Meng Yang
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Meng Yang has a total of 42 co-authors. Here are the top co-authors by number of shared publications:
- Zhengchun Zhou
School of Information Science and Technology, Southwest Jiaotong University, Chengdu, Sichuan, China
7 shared publications - Hua Meng
School of Mathematics, Southwest Jiaotong University, Chengdu, Sichuan, China
5 shared publications - Shucong Tian
School of Mathematics, Southwest Jiaotong University, Chengdu, China
5 shared publications - Yang Yang
School of Mathematics, Southwest Jiaotong University, Chengdu, China
3 shared publications - Aravinda Sridhara Rao
Department of Electrical and Electronic Engineering, The University of Melbourne, Parkville, Melbourne, VIC, Australia
2 shared publications - Avik Ranjan Adhikary
School of Mathematics, Southwest Jiaotong University, Chengdu, China
2 shared publications
Latest Publications
- 25 August 2026
- Published by Elsevier in Physical Communication
- Vol. 80
Maximum information gain reinforcement learning based on the variational information bottleneck
- 25 August 2026
- Published by Elsevier in Physical Communication
- Vol. 80
Maximum information gain reinforcement learning based on the variational information bottleneck
- 6 February 2026
- Published by Elsevier in Physical Communication
- Vol. 75
Signal classification based on multi-Scale time-Frequency transformer
- 6 February 2026
- Published by Elsevier in Physical Communication
- Vol. 75
Signal classification based on multi-Scale time-Frequency transformer
- 1 January 2025
- Published by American Institute of Mathematical Sciences (AIMS) in Advances in Mathematics of Communications
- Vol. 19 (4) , 1089-1109
Constructions of golay complementary (array) sets based on extended boolean functions
In this paper, we propose a direct construction of Golay complementary sets (GCSs) with flexible length $ aq^m $ for set size $ q^{k} $ using extended Boolean functions (EBFs), where $ a $ and $ q $ are positive integers with $ a\leq q $. As an extension of GCSs, we also propose two-dimensional (2-D) Golay complementary array sets (GCASs) using 2-D EBFs. The parameters of the proposed GCASs are more flexible as compared to the previous works.- 1 January 2025
- Published by American Institute of Mathematical Sciences (AIMS) in Advances in Mathematics of Communications
- Vol. 19 (4) , 1089-1109
Constructions of golay complementary (array) sets based on extended boolean functions
In this paper, we propose a direct construction of Golay complementary sets (GCSs) with flexible length $ aq^m $ for set size $ q^{k} $ using extended Boolean functions (EBFs), where $ a $ and $ q $ are positive integers with $ a\leq q $. As an extension of GCSs, we also propose two-dimensional (2-D) Golay complementary array sets (GCASs) using 2-D EBFs. The parameters of the proposed GCASs are more flexible as compared to the previous works.- 1 January 2025
- Published by American Institute of Mathematical Sciences (AIMS) in Advances in Mathematics of Communications
- Vol. 19 (2) , 572-587
A unified construction of type-Ⅰ even length Z-complementary pairs based on generalized Boolean function
Constructions of Z-complementary pairs (ZCPs) using generalized Boolean functions (GBFs) have a very rich literature. However, these works only focus on ZCPs having lengths with specific forms. In this paper, we propose a unified framework to construct $ q $-phase ZCPs based on GBFs, which can obtain the algebraic structure of Type-Ⅰ ZCPs of any even length. Through the proposed unified framework, we can propose several direct constructions of even-length ZCPs, which are not reported before. Also, all the previous constructions of Type-Ⅰ ZCPs of even length based on GBFs, can be seen as special cases of the proposed construction.
- 1 January 2025
- Published by American Institute of Mathematical Sciences (AIMS) in Advances in Mathematics of Communications
- Vol. 19 (2) , 572-587
A unified construction of type-Ⅰ even length Z-complementary pairs based on generalized Boolean function
Constructions of Z-complementary pairs (ZCPs) using generalized Boolean functions (GBFs) have a very rich literature. However, these works only focus on ZCPs having lengths with specific forms. In this paper, we propose a unified framework to construct $ q $-phase ZCPs based on GBFs, which can obtain the algebraic structure of Type-Ⅰ ZCPs of any even length. Through the proposed unified framework, we can propose several direct constructions of even-length ZCPs, which are not reported before. Also, all the previous constructions of Type-Ⅰ ZCPs of even length based on GBFs, can be seen as special cases of the proposed construction.
