Sajjad Hussain
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
Sajjad Hussain has a total of 17 co-authors. Here are the top co-authors by number of shared publications:
Conor Brennan
School of Electronic Engineering, Dublin City University, Glasnevin, Dublin 9, Republic of Ireland
5 shared publications- Conor Brennan
Dublin City University
5 shared publications - Adnan Ahmad Cheema
School of Engineering, Ulster University, Belfast, U.K.
1 shared publication - Ali Hassan
School of Electrical Engineering and Computer Sciences (SEECS), National University of Sciences and Technology (NUST), Islamabad, Pakistan
1 shared publication - Asif Mahmood
New York Institute of Technology, Old Westbury, NY, USA
1 shared publication - Berk Canberk
School of Computing, Engineering and Built Environment, Edinburgh Napier University, Edinburgh, U.K.
1 shared publication
Latest Publications
- 10 March 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Wireless Communications
- Vol. 25 (15361276) , 13260-13273
A Multi-Scale Feature Extraction and Fusion U-Net for Pathloss Prediction in UAV-Assisted mmWave Radio Networks
Accurate pathloss prediction is essential for the design and optimization of unmanned aerial vehicles (UAV) assisted millimeter-wave (mmWave) networks. While deep learning approaches have shown strong potential, their generalization across diverse environments, robustness to noisy inputs, and sensitivity to UAV altitude remain underexplored. To address these challenges, we propose a U-Net-based deep learning architecture that combines multi-scale feature extraction, convolution-based feature fusion, and an atrous spatial pyramid pooling (ASPP) bottleneck for efficient context aggregation. The model predicts pathloss maps from log-distance, line-of-sight (LOS) mask, and building mask inputs. In addition, we develop a fully vectorized LOS mask computation algorithm that significantly accelerates pre-processing and enables large-scale dataset generation. Extensive evaluations on both in-house ray-tracing data and the RadioMapSeer benchmark demonstrate that the proposed model outperforms several state-of-the-art baselines in accuracy and efficiency. All source code is publicly released to support reproducibility and future research.
- 10 March 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Wireless Communications
- Vol. 25 (15361276) , 13260-13273
A Multi-Scale Feature Extraction and Fusion U-Net for Pathloss Prediction in UAV-Assisted mmWave Radio Networks
Accurate pathloss prediction is essential for the design and optimization of unmanned aerial vehicles (UAV) assisted millimeter-wave (mmWave) networks. While deep learning approaches have shown strong potential, their generalization across diverse environments, robustness to noisy inputs, and sensitivity to UAV altitude remain underexplored. To address these challenges, we propose a U-Net-based deep learning architecture that combines multi-scale feature extraction, convolution-based feature fusion, and an atrous spatial pyramid pooling (ASPP) bottleneck for efficient context aggregation. The model predicts pathloss maps from log-distance, line-of-sight (LOS) mask, and building mask inputs. In addition, we develop a fully vectorized LOS mask computation algorithm that significantly accelerates pre-processing and enables large-scale dataset generation. Extensive evaluations on both in-house ray-tracing data and the RadioMapSeer benchmark demonstrate that the proposed model outperforms several state-of-the-art baselines in accuracy and efficiency. All source code is publicly released to support reproducibility and future research.
- Open Access
RadioSim Agent: Combining Large Language Models and Deterministic EM Simulators for Interactive Radio Map Analysis
Deterministic electromagnetic (EM) simulators provide accurate radio propagation modeling but often require expert configuration and lack interactive flexibility. We present RadioSim Agent, an agentic framework that integrates large language models (LLMs) with physics-based EM solvers and vision-enabled reasoning to enable interactive and explainable radio map generation. The framework encapsulates ray-tracing models as callable simulation tools, orchestrated by an LLM capable of interpreting natural language objectives, managing simulation workflows, and visually analyzing resulting radio maps. Demonstrations in urban UAV communication scenarios show that the agent autonomously selects appropriate propagation mechanisms, executes deterministic simulations, and provides semantic and visual summaries of pathloss behavior. The results indicate that RadioSim Agent provides multimodal interpretability and intuitive user interaction, paving the way for intelligent EM simulation assistants in next-generation wireless system design.
