenge, we propose a Coded Neural BP (CNBP) scheme to accelerate the NBP decoder in distributed environments, while considering storage constraints and providing resilience to stragglers. The key idea is to reformulate the primary operations of the NBP decoder as matrix- vector multiplications by introducing weight matrices and transformations. Based on this, the acceleration of the NBP decoder is achieved by speeding up matrix- vector multiplications using coded distributed computing. Extensive experiments conducted on Amazon EC2 cluster demonstrate that CNBP achieves notable acceleration and scalability performance without any loss in error correction performance.