Abdullah TahirAbdullah Tahir

Mechatronics Engineer
Ph.D. Student | Aalborg University, Denmark
Email: engrabxis(at)gmail.com | atah(at)es.aau.dk

Abdullah Tahir is a Ph.D. researcher at AI RF research group in the Department of Electronic Systems at Aalborg University (AAU, Denmark) since 2022. He received his B.Sc. and M.Sc. degrees in Mechatronics Engineering from University of Engineering and Technology Lahore (UET, Pakistan) and Gwangju Institute of Science and Technology (GIST, South Korea) in 2013 and 2017, respectively. Additinally, he also has one year of field experience as a Site Engineer at Arden Engineering & Automation, where he worked on Building Management Systems (BMS) and Low Current Systems (LCS). From 2017 till 2021, he remained the part of Human Centered Robotics (HCR) lab in Pakistan. His research focuses on artificial intelligence, human robot interaction, and medical robotics.

Highlights

Educational Background

Skills

Publications

Journal Publications

  1. Physics-informed scattering transform network for modulation recognition in 5G industrial cognitive communications considering nonlinear impairments in active phased arrays
    Zeliang An, Yuqing Xu, Abdullah Tahir, Jun Wang, Baoze Ma, Gert Frølund Pedersen, Ming Shen
    IEEE Transactions on Industrial Informatics, 2024

    Modulation recognition (MR) plays a pivotal role due to its application in the spectrum sensing of 5G industrial cognitive communications and radio interference detection at the physical layer of the Internet of Things (IoT). Previous works have mainly focused on simulated fourth-generation (4G) multicarrier systems and ideal radio frequency (RF) scenarios. To bridge the gap between practice and theory, we propose a viable MR algorithm on all-physical testbeds, with nonlinear impairments of 28 GHz active phased arrays (APA). Specifically, our testbed is built on the Rohde&Schwarz (R&S) vector signal generation R&S-SMBV100B and spectrum analyzer R&S-FSW 67 GHz. To extract salient modulation patterns, we develop a physical-informed scattering transform (SCT) MR network (SCTMR-Net). With SCT modules, SCTMR-Net produces the translation-invariant and deformation-stable representations of 5-G signals by wavelet convolution, nonlinear modulus and low-pass filters. Extensive experiments on real-world measurement verify the viability of SCTMR-Net for high robustness to APA impairments.

    An, Z., Xu, Y., Tahir, A., Wang, J., Ma, B., Pedersen, G. F., & Shen, M. (2024). Physics-Informed Scattering Transform Network for Modulation Recognition in 5G Industrial Cognitive Communications Considering Nonlinear Impairments in Active Phased Arrays. IEEE Transactions on Industrial Informatics.

  2. Collaborative learning-based modulation recognition for 6 G multibeam satellite communication systems via blind and semi-blind channel equalization
    Zeliang An, Yuqing Xu, Abdullah Tahir, Jun Wang, Baoze Ma, Gert Frølund Pedersen, Ming Shen
    IEEE Transactions on Aerospace and Electronic Systems, 2024

    Blind modulation recognition (BMR) involves identifying the modulation scheme of intercepted signals, an essential component of terrestrial radio management. While deep learning (DL) has advanced BMR research in terrestrial contexts, 6 G multi-beam mobile satellite (MMS) systems present many challenges. One pivotal hurdle is the reliance on perfect channel-state information (CSI) for signal equalization in terrestrial MIMO BMR literature. In MMS systems, the vast satellite-ground distance can lead to outdated CSI, making traditional equalization techniques less effective. Addressing this, we introduce a viable BMR algorithm that leverages blind and semi- blind channel equalization to overcome challenges like inter-beam interference (IBI), limited CSI, and Shadowed-Rician (SR) fading. By transforming the MMS system into an equivalent MIMO satellite system, we enable enhanced MIMO channel equalization to counteract IBI and SR attenuations. Our proposed locality-globality convolutional-transformer deep neural network (LG-CTDNet) merges the strengths of CNNs in local feature extraction with Transformers in global feature discernment. A decision fusion strategy is then incorporated to utilize cooperative diversity across multiple eavesdroppers. Numerical tests validate the effectiveness of our blind and semi- blind equalization techniques, emphasizing their superiority over existing DL-based methods in satellite modulation signal identification even in conditions with limited or absent CSI.

