Laxima Kandel

Dr. Laxima Kandel

Electrical Engineering and Computer Science, Embry–Riddle Aeronautical University, Daytona Beach, Florida, USA
orcid

Research Focus & Metrics

Academic Career Length

2Years (2025-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
22 August 2025 in IoT
Verified
Journal Article
Open access
PDF file
RDRaju Dhakal
Raju Dhakal
LKLaxima Kandel
Laxima Kandel
PSPrashant Shekhar
Prashant Shekhar
As IoT (internet of things) devices grow in prominence, safeguarding them from cyberattacks is becoming a pressing challenge. To bootstrap IoT security, device identification or authentication is crucial for establishing trusted connections among devices without prior trust. In this regard, radio frequency fingerprinting (RFF) is gaining attention because it is more efficient and requires fewer computational resources compared to resource-intensive cryptographic methods, such as digital signatures. RFF works by identifying unique manufacturing defects in the radio circuitry of IoT devices by analyzing over-the-air signals that embed these imperfections, allowing for the identification of the transmitting hardware. Recent studies on RFF often leverage advanced classification models, including classical machine learning techniques such as K-Nearest Neighbor (KNN) and Support Vector Machine (SVM), as well as modern deep learning architectures like Convolutional Neural Network (CNN). In particular, CNNs are well-suited as they use multidimensional mapping to detect and extract reliable fingerprints during the learning process. However, a significant limitation of these approaches is that they require large datasets and necessitate retraining when new devices not included in the initial training set are added. This retraining can cause service interruptions and is costly, especially in large-scale IoT networks. In this paper, we propose a novel solution to this problem: RFF using Siamese networks, which eliminates the need for retraining and allows for seamless authentication in IoT deployments. The proposed Siamese network is trained using in-phase and quadrature (I/Q) samples from 10 different Software-Defined Radios (SDRs). Additionally, we present a new algorithm, the Similarity-Based Embedding Classification (SBEC) for RFF. We present experimental results that demonstrate that the Siamese network effectively distinguishes between malicious and trusted devices with a remarkable 98% identification accuracy.

14 December 2025 in Systems
Verified
Journal Article
Open access
PDF file
SASahithi Angara
Sahithi Angara
LKLaxima Kandel
Laxima Kandel
RDRaju Dhakal
Raju Dhakal
The modern power grid is considered a Smart Grid (SG) when it relies extensively on technologies that integrate traditional power infrastructure with Information and Communication Technologies (ICTs). The dependence on Internet of Things (IoT)-based communication systems to operate physical power devices transforms the grid into a complex system of systems (SoS), introducing cybersecurity vulnerabilities across various SG layers. Several surveys have addressed SG cybersecurity, but none have correlated recent developments with the NIST seven-domain framework, a comprehensive model covering all major SG domains and crucial for domain-level trend analysis. To bridge this gap, we systematically review and classify studies by impacted NIST domain, threat type, and methodology (including tools/platforms used). We note that the scope of applicability of this study is 60 studies (2011–2024) selected exclusively from IEEE Xplore. Unlike prior reviews, this work maps contributions to the NIST domain architecture, examines temporal trends in research, and synthesizes cybersecurity defenses and their limitations. The analysis reveals that research is unevenly distributed: the Operations domain accounts for ~35% of all studies, followed by Generation ~25% and Distribution ~14%, while domains like Transmission (~9%) and Service Provider (5%) are comparatively under-studied. We find a heavy reliance on simulation-based tools (~46% of studies) such as MATLAB/Simulink, and False Data Injection (FDI) attacks are predominantly studied, comprising approximately 36% of analyzed attacks. The broader objective of this work is to guide researchers and SG stakeholders (e.g., utilities, policy-makers) toward understanding and coordinating strategies for improving system-level cyber-resilience, which is crucial for future SGs, while avoiding any overstatement of findings beyond the reviewed evidence.

