5G SLAM Using the Clustering and Assignment Approach with Diffuse Multipath.

Area of Science:

• Electrical Engineering
• Computer Science
• Signal Processing

Background:

• 5G communication systems operating above 24 GHz offer potential for precise user localization and environment mapping.
• Existing methods often use simplified models or direct positioning, limiting accuracy and robustness.
• There is a need for advanced techniques that handle practical signal propagation and measurement complexities.

Purpose of the Study:

• To propose a novel, intermediate approach for simultaneous localization and mapping (SLAM) using 5G signals above 24 GHz.
• To decompose the complex SLAM problem into manageable phases for reduced computational complexity.
• To develop a robust method that accounts for real-world signal impairments in 5G localization.

Main Methods:

• The proposed method involves four distinct phases: downlink data transmission, multi-dimensional channel estimation, channel parameter clustering, and SLAM.
• A novel likelihood function is introduced for the SLAM phase to enhance accuracy.
• The approach considers an end-to-end processing chain to integrate various signal processing steps.

Main Results:

• Simulation results indicate the effectiveness of the proposed four-phase approach.
• The method demonstrates improved performance in simultaneous localization and mapping tasks compared to existing techniques.
• The decomposition into phases leads to lower computational complexity while maintaining accuracy.

Conclusions:

• The developed four-phase approach provides an efficient and robust solution for 5G-based localization and mapping above 24 GHz.
• This method effectively addresses practical signal impairments, enhancing the reliability of 5G positioning systems.
• The findings pave the way for more advanced applications of 5G technology in robotics, augmented reality, and autonomous systems.