Method and device for determining at least one parameter of an object

1. A method for determining at least one parameter of a moving object having spatially separated reflectance surfaces, comprising:

receiving, at a receiving antenna, reception signals that are reflections of transmitted radar signals, from the spatially separated reflectance surfaces;

processing the reflection signals radar signal to provide a range-Doppler matrix;

determining, within the range-Doppler matrix, at least two local maximum cells, each corresponding to and indicating a respective range-radial velocity of a respective different one among the spatially separated reflectance surfaces of the object;

transferring at least a part of the range-Doppler matrix to a neural network, wherein the at least the part includes the at least two local maximum cells; and

determining at least one parameter of the object by the neural network, based at least on the respective range-radial velocity indicated by each of the at least two local maximum cells,

wherein the object is a road user.

2. The method according to claim 1 , wherein the at least one parameter is selected from the group consisting of

a distance and/or a radial velocity of the object relative to a measurement point,

an expansion of the object in at least one spatial direction,

a velocity of the object, and

at least one classification property and/or a classification of the object and/or an allocation of a zone of reflected energy to a physical object different from the object.

3. The method according to claim 1 , further comprising:

determining a zone of reflection energy in the range-Doppler matrix, the zone of reflection energy comprising a plurality of cells of the range-Doppler matrix indicating reflection energy, wherein the determining the zone of reflection energy is configured to include in the zone of reflection energy the at least two local maximum cells; and

selecting the zone of reflection energy from the range-doppler matrix,

wherein

transferring the at least two local maximum cells comprises transferring the selected zone of reflection energy to the neural network.

4. The method according to claim 1 , further comprising transmitting the radar signals, and wherein processing the reception signals comprises mixing the reception signals with transmission signals to form baseband signals, and calculating a range-Doppler matrix using the baseband signals.

5. The method according to claim 4 wherein transmitting the radar signals comprises transmitting at least two different types of frequency ramps, the transmitting being successively or simultaneously.

6. The method according to claim 5 , wherein:

the at least one parameter of the object is a radial velocity,

the determining the radial velocity of the object by the neural network is based at least in part on an evaluating, by the neural network, of the respective Doppler frequencies and/or phase information of baseband signals corresponding to the at least two local maximum cells,

and

determining the radial velocity of the object by the neural network further comprises eliminating ambiguities.

7. The method according to claim 4 , wherein transmitting the radar signal comprises transmitting the radar signals from a plurality of transmitter antennas and/or the receiving antenna is among a plurality of receiving antennas, and receiving the reception signals comprise receiving the reception signals at other among the plurality of receiving antennas.

8. The method according to claim 1 further comprising estimating a quality of the determined parameter.

9. A device for conducting a method according to claim 1 , wherein the device comprises at least one electronic data processing device which is configured to conduct the method.

10. The device according to claim 9 , further comprising at least one transmission antenna and at least one reception antenna for radar waves.