Yin Lu
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Yin Lu has a total of 47 co-authors. Here are the top co-authors by number of shared publications:
Haowei Bian
2 shared publications- Hongbo Zhu
Nanjing University of Posts and Telecommunications
2 shared publications Kai Mao
Nanjing University of Posts and Telecommunications
2 shared publicationsXiumei Wang
2 shared publicationsBao Kuanxin
Nanjing University of Posts and Telecommunications
1 shared publicationBian Haowei
Nanjing University of Information Science and Technology
1 shared publication
Latest Publications
- 29 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Vehicular Technology
- Vol. PP (00189545) , 1-6
A Constrained Decision Transformer Framework for Fast Adaptive Optimization in RIS-Assisted Wireless Communications
Reconfigurable intelligent surface (RIS) has been recognized as a revolutionary technique for reconfiguring wireless environments with low cost and high array gain. However, achieving a high passive beamforming gain requires a large number of activated reflecting elements, incurring non-negligible power consumption. To balance the fundamental tradeoff between transmission rate and energy consumption, we jointly optimize the base station (BS) transmit power, RIS beamforming, and RIS element switching control. The objective is to maximize the average achievable rate while satisfying constraints on the transmit power of the BS, the average power of the RIS, and the unit modulus of the RIS phase shifts. We formulate this problem as a constrained Markov decision process (CMDP) and propose the proximal policy optimization-Lagrangian (PPO-Lag) algorithm. Furthermore, to improve the adaptability of decision-making models across various scenarios, we develop a constrained Decision Transformer (CDT) framework that integrates offline pre-training with online fine-tuning. Specifically, we first propose a PPO-Lag algorithm to solve the resource allocation for RIS systems. Then we collect samples based on PPO-Lag in varying scenarios, i.e., different power constraints and numbers of RIS elements. Afterwards, CDT is pre-trained based on the collected offline samples and fine-tuned online utilizing a limited number of samples under the new scenario to enhance its generalization ability. Simulation results show that the proposed CDT framework can improve the achievable rate by 4.4% and converges four times faster compared to the benchmark algorithm.
- 29 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Vehicular Technology
- Vol. PP (00189545) , 1-6
A Constrained Decision Transformer Framework for Fast Adaptive Optimization in RIS-Assisted Wireless Communications
Reconfigurable intelligent surface (RIS) has been recognized as a revolutionary technique for reconfiguring wireless environments with low cost and high array gain. However, achieving a high passive beamforming gain requires a large number of activated reflecting elements, incurring non-negligible power consumption. To balance the fundamental tradeoff between transmission rate and energy consumption, we jointly optimize the base station (BS) transmit power, RIS beamforming, and RIS element switching control. The objective is to maximize the average achievable rate while satisfying constraints on the transmit power of the BS, the average power of the RIS, and the unit modulus of the RIS phase shifts. We formulate this problem as a constrained Markov decision process (CMDP) and propose the proximal policy optimization-Lagrangian (PPO-Lag) algorithm. Furthermore, to improve the adaptability of decision-making models across various scenarios, we develop a constrained Decision Transformer (CDT) framework that integrates offline pre-training with online fine-tuning. Specifically, we first propose a PPO-Lag algorithm to solve the resource allocation for RIS systems. Then we collect samples based on PPO-Lag in varying scenarios, i.e., different power constraints and numbers of RIS elements. Afterwards, CDT is pre-trained based on the collected offline samples and fine-tuned online utilizing a limited number of samples under the new scenario to enhance its generalization ability. Simulation results show that the proposed CDT framework can improve the achievable rate by 4.4% and converges four times faster compared to the benchmark algorithm.
- Open Access
Enhanced Simultaneous Localization and Mapping Algorithm Based on Deep Learning for Highly Dynamic Environment
Visual simultaneous localization and mapping (SLAM) is a critical technology for autonomous navigation in dynamic environments. However, traditional SLAM algorithms often struggle to maintain accuracy in highly dynamic environments, where elements undergo significant, rapid, and unpredictable changes, leading to asymmetric information acquisition. Aiming to improve the accuracy of the SLAM algorithm in a dynamic environment, a dynamic SLAM algorithm based on deep learning is proposed. Firstly, YOLOv10n is used to improve the front end of the system, and semantic information is added to each frame of the image. Then, ORB-SLAM2 is used to extract feature points in each region of each frame and retrieve semantic information from YOLOv10n. Finally, through the map construction thread, the dynamic object feature points extracted by YOLOv10n are eliminated, and the construction of a static map is realized. The experimental results show that the accuracy of the proposed algorithm is improved by more than 96% compared with the state-of-the-art ORB-SLAM2 in a highly dynamic environment. Compared with other dynamic SLAM algorithms, the proposed algorithm has improved both accuracy and runtime.
