System and method for multiple-input and multiple-output (MIMO) full-duplex precoding algorithms
Embodiments are provided to enable effective cancellation or reduction of the self-interference (SI) introduced when applying full-duplex transmission to MIMO systems. A method embodiment includes forming, using a first component matrix of a precoding matrix generated in accordance with signal channel conditions, a plurality of beams for a plurality of transmit signals, and forming, using a second component matrix of the precoding matrix, a plurality of SI cancellation signals corresponding to the transmit signals. The first component matrix is arranged to maximize a sum-rate of transmissions in the transmit signals assuming zero SI in the network component. The second component matrix is arranged to maximize SI suppression in a plurality of receive signals in the network component. The beams are transmitted at a plurality of antennas, also used to receive the receive signals. The SI cancellation signals are added to the receive signals to detect corrected receive signals.
1. A method performed by a network component for full-duplex communications in a Multiple-Input and Multiple-Output (MIMO) system, the method comprising:
initializing, at a precoder, a transmit covariance matrix for precoding a plurality of transmit signals;
calculating a sequential convex approximation function in accordance with an objective sum-rate for the transmit signals;
decomposing the transmit covariance matrix using Cholesky decomposition;
determining a full-duplex precoding (FDP) matrix in accordance with the decomposing of the transmit covariance matrix;
self-interference (SI) signals corresponding to the transmit signals; and
adding the SI cancellation signals to a plurality of receive signals.
2. The method of claim 1 , wherein calculating the sequential convex approximation function includes:
updating the transmit covariance matrix by reducing the sequential convex approximation function below a threshold; and
repeating the calculating and updating until the sequential convex approximation function reaches convergence.
3. The method of claim 1 , wherein each one of the SI cancellation signals corresponds to one of the transmit signals.
4. The method of claim 1 , wherein the plurality of transmit signals are M transmit signals, and wherein each one of the transmit covariance matrix and the FDP matrix has a size of 2M×2M.
5. The method of claim 1 , wherein the network component is one of a wireless base station, a WiFi access point, or a user equipment.
6. The method of claim 1 , wherein the sequential convex approximation function is a function of the transmit covariance matrix, a channel gain matrix, and a self-interference (SI) matrix corresponding to the transmit signals.
7. The method of claim 1 further comprising calculating transmission rates of the transmit signals as a function of the transmit covariance matrix, a channel gain matrix, and a self-interference (SI) matrix corresponding to the transmit signals.
8. A network component for full-duplex communications in a Multiple-Input and Multiple-Output (MIMO) system, the network component comprising:
a non-transitory memory storage comprising instructions; and
one or more processors in communication with the memory storage, wherein the one or more processors execute the instructions to:
initialize a transmit covariance matrix for precoding a plurality of transmit signals;
calculate a sequential convex approximation function in accordance with an objective sum-rate for the transmit signals, wherein the sequential convex approximation function is a function of the transmit covariance matrix, a channel gain matrix, and a self-interference (SI) matrix corresponding to the transmit signals;
decompose the transmit covariance matrix using Cholesky decomposition;
determine a full-duplex precoding (FDP) matrix in accordance with the decomposing of the transmit covariance matrix; and
generate, using the FDP matrix, the transmit signals and a plurality of SI cancellation signals corresponding to the transmit signals.
9. The network component of claim 8 further comprising:
a plurality of receivers coupled to a plurality of antennas;
a plurality of combiners configured to add the SI cancellation signals to a plurality of receive signals;
a circulator or isolator positioned on each transmit path between the one or more processors and each antenna; and
an auxiliary path between the one or more processors and each combiner.
10. The network component of claim 9 , wherein each transmit path includes a digital to analog converter (DAC) and an amplifier, wherein each combiner is positioned on a receive path between the circulator or isolator of each transmit path and a corresponding receiver of the plurality of receivers, wherein the receive path includes a low noise amplifier (LNA) and an analog to digital converter (ADC), and wherein the auxiliary path includes a digital to analog converters (DAC) and a second amplifier.
11. The network component of claim 8 , wherein the network component is one of a wireless base station, a WiFi access point, or a user equipment.
