CBL - Campus del Baix Llobregat

Projecte llegit

Títol: Phase-codification of FMCW Radars for interference mitigation in autonomous automotive or aerial vehicles

Estudiants que han llegit aquest projecte:

Director/a: ÚBEDA FARRE, EDUARD

Departament: TSC

Títol: Phase-codification of FMCW Radars for interference mitigation in autonomous automotive or aerial vehicles

Data inici oferta: 02-02-2025     Data finalització oferta: 02-10-2025


Estudis d'assignació del projecte:
    DG ENG AERO/SIS TEL
Tipus: Individual
 
Lloc de realització: EETAC
 
Paraules clau:
FMCW Radar, Chirp, Atonomous vehicle, Unmanned vehicle, Interference
 
Descripció del contingut i pla d'activitats:
In recent years, the advanced design of frequency modulated (FM) continuous wave (CW) Radars has gained a lot of interest because they are very well suited for detection and tracking in traffic environments to enable safe driving for automotive or aerial vehicles because, unlike other available technologies, such as Lidar or video cameras, they can operate under diverse weather conditions. Moreover, the short-range scope of such sensors boarded in vehicles, of several hundreds of meters, prevent the adoption of pulsed Radars, like the ones used in airports for air traffic control, which mainly focus on long distances and cannot detect targets in the very near range. Most of these boarded radars use linear frequency modulated continuous wave schemes (also called Chirp modulation) because the hardware structure of the sensor becomes simple and the bandwidth of the receiver small. However, multiple FMCW radars operating at the same time within the same frequency band are prone to cause mutual interference. Moreover, this mutual interference problem is expected to grow in the future as the vehicles require more radar sensors to better learn the environment around.

One option for mitigating this interference is the application of a phase-codification to the transmitted FMCW waveforms so that they can be distinguished by the receiver from the signals generated by other sensors around. These phase-sequences need to be mutually orthogonal in order to avoid ghost-target detection. In this work, several phase-sequences, such as Hadamard-Walsh or Gold frames, among others, will be programmed and inserted in the FMCW Radar transmitting and matched-filter receiving schemes. A relative comparison of the properties between several orthogonal phase-sequences will be carried out through the Ambiguity function and the main-to-side-lobe level. Furthermore, a realistic simulation on the degradation of the sensor detection of low-level echoes associated with interfering signals with the presence of Gaussian noise.
 
Overview (resum en anglès):
Radar technology, particularly Frequency Modulated Continuous Wave (FMCW) radar, is a cornerstone of modern Advanced Driver-Assistance Systems (ADAS) and the progression toward fully autonomous vehicles. However, the rapid proliferation of these radar-equipped vehicles creates an increasingly crowded spectrum, leading to a critical and growing problem of mutual interference. This interference can degrade sensor performance, raise the noise floor, and create dangerous "ghost targets," thereby compromising the reliability of safety-critical functions.

This work investigates Phase-Coded FMCW (PC-FMCW) as a robust strategy to mitigate this interference. The study conducts a comprehensive experimental analysis to evaluate and compare the performance of two established mutually quasi-orthogonal pseudo-random code families (Gold and Kasami) in the BPSK phase-codification of FMCW signals.

A MATLAB-based simulation is used to evaluate this interference potential, quantifying performance through a degradation factor defined as the ratio of unwanted interference power to the desired signal power. The results reveal a critical trend: as the number of codes in the family increases, Gold code performance becomes highly bounded and predictable. This convergence presents a clear design trade-off: families with a large number of codes offer reliability by improving the worst-case interference scenario, but at the cost of eliminating the exceptionally low-interference pairs found in families with a smaller number of codes.

Although this work focuses on the signal-to-noise-ratio deterioration in Short-, Mid-, and Long-Range Radars mounted on automotive vehicles, where the FMCW Radar technology is already available for the adaptive cruise control or lane keeping, the observations may be extended for the assistance of unmanned aerial vehicles, where FMCW Radars are readily available for collision avoidance and precision landing purposes.

© CBLTIC Campus del Baix Llobregat - UPC