Ramin Khosravi
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
Ramin Khosravi has a total of 9 co-authors. Here are the top co-authors by number of shared publications:
- Rashid Mirzavand
University of Alberta
2 shared publications - Ashwin K. Iyer
University of Alberta
1 shared publication - Hamidreza Bakhshi
Department of Electrical Engineering, Shahed University, Tehran, Iran
1 shared publication - Mohamad Dosaranian-Moghadam
Department of Electrical Engineering, Qazvin Branch, Islamic Azad University, Qazvin, Iran
1 shared publication - Rouzbeh Moini
Safe Engineering Services & Technologies Ltd., Laval, QC, Canada
1 shared publication - Seyed Hossein Hesamedin Sadeghi
Department of Electrical Engineering, Amirkabir University of Technology, Tehran, Iran
1 shared publication
Latest Publications
- 19 May 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Internet of Things Journal
- Vol. PP (15) , 1
Mobility-Aware Wireless Path Optimization with Physics-Guided Surrogates and Reinforcement Learning
Reliable, low-latency wireless links for mobile Internet of Things (IoT) in complex indoor environments are critical for digital twins and augmented reality, where motion and blockage rapidly change line-of-sight and reflected path quality. High-fidelity ray tracing accurately evaluates such paths but is too costly to run online, while reinforcement learning (RL) enables fast sequential path selection yet requires a physically meaningful, transfer-capable representation. In addition, practical mobility-aware operation should optimize not only received power but also propagation delay. To address these needs, this paper proposes a mobility-aware physics-guided surrogate framework for dual-objective wireless path optimization under mobility, together with controller benchmarking and hardware validation. We extend a static surrogate-guided path-optimization pipeline into a mobility-aware framework that combines deployment-aligned state design, controller benchmarking, and competence-based training under receiver motion. Crucially, we integrate a Zad-off–Chu (ZC) correlation-based Time-of-Arrival (ToA) tracker for deployment-time measurement and hardware validation of received power and propagation delay under mobility. Hardware validation confirms that the proposed framework preserves strong agreement between surrogate predictions and measured received power / propagation delay under static and mobile conditions. Experiments show strong agreement between surrogate predictions and measurements, with coefficient of determination for both predicted received power (R2 = 0.9907) and propagation delay (R2 = 0.9848). Combined with initial zero-shot transfer to a held-out environment, the results demonstrate a practical framework that substantially narrows the sim-to-real gap for mobility-aware digital-twin and IoT systems.
- 19 May 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Internet of Things Journal
- Vol. PP (15) , 1
Mobility-Aware Wireless Path Optimization with Physics-Guided Surrogates and Reinforcement Learning
Reliable, low-latency wireless links for mobile Internet of Things (IoT) in complex indoor environments are critical for digital twins and augmented reality, where motion and blockage rapidly change line-of-sight and reflected path quality. High-fidelity ray tracing accurately evaluates such paths but is too costly to run online, while reinforcement learning (RL) enables fast sequential path selection yet requires a physically meaningful, transfer-capable representation. In addition, practical mobility-aware operation should optimize not only received power but also propagation delay. To address these needs, this paper proposes a mobility-aware physics-guided surrogate framework for dual-objective wireless path optimization under mobility, together with controller benchmarking and hardware validation. We extend a static surrogate-guided path-optimization pipeline into a mobility-aware framework that combines deployment-aligned state design, controller benchmarking, and competence-based training under receiver motion. Crucially, we integrate a Zad-off–Chu (ZC) correlation-based Time-of-Arrival (ToA) tracker for deployment-time measurement and hardware validation of received power and propagation delay under mobility. Hardware validation confirms that the proposed framework preserves strong agreement between surrogate predictions and measured received power / propagation delay under static and mobile conditions. Experiments show strong agreement between surrogate predictions and measurements, with coefficient of determination for both predicted received power (R2 = 0.9907) and propagation delay (R2 = 0.9848). Combined with initial zero-shot transfer to a held-out environment, the results demonstrate a practical framework that substantially narrows the sim-to-real gap for mobility-aware digital-twin and IoT systems.
