Chunxiao Chigan
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
Chunxiao Chigan has a total of 11 co-authors. Here are the top co-authors by number of shared publications:
- Xin Ma
Department of Electrical and Computer Engineering, University of Massachusetts Lowell, Lowell, MA, USA
4 shared publications - Mahdi Shahbakhti
Department of Mechanical Engineering, University of Alberta, Edmonton, AB, Canada
2 shared publications - Jialiang Li
State Key Laboratory of Inorganic Synthesis and Preparative Chemistry, College of Chemistry, Jilin University, 2699 Qianjin Street, Changchun 130012, P. R. China
1 shared publication Lei Li
Department of Electrical and Computer Engineering, University of Massachusetts Lowell, Lowell, MA, USA
1 shared publication- Lei Li
Electrical and Computer Engineering, University of Massachusetts Lowell, 14710 Lowell, Massachusetts United States
1 shared publication - Sheikh Habibul Islam
Department of Electrical and Computer Engineering, University of Massachusetts Lowell, Lowell, MA, USA
1 shared publication
Latest Publications
- Open Access
- 2 December 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Machine Learning in Communications and Networking
- Vol. 4, 60-75
Adaptive Nonlinear Digital Self-Interference Cancellation for Full-Duplex Wireless Systems Using Hypernetwork-Based Incremental Learning
Achieving effective self-interference cancellation (SIC) in full-duplex (FD) wireless communication systems under time-varying channel conditions remains a significant challenge. To address this challenge, we propose a novel adaptive SIC solution through leveraging Hyper Neural Networks (HyperNet) and incremental learning (IL). Unlike the existing methods that rely on offline training or lack real-time adaptability, our approach enables autonomous learning and fast adaptation to the complex, nonlinear, and time-varying nature of self-interference (SI) channels. It effectively addresses dynamic adaptation challenges, such as catastrophic forgetting, through the use of experience replay (ER). Our experimental results show that traditional model-based methods exhibit limited adaptability under dynamic channel conditions, while conventional data-driven models fail to maintain consistent performance without the adaptive capabilities provided by IL. In contrast, the proposed HyperNet-based IL model reduces training time by 33% and achieves three times faster convergence compared to a standalone HyperNet trained separately for each static condition. Extensive evaluations using simulated datasets that emulate real-world scenarios demonstrate that our approach consistently achieves SI suppression down to the noise floor. It also delivers significantly lower computational complexity and training time. These improvements collectively enhance the efficiency and reliability of FD communication systems operating in dynamic wireless environments.
- 2 December 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Machine Learning in Communications and Networking
- Vol. 4, 60-75
Open AccessAdaptive Nonlinear Digital Self-Interference Cancellation for Full-Duplex Wireless Systems Using Hypernetwork-Based Incremental Learning
Achieving effective self-interference cancellation (SIC) in full-duplex (FD) wireless communication systems under time-varying channel conditions remains a significant challenge. To address this challenge, we propose a novel adaptive SIC solution through leveraging Hyper Neural Networks (HyperNet) and incremental learning (IL). Unlike the existing methods that rely on offline training or lack real-time adaptability, our approach enables autonomous learning and fast adaptation to the complex, nonlinear, and time-varying nature of self-interference (SI) channels. It effectively addresses dynamic adaptation challenges, such as catastrophic forgetting, through the use of experience replay (ER). Our experimental results show that traditional model-based methods exhibit limited adaptability under dynamic channel conditions, while conventional data-driven models fail to maintain consistent performance without the adaptive capabilities provided by IL. In contrast, the proposed HyperNet-based IL model reduces training time by 33% and achieves three times faster convergence compared to a standalone HyperNet trained separately for each static condition. Extensive evaluations using simulated datasets that emulate real-world scenarios demonstrate that our approach consistently achieves SI suppression down to the noise floor. It also delivers significantly lower computational complexity and training time. These improvements collectively enhance the efficiency and reliability of FD communication systems operating in dynamic wireless environments.
