Zubair Khalid
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
Zubair Khalid has a total of 44 co-authors. Here are the top co-authors by number of shared publications:
- Rodney Andrew Kennedy
Research School of Engineering, College of Engineering and Computer Science, The Australian National University, Canberra, Australia
44 shared publications - Salman Durrani
School of Engineering, The Australian National University, Canberra, Australia
18 shared publications - Parastoo Sadeghi
School of Engineering and Technology, University of New South Wales, Canberra, Australia
15 shared publications - Adeem Aslam
Electrical Engineering, Lahore University of Management Sciences, Lahore, Punjab, Pakistan, 54792
9 shared publications - Jason D. McEwen
Mullard Space Science Laboratory (MSSL), University College London (UCL), London, U.K.
9 shared publications Alice P. Bates
Research School of Engineering, The Australian National University, Canberra, ACT, Australia
6 shared publications
Latest Publications
- 10 June 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Signal Processing
- Vol. 74 (1053587X) , 2595-2605
Slepian Basis for Efficient Representation of Order-Limited Signals on the Sphere
The Slepian concentration problem is a classical problem in signal processing concerned with finding basis functions that are simultaneously concentrated in spatial and spectral domains. Unlike spherical harmonic basis functions, Slepian functions enable localized analysis and have been widely used for representing data restricted to regions on the sphere. However, existing Slepian representations retain all orders of the basis, which leads to redundancy, particularly when the region of interest exhibits plausible azimuthal symmetry. In this work, we propose an order-limited Slepian basis that systematically excludes redundant higher-order components while preserving accuracy. We provide an analytical formulation of this order-limited basis, establish its properties, and derive expressions for two special cases: polar cap regions and limited colatitude–longitude regions. We further establish key mathematical properties of the proposed order-limited basis, including orthogonality, completeness, and concentration characteristics within the defined subspace. Using Earth’s topographic map and climate data as applications, we demonstrate that the proposed formulation achieves efficient signal representation with substantially fewer basis functions compared to conventional Slepian approaches. This highlights order-limited Slepian functions as an efficient alternative for localized signal analysis on the sphere.
- 10 June 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Signal Processing
- Vol. 74 (1053587X) , 2595-2605
Slepian Basis for Efficient Representation of Order-Limited Signals on the Sphere
The Slepian concentration problem is a classical problem in signal processing concerned with finding basis functions that are simultaneously concentrated in spatial and spectral domains. Unlike spherical harmonic basis functions, Slepian functions enable localized analysis and have been widely used for representing data restricted to regions on the sphere. However, existing Slepian representations retain all orders of the basis, which leads to redundancy, particularly when the region of interest exhibits plausible azimuthal symmetry. In this work, we propose an order-limited Slepian basis that systematically excludes redundant higher-order components while preserving accuracy. We provide an analytical formulation of this order-limited basis, establish its properties, and derive expressions for two special cases: polar cap regions and limited colatitude–longitude regions. We further establish key mathematical properties of the proposed order-limited basis, including orthogonality, completeness, and concentration characteristics within the defined subspace. Using Earth’s topographic map and climate data as applications, we demonstrate that the proposed formulation achieves efficient signal representation with substantially fewer basis functions compared to conventional Slepian approaches. This highlights order-limited Slepian functions as an efficient alternative for localized signal analysis on the sphere.
- Open Access
Robust Spatiotemporal Forecasting Using Adaptive Deep-Unfolded Variational Mode Decomposition
Accurate spatiotemporal forecasting is critical for numerous complex systems but remains challenging due to complex volatility patterns and spectral entanglement in conventional graph neural networks (GNNs). While decomposition-integrated approaches like variational mode graph convolutional network (VMGCN) improve accuracy through signal decomposition, they suffer from computational inefficiency and manual hyperparameter tuning. To address these limitations, we propose the mode adaptive graph network (MAGN) that transforms iterative variational mode decomposition (VMD) into a trainable neural module. Our key innovations include (1) an unfolded VMD (UVMD) module that replaces iterative optimization with a fixed-depth network to reduce the decomposition time (by 250x for the LargeST benchmark), and (2) mode-specific learnable bandwidth constraints (αk ) adapt spatial heterogeneity and eliminate manual tuning while preventing spectral overlap. Evaluated on the LargeST benchmark (6,902 sensors, 241M observations), MAGN achieves an 85-95% reduction in the prediction error over VMGCN and outperforms state-of-the-art baselines.
