publications

2026

1. ACMSE2026

   Practice-Informed Pedagogy for Linux Security: Perspectives on ML-Driven Solutions for Daemon Security

   Sheikh Muhammad Farjad and Robin Gandhi

   in Proceedings of the 2026 ACM Southeast Conference, Troy, AL, USA, Apr 2026

   Abs DOI PDF

   As machine learning (ML) continues to shape modern cybersecurity tools, computing education must evolve to help students critically engage with these technologies. One underrepresented topic in current curricula is daemon security, which involves protecting the long-running background processes common in Linux systems. These background processes (i.e., daemons) pose significant security risks in enterprise and cloud environments, yet receive little attention in operating systems or security courses. This paper reports on a semi-structured interview study with 22 academic and industry stakeholders. Using a Business Model Canvas-informed framework, we explored perspectives on ML-based approaches to daemon security and their educational implications. Participants conveyed both enthusiasm and apprehension. Although many acknowledged the potential of machine learning for real-time threat detection, concerns were also raised regarding automation, trust, and the adequacy of educational preparedness. Thematic analysis highlighted persistent gaps in instruction related to daemon awareness, system-level monitoring, and intelligent defense strategies. We propose practice-informed curriculum changes that directly address these instructional gaps, positioning this work as a concrete step toward aligning cybersecurity education with emerging system-level defense challenges.

2. SSRN

   Structural Health Monitoring with Machine Learning: A Systematic Literature Review and Adversarial Attack Taxonomy

   Sheikh Muhammad Farjad, Robin Gandhi, and George Grispos

   SSRN Preprint, Feb 2026

   Abs PDF

   Interest in applying machine learning (ML) to structural health monitoring (SHM) problems is growing rapidly due to its ability to automate feature extraction and recognize patterns in sensor data. However, as ML becomes more integrated into SHM, concerns arise about the security of these systems, particularly their vulnerability to adversarial attacks. This paper investigates the extent to which existing SHM research addresses security risks introduced by ML, with a focus on adversarial attacks on ML. We propose a taxonomy of adversarial attacks relevant to SHM, enabling a structured and quantified analysis of potential threats and associated risks. Using this taxonomy, we review and synthesize knowledge from 158 SHM-related research articles identified through a systematic literature review. Our findings reveal a significant gap in addressing security considerations within current SHM literature, even in studies focused on critical infrastructure applications. To bridge this gap, we offer a systematization of knowledge for adversarial attacks targeting each component of the ML pipeline used in SHM with relevant risk metrics designed to guide researchers and system designers. Finally, we highlight emerging trends and outline open problems to inform and direct future cybersecurity research.

2025

1. BuildSys ’25

   PhD Forum Abstract: Towards Secure Edge AI for Structural Health Monitoring

   Sheikh Muhammad Farjad

   in Proceedings of the 12th ACM International Conference on Systems for Energy-Efficient Buildings, Cities, and Transportation (BuildSys ’25), Golden, CO, USA, Nov 2025

   Abs DOI PDF

   On-device deep learning (DL) is reshaping structural health monitoring (SHM), but deploying models at the edge makes them increasingly vulnerable to adversarial attacks, particularly model extraction. This threat is severe for critical infrastructure, given the limited security focus of Edge AI systems. This dissertation develops a secure Edge AI system for SHM through four key contributions: (a) an adversarial risk taxonomy for assessing SHM system susceptibility to adversarial attacks, (b) a secure edge architecture and testbed for deployment, (c) an investigation of model extraction attacks using eBPF-collected metrics, and (d) a novel kernel-level defense to secure deployment.

2. BuildSys ’25

   Zeki: A Containerized Pipeline for Deep Learning Deployment for Edge-based Structural Health Monitoring

   Sheikh Muhammad Farjad and Robin Gandhi

   in Proceedings of the 12th ACM International Conference on Systems for Energy-Efficient Buildings, Cities, and Transportation (BuildSys ’25), Golden, CO, USA, Nov 2025

   Abs DOI PDF Code

   The deployment of deep learning (DL) models on edge devices offers significant opportunities for structural health monitoring (SHM), particularly by enabling localized, low-latency inference in built environments. However, practical deployment on resource-constrained platforms faces barriers in scalability, reliability, and security. In response, we introduce Zeki, a unified, security-aware, containerized pipeline for deploying DL models for edge-based SHM. Zeki unifies model optimization (via LiteRT and quantization), container-level hardening, and benchmarking-driven model-device co-selection into a reproducible workflow. We evaluate Zeki by deploying convolutional neural networks (CNNs) for crack detection on Raspberry Pi 4 and BeagleBone AI-64 along with a server. Results show significant improvements in inference latency and memory efficiency compared to unoptimized baselines. Beyond performance, Zeki establishes a systematic methodology for safe, resilient, and evidence-based edge deployment of DL in safety-critical SHM settings, enabling long-term infrastructure monitoring.