- 16 January 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Intelligent Transportation Systems
- Vol. 25 (7) , 7323-7338
Graph Convolutional Network Based Multi-Objective Meta-Deep Q-Learning for Eco-Routing
Route selection can greatly affect vehicle fuel consumption and emissions. Finding the most fuel/energy-efficient route is known as the eco-routing problem. Existing eco-routing solutions do not effectively consider the critical traffic signal information and rely on fuel consumption models that may not be sufficiently accurate. To address the eco-routing problem in a signalized traffic network, this paper proposes a graph convolutional network based multi-objective meta-deep Q-learning (GM $^{\bm{2}}$ DQL) method. The problem is formulated as dynamic multi-objective Markov decision processes (MOMDP) and is tackled through deep reinforcement learning and meta-learning. We identify that graph convolutional network (GCN) is an efficient and suitable feature representation for a signalized traffic network. GM $^{\bm{2}}$ DQL can explore the optimal routes with respect to drivers’ different preferences on saving fuel and travel time. Through GM $^{\bm{2}}$ DQL, the agent is trained under a series of learning environments that are characterized by historical vehicle trajectories, fuel consumption data, and traffic signal data in the remote data center. The vehicle requesting eco-routing service can download the model that represents the action value function of the historical dynamic driving conditions. The model in the vehicle can quickly adapt to the most recent driving condition through online one-shot learning and predict the optimal eco-routes for the subsequent unseen driving conditions of the signalized traffic network. Extensive proof-of-concept experiments validate that GM $^{\bm{2}}$ DQL can effectively discover optimal eco-routes. It saves up to 71% travel time and 62% fuel, compared to the conventional shortest-path routing strategy that is widely used in navigation systems.
- 16 January 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Intelligent Transportation Systems
- Vol. 25 (7) , 7323-7338
Graph Convolutional Network Based Multi-Objective Meta-Deep Q-Learning for Eco-Routing
Route selection can greatly affect vehicle fuel consumption and emissions. Finding the most fuel/energy-efficient route is known as the eco-routing problem. Existing eco-routing solutions do not effectively consider the critical traffic signal information and rely on fuel consumption models that may not be sufficiently accurate. To address the eco-routing problem in a signalized traffic network, this paper proposes a graph convolutional network based multi-objective meta-deep Q-learning (GM $^{\bm{2}}$ DQL) method. The problem is formulated as dynamic multi-objective Markov decision processes (MOMDP) and is tackled through deep reinforcement learning and meta-learning. We identify that graph convolutional network (GCN) is an efficient and suitable feature representation for a signalized traffic network. GM $^{\bm{2}}$ DQL can explore the optimal routes with respect to drivers’ different preferences on saving fuel and travel time. Through GM $^{\bm{2}}$ DQL, the agent is trained under a series of learning environments that are characterized by historical vehicle trajectories, fuel consumption data, and traffic signal data in the remote data center. The vehicle requesting eco-routing service can download the model that represents the action value function of the historical dynamic driving conditions. The model in the vehicle can quickly adapt to the most recent driving condition through online one-shot learning and predict the optimal eco-routes for the subsequent unseen driving conditions of the signalized traffic network. Extensive proof-of-concept experiments validate that GM $^{\bm{2}}$ DQL can effectively discover optimal eco-routes. It saves up to 71% travel time and 62% fuel, compared to the conventional shortest-path routing strategy that is widely used in navigation systems.
- 26 January 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Vehicular Technology
- Vol. 72 (6) , 7251-7264
Deep Learning Based Distributed Meta-Learning for Fast and Accurate Online Adaptive Powertrain Fuel Consumption Modeling
Connected vehicle-based distributed meta-regression (CV-DMR) algorithms from our previous work facilitate online adaptive microscopic fuel consumption modeling, which can support model-based vehicle and powertrain control (MB-VPC) for fuel-efficient vehicle operations. By taking advantage of accessible online computational resources from connected vehicles (CVs) and CV remote data center, CV-DMR can adapt the microscopic fuel consumption model to an unseen driving condition with limited training data. However, the computational complexity of obtaining the normalized feature map in CV-DMR during training is high. It can take long time to adapt the model to a new driving condition, especially when the new driving condition is considerably different from the pre-learnt driving conditions. To reduce the computational complexity of learning/training, we propose to apply deep neural networks as the model representation method and newly design efficient deep learning based CV-supported distributed meta-learning (CV-DML) algorithms for adaptive fuel consumption modeling. To further generalize the model and reduce model adaptation time, the concept of meta-suggestion is newly proposed in CV-DML. Substantial proof-of-concept experiments are conducted with steady-state and transient vehicle engine data to evaluate the model performance in terms of model prediction accuracy, model adaptation speed, and fuel saving. Compared to the physical model and non-CV-supported model adaptation, the prediction accuracy of CV-DML is improved by 34% - 87% and 17% - 30%, respectively. Due to the improved accuracy, up to 9.4% of fuel is saved. To achieve the same the prediction accuracy and fuel saving, the model adaptation of CV-DML is 18 to 1,300 times faster than that of CV-DMR.