1 September 2025 in Land Forces Academy Review
Verified
Journal Article
Open access
PDF file
MSMiroslav Stefanov
Miroslav Stefanov
KSKristyan Stefanov
Kristyan Stefanov
LKLaxima Kandel
Laxima Kandel
SCSean Crouse
Sean Crouse
BJBoyan Jekov
Boyan Jekov
Security Operations Centers (SOCs) face significant challenges due to the large volume, diversity, and dynamics of incident events. Alarm fatigue, delayed initiation of response, and the high share of false positives or missed threats limit team effectiveness and increase organizational risk. This study presents a methodology for automated management of key performance indicators (KPIs) in an SOC environment through an Agentic AI architecture and machine learning. Within the project, 214 CSV files were processed, comprising over 8.6 million data rows extracted from SIEM, Incident Management, Task Tracking, and CRM systems. Sixteen specific indicators were used, grouped into four categories: detection and filtering (TTD, FNR, FPR), response and resolution (TTR, IRR, SIHR), recovery and operations (MTTR, OE), satisfaction and risk management (CSR, SIER). The system includes ten specialized Agentic AI agents with clearly defined roles ‒ monitoring time parameters, predicting false alarm probabilities, automatically triggering playbooks, calculating operational metrics, and analyzing customer satisfaction. Five machine learning models were trained: two XGBoost classifiers for FPR and FNR, two LightGBM regressors for TTR and MTTR, and a BERT model for textual feedback analysis. The results demonstrate reduced detection and response times, a lower rate of false alarms, and improved operational predictability in calculating KPI values. The methodology shows the applicability of Agentic AI for optimizing SOC processes on real and public data, without the need for manual intervention in most processing phases.

11 March 2026
Verified
Conference Paper
Open access
MSMiroslav Stefanov
Miroslav Stefanov
SDStoyan Denchev
Stoyan Denchev
KSKristyan Stefanov
Kristyan Stefanov
BJBoyan Jekov
Boyan Jekov
LKLaxima Kandel
Laxima Kandel
This paper presents an intelligent hybrid system for detection and response to cyberattacks targeting Unmanned Aerial Vehicles (UAVs). The proposed architecture integrates a supervised Multilayer Perceptron (MLP) for classifying known cyber threats and an unsupervised Autoencoder (AE) for identifying cyber-related behavioral anomalies. Both models operate onboard using edge AI hardware, enabling autonomous processing without reliance on ground infrastructure. The system processes telemetry, GPS, IMU, and network control data, enabling accurate detection of various attacks, including GPS spoofing, unauthorized MAVLink commands (MAVLink hijack), and DoS. Evaluation across simulated flight environments (PX4 SITL + AirSim) and real hardware-in-the-loop tests during field flight on a Blinory HK55 GPS drone, using the onboard Raspberry Pi 4 + Coral USB TPU payload processing GPS and IMU data in real time on a dataset 42 files combining UAV flight logs, IoT/CPS datasets, and simulation-generated data (>300k rows) demonstrates high classification accuracy (96.7%), F1-scores above 0.985, and anomaly-detection precision of 0.941 and recall of 0.903. The hybrid approach significantly reduces false positives and ensures on-device end-to-end reaction times (E2E RTD with sustained throughput of ∼49.9 Hz) below 0.5 s. The results validate the system's practical applicability for mission-critical UAV operations in isolated (no external connectivity) or adversarial environments.

22 February 2025 in arXiv
Verified
Preprint
Open access
PDF file
PSPrashant Shekhar
Prashant Shekhar
BDBidur Devkota
Bidur Devkota
DSDumindu Samaraweera
Dumindu Samaraweera
LKLaxima Kandel
Laxima Kandel
Manoj BabuManoj Babu
Manoj Babu
Adversarial attacks pose a significant threat to deep learning models, particularly in safety-critical applications like healthcare and autonomous driving. Recently, patch based attacks have demonstrated effectiveness in real-time inference scenarios owing to their 'drag and drop' nature. Following this idea for Semantic Segmentation (SS), here we propose a novel Expectation Over Transformation (EOT) based adversarial patch attack that is more realistic for autonomous vehicles. To effectively train this attack we also propose a 'simplified' loss function that is easy to analyze and implement. Using this attack as our basis, we investigate whether adversarial patches once optimized on a specific SS model, can fool other models or architectures. We conduct a comprehensive cross-model transferability analysis of adversarial patches trained on SOTA Convolutional Neural Network (CNN) models such PIDNet-S, PIDNet-M and PIDNet-L, among others. Additionally, we also include the Segformer model to study transferability to Vision Transformers (ViTs). All of our analysis is conducted on the widely used Cityscapes dataset. Our study reveals key insights into how model architectures (CNN vs CNN or CNN vs. Transformer-based) influence attack susceptibility. In particular, we conclude that although the transferability (effectiveness) of attacks on unseen images of any dimension is really high, the attacks trained against one particular model are minimally effective on other models. And this was found to be true for both ViT and CNN based models. Additionally our results also indicate that for CNN-based models, the repercussions of patch attacks are local, unlike ViTs. Per-class analysis reveals that simple-classes like 'sky' suffer less misclassification than others. The code for the project is available at: https://github.com/p-shekhar/adversarial-patch-transferability