Home - Research

Home

The number of mobile subscribers and internet users is growing dramatically worldwide. Maintaining good Quality of Experience (QoE) for each user has become a major challenge because network resources are limited and costly. One promising solution is the development of intelligent models powered by Artificial Intelligence (AI).

As AI, data, and cybersecurity continue to converge, there is an increasing need for innovative approaches that can shape a safer, more secure, and technologically resilient society for the future.

Intrusion Detection System for Internet of Medical Things Security

In online doctor consultations, patient health data is transmitted to hospitals and physicians through digital mediums, including wearable medical devices such as smartwatches and fitness trackers. The interconnected ecosystem that enables communication among medical devices, healthcare applications, and digital systems over the Internet is known as the Internet of Medical Things (IoMT). While IoMT is transforming healthcare through seamless connectivity, it also introduces significant cybersecurity and patient safety risks arising from emerging vulnerabilities and cyberattacks. To mitigate these risks, a novel Intrusion Detection System (IDS) based on the Tsetlin Machine (TM) is developed to detect a wide range of cyberattacks. The TM is a rule-based, interpretable machine learning approach that captures attack patterns using propositional logic. To improve transparency and trust, the proposed TM-based IDS provides clear insights into the most influential clauses and the dominant class contributing to the final model decision. Overall, the proposed IDS provides an effective and interpretable solution for enhancing the security of IoMT networks and devices.

A Simple Illustration of Internet of Medical Things (IoMT) Working in Healthcare

The Overall Architecture of the Proposed Intrusion Detection System (IDS)

Contributions:

Rahul Jaiswal, Per-Arne Andersen, Linga Reddy Cenkeramaddi, Lei Jiao, Ole-Christoffer Granmo, ``On-Device Interpretable Tsetlin Machine-Based Intrusion Detection for Secure IoMT'' 22nd IEEE Intelligent Conference on Intelligence and Security Informatics (ISI), pp. 48-55, 2026. (IEEE) [Cambridge, U.K.]
Rahul Jaiswal, Per-Arne Andersen, Linga Reddy Cenkeramaddi, Lei Jiao, Ole-Christoffer Granmo, ``A Tsetlin Machine-driven Intrusion Detection System for Next-Generation IoMT Security,'' 7th Silicon Valley Cybersecurity Conference (SVCC), pp. 1-8, 2026. (IEEE) [California, U.S.A.]

No-reference Speech Quality Model

Speech quality often degrades during communication over VoIP applications such as Google Meet and Apple FaceTime due to various types of background noise present in the surrounding environment, leading to a reduction in users’ perceived Quality of Experience (QoE). To address this issue, the proposed speech quality prediction model (CAQoE) first identifies the context or noise type or degradation type in the input noisy speech signal and then estimates the corresponding context-specific speech quality. This model can enable Internet service providers to monitor and assess speech quality in real time and deploy appropriate QoE measurement and control mechanisms to maintain acceptable end-user QoE levels.

The Proposed Speech Quality Prediction Model (CAQoE)

Contributions:

Rahul Jaiswal, Andrew Hines, ``Towards a Non-Intrusive Context-Aware Speech Quality Model,'' 31st Irish Signals and Systems Conference, pp. 219-223, 2020. (IEEE) [Letterkenny, Ireland]
Rahul Jaiswal, Rajesh Kumar Dubey, ``Non-Intrusive Speech Quality Assessment Using Context-Aware Neural Networks,'' International Journal on Speech Technology, vol. 25, issue 4, pp. 947-965, 2022. (Springer) [Dordrecht, Netherlands]
Rahul Jaiswal, Rajesh Kumar Dubey, ``CAQoE: A Novel No-Reference Context-Aware Speech Quality Prediction Metric,'' ACM Transactions on Multimedia Computing, Communications, and Applications, vol. 19, issue 1s, article 35, pp. 1-23, 2023. (ACM) [New York, U.S.A.]

