Comparison of non-cooperative interference suppression techniques

Radar sensors are widely used in the automotive industry to support Advanced Driver Assistance Systems (ADAS). Despite the fact that the use of multiple radar sensors is advantageous to provide a better understanding of the vehicles surroundings, the issue of interference arises when considering real traffic scenarios with several vehicles equipped with radar sensors.

According to the current regulation, automotive radar sensors are limited to work between 76 GHz and 81 GHz. If several radar sensors work at the same time and frequency, the involved radar sensors must be able to recognize between the own backscatterd signals, and the interfering signals that are transmitted by other radars in the surroundings. Nonetheless, this signal distinction cannot be easily done.

Consequently, each radar sensor alone must be able to distinguish and properly respond when one or multiple interferers are present in the environment. A very simple approach to do it, is to randomly change the central frequency were the affected radar is working. However, this random hopping of the central frequency results challenging when multiple radar sensors are involved. Another approach which do not involve further computational effort is to efficiently set fixed frequency bands according to the scenario description (direction of travel, position of the sensor around the vehicle, etc).

The purpose of this master thesis is to develop and compare different interference suppression strategies such as the mentioned above.