- 1 January 2025
- Published by American Institute of Mathematical Sciences (AIMS) in Advances in Mathematics of Communications
- Vol. 19 (1) , 202-218
Constructions of optimal Z-periodic complementary sequence sets with large zero correlation zone
Z-periodic complementary sequence sets (ZPCSSs) have potential applications in multi-carrier code-division multiple access (MC-CDMA) communication systems and MIMO channel estimation due to their favourable correlation properties. In this paper, we explore several methods to design optimal ZPCSSs. More specifically, combining complete complementary code (CCC) and optimal ZPCSS, we design optimal ZPCSSs with flexible parameters, using the Kronecker product. Besides, using product sequences, we obtain the optimal ZPCSS with a large zero correlation zone (ZCZ) and optimal ZPCSSs with flexible flock sizes, respectively. Finally, we can obtain more optimal ZPCSSs using the left shift operator. In particular, some lengths of the resultant sequence sets have not been reported before.
- 1 January 2025
- Published by American Institute of Mathematical Sciences (AIMS) in Advances in Mathematics of Communications
- Vol. 19 (1) , 202-218
Constructions of optimal Z-periodic complementary sequence sets with large zero correlation zone
Z-periodic complementary sequence sets (ZPCSSs) have potential applications in multi-carrier code-division multiple access (MC-CDMA) communication systems and MIMO channel estimation due to their favourable correlation properties. In this paper, we explore several methods to design optimal ZPCSSs. More specifically, combining complete complementary code (CCC) and optimal ZPCSS, we design optimal ZPCSSs with flexible parameters, using the Kronecker product. Besides, using product sequences, we obtain the optimal ZPCSS with a large zero correlation zone (ZCZ) and optimal ZPCSSs with flexible flock sizes, respectively. Finally, we can obtain more optimal ZPCSSs using the left shift operator. In particular, some lengths of the resultant sequence sets have not been reported before.
- 27 December 2024
- Published by Springer Nature in Applied Intelligence
- Vol. 55 (3) , 1-16
Efficient convolutional dual-attention transformer for automatic modulation recognition
- 27 December 2024
- Published by Springer Nature in Applied Intelligence
- Vol. 55 (3) , 1-16
Efficient convolutional dual-attention transformer for automatic modulation recognition
- 22 August 2024
- Published by Elsevier in Expert Systems with Applications
- Vol. 258
A sequential multi-agent reinforcement learning framework for different action spaces
- 22 August 2024
- Published by Elsevier in Expert Systems with Applications
- Vol. 258
A sequential multi-agent reinforcement learning framework for different action spaces
- 22 May 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Journal
- Vol. 24 (20) , 32455-32467
Lightweight Multifeature Fusion Network for Finger Vein Recognition
Lightweight networks with simplified structure and fewer parameters ensure fast computation and response time on mobile devices, making them well-suited for finger vein recognition tasks. However, these networks are limited by their parameter settings, leading to a lack of network generalization ability. Furthermore, they primarily focus on extracting semantic features, making it challenging to learn topological features such as the number and connectivity of veins within the image. Inspired by the remarkable performance of topology in characterizing fine-grained features of images, we propose a novel lightweight multi-feature fusion network capable of simultaneously extracting both semantic and topological features of finger veins. Specifically, a lightweight convolutional network is firstly designed to focus on the semantic features of the images. Then, a deep and shallow feature fusion method is introduced to accentuate the intrinsic connection among different network layers, enhancing the lightweight network’s feature generalization capacity. Moreover, we utilize persistent homology computations to obtain the persistence diagram (PD) as topological prior information. Through vectorized representations, we obtain topological features, which are combined with the extracted semantic features to assist the network in learning the data’s geometric structural characteristics. By integrating these two types of features, we achieve superior feature integrity, improving the overall recognition performance of the network. Experimental results on three publicly available databases show that our network outperforms other known ones in terms of recognition performance.