- Open Access
Wireless Capacitive Liquid-Level Detection Sensor Based on Zero-Power RFID-Sensing Architecture
In this paper, a new method for the wireless detection of liquid level is proposed by integrating a capacitive IDC-sensing element with a passive three-port RFID-sensing architecture. The sensing element transduces changes in the liquid level to corresponding fringe-capacitance variations, which alters the phase of the RFID backscattered signal. Variation in capacitance also changes the resonance magnitude of the sensing element, which is associated with a high phase transition. This change in the reactive phase is used as a sensing parameter by the RFID architecture for liquid-level detection. Practical measurements were conducted in a real-world scenario by placing the sensor at a distance of approximately 2 m (with a maximum range of about 7 m) from the RFID reader. The results show that the sensor node offers a high sensitivity of 2.15°/mm to the liquid-level variation. Additionally, the sensor can be used within or outside the container for the accurate measurement of conductive- or non-conductive-type liquids due to the use of polyethylene coating on the sensitive element. The proposed sensor increases the reliability of the current level sensors by eliminating the internal power source as well as complex signal-processing circuits, and it offers real-time response, linearity, high sensitivity, and excellent repeatability, which are suitable for widespread deployment of sensor node applications.
Open AccessWireless Capacitive Liquid-Level Detection Sensor Based on Zero-Power RFID-Sensing Architecture
In this paper, a new method for the wireless detection of liquid level is proposed by integrating a capacitive IDC-sensing element with a passive three-port RFID-sensing architecture. The sensing element transduces changes in the liquid level to corresponding fringe-capacitance variations, which alters the phase of the RFID backscattered signal. Variation in capacitance also changes the resonance magnitude of the sensing element, which is associated with a high phase transition. This change in the reactive phase is used as a sensing parameter by the RFID architecture for liquid-level detection. Practical measurements were conducted in a real-world scenario by placing the sensor at a distance of approximately 2 m (with a maximum range of about 7 m) from the RFID reader. The results show that the sensor node offers a high sensitivity of 2.15°/mm to the liquid-level variation. Additionally, the sensor can be used within or outside the container for the accurate measurement of conductive- or non-conductive-type liquids due to the use of polyethylene coating on the sensitive element. The proposed sensor increases the reliability of the current level sensors by eliminating the internal power source as well as complex signal-processing circuits, and it offers real-time response, linearity, high sensitivity, and excellent repeatability, which are suitable for widespread deployment of sensor node applications.
- 13 February 2022
- Published by Wiley in Transactions on Emerging Telecommunications Technologies
- Vol. 33 (7) , e4465
Handoff optimization for joint base station association and power control in uplink non‐orthogonal multiple access small cell networks
This article investigates the handoff optimization in an uplink non‐orthogonal multiple access‐based small cell network (SCN) with successive interference cancellation that takes both the base station (BS) association and power control into consideration. Due to user mobility and varying communication environment, it is important to associate a user with the proper BS and control its transmit power by considering the user mobility for the SCNs. To overcome overly frequent re‐association and handoffs between BSs, a novel joint optimization of handoff, BS association, and total transmit power is presented that maximizes the system‐wide and handoff utilities and minimizes power consumption simultaneously. We define the handoff utility function by considering the BS association in each time frame with the previous time frame. Then, we formulate two inter‐related optimization problems under the minimum average‐rate and maximum allowable transmit power of mobile users (MUs) constraints. The designed optimization problems are combined through the weighted sum method into a single‐stage optimization problem. The transformed problem is solved by adopting the coalition formation game and primal decomposition theories. We design the coalition value function by collectively considering the handoff number of each BS in every time frame, transmit power of all MUs, and the number of associated MUs to every BS. The simulation results show the efficiency of the proposed handoff algorithm in comparison to another BS association algorithm in terms of the number of handoffs.
- 13 February 2022
- Published by Wiley in Transactions on Emerging Telecommunications Technologies
- Vol. 33 (7) , e4465
Handoff optimization for joint base station association and power control in uplink non‐orthogonal multiple access small cell networks
This article investigates the handoff optimization in an uplink non‐orthogonal multiple access‐based small cell network (SCN) with successive interference cancellation that takes both the base station (BS) association and power control into consideration. Due to user mobility and varying communication environment, it is important to associate a user with the proper BS and control its transmit power by considering the user mobility for the SCNs. To overcome overly frequent re‐association and handoffs between BSs, a novel joint optimization of handoff, BS association, and total transmit power is presented that maximizes the system‐wide and handoff utilities and minimizes power consumption simultaneously. We define the handoff utility function by considering the BS association in each time frame with the previous time frame. Then, we formulate two inter‐related optimization problems under the minimum average‐rate and maximum allowable transmit power of mobile users (MUs) constraints. The designed optimization problems are combined through the weighted sum method into a single‐stage optimization problem. The transformed problem is solved by adopting the coalition formation game and primal decomposition theories. We design the coalition value function by collectively considering the handoff number of each BS in every time frame, transmit power of all MUs, and the number of associated MUs to every BS. The simulation results show the efficiency of the proposed handoff algorithm in comparison to another BS association algorithm in terms of the number of handoffs.
- Vol. 47 (2) , 27-37
Open AccessElectromagnetic Field Due to Lightning Strikes to Mountainous Ground