Institutional Repository
Technical University of Crete
EN  |  EL

Search

Browse

My Space

Passive radar with ambient illuminators and low-cost software-defned radios

Vardakis Iosif

Full record


URI: http://purl.tuc.gr/dl/dias/089926F2-B3F5-4A4A-91FF-AE49CE413029
Year 2020
Type of Item Diploma Work
License
Details
Bibliographic Citation Iosif Vardakis, "Passive radar with ambient illuminators and low-cost software-defned radios", Diploma Work, School of Electrical and Computer Engineering, Technical University of Crete, Chania, Greece, 2020 https://doi.org/10.26233/heallink.tuc.86359
Appears in Collections

Summary

This thesis presents a complete implementation of a low-cost FM passive radar. The hardware and software of the setup are explained in detail, including the low-cost, 12-bit resolution LimeSDR USB receiver. The adaptive filters Least-Squares (LS), Normalized Least-Mean-Square (NLMS), Block NLMS (BNLMS) and Fast BNLMS (FBNLMS) were implemented. Their effectiveness on suppressing the clutter reflections and their runtimes are compared. The results show that the fastest algorithm is FBNLMS, achieving the clutter removal processing in 0.7 seconds with a standard laptop computer. This is less than the Coherent Processing Interval (CPI), chosen at 1.09 seconds and sampling rate of 2.4 Msamples per second. As a result, it is the best choice for a real-time passive radar. NLMS had the best performance in terms of clutter removal, reducing the noise floor 4.3 dB compared to FBNLMS and 3.5 dB compared to LS. The LS filter proved to be a good trade-off between time efficiency and clutter suppression. In addition, an efficient algorithm to compute the cross-ambiguity function, along with a simple constant false alarm rate for target detection are included. Experimental results calculating the accuracy of the passive radar in both bistatic range and Doppler frequency are presented. The commercial flight OAL5SA that departed from the airport of Chania in the 14th of July, 2020, was used to measure the theoretical bistatic range and Doppler frequency, with a publicly available flight tracking application. Comparing the experimental outcomes with the theoretic measurements showed that in the bistatic range, there is an average error of 2795 metres and in the Doppler frequency there is an average error of 7 Hz.

Available Files

Services

Statistics