- 26 January 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Vehicular Technology
- Vol. 72 (6) , 7251-7264
Deep Learning Based Distributed Meta-Learning for Fast and Accurate Online Adaptive Powertrain Fuel Consumption Modeling
Connected vehicle-based distributed meta-regression (CV-DMR) algorithms from our previous work facilitate online adaptive microscopic fuel consumption modeling, which can support model-based vehicle and powertrain control (MB-VPC) for fuel-efficient vehicle operations. By taking advantage of accessible online computational resources from connected vehicles (CVs) and CV remote data center, CV-DMR can adapt the microscopic fuel consumption model to an unseen driving condition with limited training data. However, the computational complexity of obtaining the normalized feature map in CV-DMR during training is high. It can take long time to adapt the model to a new driving condition, especially when the new driving condition is considerably different from the pre-learnt driving conditions. To reduce the computational complexity of learning/training, we propose to apply deep neural networks as the model representation method and newly design efficient deep learning based CV-supported distributed meta-learning (CV-DML) algorithms for adaptive fuel consumption modeling. To further generalize the model and reduce model adaptation time, the concept of meta-suggestion is newly proposed in CV-DML. Substantial proof-of-concept experiments are conducted with steady-state and transient vehicle engine data to evaluate the model performance in terms of model prediction accuracy, model adaptation speed, and fuel saving. Compared to the physical model and non-CV-supported model adaptation, the prediction accuracy of CV-DML is improved by 34% - 87% and 17% - 30%, respectively. Due to the improved accuracy, up to 9.4% of fuel is saved. To achieve the same the prediction accuracy and fuel saving, the model adaptation of CV-DML is 18 to 1,300 times faster than that of CV-DMR.
- 15 June 2020
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Vehicular Technology
- Vol. 69 (9) , 9553-9565
Connected Vehicle Based Distributed Meta-Learning for Online Adaptive Engine/Powertrain Fuel Consumption Modeling
A microscopic fuel consumption model with respect to comprehensive fuel consumption key impact factors that can reflect dynamic engine operations and various real-world driving conditions is essential for fuel efficient model-based vehicle and powertrain control (MB-VPC). Existing microscopic fuel consumption models are mostly steady state models with limited vehicle parameters as dynamic variables. Data-driven solutions can make the model comprehensive by introducing broad impact factors. However, without the required accessible online computational capability and fast model adaptation mechanisms, those data-driven solutions cannot quickly adapt to unseen driving conditions and engine condition changes based on real-world vehicle data to support MB-VPC. In this paper, a connected vehicle-based data mining (CV-DM) framework is proposed to achieve online adaptive dynamic fuel consumption modeling through knowledge sharing over CVs and the CV remote data center. Based on the CV-DM framework, CV-supported Distributed Meta-regression (CV-DMR) algorithms are developed to realize a fast few-shot adaptation with limited training data. Extensive proof-of-concept experiments are conducted with steady-state and transient vehicle engine data. Compared to the baseline physical model and the existing non-adaptive grey-box model, prediction accuracy is improved by 47% - 85% and 38% - 80% respectively with only limited training data needed for the subject vehicle. Accordingly, a fuel savings of up to 9.4% can be achieved owing to the improvement of prediction accuracy.
- 15 June 2020
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Vehicular Technology
- Vol. 69 (9) , 9553-9565
Connected Vehicle Based Distributed Meta-Learning for Online Adaptive Engine/Powertrain Fuel Consumption Modeling
A microscopic fuel consumption model with respect to comprehensive fuel consumption key impact factors that can reflect dynamic engine operations and various real-world driving conditions is essential for fuel efficient model-based vehicle and powertrain control (MB-VPC). Existing microscopic fuel consumption models are mostly steady state models with limited vehicle parameters as dynamic variables. Data-driven solutions can make the model comprehensive by introducing broad impact factors. However, without the required accessible online computational capability and fast model adaptation mechanisms, those data-driven solutions cannot quickly adapt to unseen driving conditions and engine condition changes based on real-world vehicle data to support MB-VPC. In this paper, a connected vehicle-based data mining (CV-DM) framework is proposed to achieve online adaptive dynamic fuel consumption modeling through knowledge sharing over CVs and the CV remote data center. Based on the CV-DM framework, CV-supported Distributed Meta-regression (CV-DMR) algorithms are developed to realize a fast few-shot adaptation with limited training data. Extensive proof-of-concept experiments are conducted with steady-state and transient vehicle engine data. Compared to the baseline physical model and the existing non-adaptive grey-box model, prediction accuracy is improved by 47% - 85% and 38% - 80% respectively with only limited training data needed for the subject vehicle. Accordingly, a fuel savings of up to 9.4% can be achieved owing to the improvement of prediction accuracy.
