Theme: | Localization/Signal Processing |
Application: | Object/people localization |
Contact Person: | |
External collabolator: | |
Intorduction: |
|
Description: | To determine a tag’s location the received signal from each antenna is sampled using a high-speed ADC and the derived samples are processed further in the digital domain. One of the sources of the complexity and cost of the system is the high resolution of the ADCs. It not only affects the costs of ADC itself but also the large word-length of samples at the output of the ADC increases the complexity of the subsequent processing units. Decreasing the resolution of the ADC can simplify the design; however, this can potentially degrade the performance of the localization. In this project, we investigate the use of single-bit quantization for direction of arrival (DoA) estimation and DoA-based localization. Particularly, we focus on the performance of common DoA estimation algorithms (e.g., traditional beamforming, MUSIC, CS-based methods) when applied to single-bit quantized data. Since quantization is a nonlinear operation, it can introduce unwanted artifacts, especially when multiple signal sources are present. Therefore, the goal of this project is to investigate the effects arising from single-bit quantization and how they can be mitigated |
Requirements: | You need MATLAB coding skills and basic knowledge of wireless communications (or you are willing to learn these). |
Additional Info |