This study aimed to develop a radar solution that would aid the traffic calming efforts of the CSIR business campus. The Institute of Transportation Engineers defined traffic calming as "The combination of mainly physical measures that reduce the negative effects of motor vehicle use." Radar-based solutions have been proven to help reduce the speeds of motorists in areas with speed restrictions. Unfortunately, these solutions are expensive and difficult to import. Thus, this dissertation's main focus is to produce a detailed blueprint of a radar-based solution, with technical specifications that are similar to those of commercial and experimental systems at relatively low-cost. With the above mindset, the project was initiated with the user requirements being stated. Then a detailed study of current experimental and commercial radar-based traffic calming systems followed. Thereafter, the technical and non-technical requirements were derived from user requirements, and the technical specifications obtained from the literature study. A review of fundamental radar and signal processing principles was initiated to give background knowledge for the design and simulation process. Consequently, a detailed design of the system's functional components was conceptualized, which included the hardware, software, and electrical aspects of the system as well as the enclosure design. With the detailed design in mind, a data-collection system was built. The data-collection system was built to verify whether the technical specifications, which relate to the detection performance and the velocity accuracy of the proposed radar design, were met. This was done to save on buying all the components of the proposed system while proving the design's technical feasibility. The data-collection system consisted of a radar sensor, an Analogue to Digital Converter (ADC), and a laptop computer. The radar sensor was a k-band, Continuous Wave (CW) transceiver, which provided I/Q demodulated data with beat frequencies ranging from DC to 50 kHz. The ADC is an 8-bit Picoscope 2206B portable oscilloscope, capable of sampling frequencies of up to 50 MHz. The target detection and the velocity estimation algorithms were executed on a Samsung Series 7 Chronos laptop. Preliminary experiments enabled the approximation of the noise intensity of the scene in which the radar would be placed. These noise intensity values enabled the relationship between the Signal to Noise Ratio (SNR) and the velocity error to be modelled at specific ranges from the radar, which led to a series of experiments that verified the prototypes' ability to accurately detect and estimate the vehicle speed at distances of up to 40 meters from the radar. The cell-averaging constant false alarm rate (CA-CFAR) detector was chosen as an optimum detector for this application, and parameters that produced the best results were found to be 50 reference cells and 12 guard cells. The detection rate was found to be 100% for all coherent processing intervals (CPIs) tested. The prototype was able to detect vehicle speeds that ranged from 2 km/h up to 60 km/h with an uncertainty of ±0.415 km/h, ±0.276 km/h, and ±0.156 km/h using a CPI of 0.0128 s, 0.256 s, and 0.0512 s respectively. The optimal CPI was found to be 0.0512 s, as it had the lowest mean velocity uncertainty, and it produced the largest first detection SNR of the CPIs tested. These findings were crucial for the feasibility of manufacturing a low-cost traffic calming solution for the South African market.
| Identifer | oai:union.ndltd.org:netd.ac.za/oai:union.ndltd.org:uct/oai:localhost:11427/32838 |
| Date | 12 February 2021 |
| Creators | Matu, Unathi Neo |
| Contributors | Abdul Gaffar, Mohammed Yunus, Janse van Rensburg, Vanessa |
| Publisher | Faculty of Engineering and the Built Environment, Department of Electrical Engineering |
| Source Sets | South African National ETD Portal |
| Language | English |
| Detected Language | English |
| Type | Master Thesis, Masters, MSc (Eng) |
| Format | application/pdf |
Page generated in 0.0019 seconds