A Combined Model WAPI Indoor Localization Method Based on UMAP

International Journal of Communication Systems, Journal Year: 2025, Volume and Issue: 38(6)

Published: March 5, 2025

ABSTRACT With the rapid advancement of Internet, indoor localization technology has gained increasing importance across various fields. However, complexity environments presents significant challenges for achieving precise positioning using GPS or BeiDou systems. As a result, there is growing demand innovative methods that deliver high accuracy, improved security, and cost‐effectiveness. In this study, dataset comprising 9291 fingerprints collected from building was processed split into training test sets in 7:3 ratio. To facilitate feature extraction, four algorithms—UMAP, LDA, PCA, SVD—were employed. Subsequently, six machine learning models (KNN, Random Forest, ANN, SVM, GBDT, XgBoost) were trained on set evaluated to compare their performance with different extraction algorithms. The objective identify most effective method. Model assessed three metrics: average error, coefficient determination, accuracy. Finally, stacking ensemble model developed, incorporating as primary learners selecting five superior predictive secondary learners. This approach aimed enhance UMAP significantly prediction accuracy model, whereas combining KNN, XgBoost, SVM Forest learner, achieved highest an error approximately 1.48 m.

Language: Английский

---

Artificial Intelligence-Aided Next Generation Cloud Computing Networks

Advances in computational intelligence and robotics book series, Journal Year: 2025, Volume and Issue: unknown, P. 285 - 306

Published: Jan. 17, 2025

Next-generation wireless networks (NGWNs) are extremely dynamic due to the integration of communications at different scales. (NGNs) high-speed communication network that enables many services such as packet based seamless transmission data and information. The emergence artificial intelligence (AI) helped NGNs overcome their existing issues by enabling automated management real-time optimization. AI can address next-generation challenges enhancing security through intelligent threat detection, improving scalability via resource management, reducing latency with predictive analytics. paper presents various aspects discusses most potential techniques. Further, this highlights next generation how solve problems in NGNs. Finally, some recent advancement regards techniques being used specifically machine learning, deep fuzzy logic, rule modelling natural language processing.

Language: Английский

---

Performance-driven visible light sensor for indoor georeferenced positioning

Sensors and Actuators A Physical, Journal Year: 2025, Volume and Issue: 391, P. 116612 - 116612

Published: May 2, 2025

Language: Английский

---

Digital Engineering towards future Industry

Digital engineering., Journal Year: 2024, Volume and Issue: 3, P. 100029 - 100029

Published: Dec. 1, 2024

Language: Английский

---