Calibration correction for implicit beamforming in a wireless MIMO communication system
A first packet is transmitted via a forward channel corresponding to a signal direction from a first transceiver device to a second transceiver device, wherein the first packet includes a request to initiate calibration. A partial dimensional description of a reverse channel is determined based on the reception of a second packet received from the second transceiver device, wherein the second packet includes training information of spatial dimensions used for the transmission of the second packet but does not include training information for all available spatial dimensions of a reverse channel. A correction matrix is developed from the partial dimensional description of the reverse channel and a description of the forward channel, and the correction matrix and a steering matrix is used to process signals to be transmitted via the forward channel.
1. A method of beamforming within a communication system having (i) a first transceiver device having a first plurality of antennas and (ii) a second transceiver device having a second plurality of antennas, the method comprising:
transmitting a first packet via a forward channel corresponding to a signal direction from the first transceiver device to the second transceiver device, wherein the first packet includes a request to initiate calibration;
receiving, from the second transceiver device, a second packet that includes training information of spatial dimensions used for the transmission of the second packet but does not include training information for all available spatial dimensions of a reverse channel, wherein the reverse channel corresponds to a signal direction from the second transceiver device to the first transceiver device;
determining a partial dimensional description of the reverse channel based on the reception of the second packet;
developing a correction matrix from the partial dimensional description of the reverse channel and a description of the forward channel;
using the correction matrix to process signals to be transmitted via the forward channel; and
using a steering matrix to perform beamforming in the forward channel.
2. A method according to claim 1 , wherein determining the partial dimensional description of the reverse channel comprises determining the partial dimensional description of the reverse channel without determining a full dimensional description of the reverse channel.
3. A method according to claim 1 , wherein the second packet is in response to the first packet.
4. A method according to claim 1 , further comprising receiving the description of the forward channel from the second transceiver device.
5. A method according to claim 4 , wherein receiving the description of the forward channel comprises receiving channel state information via a third packet.
6. A method according to claim 5 , further comprising:
transmitting via the forward channel a fourth packet that includes training information for all available spatial dimensions of the forward channel,
wherein the third packet is in response to the fourth packet.
7. A method according to claim 1 , wherein determining the partial dimensional description of the reverse channel comprises determining a subset of columns of a matrix corresponding to a full-dimensional description of the reverse channel.
8. A method according to claim 7 , wherein developing the correction matrix comprises:
determining a respective correction matrix estimate for each column in the subset of columns of the matrix corresponding to the full-dimensional description of the reverse channel; and
generating the correction matrix using the respective correction matrix estimates.
9. A method according to claim 7 , wherein developing the correction matrix comprises:
calculating a first singular value decomposition corresponding to the subset of columns of the matrix corresponding to the full-dimensional description of the reverse channel;
determining a subset of rows of a matrix corresponding to the description of the forward channel;
calculating a second singular value decomposition corresponding to the subset of rows of the matrix corresponding to the description of the forward channel; and
generating the correction matrix based on the first singular value decomposition and the second singular value decomposition.
10. A method according to claim 9 , wherein generating the correction matrix comprises normalizing the diagonal of the correction matrix such that the correction matrix corresponds to phase shifts.
11. A method according to claim 7 , wherein developing the correction matrix comprises:
calculating a first right singular matrix corresponding to the subset of columns of the matrix corresponding to the full-dimensional description of the reverse channel;
determining a subset of rows of a matrix corresponding to the description of the forward channel;
calculating a second right singular matrix corresponding to the subset of rows of the matrix corresponding to the description of the forward channel; and
generating the correction matrix based on the first right singular matrix and the second right singular matrix.
12. A method according to claim 11 , wherein calculating the first right singular matrix comprises using a first singular value decomposition; and
wherein calculating the second right singular matrix comprises using a second singular value decomposition.
13. A method according to claim 1 , wherein using the correction matrix comprises left-multiplying the correction matrix at baseband with a signal to be transmitted via the forward channel.
14. A method according to claim 1 , wherein using the correction matrix comprises left-multiplying the correction matrix with the steering matrix.
15. A method according to claim 1 , wherein using the correction matrix comprises calibrating a measured description of the reverse channel using the correction matrix;
the method further comprising generating the steering matrix based on the calibrated measured description of the reverse channel.
16. A method according to claim 1 , further comprising:
generating an additional correction matrix based on the correction matrix; and
processing signals received via the reverse channel using the additional correction matrix.
17. A method according to claim 16 , wherein generating the additional correction matrix comprises inverting the correction matrix.
18. An apparatus, comprising:
a beamforming network;
a controller coupled to the beamforming network, the controller configured to control the beamforming network using a steering matrix, and use a correction matrix to process signals to be transmitted via a forward channel;
a correction matrix calculation unit configured to
obtain a description of the forward channel,
obtain a partial dimensional description of a reverse channel, and
develop the correction matrix from (i) the description of the forward channel and (ii) the partial dimensional description of the reverse channel; and
a steering matrix calculation unit to develop the steering matrix.
19. An apparatus according to claim 18 , wherein the correction matrix calculation unit develops the correction matrix without determining a full dimensional description of the reverse channel.
20. An apparatus according to claim 18 , wherein the controller is configured to cause a wireless transceiver to transmit via the forward channel a first packet including a calibration initiation request;
wherein the partial dimensional description of the reverse channel is determined based on reception of a second packet received in response to the first packet, wherein the second packet includes training information of spatial dimensions used for the transmission of the second packet but does not include training information for all available spatial dimensions of the reverse channel.
21. An apparatus according to claim 20 , wherein the description of the forward channel is received via the reverse channel.
22. An apparatus according to claim 21 , wherein the controller is configured to cause the wireless transceiver to transmit via the forward channel a third packet that includes training information for all available spatial dimensions of the forward channel; and
wherein the description of the forward channel is received in response to the third packet.
23. A method, comprising:
causing a first packet to be transmitted via a forward channel corresponding to a signal direction from a first transceiver device to a second transceiver device, wherein the first packet includes a request to initiate calibration;
determining a partial dimensional description of a reverse channel based on the reception of a second packet received from the second transceiver device, wherein the second packet includes training information of spatial dimensions used for the transmission of the second packet but does not include training information for all available spatial dimensions of the reverse channel, wherein the reverse channel corresponds to a signal direction from the second transceiver device to the first transceiver device;
developing a correction matrix from the partial dimensional description of the reverse channel and a description of the forward channel; and
using the correction matrix and a steering matrix to process signals to be transmitted via the forward channel.