- 16 September 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Vehicular Technology
- Vol. 68 (11) , 11283-11294
An Energy-Efficient Privacy Preserving Security-Oriented DSA With Low Latency
Conventional cryptography techniques based upper layer secure wireless communication solutions are often subject to major issues such as complex key management, high computational complexity, etc. Physical layer security mechanisms are proposed as appealing alternative solutions for secure wireless communications. However, existing physical layer security schemes each faces some major limitations including limited achievable secrecy capacity, sensitive to time-varying channel status, high power consumption and long delay. To tackle these issues, we propose a novel energy efficient security-oriented dynamic spectrum allocation/assignment (S-DSA) algorithm to provide message privacy (confidentiality) with low delay. Our S-DSA algorithm solves the joint radio resource (spectrum and power) allocation and privacy preserving problems, which is modeled as a mixed-integer multi-objective optimization problem. The branch and bound mechanism integrated with non-dominated sorting genetic algorithm-II (NSGA-II) is proposed to search the optimal channel and power allocation strategy to maximize the achievable system secrecy capacity and minimize the latency and power consumption simultaneously. Simulations confirm that the confidentiality achieved by our serenity-oriented DSA induces negligible degradation on average throughput per user compared with exiting DSA schemes.
- 16 September 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Vehicular Technology
- Vol. 68 (11) , 11283-11294
An Energy-Efficient Privacy Preserving Security-Oriented DSA With Low Latency
Conventional cryptography techniques based upper layer secure wireless communication solutions are often subject to major issues such as complex key management, high computational complexity, etc. Physical layer security mechanisms are proposed as appealing alternative solutions for secure wireless communications. However, existing physical layer security schemes each faces some major limitations including limited achievable secrecy capacity, sensitive to time-varying channel status, high power consumption and long delay. To tackle these issues, we propose a novel energy efficient security-oriented dynamic spectrum allocation/assignment (S-DSA) algorithm to provide message privacy (confidentiality) with low delay. Our S-DSA algorithm solves the joint radio resource (spectrum and power) allocation and privacy preserving problems, which is modeled as a mixed-integer multi-objective optimization problem. The branch and bound mechanism integrated with non-dominated sorting genetic algorithm-II (NSGA-II) is proposed to search the optimal channel and power allocation strategy to maximize the achievable system secrecy capacity and minimize the latency and power consumption simultaneously. Simulations confirm that the confidentiality achieved by our serenity-oriented DSA induces negligible degradation on average throughput per user compared with exiting DSA schemes.
- 19 March 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Cognitive Communications and Networking
- Vol. 5 (3) , 504-516
On MMD-Based Secure Fusion Strategy for Robust Cooperative Spectrum Sensing
Collaborative spectrum sensing (CSS) in cognitive radio-based networks (CRNs) is vulnerable to spectrum sensing data falsification (SSDF) attack. Existing defense mechanisms are commonly subject to major limitations with unrealistic assumptions which can be easily violated in future wireless networks. Such assumptions include the number of honest users is in majority, the attackers' flip rates are identical and fixed, the adoption of hard decision approach in the fusion strategy and only TV sets are considered as the primary users to be protected. Essentially, all existing representative schemes utilize certain low-dimensional human-observed metric to distinguish malicious users and honest users based on domain knowledge. Therefore, these defense mechanisms cannot perform properly under certain condition, such as sensing reports with different distributions but have equal mean and variance. In this paper, we propose a secure fusion strategy which adopts “soft decision” method and can distinguish malicious users and honest users under any distribution of sensing reports using maximum mean discrepancy (MMD). Our proposed CSS scheme is suitable for general CRN application scenarios. The simulation results reflect the effects of different kernel functions and window size on the system performance, and show our proposed defense mechanism outperforms the existing works.
- 19 March 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Cognitive Communications and Networking
- Vol. 5 (3) , 504-516
On MMD-Based Secure Fusion Strategy for Robust Cooperative Spectrum Sensing
Collaborative spectrum sensing (CSS) in cognitive radio-based networks (CRNs) is vulnerable to spectrum sensing data falsification (SSDF) attack. Existing defense mechanisms are commonly subject to major limitations with unrealistic assumptions which can be easily violated in future wireless networks. Such assumptions include the number of honest users is in majority, the attackers' flip rates are identical and fixed, the adoption of hard decision approach in the fusion strategy and only TV sets are considered as the primary users to be protected. Essentially, all existing representative schemes utilize certain low-dimensional human-observed metric to distinguish malicious users and honest users based on domain knowledge. Therefore, these defense mechanisms cannot perform properly under certain condition, such as sensing reports with different distributions but have equal mean and variance. In this paper, we propose a secure fusion strategy which adopts “soft decision” method and can distinguish malicious users and honest users under any distribution of sensing reports using maximum mean discrepancy (MMD). Our proposed CSS scheme is suitable for general CRN application scenarios. The simulation results reflect the effects of different kernel functions and window size on the system performance, and show our proposed defense mechanism outperforms the existing works.
