Self-Supervised Transformer-based Contrastive Learning for Intrusion Detection Systems

As the digital landscape becomes more interconnected, the frequency and severity of zero-day attacks, have significantly increased, leading to an urgent need for innovative Intrusion Detection Systems (IDS). Machine Learning-based IDS that learn from the network traffic characteristics and can discern attack patterns from benign traffic offer an advanced solution to traditional signature-based IDS. However, they heavily rely on labeled datasets, and their ability to generalize when encountering unseen traffic patterns remains a challenge. This paper proposes a novel self-supervised contrastive learning approach based on transformer encoders, specifically tailored for generalizable intrusion detection on raw packet sequences. Our proposed learning scheme employs a packet-level data augmentation strategy combined with a transformer-based architecture to extract and generate meaningful representations of traffic flows. Unlike traditional methods reliant on handcrafted statistical features (NetFlow), our approach automatically learns comprehensive packet sequence representations, significantly enhancing performance in anomaly identification tasks and supervised learning for intrusion detection. Our transformer-based framework exhibits better performance in comparison to existing NetFlow self-supervised methods. Specifically, we achieve up to a 3% higher AUC in anomaly detection for intra-dataset evaluation and up to 20% higher AUC scores in inter-dataset evaluation. Moreover, our model provides a strong baseline for supervised intrusion detection with limited labeled data, exhibiting an improvement over self-supervised NetFlow models of up to 1.5% AUC when pretrained and evaluated on the same dataset. Additionally, we show the adaptability of our pretrained model when fine-tuned across different datasets, demonstrating strong performance even when lacking benign data from the target domain.

eWake: A Novel Architecture for Semi-Active Wake-Up Radios Attaining Ultra-High Sensitivity at Extremely-Low Consumption

In this work we propose a new scheme for semi-passive Wake-Up Receiver circuits that exhibits remarkable sensitivity beyond -70 dBm, while state-of-the-art receivers illustrate sensitivity of up to -55 dBm. The receiver employs the typical principle of an envelope detector that harvests RF energy from its antenna, while it employs a nano-power operation amplifier to intensify the obtained signal prior to the final decoding that is realized with the aid of a comparator circuit. It operates at the 868 MHz ISM band using OOK signals propagated through LoRa transceivers, while also supporting addressing capabilities in order to awake only the specified network's nodes. The power expenditure of the developed receiver is as low as 580 nA, remaining at the same power consumption levels as the state-of-the-art implementations.