Open AccessEnhanced Simultaneous Localization and Mapping Algorithm Based on Deep Learning for Highly Dynamic Environment
Visual simultaneous localization and mapping (SLAM) is a critical technology for autonomous navigation in dynamic environments. However, traditional SLAM algorithms often struggle to maintain accuracy in highly dynamic environments, where elements undergo significant, rapid, and unpredictable changes, leading to asymmetric information acquisition. Aiming to improve the accuracy of the SLAM algorithm in a dynamic environment, a dynamic SLAM algorithm based on deep learning is proposed. Firstly, YOLOv10n is used to improve the front end of the system, and semantic information is added to each frame of the image. Then, ORB-SLAM2 is used to extract feature points in each region of each frame and retrieve semantic information from YOLOv10n. Finally, through the map construction thread, the dynamic object feature points extracted by YOLOv10n are eliminated, and the construction of a static map is realized. The experimental results show that the accuracy of the proposed algorithm is improved by more than 96% compared with the state-of-the-art ORB-SLAM2 in a highly dynamic environment. Compared with other dynamic SLAM algorithms, the proposed algorithm has improved both accuracy and runtime.
- 21 January 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Vehicular Technology
- Vol. 74 (6) , 8989-9001
Blockchain-Enabled Credible Multi-Operator Spectrum Sharing in UAV Communication Systems
With the continuous advancement of the Unmanned Aerial Vehicle (UAV) communication industry, the efficient allocation of scarce spectrum resources to UAVs has become a pressing issue. Given the unique nature of UAVs and the broadcast nature of wireless channels, there are significant instances of unregulated flight operations and malicious attacks, making it essential to ensure the security and fairness of spectrum allocation. The previous work on spectrum sharing schemes seldom considers UAV communication scenarios and fails to consider the transaction environment. This paper employs combinatorial auctions and Stackelberg games to study spectrum trading between multiple base stations (BSs) and UAVs to address these issues. A dynamic, demand-driven spectrum trading model is proposed that accounts for UAVs' flexibility and evolving spectrum needs, maximizing utility for both UAVs and BSs. Subsequently, a blockchain-based spectrum sharing framework is designed, in which the blockchain is made public to trading participants to uphold decentralization. It reflects transactional fairness and reliability through mutual evaluations between the blockchain management platform and trading participants. To simulate a realistic trading environment, we consider specific attack scenarios to assess the credibility guarantee of spectrum trading under blockchain. The simulation results validate the effectiveness of the blockchain-based spectrum trading scheme, enhancing spectrum utilization, ensuring the utility of all entities involved, and maintaining the reliability of the trading process.
- 21 January 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Vehicular Technology
- Vol. 74 (6) , 8989-9001
Blockchain-Enabled Credible Multi-Operator Spectrum Sharing in UAV Communication Systems
With the continuous advancement of the Unmanned Aerial Vehicle (UAV) communication industry, the efficient allocation of scarce spectrum resources to UAVs has become a pressing issue. Given the unique nature of UAVs and the broadcast nature of wireless channels, there are significant instances of unregulated flight operations and malicious attacks, making it essential to ensure the security and fairness of spectrum allocation. The previous work on spectrum sharing schemes seldom considers UAV communication scenarios and fails to consider the transaction environment. This paper employs combinatorial auctions and Stackelberg games to study spectrum trading between multiple base stations (BSs) and UAVs to address these issues. A dynamic, demand-driven spectrum trading model is proposed that accounts for UAVs' flexibility and evolving spectrum needs, maximizing utility for both UAVs and BSs. Subsequently, a blockchain-based spectrum sharing framework is designed, in which the blockchain is made public to trading participants to uphold decentralization. It reflects transactional fairness and reliability through mutual evaluations between the blockchain management platform and trading participants. To simulate a realistic trading environment, we consider specific attack scenarios to assess the credibility guarantee of spectrum trading under blockchain. The simulation results validate the effectiveness of the blockchain-based spectrum trading scheme, enhancing spectrum utilization, ensuring the utility of all entities involved, and maintaining the reliability of the trading process.
- Open Access
- 10 January 2025
- Published by EWA Publishing in Applied and Computational Engineering
- Vol. 121 (1) , 198-208
Automatic Pricing and Replenishment Decisions for Vegetable Products
The perishable nature of vegetable products poses significant challenges in inventory management and pricing, often leading to high waste rates and economic losses. This study addresses these issues by proposing an automated pricing and replenishment decision model that leverages statistical analysis, regression methods, and time-series forecasting. Utilizing three years of sales data from a supermarket, the research develops models to analyze the relationship between sales volume and cost-plus pricing, optimize replenishment strategies, and predict pricing for various vegetable categories. The results demonstrate that dynamic pricing and replenishment decisions can reduce waste, increase profits, and enhance supply chain efficiency. The study highlights the practicality of using machine learning models like LSTM to forecast demand and pricing while acknowledging limitations such as simplifying assumptions and the need for multidimensional data. Future research directions include integrating external factors, such as climate impacts, to improve the model’s adaptability and accuracy.