3. ACMSE2025

   Secure Edge Computing Reference Architecture for Data-driven Structural Health Monitoring: Lessons Learned from Implementation and Benchmarking 🏆

   Sheikh Muhammad Farjad, Sandeep Reddy Patllola, Yonas Kassa, and 2 more authors

   in Proceedings of the 2025 ACM Southeast Conference, Cape Girardeau, MO, USA, Apr 2025

   Best Paper Award Abs DOI PDF Code

   Received Best Full Paper Award from ACM

   Structural Health Monitoring (SHM) plays a crucial role in maintaining aging and critical infrastructure, supporting applications such as smart cities and digital twinning. These applications demand machine learning models capable of processing large volumes of real-time sensor data at the network edge. However, existing approaches often neglect the challenges of deploying machine learning models at the edge or are constrained by vendor-specific platforms. This paper introduces a scalable and secure edge-computing reference architecture tailored for data-driven SHM. We share practical insights from deploying this architecture at the Memorial Bridge in New Hampshire, US, referred to as the Living Bridge project. Our solution integrates a commercial data acquisition system with off-the-shelf hardware running an open-source edge-computing platform, remotely managed and scaled through cloud services. To support the development of data-driven SHM systems, we propose a resource consumption benchmarking framework called edgeOps to evaluate the performance of machine learning models on edge devices. We study this framework by collecting resource utilization data for machine learning models typically used in SHM applications on two different edge computing hardware platforms. edgeOps was specifically studied on off-the-shelf Linux and ARM-based edge devices. Our findings demonstrate the impact of platform and model selection on system performance, providing actionable guidance for edge-based SHM system design.

4. arXiv

   

   DaemonSec: Examining the Role of Machine Learning for Daemon Security in Linux Environments

   Sheikh Muhammad Farjad

   arXiv preprint arXiv:2504.08227, Apr 2025

   Abs DOI PDF

   DaemonSec is an early-stage startup exploring machine learning (ML)-based security for Linux daemons, a critical yet often overlooked attack surface. While daemon security remains underexplored, conventional defenses struggle against adaptive threats and zero-day exploits. To assess the perspectives of IT professionals on ML-driven daemon protection, a systematic interview study based on semi-structured interviews was conducted with 22 professionals from industry and academia. The study evaluates adoption, feasibility, and trust in ML-based security solutions. While participants recognized the potential of ML for real-time anomaly detection, findings reveal skepticism toward full automation, limited security awareness among non-security roles, and concerns about patching delays creating attack windows. This paper presents the methods, key findings, and implications for advancing ML-driven daemon security in industry.

2023

1. Investigating the Characteristics and Performance of Augmented Reality Applications on Head-Mounted Displays: A Study of the Hololens Application Store

   Pubudu Wijesooriya, Sheikh Muhammad Farjad, Nikolaos Stergiou, and 1 more author

   In 2023 IEEE International Conference on Communications Workshops (ICC Workshops), Rome, Italy, Apr 2023

   Abs DOI PDF

   Augmented Reality (AR) based on Head-Mounted Displays (HMDs) has gained significant traction over the recent years. Nevertheless, it remains unclear what AR HMD-based applications have been developed over the years and what their system performance is when they are run on HMDs. In this paper, we aim to shed light into this direction. Our study focuses on the applications available on the Microsoft Hololens application store given the wide use of the Hololens headset. Our study has two major parts: (i) we collect metadata about the applications available on the Microsoft Hololens application store to understand their characteristics (e.g., categories, pricing, permissions requested, hardware and software compatibility); and (ii) we interact with these applications while running on a Hololens 2 headset and collect data about systems-related metrics (e.g., memory and storage usage, time spent on CPU and GPU related operations) to investigate the systems performance of applications. Our study has resulted in several interesting findings, which we share with the research community.

2020

1. Cluster Analysis and Statistical Modeling: A Unified Approach for Packet Inspection

   Sheikh Muhammad Farjad and Asad Arfeen

   In 2020 International Conference on Cyber Warfare and Security (ICCWS), Islamabad, Pakistan, Apr 2020

   Abs DOI PDF

   A secure network layer capable of distinguishing between malicious and genuine traffic flows is the need of every transit service provider, edge network, corporate customer, and a common Internet user. With the emergence of advanced technologies, the demand for security has been drastically increased over the past decade. The analysis of network traffic is essential for various tasks like security, capacity planning, and visibility at various levels. In this paper, a novel architecture is proposed which exploits two powerful techniques for network traffic inspection, that is, cluster analysis and statistical modeling, and unifies them in a single framework. The proposed architecture leverages the clustering technique and statistical modeling for analyzing and inspecting the network traffic. Instead of selecting NetFlow records as the primary format, this research paper presents an approach that employs Packet Capture (PCAP) data format for network analysis. The clustering technique can be used for classifying benign and malicious traffic but there may arise many uncertainties caused by various dynamic factors due to emerging application mixture. Our proposed model uses statistical modeling for supplementing the results obtained from clustering. This unified approach for traffic analysis reduces the chances of the false alert generation that substantially deteriorates the security ecosystem. The proposed architecture inspects different parameters of network traffic to uncover any strong correlation for identifying malicious network traffic flows.

2019

1. EAI

   A Novel IoT Switching Model Based on Cloud-Centric RTDBS

   Sheikh Muhammad Farjad

   EAI Endorsed Transactions on Internet of Things, Jul 2019

   DOI PDF
2. Exploiting Real-time Database for Efficient Electrical Switching 🏆

   Sheikh Muhammad Farjad

   In Proceedings of the 34th Multi-topic International Symposium 2019, Karachi, Sindh, Pakistan, Feb 2019

   Awarded PDF

   Silver Medal

3. Exploiting Real-time Database for Efficient Electrical Switching 🏆

   Sheikh Muhammad Farjad

   In Proceedings of the 34th IEEEP All Pakistan Students’ Seminar, Karachi, Sindh, Pakistan, Jan 2019

   Awarded PDF

   Silver Paper Award