Open AccessRobust Spatiotemporal Forecasting Using Adaptive Deep-Unfolded Variational Mode Decomposition
Accurate spatiotemporal forecasting is critical for numerous complex systems but remains challenging due to complex volatility patterns and spectral entanglement in conventional graph neural networks (GNNs). While decomposition-integrated approaches like variational mode graph convolutional network (VMGCN) improve accuracy through signal decomposition, they suffer from computational inefficiency and manual hyperparameter tuning. To address these limitations, we propose the mode adaptive graph network (MAGN) that transforms iterative variational mode decomposition (VMD) into a trainable neural module. Our key innovations include (1) an unfolded VMD (UVMD) module that replaces iterative optimization with a fixed-depth network to reduce the decomposition time (by 250x for the LargeST benchmark), and (2) mode-specific learnable bandwidth constraints (αk ) adapt spatial heterogeneity and eliminate manual tuning while preventing spectral overlap. Evaluated on the LargeST benchmark (6,902 sensors, 241M observations), MAGN achieves an 85-95% reduction in the prediction error over VMGCN and outperforms state-of-the-art baselines.
- Open Access
Robust and Noise-resilient Long-Term Prediction of Spatiotemporal Data Using Variational Mode Graph Neural Networks with 3D Attention
This paper focuses on improving the robustness of spatiotemporal long-term prediction using a variational mode graph convolutional network (VMGCN) by introducing 3D channel attention. The deep learning network for this task relies on historical data inputs, yet real-time data can be corrupted by sensor noise, altering its distribution. We model this noise as independent and identically distributed (i.i.d.) Gaussian noise and incorporate it into the LargeST traffic volume dataset, resulting in data with both inherent and additive noise components. Our approach involves decomposing the corrupted signal into modes using variational mode decomposition, followed by feeding the data into a learning pipeline for prediction. We integrate a 3D attention mechanism encompassing spatial, temporal, and channel attention. The spatial and temporal attention modules learn their respective correlations, while the channel attention mechanism is used to suppress noise and highlight the significant modes in the spatiotemporal signals. Additionally, a learnable soft thresholding method is implemented to exclude unimportant modes from the feature vector, and a feature reduction method based on the signal-to-noise ratio (SNR) is applied. We compare the performance of our approach against baseline models, demonstrating that our method achieves superior long-term prediction accuracy, robustness to noise, and improved performance with mode truncation compared to the baseline models. The code of the paper is available at https://github.com/OsamaAhmad369/VMGCN.
Open AccessRobust and Noise-resilient Long-Term Prediction of Spatiotemporal Data Using Variational Mode Graph Neural Networks with 3D Attention
This paper focuses on improving the robustness of spatiotemporal long-term prediction using a variational mode graph convolutional network (VMGCN) by introducing 3D channel attention. The deep learning network for this task relies on historical data inputs, yet real-time data can be corrupted by sensor noise, altering its distribution. We model this noise as independent and identically distributed (i.i.d.) Gaussian noise and incorporate it into the LargeST traffic volume dataset, resulting in data with both inherent and additive noise components. Our approach involves decomposing the corrupted signal into modes using variational mode decomposition, followed by feeding the data into a learning pipeline for prediction. We integrate a 3D attention mechanism encompassing spatial, temporal, and channel attention. The spatial and temporal attention modules learn their respective correlations, while the channel attention mechanism is used to suppress noise and highlight the significant modes in the spatiotemporal signals. Additionally, a learnable soft thresholding method is implemented to exclude unimportant modes from the feature vector, and a feature reduction method based on the signal-to-noise ratio (SNR) is applied. We compare the performance of our approach against baseline models, demonstrating that our method achieves superior long-term prediction accuracy, robustness to noise, and improved performance with mode truncation compared to the baseline models. The code of the paper is available at https://github.com/OsamaAhmad369/VMGCN.
