Hasil Pencarian  ::  Simpan CSV :: Kembali

"Penelitian ini bertujuan untuk mengatasi kekurangan Global Positioning System (GPS) dalam penentuan posisi di dalam ruangan dengan mengembangkan sebuah indoor positioning system (IPS) yang mengkombinasikan Wi-Fi fingerprinting dengan pedestrian dead reckoning (PDR). Wi-Fi fingerprinting memanfaatkan infrastruktur dalam ruangan yang tersebar luas sehingga implementasi menjadi praktis dan meminimalkan biaya instalasi. Namun, Wi-Fi fingerprinting memiliki tantangan karena adanya overhead dalam proses pengumpulan data. PDR melengkapi Wi-Fi fingerprinting dengan melacak pergerakan pengguna melalui deteksi langkah dan arah dengan memanfaatkan sensor inersia dalam smartphone untuk mengurangi kebutuhan pemindaian fingerprint yang sering sambil tetap mempertahankan atau meningkatkan akurasi posisi. Fokus dari penelitian ini adalah penggabungan Wi-Fi fingerprinting dengan PDR untuk menciptakan sebuah IPS yang praktis dan efisien. Teknik machine learning digunakan untuk prediksi posisi pada Wi-Fi fingerprinting. Thresholding digunakan untuk deteksi langkah (step detection), fixed step length digunakan untuk panjang langkah (step length), dan magnetometer dimanfaatkan untuk deteksi arah (heading detection). Implementasi dilakukan pada perangkat Android, di mana desain antarmuka (interface) dan pengalaman pengguna (user experience) tidak termasuk dalam cakupan penelitian. Desain sistem melibatkan aplikasi sisi klien (client-side application), backend service, machine learning service, serta algoritma penggabungan Wi-Fi fingerprinting dengan PDR. Studi ini menemukan bahwa integrasi antara Wi-Fi fingerprinting dengan PDR secara signifikan mengurangi overhead, serta meningkatkan akurasi dari penentuan posisi dalam ruangan.

---

This research aims to address the shortcomings of the Global Positioning System (GPS) in indoor positioning by developing an indoor positioning system (IPS) that combines Wi-Fi fingerprinting with pedestrian dead reckoning (PDR). Wi-Fi fingerprinting leverages widely available indoor infrastructure, making implementation practical and minimizing installation costs. However, Wi-Fi fingerprinting faces challenges due to the overhead involved in data collection. PDR complements Wi-Fi fingerprinting by tracking user movement through step detection and direction using inertial sensors in smartphones, thereby reducing the need for frequent fingerprint scanning while maintaining or improving positioning accuracy. The focus of this research is the integration of Wi-Fi fingerprinting with PDR to create a practical and efficient IPS. Machine learning techniques are used for position prediction in Wi-Fi fingerprinting. Thresholding is used for step detection, fixed step length for step length measurement, and a magnetometer for heading detection. The implementation is done on Android devices, with interface design and user experience not being within the scope of this research. The system design involves a client-side application, backend services, machine learning services, and algorithms for integrating Wi-Fi fingerprinting with PDR. This study finds that the integration of Wi-Fi fingerprinting with PDR significantly reduces overhead and improves the accuracy of indoor positioning."

"This book "is designed to teach, one step at a time. Features: the classic object-oriented introduction for beginners and experts alike, fully updated for Java SE5/​6 with many new examples and chapters! ; test framework shows program output ; Design Patterns are shown with multiple examples throughout: Adapter, Bridge, Chain of Responsibility, Command, Decorator, Facade, Factory Method, Flyweight, Iterator, Data Transfer Object, Null Object, Proxy, Singleton, State, Strategy, Template Method, and Visitor ; introduction to XML for data transfer: SWT, Flash for user interfaces ; completely rewritten concurrency chapter gives you a solid grasp of threading fundamentals ; 500+ working Java programs in 700+ compiling files, rewritten for this edition &​ Java SE5/​6 ; companion website includes all source code, annotated solution guide, weblog, and multimedia seminars ; thorough coverage of fundamentals, demonstrates advanced topics ; explains sound object-orientated principles ; 'Hands-on Java Seminar CD' available online with full multimedia seminar by Bruce Eckel"