    An, Z., Xu, Y., Tahir, A., Wang, J., Ma, B., Pedersen, G. F., & Shen, M. (2024). Collaborative Learning-Based Modulation Recognition for 6 G Multibeam Satellite Communication Systems Via Blind and Semi-Blind Channel Equalization. IEEE Transactions on Aerospace and Electronic Systems.

  3. Design optimization and redundant actuation selection for an efficient assistive robotic exoskeleton
    Asim Ghaffar, Abbas A. Dehghani-Sanij, Sheng Quan Xie, Abdullah Tahir, and Awais Hafeez
    Journal of the Chinese Institute of Engineers, 2023

    Selection of an actuation system for assistive robotic exoskeletons requires careful consideration of various design factors. It is generally the requirement of the system to produce lightweight and power-efficient systems. In some cases, the torque and power requirements could be relaxed by using redundant systems. This paper involves the study of one such case in which the actuation redundancy of the system will be exploited, and the design optimization will be explored for a rigid and an elastic system. A multi-factor optimization technique will be developed for a redundant elastic actuation system. An actuator design framework will be used to evaluate the different actuator choices to determine the best motor and transmission system combination in a redundant actuation system arrangement. This will be evaluated for a rigid, parallel, and series elastic actuation system. The optimal redundant actuation system significantly reduced the power requirements of the system. The case study was virtually implemented. It was established that variable parallel elastic actuators (V-PEA) performed better as compared to variable series elastic actuators (V-SEA).

    Ghaffar, A., Dehghani-Sanij, A. A., Xie, S. Q., Tahir, A., & Hafeez, A. (2023). Design optimization and redundant actuation selection for an efficient assistive robotic exoskeleton. Journal of the Chinese Institute of Engineers, 46(5), 490-503.

  4. A wearable multi-Modal digital upper limb assessment system for automatic musculoskeletal risk evaluation
    , , and
    Sensors 2023

    Continuous ergonomic risk assessment of the human body is critical to avoid various musculoskeletal disorders (MSDs) for people involved in physical jobs. This paper presents a digital upper limb assessment (DULA) system that automatically performs rapid upper limb assessment (RULA) in real-time for the timely intervention and prevention of MSDs. While existing approaches require human resources for computing the RULA score, which is highly subjective and untimely, the proposed DULA achieves automatic and objective assessment of musculoskeletal risks using a wireless sensor band embedded with multi-modal sensors. The system continuously tracks and records upper limb movements and muscle activation levels and automatically generates musculoskeletal risk levels. Moreover, it stores the data in a cloud database for in-depth analysis by a healthcare expert. Limb movements and muscle fatigue levels can also be visually seen using any tablet/computer in real-time. In the paper, algorithms of robust limb motion detection are developed, and an explanation of the system is provided along with the presentation of preliminary results, which validate the effectiveness of the new technology.

    Tahir, A., Bai, S., & Shen, M. (2023). A Wearable Multi-Modal Digital Upper Limb Assessment System for Automatic Musculoskeletal Risk Evaluation. Sensors, 23(10), 4863.