Transfer Learning for Radio Map Estimation

Accurate radio maps are essential for a wide range of wireless communication applications, including channel modelling, resource allocation, and network planning. However, due to variations in wireless propagation characteristics, a radio map model developed for a specific wireless environment (source environment) cannot be directly applied to a different or new wireless environment (target environment). Furthermore, constructing a new radio map model for each new wireless environment typically requires a large amount of data and involves significant computational complexity. To overcome these challenges, we propose a novel data-driven transfer learning approach that transfers and fine-tunes a deep neural network (DNN)–based radio map model trained in an original (source) wireless environment for use in new but sufficiently similar target wireless environments, thereby enabling accurate radio map generation with fewer sensor measurements.

The Proposed Model: Training Phase

The Proposed Model: Execution Phase

Contributions:

Rahul Jaiswal, Mohamed Elnourani, Siddharth Deshmukh, Baltasar Beferull-Lozano, ``Deep Transfer learning Based Radio Map Estimation for Indoor Wireless Communications," 23rd IEEE International Workshop on Signal Processing Advances in Wireless Communications (SPAWC), pp. 1-5, 2022. (IEEE) [Oulu, Finland]
Rahul Jaiswal, Mohamed Elnourani, Siddharth Deshmukh, Baltasar Beferull-Lozano, ``Location-free Indoor Radio Map Estimation using Transfer learning,'' 97th IEEE Vehicular Technology Conference (VTC-Spring), pp. 1-7, 2023. (IEEE) [Florence, Italy]
Rahul Jaiswal, Mohamed Elnourani, Siddharth Deshmukh, Baltasar Beferull-Lozano, ``A Data-driven Transfer Learning Method for Indoor Radio Map Estimation,'' IEEE Transactions on Vehicular Technology (TVT), vol. 75, no. 3, pp. 4261-4277, 2025. (IEEE) [New York, U.S.A.]
Rahul Jaiswal, Mohamed Elnourani, Siddharth Deshmukh, Baltasar Beferull-Lozano, ``Leveraging Transfer Learning for Radio Map Estimation via Mixture of Experts," IEEE Transactions on Cognitive Communications and Networking (TCCN), vol. 12, pp. 846-863, 2026. (IEEE) [New York, U.S.A.]

Artificial Intelligence for Speech Quality

Contributions:

Rahul Jaiswal, ``Influence of Silence and Noise Filtering on Speech Quality Monitoring,'' 11th International Conference on Speech Technology and Human-Computer Dialogue (SpeD), pp. 109-113, 2021. (IEEE) [Bucharest, Romania]
Rahul Jaiswal, ``Performance Analysis of Deep Learning Based Speech Quality Model with Mixture of Features,'' 24th IEEE International Symposium on Multimedia (ISM), pp. 240-244, 2023. (IEEE) [Received travel grant] [Naples, Italy]
Rahul Jaiswal, Rajesh Kumar Dubey, ``Multiple Time-instances Features Based Approach for Reference-Free Speech Quality Measurement,'' Elsevier Journal on Computer Speech and Language (CSL), vol. 79, article number 101478, pp. 1-19, 2023. (Elsevier) [Amsterdam, Netherlands]

Other Research Interests:

Speech Synthesis
Image Processing
AI Ethics

Contributions:

Rahul Jaiswal, Gaurav Saxena, ``Biometric Recognition System (Algorithm),'' 5th International Conference on Advances in Electrical Measurements and Instrumentation Engineering, pp. 40-47, 2016. (McGraw-Hill) [Cochin, India]
Rahul Jaiswal, Rajesh Kumar Dubey, ``Concatenative Text-to-Speech Synthesis System for Communication Recognition,'' 5th International Conference on Electronics, Communication and Aerospace Technology, pp. 867-872, 2021. (IEEE) [Coimbatore, India]
Rahul Jaiswal, ``AI Ethics in Next Generation Wireless Networks: A Philosophical Outlook,'' 12th Asian Conference on Ethics, Religion and Philosophy, pp. 135-144, 2022. [Tokyo, Japan]

Google Sites

Report abuse