Open AccessRadioSim Agent: Combining Large Language Models and Deterministic EM Simulators for Interactive Radio Map Analysis
Deterministic electromagnetic (EM) simulators provide accurate radio propagation modeling but often require expert configuration and lack interactive flexibility. We present RadioSim Agent, an agentic framework that integrates large language models (LLMs) with physics-based EM solvers and vision-enabled reasoning to enable interactive and explainable radio map generation. The framework encapsulates ray-tracing models as callable simulation tools, orchestrated by an LLM capable of interpreting natural language objectives, managing simulation workflows, and visually analyzing resulting radio maps. Demonstrations in urban UAV communication scenarios show that the agent autonomously selects appropriate propagation mechanisms, executes deterministic simulations, and provides semantic and visual summaries of pathloss behavior. The results indicate that RadioSim Agent provides multimodal interpretability and intuitive user interaction, paving the way for intelligent EM simulation assistants in next-generation wireless system design.
- Open Access
- 26 August 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Open Journal of the Communications Society
- Vol. 5, 5667-5679
Geometrical Features Based-mmWave UAV Path Loss Prediction Using Machine Learning for 5G and Beyond
Unmanned aerial vehicles (UAVs) are envisioned to play a pivotal role in modern telecommunication and wireless sensor networks, offering unparalleled flexibility and mobility for communication and data collection in diverse environments. This paper presents a comprehensive investigation into the performance of supervised machine learning (ML) models for path loss (PL) prediction in UAV-assisted millimeter-wave (mmWave) radio networks. Leveraging a unique set of interpretable geometrical features, six distinct ML models–linear regression (LR), support vector regressor (SVR), K nearest neighbors (KNN), random forest (RF), extreme gradient boosting (XGBoost), and deep neural network (DNN)–are rigorously evaluated using a massive dataset generated from extensive ray-tracing (RT) simulations in a typical urban environment. Our results demonstrate that the RF algorithm outperforms other models showcasing superior predictive performance for the test dataset with a root mean square error (RMSE) of 2.38 dB. The proposed ML models demonstrate superior accuracy compared to 3GPP and ITU-R models for mmWave radio networks. This study thoroughly investigates the adaptability of these models to unseen environments and examines the feasibility of training them with sparse datasets to improve accuracy. The reduction in computation time achieved by using ML models instead of extensive RT computations for sparse training datasets is evaluated, and an efficient algorithm for training such models is proposed. Additionally, the sensitivity of ML models to noisy input features is analyzed. We also assess the importance of geometrical features and the impact of sequentially increasing the number of these features on model performance. The results emphasize the significance of the proposed geometrical features and demonstrate the potential of ML models to provide computationally efficient and relatively accurate PL predictions in diverse urban environments.
- 26 August 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Open Journal of the Communications Society
- Vol. 5, 5667-5679
Open AccessGeometrical Features Based-mmWave UAV Path Loss Prediction Using Machine Learning for 5G and Beyond
Unmanned aerial vehicles (UAVs) are envisioned to play a pivotal role in modern telecommunication and wireless sensor networks, offering unparalleled flexibility and mobility for communication and data collection in diverse environments. This paper presents a comprehensive investigation into the performance of supervised machine learning (ML) models for path loss (PL) prediction in UAV-assisted millimeter-wave (mmWave) radio networks. Leveraging a unique set of interpretable geometrical features, six distinct ML models–linear regression (LR), support vector regressor (SVR), K nearest neighbors (KNN), random forest (RF), extreme gradient boosting (XGBoost), and deep neural network (DNN)–are rigorously evaluated using a massive dataset generated from extensive ray-tracing (RT) simulations in a typical urban environment. Our results demonstrate that the RF algorithm outperforms other models showcasing superior predictive performance for the test dataset with a root mean square error (RMSE) of 2.38 dB. The proposed ML models demonstrate superior accuracy compared to 3GPP and ITU-R models for mmWave radio networks. This study thoroughly investigates the adaptability of these models to unseen environments and examines the feasibility of training them with sparse datasets to improve accuracy. The reduction in computation time achieved by using ML models instead of extensive RT computations for sparse training datasets is evaluated, and an efficient algorithm for training such models is proposed. Additionally, the sensitivity of ML models to noisy input features is analyzed. We also assess the importance of geometrical features and the impact of sequentially increasing the number of these features on model performance. The results emphasize the significance of the proposed geometrical features and demonstrate the potential of ML models to provide computationally efficient and relatively accurate PL predictions in diverse urban environments.
