Publications

Conventional asymmetric cryptography is threatened by the ongoing development of quantum computers. A mandatory countermeasure in the context of virtual private networks (VPNs) is to use post-quantum cryptography (PQC) as a drop-in replacement for the authenticated key exchange in the Internet Key Exchange (IKE) protocol. However, the results of the ongoing cryptanalysis of PQC cannot be predicted. Consequently, this article discusses orthogonal methods for quantum-resistant key exchanges, like quantum key distribution (QKD) and multipath key reinforcement (MKR). As each method has limitations when used on its own, we conclude that it is best to maximize security by combining all available sources of symmetric key material to protect traffic inside a VPN. As one possible realization, we propose a lightweight proxy concept that uses available symmetric keys, like QKD and MKR keys, to implement a transparent cryptographic tunnel for all IKE packets, and consequently for PQC key exchang es. In contrast to combining PQC and symmetric key material within the IKE protocol, our approach provides security in depth: If secure symmetric keys are available, attacks on IKE and hence on PQC algorithms are infeasible. But even otherwise, the security properties of IKE and thus PQC are not weakened, so the overall security of the VPN is guaranteed to increase.

Current network functions build heavily on fixed programmed rules and lack capacity to support more expressive learning models, e.g. brain-inspired Cognitive computational models using neuromorphic computations. The major reason for this shortcoming is the huge energy consumption and limitation in expressiveness by the underlying TCAM-based digital packet processors. In this research, we show that recent emerging technologies from the analog domain have a high potential in supporting network functions with energy efficiency and more expressiveness, so called cognitive functions. We propose an analog packet processing architecture building on a novel technology named Memristors. We develop a novel analog match-action memory called Probabilistic Content-Addressable Memory (pCAM) for supporting deterministic and probabilistic match functions. We develop the programming abstractions and show the support of pCAM for an active queue management-based analog network function. The analysis over an experimental dataset of a memristor chip showed only 0.01 fJ/bit/cell of energy consumption for corresponding analog computations which is 50 times less than digital computations.

Point cloud LiDAR data are increasingly used for detecting road situations for autonomous driving. As data acquisition is much cheaper than data annotation, the classification of the collected data and selecting a part of it for annotation is an important pre-processing step. In this study, we propose a self-supervised method and a corresponding package (PSCO) to classify driving scenes in large-scale point cloud data. We propose a contrastive learning approach which encodes each single frame of the collected raw data into a feature vector that can be effectively classified. Using this approach, the number of necessary frames for annotation to train scene-based models can be significantly reduced. In addition, our method can also select specific scenario data for different individual training stages.

Rational relations on words form a well-studied and often applied notion. While the definition in trace monoids is immediate, they have not been studied in this more general context. A possible reason is that they do not share the main useful properties of rational relations on words. To overcome this unfortunate limitation, this paper proposes a restricted class of rational relations, investigates its properties, and applies the findings to systems equipped with a pushdown that does not hold a word but a trace.

N-Interval Fourier Transform Analysis (N-FTA) allows for spectral separation of a periodic target signal from uncorrelated background interference. A N-FTA pseudo-code is presented. The spectral resolution is defined by the repetition rate of the near periodic signal. Acceptance criteria for spectral targets were defined such that the probability of accepting false positives is less than 1/1500. Simulated and recorded neural compound action potentials (CAPs) were investigated. Simulated data allowed for comparison with reference solutions demonstrating the stability of N-FTA at conditions being comparable to real world data. Background activity was assessed with small errors. Evoked target components were assessed down to power spectral density being approximately N times below the background level. Validation was completed investigating a measured CAP. In neurophysiological recordings, this approach allows for accurate separation of near periodic evoked activity from uncorrelated background activities for frequencies below 1kHz. • N-FTA allows for spectral separation of a periodic target signal from uncorrelated interference by analyzing a segment containing N target signal repetitions. • A MATLAB implementation of the algorithm is provided along with simulated and recorded data. • N-FTA was successfully validated using simulated and measured data for CAPs.