- Open Access
- 19 February 2025
- Published by E-palli in American Journal of Education and Technology
- Vol. 4 (1) , 118-130
Investigation into Rural-Urban Gap in Readiness of Universities for Online Teaching: A Case of Pakistan
Higher education institutions worldwide faced significant disruption and chaos as COVID-19 swept away the normal in 2020, catapulting them into online mode of instruction with hardly any transitional time. It can be argued that the challenge was particularly daunting for universities in developing country contexts of South Asia, like Pakistan, with limited resources and a large student body coming from socioeconomically disadvantaged backgrounds. While research evidence piled up around the world about different challenges that specific universities faced within or across countries, there has been little systematic work done to explore the comparative challenges faced by universities in urban (large metropolitan) and rural (smaller/semi-urban /cities) in a single country context. This paper presents the findings of a large-scale survey sent to all the universities in Pakistan during COVID to understand the intra-country variation that one needs to keep in mind as we prepare for online instruction to meet similar challenges in the future. A quantitative survey was sent to 196 universities in Pakistan. The survey had questions under the following categories, namely: university readiness, faculty readiness, student readiness, technology readiness and laboratory readiness. The results presented here are from 170 universities. The results reveal that urban universities were much more successful in their transition to online instruction than rural universities. The disparities were across: a) the extent of online policy adoption; b) infrastructure and support; c) operationalization of online teaching processes; and d) student and faculty readiness for online instruction. The study in the developing country context of the global South contributes to the relevant literature by providing insights into the unique challenges faced by universities because of their location.
- 19 February 2025
- Published by E-palli in American Journal of Education and Technology
- Vol. 4 (1) , 118-130
Open AccessInvestigation into Rural-Urban Gap in Readiness of Universities for Online Teaching: A Case of Pakistan
Higher education institutions worldwide faced significant disruption and chaos as COVID-19 swept away the normal in 2020, catapulting them into online mode of instruction with hardly any transitional time. It can be argued that the challenge was particularly daunting for universities in developing country contexts of South Asia, like Pakistan, with limited resources and a large student body coming from socioeconomically disadvantaged backgrounds. While research evidence piled up around the world about different challenges that specific universities faced within or across countries, there has been little systematic work done to explore the comparative challenges faced by universities in urban (large metropolitan) and rural (smaller/semi-urban /cities) in a single country context. This paper presents the findings of a large-scale survey sent to all the universities in Pakistan during COVID to understand the intra-country variation that one needs to keep in mind as we prepare for online instruction to meet similar challenges in the future. A quantitative survey was sent to 196 universities in Pakistan. The survey had questions under the following categories, namely: university readiness, faculty readiness, student readiness, technology readiness and laboratory readiness. The results presented here are from 170 universities. The results reveal that urban universities were much more successful in their transition to online instruction than rural universities. The disparities were across: a) the extent of online policy adoption; b) infrastructure and support; c) operationalization of online teaching processes; and d) student and faculty readiness for online instruction. The study in the developing country context of the global South contributes to the relevant literature by providing insights into the unique challenges faced by universities because of their location.
- 21 January 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Signal Processing Letters
- Vol. 32 (10709908) , 746-750
Efficient Sampling and Accurate Reconstruction of Order-Limited Spherical Signals
In this work, we propose an efficient sampling design on the sphere for sampling order-limited spherical signals. The current sampling schemes on the sphere do not take into account the plausible rotational symmetry that is naturally exhibited by signals in many applications (e.g., climate data and antenna radiation pattern). These nearly rotationally symmetric signals have negligible dependence on the spherical harmonics of higher orders, and therefore can be modeled as order-limited signals. For sampling of spherical signals band-limited at L and order-limited at M, we present a sampling design for accurate computation of spherical harmonic transform and accurate signal reconstruction. The proposed sampling scheme achieves optimal dimensionality, requiring M2 + (L − M)(2M + 1) samples, which match the degrees of freedom in harmonic space.