  5. Robust payload recognition based on sensor-over-muscle-independence deep learning for the control of exoskeletons
    , , , and
    IEEE Transactions on Circuits and Systems II: Express Briefs 2023

    Force myography (FMG) can detect changes in the muscle volume which can be interpreted to recognize human intention. FMG data, however, is highly dependent on the placement of sensors over muscles, and interpretation becomes challenging if the sensors get displaced. This brief presents a robust sensor over muscle independence (SOMI) preprocessing algorithm combined with lightweight deep neural network (DNN) which shows high classification accuracy of the FMG data. SOMI organizes the irregular sensory data into regular patterns. Proposed algorithm makes the DNN insensitive to not only position and rotation shift but also to the flip of the sensors arrangement. A custom designed FMG band is used for payload recognition to experimentally validate the proposed method with five payload statuses and eight subjects. A five-fold cross validation comparative study demonstrated that the proposed method is 19.8% and 7.1% more accurate than support vector machine (SVM) and DNN without SOMI, respectively, and showed superior performance against k-nearest neighbors (KNN) and decision tree (DT). SOMI empowered a lightweight DNN to maintain the accuracy over 98% for different arbitrary wearing schemes of the FMG band over both left and right upper arms.

    Tahir, A., Bai, S., & Shen, M. (2023). A Wearable Multi-Modal Digital Upper Limb Assessment System for Automatic Musculoskeletal Risk Evaluation. Sensors, 23(10), 4863.

  6. A high-quality data acquisition method for machine-learning-based design and analysis of electromagnetic structures
    Zhao Zhou, Zhaohui Wei, Abdullah Tahir, Jian Ren, Yingzeng Yin, Gert Frolund Pedersen, Ming Shen
    IEEE Transactions on Microwave Theory and Techniques 2023

    Electromagnetic (EM) structures play a significant role in wireless communication, radar detection, medical imaging, and so on. Machine learning (ML) has been increasingly applied to facilitate the design and analysis of EM structures. Data acquisition is a major bottleneck. Conventional methods blindly sweep geometric parameters on a uniform grid and acquire corresponding responses via simulation. Acquired data have unstable quality due to inconsistent informativeness of responses, leading to a low ratio of model performance to data amount. This article proposes a high-quality data acquisition method to increase the ratio of model performance to data amount. It anticipates and generates high-quality data by analyzing the distribution of existing data iteratively. Comparative analysis of four implementations proves that the proposed method reduces the required data amount by around 40% for the same model performance and hence saves around 40% simulation and computing resources. The proposed method benefits ML applications of metasurfaces, antennas, and many other microwave structures.

    Zhou, Z., Wei, Z., Tahir, A., Ren, J., Yin, Y., Pedersen, G. F., & Shen, M. (2023). A High-Quality Data Acquisition Method for Machine-Learning-Based Design and Analysis of Electromagnetic Structures. IEEE Transactions on Microwave Theory and Techniques.

  7. An intelligent health monitoring and diagnosis system based on the internet of things and fuzzy logic for cardiac arrhythmia COVID-19 patients
    Muhammad Zia Rahman, Muhammad Azeem Akbar, Víctor Leiva, Abdullah Tahir, Muhammad Tanveer Riaz, Carlos Martin-Barreiro
    Computers in Biology and Medicine 2023

    Objective: To design and implement an intelligent health monitoring and diagnosis system for critical cardiac arrhythmia COVID-19 patients.
    Methodology: We use artificial intelligence tools divided into two parts: (i) IoT-based health monitoring; and (ii) fuzzy logic-based medical diagnosis. The intelligent diagnosis of heart conditions and IoT-based health surveillance by doctors is offered to critical COVID-19 patients or isolated in remote locations. Sensors, cloud storage, as well as a global system for mobile texts and emails for communication with doctors in case of emergency are employed in our proposal.
    Results: Our implemented system favors remote areas and isolated critical patients. This system utilizes an intelligent algorithm that employs an ECG signal pre-processed by moving through six digital filters. Then, based on the processed results, features are computed and assessed. The intelligent fuzzy system can make an autonomous diagnosis and has enough information to avoid human intervention. The algorithm is trained using ECG data from the MIT-BIH database and achieves high accuracy. In real-time validation, the fuzzy algorithm obtained almost 100% accuracy for all experiments.
    Conclusion: Our intelligent system can be helpful in many situations, but it is particularly beneficial for isolated COVID-19 patients who have critical heart arrhythmia and must receive intensive care.

