Karim Khamaisi

Thumbnail Image
Last Name
Khamaisi
First name
Karim
Email
karim.khamaisi@unisg.ch

Settings

Now showing 1 - 5 of 5
  • Thumbnail Image
    Some of the metrics are blocked by your 
    consent settings
    Item type:Publication,
  • Thumbnail Image
    Some of the metrics are blocked by your 
    consent settings
    Item type:Publication,
    OpenCSI: Self-Calibration Layer for Heterogeneous Mesh Wireless Sensor Networks
    (IEEE, 2026-09-07)
    Khamaisi, Karim  
    ;
    Sigg, Simon
    ;
    Rodrigues, Bruno  
    WiFi CSI sensing models trained in one environment usually fail in another because standard per-session normalization bakes chip- and room-specific artifacts into feature space, requiring fresh calibration for every new room or radio. We propose OpenCSI, an abstraction layer that hides these artifacts by exposing each mesh link as a single dimensionless Z-score against its own quiet-period temporal standard deviation. The denominator is learned online from a short empty-room bootstrap and reported with a maturity tag, enabling downstream logic to detect drift and abstain when baselines become unreliable. We evaluate OpenCSI on binary occupancy across three distinct rooms and three ESP32 generations (S3, C3, C6, spanning 802.11n HT20 and 802.11ax HE20), including a same-room chip swap isolating hardware from geometry. A model trained on one deployment holds a single empty-versus-occupied decision threshold zero-shot across nearly all transfer cells, reaching binary F1 up to 0.99 where standard normalization drops to 0.87 or fails outright, with no target-domain data or retraining. The transfer is scoped to binary presence by construction, as distinguishing static from moving motion requires the absolute magnitude that temporal standard deviation removes. We release the source code and dataset to support reproducible cross-environment CSI research.
    Type:conference paper
    Journal:2026 IEEE 51st Conference on Local Computer Networks (LCN)
    DOI:10.1109/LCN67947.2026.11660791
    URI:https://alexandria.unisg.ch/handle/20.500.14171/137316
  • Thumbnail Image
    Some of the metrics are blocked by your 
    consent settings
    Item type:Publication,
    Moving Target Offense: RL-Based Adaptive Evasion Strategies for C2 Frameworks
    (2025-07-23)
    Karim Khamaisi  
    ;
    Anton Crazzolara
    ;
    Aleksandar Ristic
    ;
    Samuel Brugger
    ;
    Bruno Rodrigues  
    Modern Intrusion Detection Systems (IDS) that rely on signature-based detection struggle to detect novel threats like AI-based Command-and-Control (C2) frameworks, highlighting a gap in the ability of current systems to detect and mitigate emerging botnet threats. This work addresses this gap by investigating NimPlant's detection and evasion capabilities. This paper (1) shows strategies to detect NimPlant bots using network traffic analysis, (2) enhances their evasion capabilities using Reinforcement Learning (RL), and (3) evaluates AI-driven evasion against IDS. Using a controlled testing environment, an infected device is simulated by implementing evasion strategies and integrating them into an RL-based AI system. Herewith, AI significantly improves evasion, with the RL-enhanced NimPlant achieving higher detection bypass rates. However, the IDS configuration heavily impacts AI effectiveness, underscoring the need for robust security setups. Thus, recommendations are provided for detecting bot infections and countering AI-enhanced botnets.
    Type:conference paper
    URL:https://ieeexplore.ieee.org/abstract/document/11146346/
    URI:https://www.alexandria.unisg.ch/handle/20.500.14171/123178
  • Thumbnail Image
    Some of the metrics are blocked by your 
    consent settings
    Item type:Publication,
    From Noise to Knowledge: A Comparative Study of Acoustic Anomaly Detection Models in Pumped-storage Hydropower Plants
    (2025)
    Karim Khamaisi  
    ;
    Keller, Nicolas
    ;
    Stefan Krummenacher
    ;
    Valentin Huber
    ;
    Fäßler, Bernhard
    In the context of industrial factories and energy producers, unplanned outages are highly costly and difficult to service. However, existing acoustic-anomaly detection studies largely rely on generic industrial or synthetic datasets, with few focused on hydropower plants due to limited access. This paper presents a comparative analysis of acoustic-based anomaly detection methods, as a way to improve predictive maintenance in hydropower plants. We address key challenges in the acoustic preprocessing under highly noisy conditions before extracting time- and frequency-domain features. Then, we benchmark three machine learning models: LSTM AE, K-Means, and OC-SVM, which are tested on two real-world datasets from the Rodundwerk II pumped-storage plant in Austria, one with induced anomalies and one with real-world conditions. The One-Class SVM achieved the best trade-off of accuracy (ROC AUC 0.966-0.998) and minimal training time, while the LSTM autoencoder delivered strong detection (ROC AUC 0.889-0.997) at the expense of higher computational cost.
    Type:conference paper
    URL:http://arxiv.org/abs/2509.22881
    URI:https://www.alexandria.unisg.ch/handle/20.500.14171/123675
  • Thumbnail Image
    Some of the metrics are blocked by your 
    consent settings
    Item type:Publication,
    Scopus© Citations 1