- 21 January 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Signal Processing Letters
- Vol. 32 (10709908) , 746-750
Efficient Sampling and Accurate Reconstruction of Order-Limited Spherical Signals
In this work, we propose an efficient sampling design on the sphere for sampling order-limited spherical signals. The current sampling schemes on the sphere do not take into account the plausible rotational symmetry that is naturally exhibited by signals in many applications (e.g., climate data and antenna radiation pattern). These nearly rotationally symmetric signals have negligible dependence on the spherical harmonics of higher orders, and therefore can be modeled as order-limited signals. For sampling of spherical signals band-limited at L and order-limited at M, we present a sampling design for accurate computation of spherical harmonic transform and accurate signal reconstruction. The proposed sampling scheme achieves optimal dimensionality, requiring M2 + (L − M)(2M + 1) samples, which match the degrees of freedom in harmonic space.
- Open Access
Spatiotemporal Air Quality Mapping in Urban Areas Using Sparse Sensor Data, Satellite Imagery, Meteorological Factors, and Spatial Features
Monitoring air pollution is crucial for protecting human health from exposure to harmful substances. Traditional methods of air quality monitoring, such as ground-based sensors and satellite-based remote sensing, face limitations due to high deployment costs, sparse sensor coverage, and environmental interferences. To address these challenges, this paper proposes a framework for high-resolution spatiotemporal Air Quality Index (AQI) mapping using sparse sensor data, satellite imagery, and various spatiotemporal factors. By leveraging Graph Neural Networks (GNNs), we estimate AQI values at unmonitored locations based on both spatial and temporal dependencies. The framework incorporates a wide range of environmental features, including meteorological data, road networks, points of interest (PoIs), population density, and urban green spaces, which enhance prediction accuracy. We illustrate the use of our approach through a case study in Lahore, Pakistan, where multi-resolution data is used to generate the air quality index map at a fine spatiotemporal scale.
Open AccessSpatiotemporal Air Quality Mapping in Urban Areas Using Sparse Sensor Data, Satellite Imagery, Meteorological Factors, and Spatial Features
Monitoring air pollution is crucial for protecting human health from exposure to harmful substances. Traditional methods of air quality monitoring, such as ground-based sensors and satellite-based remote sensing, face limitations due to high deployment costs, sparse sensor coverage, and environmental interferences. To address these challenges, this paper proposes a framework for high-resolution spatiotemporal Air Quality Index (AQI) mapping using sparse sensor data, satellite imagery, and various spatiotemporal factors. By leveraging Graph Neural Networks (GNNs), we estimate AQI values at unmonitored locations based on both spatial and temporal dependencies. The framework incorporates a wide range of environmental features, including meteorological data, road networks, points of interest (PoIs), population density, and urban green spaces, which enhance prediction accuracy. We illustrate the use of our approach through a case study in Lahore, Pakistan, where multi-resolution data is used to generate the air quality index map at a fine spatiotemporal scale.
- 1 January 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Signal Processing Letters
- Vol. 31, 1825-1829
Fast and Accurate Spherical Harmonic Transform for Spatio-Temporal Regular Grid Data
- 1 January 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Signal Processing Letters
- Vol. 31, 1825-1829
Fast and Accurate Spherical Harmonic Transform for Spatio-Temporal Regular Grid Data
- 8 June 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Signal Processing
- Vol. 71 (1053587X) , 2532-2547
Overcomplete Multiscale Dictionary of Slepian Functions for HEALPix on the Sphere
We present a framework for exact analytical computation of bandlimited Slepian functions for Hierarchical Equal Area iso-Latitude Pixelization (HEALPix) scheme on the sphere. Slepian functions are bandlimited eigenfunctions obtained by solving the spatial-spectral concentration problem on the sphere. Utilizing rotational symmetries between the HEALPix pixels, we employ Wigner- $D$ functions to efficiently compute the bandlimited Slepian functions at different resolutions of the HEALPix partitioning scheme. We present convergence criteria for the infinite series expansions involved in the analytical expressions, analyze the complexity of computing bandlimited Slepian functions and construct an overcomplete multiscale dictionary of bandlimited Slepian functions on the sphere. We show that the dictionary spans the space of bandlimited functions, which are well-optimally (energy) concentrated within a region on the sphere, and that its elements exhibit small mutual coherence. As a demonstration of the utility of the dictionary, we present localized representation of a bandlimited Earth topography map over the South American continent, and conduct performance comparison with the representations obtained from the Slepian functions for a rotationally symmetric region, centered on South America, and for the South American coastal region, extracted from the coastlines data available in MATLAB .