    Rahman, M. Z., Akbar, M. A., Leiva, V., Tahir, A., Riaz, M. T., & Martin-Barreiro, C. (2023). An intelligent health monitoring and diagnosis system based on the internet of things and fuzzy logic for cardiac arrhythmia COVID-19 patients. Computers in Biology and Medicine, 154, 106583.

  8. Efficient approach for extracting high-level B-spline features from LIDAR data for light-weight mapping
    Muhammad Usman, Ahmad Ali, Abdullah Tahir, Muhammad Zia Ur Rahman, Abdul Manan Khan
    Sensors 2022

    Light-weight and accurate mapping is made possible by high-level feature extraction from sensor readings. In this paper, the high-level B-spline features from a 2D LIDAR are extracted with a faster method as a solution to the mapping problem, making it possible for the robot to interact with its environment while navigating. The computation time of feature extraction is very crucial when mobile robots perform real-time tasks. In addition to the existing assessment measures of B-spline feature extraction methods, the paper also includes a new benchmark time metric for evaluating how well the extracted features perform. For point-to-point association, the most reliable vertex control points of the spline features generated from the hints of low-level point feature FALKO were chosen. The standard three indoor and one outdoor data sets were used for the experiment. The experimental results based on benchmark performance metrics, specifically computation time, show that the presented approach achieves better results than the state-of-the-art methods for extracting B-spline features. The classification of the methods implemented in the B-spline features detection and the algorithms are also presented in the paper.

    Usman, M., Ali, A., Tahir, A., Rahman, M. Z. U., & Khan, A. M. (2022). Efficient approach for extracting high-level B-spline features from LIDAR data for light-weight mapping. Sensors, 22(23), 9168.

  9. Cardiac X-ray image-based haptic guidance for robot-assisted coronary intervention: A feasibility study
    Abdullah Tahir, Hashim Iqbal, Muhammad Usman, Asim Ghaffar, Awais Hafeez
    International Journal of Computer Assisted Radiology and Surgery 2022

    Purpose: Effective and efficient haptic guidance is desirable for tele-operated robotic surgery because it has a potential to enhance surgeon’s skills, especially in coronary interventions where surgeon loses both an eye–hand coordination and a direct sight to the organ. This paper proposes a novel haptic guidance procedure—both kinesthetic and cutaneous, which solely depends upon X-ray images, for tele-robotic system that assists an efficient navigation of the guidewire towards the target location during a coronary intervention.
    Methods: Proposed methodology requires cardiologists to draw virtual fixtures (VFs) on angiograms as a preoperative procedure. During an operation, these VFs direct the guidewire to the desired coronary vessel. For this, the position and orientation of guidewire tip are calculated with respect to VFs’ anatomy, using image processing on the real-time 2D fluoroscopic images. The haptic feedbacks are then rendered on to the master device depending on the interaction with attractive and repulsive, guidance and forbidden region VFs.
    Results: A feasibility study in the laboratory environment is performed by using a webcam as an image acquisition device and a phantom-based coronary vessel model. The subsequent statistical analysis shows that, on an average, a decrease of approx. 37% in task completion time is observed with haptic feedback. Moreover, haptic guidance is found effective for most difficult branch, whereas there is a minimal significance of such haptics for the easiest branch.
    Conclusions: The proposed haptic guidance procedure may assist cardiologists for an efficient and effective guidewire navigation during a surgical procedure. The cutaneous haptics (vibration feedback) is found more helpful in coronary interventions compared with kinesthetic haptics (force feedback).

    Tahir, A., Iqbal, H., Usman, M., Ghaffar, A., & Hafeez, A. (2022). Cardiac X-ray image-based haptic guidance for robot-assisted coronary intervention: A feasibility study. International Journal of Computer Assisted Radiology and Surgery, 17(3), 531-539.