- 10 January 2025
- Published by EWA Publishing in Applied and Computational Engineering
- Vol. 121 (1) , 198-208
Open AccessAutomatic Pricing and Replenishment Decisions for Vegetable Products
The perishable nature of vegetable products poses significant challenges in inventory management and pricing, often leading to high waste rates and economic losses. This study addresses these issues by proposing an automated pricing and replenishment decision model that leverages statistical analysis, regression methods, and time-series forecasting. Utilizing three years of sales data from a supermarket, the research develops models to analyze the relationship between sales volume and cost-plus pricing, optimize replenishment strategies, and predict pricing for various vegetable categories. The results demonstrate that dynamic pricing and replenishment decisions can reduce waste, increase profits, and enhance supply chain efficiency. The study highlights the practicality of using machine learning models like LSTM to forecast demand and pricing while acknowledging limitations such as simplifying assumptions and the need for multidimensional data. Future research directions include integrating external factors, such as climate impacts, to improve the model’s adaptability and accuracy.
- Open Access
Improved Scheme for Data Aggregation of Distributed Oracle for Intelligent Internet of Things
Oracle is a data supply mechanism that provides real-world data for blockchain. It serves as a bridge between blockchain and the IoT world, playing a crucial role in solving problems such as data sharing and device management in the IoT field. The main challenge at this stage is determining how to achieve data privacy protection in distributed Oracle machines to safeguard the value hidden in data on the blockchain. In this paper, we propose an improved scheme for distributed Oracle data aggregation based on Paillier encryption algorithm, which achieves end-to-end data privacy protection from devices to users. To address the issue of dishonest distributed Oracle machines running out of funds, we have designed an algorithm called PICA (Paillier-based InChain Aggregation). Based on the aggregation on the Chainlink chain and the Paillier encryption algorithm, random numbers are introduced to avoid the problem of dishonest Oracle machines running out of funds. We use the traffic coverage method to solve the problem of exposed request paths in distributed Oracle machines. Simulation and experimental results show that in small and medium-sized IoT application scenarios with 10,000 data nodes, each additional false request in a single request will result in a delay of about 2 s in data acquisition and can achieve a request response time of 20 s. The proposed method can achieve user data privacy protection.
Open AccessImproved Scheme for Data Aggregation of Distributed Oracle for Intelligent Internet of Things
Oracle is a data supply mechanism that provides real-world data for blockchain. It serves as a bridge between blockchain and the IoT world, playing a crucial role in solving problems such as data sharing and device management in the IoT field. The main challenge at this stage is determining how to achieve data privacy protection in distributed Oracle machines to safeguard the value hidden in data on the blockchain. In this paper, we propose an improved scheme for distributed Oracle data aggregation based on Paillier encryption algorithm, which achieves end-to-end data privacy protection from devices to users. To address the issue of dishonest distributed Oracle machines running out of funds, we have designed an algorithm called PICA (Paillier-based InChain Aggregation). Based on the aggregation on the Chainlink chain and the Paillier encryption algorithm, random numbers are introduced to avoid the problem of dishonest Oracle machines running out of funds. We use the traffic coverage method to solve the problem of exposed request paths in distributed Oracle machines. Simulation and experimental results show that in small and medium-sized IoT application scenarios with 10,000 data nodes, each additional false request in a single request will result in a delay of about 2 s in data acquisition and can achieve a request response time of 20 s. The proposed method can achieve user data privacy protection.
- Open Access
Adaptive UAV Navigation Method Based on AHRS
To address the inaccuracy of the Constant Acceleration/Constant Velocity (CA/CV) model as the state equation in describing the relative motion state in UAV relative navigation, an adaptive UAV relative navigation method is proposed, which is based on the UAV attitude information provided by Attitude and Heading Reference System (AHRS). The proposed method utilizes the AHRS output attitude parameters as the benchmark for dead reckoning and derives a relative navigation state equation with attitude error as process noise. By integrating the extended Kalman filter output for relative state estimation and employing an adaptive decision rule designed using the innovation of the filter update phase, the proposed method recalculates motion states deviating from the actual motion using the Tasmanian Devil Optimization (TDO) algorithm. The simulation results show that, compared with the CA/CV model, the proposed method reduces the relative position errors by 12%, 23%, and 32% in the X, Y, and Z directions, respectively, and that it reduces the relative velocity errors by 350%, 330%, and 300%, respectively. There is a significant improvement in the relative navigation accuracy.