- 8 June 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Signal Processing
- Vol. 71 (1053587X) , 2532-2547
Overcomplete Multiscale Dictionary of Slepian Functions for HEALPix on the Sphere
We present a framework for exact analytical computation of bandlimited Slepian functions for Hierarchical Equal Area iso-Latitude Pixelization (HEALPix) scheme on the sphere. Slepian functions are bandlimited eigenfunctions obtained by solving the spatial-spectral concentration problem on the sphere. Utilizing rotational symmetries between the HEALPix pixels, we employ Wigner- $D$ functions to efficiently compute the bandlimited Slepian functions at different resolutions of the HEALPix partitioning scheme. We present convergence criteria for the infinite series expansions involved in the analytical expressions, analyze the complexity of computing bandlimited Slepian functions and construct an overcomplete multiscale dictionary of bandlimited Slepian functions on the sphere. We show that the dictionary spans the space of bandlimited functions, which are well-optimally (energy) concentrated within a region on the sphere, and that its elements exhibit small mutual coherence. As a demonstration of the utility of the dictionary, we present localized representation of a bandlimited Earth topography map over the South American continent, and conduct performance comparison with the representations obtained from the Slepian functions for a rotationally symmetric region, centered on South America, and for the South American coastal region, extracted from the coastlines data available in MATLAB .
- 21 April 2023
- Published by Springer Nature in Wireless Networks
- Vol. 29 (6) , 2609-2617
On the performance of hybrid beamforming for closely-spaced and randomly located users
- 21 April 2023
- Published by Springer Nature in Wireless Networks
- Vol. 29 (6) , 2609-2617
On the performance of hybrid beamforming for closely-spaced and randomly located users
- 30 September 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Signal Processing Letters
- Vol. 29 (10709908) , 2048-2052
A Deep Unfolded Prior-Aided RPCA Network for Cloud Removal
Clouds, together with their shadows, usually occlude ground-cover features in optical remote sensing images. This hinders the utilization of these images for a range of applications such as earth observation, land-cover classification and urban planning. In this work, we propose a deep unfolded and prior-aided robust principal component analysis (DUPA-RPCA) network for removing clouds and recovering ground-cover information in multi-temporal satellite images. We model these cloud-contaminated images as a sum of low rank and sparse elements and then unfold an iterative RPCA algorithm that has been designed for reweighted $\ell _{1}$ minimization. As a result, the activation function in DUPA-RPCA adapts for every input at each layer of the network. Our experimental results on both Landsat and Sentinel images indicate that our method gives better accuracy and efficiency when compared with existing state of the art methods.
- 30 September 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Signal Processing Letters
- Vol. 29 (10709908) , 2048-2052
A Deep Unfolded Prior-Aided RPCA Network for Cloud Removal
Clouds, together with their shadows, usually occlude ground-cover features in optical remote sensing images. This hinders the utilization of these images for a range of applications such as earth observation, land-cover classification and urban planning. In this work, we propose a deep unfolded and prior-aided robust principal component analysis (DUPA-RPCA) network for removing clouds and recovering ground-cover information in multi-temporal satellite images. We model these cloud-contaminated images as a sum of low rank and sparse elements and then unfold an iterative RPCA algorithm that has been designed for reweighted $\ell _{1}$ minimization. As a result, the activation function in DUPA-RPCA adapts for every input at each layer of the network. Our experimental results on both Landsat and Sentinel images indicate that our method gives better accuracy and efficiency when compared with existing state of the art methods.