Techniques to improve computational efficiency in massive multiple input multiple output (MIMO) systems
Various embodiments herein provide techniques for reducing computational complexity in multiple input multiple output (MIMO, e.g., massive MIMO (mMIMO)) communications, such as efficient decomposition of a channel covariance matrix. Embodiments enable accurate mMIMO channel estimation with relatively low computational complexity compared with prior techniques. Additionally, embodiments may provide efficient beamforming and/or spatial compression. Other embodiments may be described and claimed.
1. One or more non-transitory computer-readable media (NTCRM) having instructions stored thereon, that when executed by one or more processors cause a device to:
obtain a channel covariance matrix associated with a channel for multiple input multiple output (MIMO) communication;
select a first number of columns of the channel covariance matrix to obtain a sub-matrix, wherein the first number is less than a total number of columns of the channel covariance matrix;
compute a QR decomposition of the sub-matrix to obtain a matrix Q; and
generate a compressed matrix that includes a second number of columns of the matrix Q, wherein the second number is equal to or less than the first number.
2. The one or more NTCRM of claim 1 , wherein the first number of columns are selected randomly or deterministically.
3. The one or more NTCRM of claim 1 , wherein the QR decomposition is a sorted QR decomposition.
4. The one or more NTCRM of claim 1 , wherein the channel covariance matrix is associated with received sounding reference signals (SRSs) from respective user equipments (UEs).
5. The one or more NTCRM of claim 4 , wherein the device is a baseband unit (BBU), and wherein the instructions, when executed, are further to cause the BBU to send the compressed matrix to a radio unit (RU) for MIMO spatial compression by the RU of data or control signals received on the channel.
6. The one or more NTCRM of claim 1 , wherein the channel covariance matrix is combined across individual user equipments (UEs), and wherein the instructions, when executed, are further to cause the device to:
determine beamforming vectors based on the compressed matrix; and
receive MIMO signals from a plurality of UEs based on the determined beamforming vectors.
7. The one or more NTCRM of claim 1 , wherein the channel covariance matrix is associated with a UE, and wherein the instructions, when executed, are further to cause the device to:
determine beamforming weights based on the compressed matrix; and
receive an uplink signal from the UE or transmit a downlink signal to the UE based on the determined beamforming weights.
8. The one or more NTCRM of claim 1 , wherein the device is an open radio unit (O-RU).
9. The one or more NTCRM of claim 1 , wherein the device is an open distributed unit (O-DU).
10. The one or more NTCRM of claim 1 , wherein the MIMO communication is massive MIMO communication.
11. One or more non-transitory computer-readable media (NTCRM) having instructions stored thereon, that when executed by one or more processors cause a distributed unit (DU) to:
generate a matrix C′ that includes a first number of columns of an overall covariance matrix that is combined across individual user equipments (UEs) of a plurality of UEs, wherein the first number is less than a total number of columns of the overall covariance matrix;
compute a QR decomposition of the matrix C′ to obtain a matrix Q;
generate a compressed matrix that includes a second number of columns of the matrix Q, wherein the second number is equal to or less than the first number;
determine beamforming vectors based on the compressed matrix; and
receive signals from the plurality of UEs based on the determined beamforming vectors.
12. The one or more NTCRM of claim 11 , wherein the instructions, when executed, are further to cause the DU to compute only the first number of columns of the overall covariance matrix.
13. The one or more NTCRM of claim 11 , wherein the second number is less than the first number.
14. The one or more NTCRM of claim 12 , wherein the instructions, when executed, are further to cause the DU to provide the beamforming vectors to a radio unit (RU) for uplink compression.
15. The one or more NTCRM of claim 14 , wherein to receive signals from the plurality of UEs based on the determined beamforming vectors includes to receive compressed signals from the RU based on the beamforming vectors.
16. The one or more NTCRM of claim 11 , wherein the DU is an open DU (O-DU) of an open radio access network (O-RAN).
17. An apparatus to be implemented in a device of a radio access network, the apparatus comprising:
processor circuitry to:
obtain a covariance matrix associated with a channel for massive multiple input multiple output (mMIMO) communication;
select a first number of columns of the channel covariance matrix to obtain a sub-matrix, wherein the first number is less than a total number of columns of the covariance matrix;
compute a QR decomposition of the sub-matrix to obtain a matrix Q;
generate a compressed matrix that includes a second number of columns of the matrix Q wherein the second number is equal to or less than the first number; and
receive uplink signals from a plurality of user equipments (UEs) or encode downlink signals for transmission to the plurality of UEs based on the compressed matrix; and
a memory to store the compressed matrix.
18. The apparatus of claim 17 , wherein the first number of columns are selected randomly or deterministically.
19. The apparatus of claim 17 , wherein the QR decomposition is a sorted QR decomposition.
20. The apparatus of claim 17 , wherein the processing circuitry is further to:
compute the covariance matrix based on received sounding reference signals (SRSs) from respective user equipments (UEs); and
send the compressed matrix to a radio unit (RU) for mMIMO spatial compression by the RU of data or control signals received on the channel.
21. The apparatus of claim 17 , wherein the covariance matrix is combined across individual UEs of the plurality of UEs, and wherein the processing circuitry is further to:
determine beamforming vectors based on the compressed matrix; and
receive mMIMO signals from the plurality of UEs based on the determined beamforming vectors.
22. The apparatus of claim 17 , wherein the processing circuitry is to determine beamforming weights based on the compressed matrix, and wherein the uplink signals are received or the downlink signals are encoded based on the determined beamforming weights.
23. The apparatus of claim 17 , wherein the device is remote radio unit (RRU).
24. The apparatus of claim 17 , wherein the device is baseband unit (BBU).
25. The apparatus of claim 17 , wherein the RAN is an open RAN (O-RAN).