- 24 August 2010
- Published by Elsevier in Ad Hoc Networks
- Vol. 9 (3) , 418-429
LEAPER: A lightweight reliable and faithful data packet relaying framework for VANETs
- 24 August 2010
- Published by Elsevier in Ad Hoc Networks
- Vol. 9 (3) , 418-429
LEAPER: A lightweight reliable and faithful data packet relaying framework for VANETs
- 1 May 2010
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- No. 15503607,p. 1-6
On Resource-Aware Message Verification in VANETs
In Vehicular Ad Hoc Networks (VANETs), the massive messages are periodically broadcasted by each node and need to be processed by each receiver. Due to their enormous quantity, message verification in this context threatens excessive resource consumption for VANET nodes. Thus, in this paper a novel Resource-Aware Message Verification (RAMV) scheme is proposed to efficiently and securely verify the massive messages in VANETs. Streamlining the existing message verification schemes, RAMV keeps their resource consumption under control in face of massive messages exchanges. Meanwhile, it also ensures that the security requirements for message verification are all properly met, endorsed by both the theoretical analysis and simulation results.
- 1 May 2010
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- No. 15503607,p. 1-6
On Resource-Aware Message Verification in VANETs
In Vehicular Ad Hoc Networks (VANETs), the massive messages are periodically broadcasted by each node and need to be processed by each receiver. Due to their enormous quantity, message verification in this context threatens excessive resource consumption for VANET nodes. Thus, in this paper a novel Resource-Aware Message Verification (RAMV) scheme is proposed to efficiently and securely verify the massive messages in VANETs. Streamlining the existing message verification schemes, RAMV keeps their resource consumption under control in face of massive messages exchanges. Meanwhile, it also ensures that the security requirements for message verification are all properly met, endorsed by both the theoretical analysis and simulation results.
- 1 November 2006
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Globecom 2006
- No. 1930529X,p. 1-6
WSN07-3: RPB-MACn: A Relative Position Based Collision-free MAC Nucleus for Vehicular Ad Hoc Networks
Vehicular ad hoc networks (VANETs) will play a vital role in reducing the number of road accidents and fatalities as well as vehicular traffic optimization. Though an instance of mobile ad hoc networks (MANETs), the highly mobile and dynamic environment of VANET introduces new research challenges. To quickly adapt to the highly mobile and constantly changing topology and still meet the desired application requirements, it is vital for the VANET media access control (MAC) protocol guarantees wireless channel access with minimal delay and minimal exchange of control messages. This paper proposes an innovative relative-position-based MAC Nucleus (RPB- MACn). By combining the dedicated communication channel pair and the dedicated directional antenna associated to the relative position of vehicles, a conceptually contention-free RPB-MACn mechanism for VANET is realized (when coupling with dynamic power control mechanism). Consequently, RPB-MACn can meet all the key requirements of the MAC protocol in a VANET system.
- 1 November 2006
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Globecom 2006
- No. 1930529X,p. 1-6
WSN07-3: RPB-MACn: A Relative Position Based Collision-free MAC Nucleus for Vehicular Ad Hoc Networks
Vehicular ad hoc networks (VANETs) will play a vital role in reducing the number of road accidents and fatalities as well as vehicular traffic optimization. Though an instance of mobile ad hoc networks (MANETs), the highly mobile and dynamic environment of VANET introduces new research challenges. To quickly adapt to the highly mobile and constantly changing topology and still meet the desired application requirements, it is vital for the VANET media access control (MAC) protocol guarantees wireless channel access with minimal delay and minimal exchange of control messages. This paper proposes an innovative relative-position-based MAC Nucleus (RPB- MACn). By combining the dedicated communication channel pair and the dedicated directional antenna associated to the relative position of vehicles, a conceptually contention-free RPB-MACn mechanism for VANET is realized (when coupling with dynamic power control mechanism). Consequently, RPB-MACn can meet all the key requirements of the MAC protocol in a VANET system.