This paper proposes a hardware-based parallel pattern matching engine using a memory-based bit-split string matcher architecture. The proposed bit-split string matcher separates the transition table from the state table, so that state transitions towards the initial state are not stored. Therefore, total memory requirements can be minimized. key words: computer network security, deep packet ins...

Implementing accurate Distribution System State Estimation (DSSE) faces several challenges, among which the lack of observability and high density distribution system. While data-driven alternatives based on Machine Learning models could be a choice, they suffer in DSSE because labeled data. In fact, measurements system are often noisy, corrupted, unavailable. To address these issues, we propos...

Abstract Deep state space models (SSMs) are an actively researched model class for temporal developed in the deep learning community which have a close connection to classic SSMs. The use of SSMs as black-box identification can describe wide range dynamics due flexibility neural networks. Additionally, probabilistic nature allows uncertainty system be modelled. In this work SSM and its paramete...