Open AccessAdaptive UAV Navigation Method Based on AHRS
To address the inaccuracy of the Constant Acceleration/Constant Velocity (CA/CV) model as the state equation in describing the relative motion state in UAV relative navigation, an adaptive UAV relative navigation method is proposed, which is based on the UAV attitude information provided by Attitude and Heading Reference System (AHRS). The proposed method utilizes the AHRS output attitude parameters as the benchmark for dead reckoning and derives a relative navigation state equation with attitude error as process noise. By integrating the extended Kalman filter output for relative state estimation and employing an adaptive decision rule designed using the innovation of the filter update phase, the proposed method recalculates motion states deviating from the actual motion using the Tasmanian Devil Optimization (TDO) algorithm. The simulation results show that, compared with the CA/CV model, the proposed method reduces the relative position errors by 12%, 23%, and 32% in the X, Y, and Z directions, respectively, and that it reduces the relative velocity errors by 350%, 330%, and 300%, respectively. There is a significant improvement in the relative navigation accuracy.
- Open Access
A Sketch-Based Fine-Grained Proportional Integral Queue Management Method
The phenomenon “bufferbloat” occurs when the buffers of the network intermediary nodes fill up, causing long queuing delays. This has a significant negative impact on the quality of service of network applications, particularly those that are sensitive to time delay. Many active queue management (AQM) algorithms have been proposed to overcome this problem. Those AQMs attempt to maintain minimal queuing delays and good throughput by purposefully dropping packets at network intermediary nodes. However, the existing AQM algorithms mostly drop packets randomly based on a certain metric such as queue length or queuing delay, which fails to achieve fine-grained differentiation of data streams. In this paper, we propose a fine-grained sketch-based proportional integral queue management algorithm S-PIE, which uses an additional measurement structure Sketch for packet frequency share judgment based on the existing PIE algorithm for the fine-grained differentiation between data streams and adjust the drop policy for a differentiated packet drop. Experimental results on the NS-3 simulation platform show that the S-PIE algorithm achieves lower average queue length and RTT and higher fairness than PIE, RED, and CoDel algorithms while maintaining a similar throughput performance, maintaining network availability and stability, and improving network quality of service.
Open AccessA Sketch-Based Fine-Grained Proportional Integral Queue Management Method
The phenomenon “bufferbloat” occurs when the buffers of the network intermediary nodes fill up, causing long queuing delays. This has a significant negative impact on the quality of service of network applications, particularly those that are sensitive to time delay. Many active queue management (AQM) algorithms have been proposed to overcome this problem. Those AQMs attempt to maintain minimal queuing delays and good throughput by purposefully dropping packets at network intermediary nodes. However, the existing AQM algorithms mostly drop packets randomly based on a certain metric such as queue length or queuing delay, which fails to achieve fine-grained differentiation of data streams. In this paper, we propose a fine-grained sketch-based proportional integral queue management algorithm S-PIE, which uses an additional measurement structure Sketch for packet frequency share judgment based on the existing PIE algorithm for the fine-grained differentiation between data streams and adjust the drop policy for a differentiated packet drop. Experimental results on the NS-3 simulation platform show that the S-PIE algorithm achieves lower average queue length and RTT and higher fairness than PIE, RED, and CoDel algorithms while maintaining a similar throughput performance, maintaining network availability and stability, and improving network quality of service.
- 23 April 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 836-840
Performance Analysis for Millimeter Wave Array Antenna Systems Based on Influence of Amplitude Quantization Error
For phased array antenna, the required pattern and its beam pointing are generated by different excitation amplitude and phase of each array element. Various kinds of errors will be introduced in manufacturing of each array element and assembly of these elements. This paper, mainly focusing on the millimeter wave phased array antenna. Firstly, the theoretical model of amplitude quantization error is established. Secondly, the simulation of influence on the electrical performance of the antenna array caused by this kind of error shows that the technical indicators change with amplitude quantification. During design process, the influences of all kinds of errors must be considered.
- 23 April 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 836-840
Performance Analysis for Millimeter Wave Array Antenna Systems Based on Influence of Amplitude Quantization Error
For phased array antenna, the required pattern and its beam pointing are generated by different excitation amplitude and phase of each array element. Various kinds of errors will be introduced in manufacturing of each array element and assembly of these elements. This paper, mainly focusing on the millimeter wave phased array antenna. Firstly, the theoretical model of amplitude quantization error is established. Secondly, the simulation of influence on the electrical performance of the antenna array caused by this kind of error shows that the technical indicators change with amplitude quantification. During design process, the influences of all kinds of errors must be considered.
- 1 October 2020
- Published by Elsevier in Physical Communication
- Vol. 42
A user matching and power allocation scheme for downlink MIMO-NOMA communication system
- 1 October 2020
- Published by Elsevier in Physical Communication
- Vol. 42
A user matching and power allocation scheme for downlink MIMO-NOMA communication system
- 4 April 2020
- Published by Springer Nature
- p. 2208-2216
An Intelligent Garbage Bin Based on NB-IoT