Meng Shi

The increasing quantity and complexity of code in vehicles have imposed a heavy burden on traditional integration testing methods. However, applying new testing methods, such as eliminating redundant test cases, prioritizing based on risk, and optimizing test matching, requires the testing team to possess sufficient information, which entails communication and time costs. In this study, an information management framework is developed for the integration testing phase of automotive software to assess the importance and acquisition difficulty of specific information. The framework mainly includes three core parts: (1) classifying 37 types of test-related information into five hierarchical levels (requirement level, architecture level, function level, component level, and source code level) based on risk theory; (2) designing a scale to evaluate the difficulty of information usage from three dimensions (acquisition, transmission, and evaluation); (3) providing an operational guide for integration testing departments to match information with testing strategies. This framework assists enterprises in making wiser decisions regarding testing methods and provides guidance for future collaboration between original equipment manufacturers and suppliers.

Integration compatibility has emerged as a prominent challenge in the development of digital vehicles. This paper provides a comprehensive review of previous research on automotive integration compatibility, categorizing the relevant challenges into three main categories: technical, organizational, and methodological. Furthermore, the paper distinguishes between challenges encountered during the initial integration phase and those encountered during continuous integration according to the nodes of the start of production. By engaging in discussions with internal experts from an original equipment manufacturer (OEM) and other related enterprises regarding these challenges, the paper identifies the most pressing issues that require novel solutions, which are the current practical pain points of OEMs, thereby providing directions for future research.

GNSS spoofing has become a significant security vulnerability threatening remote sensing systems. Hardware fingerprint-based GNSS receiver identification is one of the solutions to address this security issue. However, existing research has not provided a solution for distinguishing GNSS receivers of the same specification. This paper first theoretically proves that the CSACs (Chip-Scale Atomic Clocks) used in GNSS receivers have unique hardware noise and then proposes a fingerprinting scheme based on this hardware noise. Experiments based on the neural network method demonstrate that this fingerprint achieved an identification accuracy of 94.60% for commercial GNSS receivers of the same specification and performed excellently in anomaly detection, confirming the robustness of the fingerprinting method. This method shows a new real-time GNSS security monitoring method based on CSACs and can be easily used with any commercial GNSS receivers.

Timely detection and response to Intraoperative Hypotension (IOH) during surgery is crucial to avoid severe postoperative complications. Although several methods have been proposed to predict IOH using machine learning, their performance still has space for improvement. In this paper, we propose a ResNet-BiLSTM model based on multitask training and attention mechanism for IOH prediction. We trained and tested our proposed model using bio-signal waveforms obtained from patient monitoring of non-cardiac surgery. We selected three models (WaveNet, CNN, and TCN) that process time-series data for comparison. The experimental results demonstrate that our proposed model has optimal MSE (43.83) and accuracy (0.9224) compared to other models, including WaveNet (51.52, 0.9087), CNN (318.52, 0.5861), and TCN (62.31, 0.9045), which suggests that our proposed model has better regression and classification performance. We conducted ablation experiments on the multitask and attention mechanisms, and the experimental results demonstrated that the multitask and attention mechanisms improved MSE and accuracy. The results demonstrate the effectiveness and superiority of our proposed model in predicting IOH.

We apply the Empirical Mode Decomposition (EMD) algorithm and the Time Convolutional Network (TCN) structure, predicated on Convolutional Neural Networks, to successfully enable feature extraction within high-precision optical time-frequency signals, and provide effective identification and alerts for abnormal link states. Experimental validation confirms that the proposed method not only delivers an efficacy on par with traditional manual techniques, but also excels in swiftly identifying anomalies that typically elude conventional approaches. This investigation furnishes novel theoretical backing and forecasting tools for high-precision optical transmission.

The reliable monitoring of the depth of anesthesia (DoA) is essential to control the anesthesia procedure. Electroencephalography (EEG) has been widely used to estimate DoA since EEG could reflect the effect of anesthetic drugs on the central nervous system (CNS). In this study, we propose that a deep learning model consisting mainly of a deep residual shrinkage network (DRSN) and a 1 × 1 convolution network could estimate DoA in terms of patient state index (PSI) values. First, we preprocessed the four raw channels of EEG signals to remove electrical noise and other physiological signals. The proposed model then takes the preprocessed EEG signals as inputs to predict PSI values. Then we extracted 14 features from the preprocessed EEG signals and implemented three conventional feature-based models as comparisons. A dataset of 18 patients was used to evaluate the models’ performances. The results of the five-fold cross-validation show that there is a relatively high similarity between the ground-truth PSI values and the predicted PSI values of our proposed model, which outperforms the conventional models, and further, that the Spearman’s rank correlation coefficient is 0.9344. In addition, an ablation experiment was conducted to demonstrate the effectiveness of the soft-thresholding module for EEG-signal processing, and a cross-subject validation was implemented to illustrate the robustness of the proposed method. In summary, the procedure is not merely feasible for estimating DoA by mimicking PSI values but also inspired us to develop a precise DoA-estimation system with more convincing assessments of anesthetization levels.