Bioacoustic analyses of animal vocalizations are predominantly accomplished through manual scanning, a highly subjective and time-consuming process. Thus, validated automated analyses are needed that are usable for a variety of animal species and easy to handle by non-programing specialists. This study tested and validated whether DeepSqueak, a user-friendly software, developed for rodent ultrasonic vocalizations, can be generalized to automate the detection/segmentation, clustering and classification of high-frequency/ultrasonic vocalizations of a primate species. Our validation procedure showed that the trained detectors for vocalizations of the gray mouse lemur (Microcebus murinus) can deal with different call types, individual variation and different recording quality. Implementing additional filters drastically reduced noise signals (4225 events) and call fragments (637 events), resulting in 91% correct detections (Ntotal = 3040). Additionally, the detectors could be used to detect the vocalizations of an evolutionary closely related species, the Goodman's mouse lemur (M. lehilahytsara). An integrated supervised classifier classified 93% of the 2683 calls correctly to the respective call type, and the unsupervised clustering model grouped the calls into clusters matching the published human-made categories. This study shows that DeepSqueak can be successfully utilized to detect, cluster and classify high-frequency/ultrasonic vocalizations of other taxa than rodents, and suggests a validation procedure usable to evaluate further bioacoustics software.

Digitization of the industry comes along with improvements for modern production, because the processes can be influenced, monitored and coordinated. A digitized facility needs the possibility of communication between distributed nodes, e.g. to react to events or to provide useful information to adjust the production process. However, processes of communication can be misused by attackers. Security holes in different information systems can be found by third parties and exploited. Thus, growing data exchange needs growing security of communication. Modern intrusion detection systems (IDS) often do not fulfill the requirements of industrial systems, because they either neglect safety aspects or are not failure resistant or interrupt the data flow. The aim of this paper is to propose improvements regarding all those issues. In this paper, an online intrusion detection system architecture for industrial Ethernet is being researched on an industrial line testbed. In the current work, the requirements for intrusion detection in an industrial environment are analyzed and a hardware architecture to carry out online intrusion detection for Ethernet-based connections using a passive sniffer approach is proposed. The data is being processed in-place in a microcontroller. For the developed platform an intrusion detection algorithm using self-organizing map algorithm was implemented. The model has to be trained with normal vectors in a semi-supervised way. A prototype of the proposed architecture is evaluated on an industrial line testbed (cyber-physical factory) using TCP/IP/Ethernet header analysis. The proposed IDS, which is based on two microcontrollers, monitors an Ethernet 100-BaseTX cable and was able to detect TCP port scans, remote denial-of-service exploits and ARP cache poisoning which targeted the programmable logic controller in an industrial testbed. The proposed architecture can be used for online intrusion detection under speed restrictions.

Intrusion detection systems (IDS) based on machine learning (ML) can be used to detect anomalies in data traffic. Common challenges for IDSs are low detection rates, high false alarm rates, and the need to process large amount of data. In order to overcome these challenges various types of supervised, semi-supervised and unsupervised ML methods are being widely researched. However, the need for high-performance processing capabilities in order to perform the calculations restricts the use in industrial control systems (ICS) with usually small embedded processing elements. In this paper, a comparison of key ML methods is carried out that focuses on data traffic and processing requirements for the application to ICS. Furthermore, a variant of growing hierarchical self-organizing maps (GHSOM) is described that proposes a new method to detect anomalies based on the GHSOM behaviour. The influence of the GHSOM parameters on the detection ratio of different kinds of attacks on a production line for the proposed method is examined. To detect attacks on the ICS, growth criteria of the GHSOM have been applied as alarm generating conditions. The comparison of the performance of GHSOM, Feedforward Artificial Neural Network (ANN), Support Vector Machine (SVM) and Local Outlier Factor (LOF) took place based on receiver operating characteristics of the classification between normal and anomalous data captured from a real industrial line testbed. The proposed method has shown comparable classification results with one-class SVM and LOF as novelty detection method in the studied case scenario.

Complex production systems are increasingly using Industrial Ethernet for connecting MES, PLCs, Touch Control, and even sensors and actors within the industrial control network. While today’s production is still process driven under MES control, digitalization requires a data driven approach with cyber physical systems acting autonomously in a connected production world. Moving away from a centralized control architecture has the advantage of more flexibility but eliminates the implicit knowledge on global parameters such as the current condition or state of the overall machinery. This paper describes a methodology to retrieve these global parameters independently from any control system and fully transparent to the control network. A data sensor device is introduced that can listen to any Ethernet data traffic. Together with a specialized packet rules engine it is used to extract and combine information out of a raw Ethernet data stream to build up a virtual sensor device. A production state identification sensor is described as an example application of the virtual data sensor device.