Conference Publications

  1. Haptic guidance based only on cardiac X-ray images for robot assisted coronary intervention
    Abdullah Tahir, Sehun Park, Dongkue Kim, Ryu Jeha
    International Journal of Computer Assisted Radiology and Surgery Proceedings of 31st International Congress and Exhibition, Spain 2017

    Tele-operated robot assisted surgery is of huge interest as it enhances the surgeon’s maneuverability, particularly during Minimally Invasive Surgery (MIS), such as coronary intervention, where a surgeon loses direct sight of organs and eye-hand coordination. An efficient and effective haptic guidance method in a tele-robotic surgical system can be utilized to lessen the surgeon’s effort, and hence decrease task completion time and/or to reduce errors in navigating the guidewire towards a target location inside a patient’s heart. Previous guidewire navigation procedures usually require time consuming preoperative procedures and/or require an expensive and hard-to-realize sensorized guidewire/catheter. In order to provide efficient and effective haptic guidance in robot-assisted coronary intervention, this paper proposes a haptic guidance system that uses only two X-ray images, which are currently available from C-arm X-ray machines.

    Tahir, A., Park, S., Kim, D., Jeha, R., International Journal of Computer Assisted Radiology and Surgery Proceedings of 31st International Congress and Exhibition, Spain 2017.

  2. Real time 3D representation and tracking of guidewire for image guided cardiovascular interventions
    Rashid Mehmood, Naveed Iqbal, Abdullah Tahir, M Mohsin Riaz, Rab Nawaz
    GeNeDis 2016: Geriatrics

    Visual tracking and 3D representation of guidewire in fluoroscopic image sequence for beating heart image guided interventions is very challenging task. The degraded image quality due to low dose fluoroscopy further complicates the problem. In this paper a robust guidewire tracking is proposed for mean shift algorithm using integrated colour, texture and depth features. The target colour, texture and depth features are encoded into gray level intensity histogram, filtered local binary pattern histogram and filtered local depth pattern histograms respectively. For depth features a 3D image acquisition system for C-Arm, X-Ray imaging system is simulated for real time three dimensional shape recovery of guidewire and associated vessels for vertical beating heart motion using shape from focus technique. The proposed technique provides 3D visualization of guide wire and vessels to the physician as well as real time robust guidewire tip tracking. Experimental results of guidewire tip tracking and 3D shape recovery on image sequence acquired through beating heart simulated phantom show the significance of the proposed technique.

    Mehmood, R., Iqbal, N., Tahir, A., Riaz, M. M., & Nawaz, R. (2017). Real Time 3D Representation and Tracking of Guidewire for Image Guided Cardiovascular Interventions. In GeNeDis 2016: Geriatrics (pp. 165-176). Springer International Publishing.

  3. Single object tracking system using fast compressive tracking
    Abdullah Tahir, Shoaib Azam, Sujani Sagabala, Moongu Jeon, Ryu Jeha
    IEEE International Conference on Consumer Electronics-Asia (ICCE-Asia), 2016

    In this work we focused on the application aspect of object tracking for pan-tilt-zoom (PTZ) camera using ordinary webcam mounted on custom made motor-assembly and found that our system is not only robust to illumination conditions but also cost-effective in comparison with PTZ cameras. For object tracking we utilized Fast Compressive Tracking (FCT) algorithm because of its attractive features e.g. online learning, fast computation and robust performance. A PC program interfaced with embedded system through serial RS232 commands motors, hence camera, to real time track desired object in world such that object being tracked remains in the center of image.

    Tahir, A., Azam, S., Sagabala, S., Jeon, M., & Jeha, R. (2016, October). Single object tracking system using fast compressive tracking. In 2016 IEEE International Conference on Consumer Electronics-Asia (ICCE-Asia) (pp. 1-3). IEEE.

Honors and Awards

Hobby Projects

Besides working on AI and mechatronics systems, various other projects can be seen in my github repositories:

Hobbies

Badminton, cooking, computer gaming, and music.

Updated: October 2024