Abstract
Indoor positioning has garnered significant attention from researchers and developers due to the rapid advancement of technology and artificial intelligence. However, this field still encounters numerous challenges and limitations, particularly in achieving a balance between efficiency and accuracy.Indoor positioning systems significantly benefit various sectors, such as healthcare, smart homes, and smart transportation systems. Nonetheless, the intricate and ever-changing infrastructure of the indoor environment necessitates continuous improvement and development.
The indoor environment is rich in signals, including Bluetooth and WiFi, which can be detected using different techniques such as received signal strength (RSS) and channel state information (CSI). Each technique possesses unique features that can be extracted. Through the utilization of machine learning algorithms like support vector machine, decision tree, and other techniques, as well as more advanced methods such as deep learning, the system can extract features and establish relationships while accounting for noise and signal distortion to achieve indoor positioning and environment identification.
The accuracy of an indoor positioning system hinges on the quality and quantity of data, as well as how the data is processed and presented to enhance the feature extraction process. This paper will discuss different localization technologies and techniques, explore various artificial intelligence approaches, and examine different preprocessing techniques to enhance the output of indoor positioning systems.
| Date of Award | 17 Dec 2024 |
|---|---|
| Original language | American English |
| Supervisor | NAZAR Ali (Supervisor) |
Keywords
- Indoor Positioning Systems (IPS)
- Machine Learning Algorithms
- Received Signal Strength (RSS)
- Channel State Information (CSI)
- Deep Learning for Indoor Localization