- 22 May 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Journal
- Vol. 24 (20) , 32455-32467
Lightweight Multifeature Fusion Network for Finger Vein Recognition
Lightweight networks with simplified structure and fewer parameters ensure fast computation and response time on mobile devices, making them well-suited for finger vein recognition tasks. However, these networks are limited by their parameter settings, leading to a lack of network generalization ability. Furthermore, they primarily focus on extracting semantic features, making it challenging to learn topological features such as the number and connectivity of veins within the image. Inspired by the remarkable performance of topology in characterizing fine-grained features of images, we propose a novel lightweight multi-feature fusion network capable of simultaneously extracting both semantic and topological features of finger veins. Specifically, a lightweight convolutional network is firstly designed to focus on the semantic features of the images. Then, a deep and shallow feature fusion method is introduced to accentuate the intrinsic connection among different network layers, enhancing the lightweight network’s feature generalization capacity. Moreover, we utilize persistent homology computations to obtain the persistence diagram (PD) as topological prior information. Through vectorized representations, we obtain topological features, which are combined with the extracted semantic features to assist the network in learning the data’s geometric structural characteristics. By integrating these two types of features, we achieve superior feature integrity, improving the overall recognition performance of the network. Experimental results on three publicly available databases show that our network outperforms other known ones in terms of recognition performance.
- 29 March 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Aerospace and Electronic Systems
- Vol. 60 (4) , 4463-4478
Reinforcement Learning-Based MIMO Radar Multitarget Detection Assisted by Bayesian Inference
Reinforcement Learning (RL) has been used to implement the perception-action cycle of multi-input-multi-output (MIMO) cognitive radar (CR). This allows for the adaptive optimization of the radar's beampattern, which is guided by information from echoes and an appropriate reward signal. However, the present approaches rely on a greedy DOA estimation to select candidate angles, which means that decisions are primarily based on detection results from the last pulse. If the system misses a target, it can be time-consuming to recapture it, and this can limit the detection performance. To address this issue, we develop a data-driven method for executing DOA estimation by utilizing the Bayesian inference to evaluate the likelihood of an angle containing a target based on historical detection information. Furthermore, to accommodate dynamic scenarios, a decay method for historical experience is proposed, allowing the system to adapt to environmental changes dynamically. The simulation results show that the RL-based MIMO radar with our refined DOA estimation module outperforms existing RL-based detector, providing the SOTA detection performance by focusing more frequently on essential angles, even in small-scale system setups.
- 29 March 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Aerospace and Electronic Systems
- Vol. 60 (4) , 4463-4478
Reinforcement Learning-Based MIMO Radar Multitarget Detection Assisted by Bayesian Inference
Reinforcement Learning (RL) has been used to implement the perception-action cycle of multi-input-multi-output (MIMO) cognitive radar (CR). This allows for the adaptive optimization of the radar's beampattern, which is guided by information from echoes and an appropriate reward signal. However, the present approaches rely on a greedy DOA estimation to select candidate angles, which means that decisions are primarily based on detection results from the last pulse. If the system misses a target, it can be time-consuming to recapture it, and this can limit the detection performance. To address this issue, we develop a data-driven method for executing DOA estimation by utilizing the Bayesian inference to evaluate the likelihood of an angle containing a target based on historical detection information. Furthermore, to accommodate dynamic scenarios, a decay method for historical experience is proposed, allowing the system to adapt to environmental changes dynamically. The simulation results show that the RL-based MIMO radar with our refined DOA estimation module outperforms existing RL-based detector, providing the SOTA detection performance by focusing more frequently on essential angles, even in small-scale system setups.
- 30 January 2023
- Published by Springer Nature in The Visual Computer
- Vol. 40 (1) , 303-318
Evolving graph-based video crowd anomaly detection
- 30 January 2023
- Published by Springer Nature in The Visual Computer
- Vol. 40 (1) , 303-318
Evolving graph-based video crowd anomaly detection