- Open Access
- 13 March 2024
- Published by Springer Nature in Discover Computing
- Vol. 27 (1) , 1-14
Arithmetic N-gram: an efficient data compression technique
Due to the increase in the growth of data in this era of the digital world and limited resources, there is a need for more efficient data compression techniques for storing and transmitting data. Data compression can significantly reduce the amount of storage space and transmission time to store and transmit given data. More specifically, text compression has got more attention for effectively managing and processing data due to the increased use of the internet, digital devices, data transfer, etc. Over the years, various algorithms have been used for text compression such as Huffman coding, Lempel-Ziv-Welch (LZW) coding, arithmetic coding, etc. However, these methods have a limited compression ratio specifically for data storage applications where a considerable amount of data must be compressed to use storage resources efficiently. They consider individual characters to compress data. It can be more advantageous to consider words or sequences of words rather than individual characters to get a better compression ratio. Compressing individual characters results in a sizeable compressed representation due to their less repetition and structure in the data. In this paper, we proposed the ArthNgram model, in which the N-gram language model coupled with arithmetic coding is used to compress data more efficiently for data storage applications. The performance of the proposed model is evaluated based on compression ratio and compression speed. Results show that the proposed model performs better than traditional techniques.
- 13 March 2024
- Published by Springer Nature in Discover Computing
- Vol. 27 (1) , 1-14
Open AccessArithmetic N-gram: an efficient data compression technique
Due to the increase in the growth of data in this era of the digital world and limited resources, there is a need for more efficient data compression techniques for storing and transmitting data. Data compression can significantly reduce the amount of storage space and transmission time to store and transmit given data. More specifically, text compression has got more attention for effectively managing and processing data due to the increased use of the internet, digital devices, data transfer, etc. Over the years, various algorithms have been used for text compression such as Huffman coding, Lempel-Ziv-Welch (LZW) coding, arithmetic coding, etc. However, these methods have a limited compression ratio specifically for data storage applications where a considerable amount of data must be compressed to use storage resources efficiently. They consider individual characters to compress data. It can be more advantageous to consider words or sequences of words rather than individual characters to get a better compression ratio. Compressing individual characters results in a sizeable compressed representation due to their less repetition and structure in the data. In this paper, we proposed the ArthNgram model, in which the N-gram language model coupled with arithmetic coding is used to compress data more efficiently for data storage applications. The performance of the proposed model is evaluated based on compression ratio and compression speed. Results show that the proposed model performs better than traditional techniques.
- 9 February 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Antennas and Wireless Propagation Letters
- Vol. 22 (6) , 1401-1405
An Efficient Ray Validation Technique for Ray-Tracing in Urban Microcellular Environments
Ray-tracing propagation models for urban microcellular environments become time consuming as the validation of higher order of ray interactions presents a significant computational overhead. This letter presents a time efficient ray validation algorithm for ray-tracing models. The model presented in this letter uses a pre-computed database of building faces that are completely or partially visible to each face and vertical edge of all the buildings in the radio environment. This database is used to readily validate the ray-segments between the transmitter and receiver. The validation results show considerable time saving over previous models.
