Positioning, coverage, and connectivity play important roles in next-generation wireless network applications. The coverage in a wireless sensor network (WSN) is a measure of how effectively a region of interest (ROI) is monitored and targets are detected by the sensor nodes. The random deployment of sensor nodes results in poor coverage in WSNs. Additionally, battery depletion at the sensor nodes creates coverage holes in the ROI and affects network coverage. To enhance the coverage, determining the optimal position of the sensor nodes in the ROI is essential. The objective of this study is to define the optimal locations of sensor nodes prior to their deployment in the given network terrain and to increase the coverage area using the proposed version of an enhanced particle swarm optimization (EPSO) algorithm for different frequency bands. The EPSO algorithm avoids the deployment of sensor nodes in close proximity to each other and ensures that every target is covered by at least one sensor node. It applies a probabilistic coverage model based on the Euclidean distances to detect the coverage holes in the initial deployment of sensor nodes and guarantees a higher coverage probability. Delaunay triangulation (DT) helps to enhance the coverage of a given network terrain in the presence of targets. The combination of EPSO and DT is applied to cover the holes and optimize the position of the remaining sensor nodes in the WSN. The fitness function of the EPSO algorithm yielded converged results with the average number of iterations of 78, 82, and 80 at 3.6 GHz, 26 GHz, and 38 GHz frequency bands, respectively. The results of the sensor deployment and coverage showed that the required coverage conditions were met with a communication radius of 4 m compared with 6–120 m with the existing works.

Large language models (LLMs) now generate fluent natural language and source code, creating challenges for authorship attribution, academic integrity, and software supply-chain security. Most existing detectors for AI-generated content are evaluated separately on natural language or source code, often under matched train–test conditions that can overestimate real-world reliability. We present a paired-prompt benchmark for human-versus-machine detection across English text, Python code, and mixed text–code documents. The benchmark includes $\mathrm{22,141}$ instances from HC3, CodeSearchNet, MBPP, and HumanEval across training, validation, and test partitions, plus Mix-Eval, a mixed-content set of 997 Jupyter-notebook-style samples. We evaluate RoBERTa-large for text, GraphCodeBERT and CodeBERT-base for code, a unified RoBERTa-base detector trained on both modalities, and zero-shot baselines. Fine-tuned detectors achieve near-perfect in-distribution performance, with AUROC $1.0000\pm 0.0000$ and accuracy above $99.5\%$. Across five instruction-tuned generator families of varying size (3.8B–7B) and architecture, with the human and problem distributions held fixed, cross-generator transfer causes negligible degradation (AUROC spread $0.0002$; drops of at most $0.0003$). In contrast, domain shift is the main failure mode: on MBPP+HumanEval, GraphCodeBERT drops to $0.85\pm 0.02$ AUROC and CodeBERT-base to $0.67\pm 0.02$. On Mix-Eval, the unified detector outperforms a routed text–code pipeline by 21 AUROC points ($0.96$ vs. $0.75$), largely because of router failures on mixed inputs. Training-time augmentation improves low-false-positive performance, while legacy supervised detectors show systematic class inversion on modern LLM outputs. These results show that reliable deployment requires cross-domain evaluation, mixed-content testing, and calibration beyond in-distribution accuracy.