A method of operating a radar system, wherein position information of detected targets are determined by use of machine learning, radar system, driver assistance system and vehicle

A method of operating a radar system in which at least one electromagnetic radar signal is transmitted from at least one transmit antenna element of the radar system, comprising:

• receiving at least one electromagnetic echo signal resulting from at least one radar signal reflected from at least one target in a field of view of the radar system by at least one receiving antenna element and converting the at least one received electromagnetic echo signal into received data suitable for signal processing;
• determining at least one magnitude information and at least one phase information corresponding to the at least one received electromagnetic echo signal from at least a part of the received data;
• determining at least one position information by use of machine learning, the at least one position information characterizing at least a direction of the at least one target relative to a reference system, which is related to the radar system,
• wherein the radar system is operated as a multiple-input multiple-output radar comprising multiple transmit antenna elements and multiple receiving antenna elements,
• wherein the multiple transmit antenna elements and the multiple receiving antenna elements generate a virtual antenna array with multiple virtual antenna elements for receiving electromagnetic echo signals during a multiple-input multiple-output operation of the radar system;
• determining for at least a part of the virtual antenna elements each a magnitude information and a phase information from the at least one part of the received data by performing at least one two-dimensional fast Fourier transform;
• determining at least one array data set, which comprises at least the magnitude information and the phase information of the at least one part of the virtual antenna elements;
• and
• feeding at least a part of data of the at least one array data set to at least one neural network, with which at least one position information for the at least one target is determined.