- 9 February 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Antennas and Wireless Propagation Letters
- Vol. 22 (6) , 1401-1405
An Efficient Ray Validation Technique for Ray-Tracing in Urban Microcellular Environments
Ray-tracing propagation models for urban microcellular environments become time consuming as the validation of higher order of ray interactions presents a significant computational overhead. This letter presents a time efficient ray validation algorithm for ray-tracing models. The model presented in this letter uses a pre-computed database of building faces that are completely or partially visible to each face and vertical edge of all the buildings in the radio environment. This database is used to readily validate the ray-segments between the transmitter and receiver. The validation results show considerable time saving over previous models.
- Open Access
- 1 November 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 10, 113690-113698
Machine Learning Approaches for Radio Propagation Modeling in Urban Vehicular Channels
The use of vehicular communications is anticipated to improve safety in road traffic. The traditional radio channel models that describe the effects of radio wave propagation in dynamic vehicular environments have their own limitations. In this paper, machine learning techniques are applied for radio channel modeling in urban vehicular environments. A large data set of path loss and RMS delay spread is computed using ray-tracing for a Line-of-Sight (LOS) straight road and a Non Line-of-Sight (NLOS) intersection road scenario. Fourteen input features are used to train three machine learning models for vehicular channel prediction. The models considered in this work include Multi-layer perceptron (MLP), Convolutional neural network (CNN) and Random forest (RF). The results show that RF gives better performance than MLP and CNN models in prediction of path loss and RMS delay spread in urban vehicular channels.
- 1 November 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 10, 113690-113698
Open AccessMachine Learning Approaches for Radio Propagation Modeling in Urban Vehicular Channels
The use of vehicular communications is anticipated to improve safety in road traffic. The traditional radio channel models that describe the effects of radio wave propagation in dynamic vehicular environments have their own limitations. In this paper, machine learning techniques are applied for radio channel modeling in urban vehicular environments. A large data set of path loss and RMS delay spread is computed using ray-tracing for a Line-of-Sight (LOS) straight road and a Non Line-of-Sight (NLOS) intersection road scenario. Fourteen input features are used to train three machine learning models for vehicular channel prediction. The models considered in this work include Multi-layer perceptron (MLP), Convolutional neural network (CNN) and Random forest (RF). The results show that RF gives better performance than MLP and CNN models in prediction of path loss and RMS delay spread in urban vehicular channels.
- 2 June 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Antennas and Propagation
- Vol. 70 (10) , 9977-9982
A Visibility Matching Technique for Efficient Millimeter-Wave Vehicular Channel Modeling
Site specific propagation models for highly dynamic vehicle-to-vehicle scenarios become time consuming as the identification of visible surfaces must be performed independently for each transmitter location. Moreover the validation of higher order of ray interactions presents a significant computational overhead. This paper presents an efficient ray-tracing algorithm for vehicle-to-vehicle propagation prediction. The model presented in this paper extends a previous model that uses a pre-computed database of intra-visibility of walls and edges in the environment. The model computes the list of faces and edges that are directly visible to the mobile receiver. This visibility information is then used to reduce the image-tree thereby accelerating the ray validation. The paper also presents an efficient technique to compute diffuse scattered rays in a timely manner. The validation results show considerable time saving over previous models.
- 2 June 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Antennas and Propagation
- Vol. 70 (10) , 9977-9982
A Visibility Matching Technique for Efficient Millimeter-Wave Vehicular Channel Modeling
Site specific propagation models for highly dynamic vehicle-to-vehicle scenarios become time consuming as the identification of visible surfaces must be performed independently for each transmitter location. Moreover the validation of higher order of ray interactions presents a significant computational overhead. This paper presents an efficient ray-tracing algorithm for vehicle-to-vehicle propagation prediction. The model presented in this paper extends a previous model that uses a pre-computed database of intra-visibility of walls and edges in the environment. The model computes the list of faces and edges that are directly visible to the mobile receiver. This visibility information is then used to reduce the image-tree thereby accelerating the ray validation. The paper also presents an efficient technique to compute diffuse scattered rays in a timely manner. The validation results show considerable time saving over previous models.
