Wireless communications device and methods for beam training in a wireless communications network
A method, performed by a wireless communications device, for beam training with a radio access network of a wireless communications system. The wireless communications device is configured with a first set of multiple beams for a first beam training and a second set of multiple beams for a second beam training. The second set is different from the first set, the method including: performing the first beam training with the first set of multiple beams; calculating a channel metric associated with the wireless communications device based on the performed first beam training with the first set; selecting, for the second beam training, a subset of beams of the second set based on the calculated channel metric; and performing the second beam training with the selected subset.
1 . A method, performed by a wireless communications device, for beam training with a radio access network of a wireless communications system, wherein the wireless communications device is configured with a first set of multiple beams for a first beam training and a second set of multiple beams for a second beam training and wherein the second set is different from the first set,
the method comprising:
performing the first beam training with the first set of multiple beams;
calculating a channel metric associated with the wireless communications device based on the performed first beam training with the first set;
selecting, for the second beam training, a subset of beams of the second set based on the calculated channel metric and a learned function relating the channel metric to the subset beams of the second set; and
performing the second beam training with the selected subset;
the method further comprises a learning phase comprising:
performing multiple first beam trainings with a first learning set of multiple beams;
performing multiple second beam trainings with a second learning set of multiple beams; each beam training with the second learning set corresponding to a respective beam training with the first learning set; and
computing the learned function based on a supervised learning algorithm and based on a training data set comprising multiple instances of:
a channel metric associated with the wireless communications device and based on the respective first beam training with the first learning set in the learning phase and
a corresponding indication of beam quality of a respective beam out of one or more beams of the second learning set based on the respective second beam training in the learning phase with the second learning set.
2 . The method according to a claim 1 , wherein the channel metric comprises a spatio-temporal channel fingerprint, wherein the spatio-temporal channel fingerprint comprises one or more spatial and/or temporal characteristics of a wireless propagation channel used for the first beam training with the first set.
3 . The method according to claim 1 , wherein the function relates the channel metric to an estimated measure of a beam quality of the respective beam of the subset of beams and wherein selecting the subset of beams is further based on the estimated measure of the beam quality.
4 . The method according to claim 1 , wherein the second beam training with the subset comprises: measuring beam quality of the respective beam of the selected subset.
5 . The method according to claim 4 , further comprising: terminating measuring beam quality of the respective beam of the selected subset in response to a measured beam quality fulfilling a beam quality criterion.
6 . The method according to claim 1 , wherein the first set is associated with a first Radio Access Technology, RAT, and the second set is associated with a second RAT which differs from the first RAT and/or wherein the first set is associated with a first frequency range, and the second set is associated with a second frequency range, wherein the second frequency range differs from the first frequency range.
7 . The method according to claim 1 , wherein calculating the channel metric is based on baseband processing of one or more reference signals received with the first set during the first beam training with the first set.
8 . The method according to claim 1 , wherein selecting the subset is performed in response to a constraint on power consumption of the wireless communications device or a constraint on dedicated radio access resources used for beam training.
9 . The method according to claim 1 , further comprising determining to not buffer second reference signals received on beams of the second set which were not selected for beam training.
10 . The method according to claim 5 , further comprising, in response to the measured beam quality meeting the beam quality criterion, reducing a supplied power to at least a part of an antenna circuitry associated with beam training on beams of the second set which were not selected for the second beam training.
11 . The method according to claim 1 , wherein selecting the subset of beams out of the second set is further based on a cardinality of a high-quality subset of beams of the first set, wherein the high-quality subset comprises beams meeting a high-quality beam quality criterion.
12 . The method according to claim 1 , wherein the second set of multiple beams comprises combinations of beams comprised in a codebook, and wherein selecting the subset of beams out of the second set of multiple beams comprises selecting a subset of the combinations of beams comprised in the codebook.
13 . The method according to claim 1 , wherein the beams of the first set and the beams of the second set are receive beams.
14 . A wireless communications device for beam training with a radio access network of a wireless communications system, wherein the wireless communications device is configured with a first set of multiple beams for a first beam training and a second set of multiple beams for a second beam training and wherein the second set is different from the first set, wherein the wireless communications device comprises a processing module, which is further configured to: perform the first beam training with the first set of multiple beams; calculate a channel metric associated with the wireless communications device based on the performed beam training with the first set; select, for the second beam training, a subset of beams of the second set based on the calculated channel metric and a learned function relating the channel metric to the subset of beams of the second set; and perform the second beam training with the selected subset the wireless communications device is further configured to, during a learning phase, to: perform multiple first beam trainings with a first learning set of multiple beams; perform multiple second beam trainings with a second learning set of multiple beams; each beam training with the second learning set corresponding to a respective beam training with the first learning set; and compute the learned function based on a supervised learning algorithm and based on a training data set comprising multiple instances of: a channel metric associated with the wireless communications device and based on the respective first beam training with the first learning set in the learning phase and a corresponding indication of beam quality of a respective beam out of one or more beams of the second learning set based on the respective second beam training in the learning phase with the second learning set.