In this study, a new volumetric array design approach is presented for both azimuth and elevation isotropic direction of arrival (DOA) estimation. The presented method uses a minimum number of additional sensors to extend the given arrays to a volumetric array which has equal values for the Cramér-Rao lower bound (CRLB) for the entire azimuth and elevation angles. This approach can be applied to all arbitrary linear and planar arrays. Moreover, the design approach also takes into account the electromagnetic mutual coupling (MC) effect between the array elements and both the azimuth and elevation angular ambiguity uncertainties in an optimum manner while extending the array. It has been verified both analytically and with simulations that the designed volumetric arrays satisfy the two dimensional (2-D) isotropic DOA estimation conditions. The proposed approach is applied to Uniform Linear Arrays (ULA), Uniform Circular Arrays (UCA), V-shaped Arrays, and an arbitrary planar array to extend to optimum volumetric arrays with considerable results.

This paper presents a methodology for generating realistic driving cycles through a combination of Markov chain modeling, Monte Carlo simulation, and dynamic time warping. The study is focused on the construction of a representative driving cycle for the city of Iğdır in Turkey, taking into account its unique traffic characteristics. The methodology involves two main stages: first, determining reference segments partitioned from original driving datasets based on traffic conditions and road types, using the dynamic time warping technique based on the similarity between each segment time series. The second stage is to stochastically generate a representative driving cycle by employing a combination of Markov chain and Monte Carlo simulation, producing variability and randomness. In this stage, the best driving cycle segment of each segment group from among the generated driving segments utilizing Markov chain modeling and Monte Carlo simulation was selected using the dynamic time warping techniques, considering the reference segments. Finally, a representative driving cycle was constructed by stitching each segment. To assess the generated representative cycle, commonly used kinematic parameters were compared with real-world driving cycle data for Iğdır. The results show that the proposed methodology provides an advanced algorithm for generating a reasonable representative driving cycle, which can contribute to energy consumption analysis, vehicle performance, and emission evaluation. The comprehensive approach provided by the proposed methodology enables an accurate understanding of driving patterns, promoting the development of sustainable mobility solutions.

Large language models (LLMs) are rapidly moving from research prototypes to socio-technical infrastructure. This chapter translates the National Institute of Standards and Technology Artificial Intelligence Risk Management Framework (NIST AI RMF) into concrete engineering practices for LLM applications, with emphasis on decoder-only architectures, retrieval-augmented generation (RAG), and secure, privacy-preserving deployment. Each NIST AI RMF function, “GOVERN, MAP, MEASURE, MANAGE,” is instantiated with implementable controls and evidence artifacts (policy maps, software bill of materials like model/data inventories, risk registers, test, evaluation, verification, and validation plans, guardrail policies, telemetry schemas, and incident playbooks). We provide metrics and release thresholds for task quality, grounding, privacy leakage, jailbreak/prompt-injection resilience, and latency/cost, and we show how these artifacts support regulatory and liability needs. A worked case study on enterprise RAG demonstrates before/after results and an incident vignette, linking technical choices to legal/policy outcomes. A concise EU crosswalk maps RMF artifacts to the EU AI Act technical documentation/postmarket monitoring and to ISO/IEC 23894/42001. A short Türkiye application note shows compatibility with the Law on the Protection of Personal Data and current national strategy, highlighting mitigations for logging/retention and cross-border transfers. The contribution is an engineering-first operationalization of ethics, law, and policy goals packaged as reusable checklists, tables, and figure templates for researchers and practitioners.

The Internet of Things (IoT) is now used across many connected settings, yet a large number of edge devices and gateways still have limited memory, processing capacity, and energy. Because of this, deployability becomes a central issue in intrusion detection, not something secondary to model accuracy. A detector that performs well in experiments may still be difficult to use in practice if it is too heavy or behaves unreliably at the edge. In this work, we study a lightweight intrusion detection framework using the CICIoT2023 benchmark. The framework includes leakage-aware preprocessing, aligned binary and multiclass labeling, several lightweight supervised models, compression for the neural model, and post hoc confidence calibration. We evaluate it under both a standard in-dataset setting and a more difficult within-dataset shift setting, where selected attack categories where selected attack categories are left out during training. The experiments cover Logistic Regression, Random Forest, LightGBM, and a compact multilayer perceptron, along with INT8 quantization, structured pruning, feature-subset ablation, and threshold-based decision control. In the standard setting, the tree-based models give the strongest overall results, while the compact neural model has the smallest footprint. The results are less reassuring under shift. When some attack families are absent from training, performance drops in a number of cases, most notably for reconnaissance-related traffic, even though the random-split results remain strong. A similar tradeoff appears in the calibration analysis: stricter thresholds reduce false alarms, but they also lower recall for malicious traffic. Overall, the results suggest that IoT intrusion detection should be evaluated with deployment in mind, considering predictive performance, runtime cost, confidence behavior, and robustness to distribution shift together.