- 1 October 2021
- Published by Elsevier in Physical Communication
- Vol. 48, 101445
Investigation of corner diffraction for different materials in vehicular channels at urban street intersection
- 1 October 2021
- Published by Elsevier in Physical Communication
- Vol. 48, 101445
Investigation of corner diffraction for different materials in vehicular channels at urban street intersection
- Open Access
- 31 March 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 9, 52851-52859
Performance Analysis of Multi-User Downlink PD-NOMA Under SUI Fading Channel Models
Power Domain Non Orthogonal Multiple Access (PD-NOMA) is a multiple access technique that offers spectral efficiency, low latency and user fairness in 5G networks. This paper evaluates the performance of multi-user downlink PD-NOMA in suburban environments under SUI fading channels. Baseband modulated user signals with phase shifts are allocated power levels in accordance with users distance before the superposition coding is performed at the base station. A multi-level successive interference cancellation (SIC) is performed at the receivers. Bit error rate (BER) performance against signal-to-noise ratio (SNR) for up to four users per cluster is compared for all the SUI models corresponding to different suburban terrains with varying vegetation densities. An upper limit on number of users per cluster that are supported at a benchmark BER and SNR values is also evaluated.
- 31 March 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 9, 52851-52859
Open AccessPerformance Analysis of Multi-User Downlink PD-NOMA Under SUI Fading Channel Models
Power Domain Non Orthogonal Multiple Access (PD-NOMA) is a multiple access technique that offers spectral efficiency, low latency and user fairness in 5G networks. This paper evaluates the performance of multi-user downlink PD-NOMA in suburban environments under SUI fading channels. Baseband modulated user signals with phase shifts are allocated power levels in accordance with users distance before the superposition coding is performed at the base station. A multi-level successive interference cancellation (SIC) is performed at the receivers. Bit error rate (BER) performance against signal-to-noise ratio (SNR) for up to four users per cluster is compared for all the SUI models corresponding to different suburban terrains with varying vegetation densities. An upper limit on number of users per cluster that are supported at a benchmark BER and SNR values is also evaluated.
- 31 January 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Antennas and Propagation
- Vol. 67 (5) , 3323-3333
Efficient Preprocessed Ray Tracing for 5G Mobile Transmitter Scenarios in Urban Microcellular Environments
There has been a great deal of interest in radio channel modeling for upcoming vehicular and device-to-device 5G applications. Site-specific models for such mobile transmitter scenarios become time-consuming as the visibility computations need to be independently performed for each transmitter location. This paper presents an efficient ray-tracing algorithm for a mobile transmitter radiating in an urban environment. The model presented uses a precomputed database of intravisibility of walls and edges in the environment. A visibility table that contains a list of walls and edges visible to multiple transmitters along a linear trajectory is computed for the given order of ray interaction. The walls and edges visible at a given location along the transmitter route are then readily determined using the visibility table thereby accelerating the ray-tracing algorithm. The validation results show considerable time-saving over unaccelerated ray tracing.
- 31 January 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Antennas and Propagation
- Vol. 67 (5) , 3323-3333
Efficient Preprocessed Ray Tracing for 5G Mobile Transmitter Scenarios in Urban Microcellular Environments
There has been a great deal of interest in radio channel modeling for upcoming vehicular and device-to-device 5G applications. Site-specific models for such mobile transmitter scenarios become time-consuming as the visibility computations need to be independently performed for each transmitter location. This paper presents an efficient ray-tracing algorithm for a mobile transmitter radiating in an urban environment. The model presented uses a precomputed database of intravisibility of walls and edges in the environment. A visibility table that contains a list of walls and edges visible to multiple transmitters along a linear trajectory is computed for the given order of ray interaction. The walls and edges visible at a given location along the transmitter route are then readily determined using the visibility table thereby accelerating the ray-tracing algorithm. The validation results show considerable time-saving over unaccelerated ray tracing.