12. A network component for full-duplex communications in a Multiple-Input and Multiple-Output (MIMO) system, the network component comprising:
a non-transitory memory storage comprising instructions; and
one or more processors in communication with the memory storage, wherein the one or more processors execute the instructions to:
initialize a transmit covariance matrix for precoding a plurality of transmit signals;
calculate transmission rates of the transmit signals as a function of the transmit covariance matrix, a channel gain matrix, and a self-interference (SI) matrix corresponding to the transmit signals;
calculate a sequential convex approximation function in accordance with an objective sum-rate for the transmit signals;
decompose the transmit covariance matrix using Cholesky decomposition;
determine a full-duplex precoding (FDP) matrix in accordance with the decomposing of the transmit covariance matrix; and
generate, using the FDP matrix, the transmit signals and a plurality of SI cancellation signals corresponding to the transmit signals.
13. The network component of claim 12 further comprising:
a plurality of receivers coupled to a plurality of antennas;
a plurality of combiners configured to adding the SI cancellation signals to a plurality of receive signals;
a circulator or isolator positioned on each transmit path between the one or more processors and each antenna; and
an auxiliary path between the one or more processors and each combiner.
14. The network component of claim 13 , wherein each transmit path includes a digital to analog converter (DAC) and an amplifier, wherein each combiner is positioned on a receive path between the circulator or isolator of each transmit path and a corresponding receiver of the plurality of receivers, wherein the receive path includes a low noise amplifier (LNA) and an analog to digital converter (ADC), and wherein the auxiliary path includes a digital to analog converter (DAC) and a second amplifier.
15. The network component of claim 12 , wherein the network component is one of a wireless base station, a WiFi access point, or a user equipment.
16. A network component for full-duplex communications in a Multiple-Input and Multiple-Output (MIMO) system, the network component comprising:
a non-transitory memory storage comprising instructions; and
one or more processors in communication with the memory storage, wherein the one or more processors execute the instructions to:
initialize a transmit covariance matrix for precoding a plurality of transmit signals;
calculate a sequential convex approximation function in accordance with an objective sum-rate for the transmit signals;
decompose the transmit covariance matrix using Cholesky decomposition;
determine a full-duplex precoding (FDP) matrix in accordance with the decomposing of the transmit covariance matrix;
self-interference (SI) cancellation signals corresponding to the transmit signals; and
add the SI cancellation signals to a plurality of receive signals.
17. The network component of claim 16 , wherein the one or more processors execute the instructions to calculate the sequential convex approximation function includes the one or more processors execute the instructions to:
update the transmit covariance matrix by reducing the sequential convex approximation function below a threshold; and
repeat the calculate and update until the sequential convex approximation function reaches convergence.
18. The network component of claim 16 , wherein each one of the SI cancellation signals corresponds to one of the transmit signals.
19. The network component of claim 16 , wherein the plurality of transmit signals are M transmit signals, and wherein each one of the transmit covariance matrix and the FDP matrix has a size of 2M×2M.
20. The network component of claim 16 , wherein the network component is one of a wireless base station, a WiFi access point, or a user equipment.
21. The network component of claim 16 , wherein the sequential convex approximation function is a function of the transmit covariance matrix, a channel gain matrix, and a self-interference (SI) matrix corresponding to the transmit signals.
22. The network component of claim 16 , wherein the one or more processors execute the instructions to calculate transmission rates of the transmit signals as a function of the transmit covariance matrix, a channel gain matrix, and a self-interference (SI) matrix corresponding to the transmit signals.
23. A method performed by a network component for full-duplex communications in a Multiple-Input and Multiple-Output (MIMO) system, the method comprising:
initializing a transmit covariance matrix for precoding a plurality of transmit signals;
calculating a sequential convex approximation function in accordance with an objective sum-rate for the transmit signals, the sequential convex approximation function being a function of the transmit covariance matrix, a channel gain matrix, and a self-interference (SI) matrix corresponding to the transmit signals;
decomposing the transmit covariance matrix using Cholesky decomposition;
determining a full-duplex precoding (FDP) matrix in accordance with the decomposing of the transmit covariance matrix; and
generating, using the FDP matrix, the transmit signals and a plurality of SI cancellation signals corresponding to the transmit signals.
24. The method of claim 23 , further comprising adding the SI cancellation signals to a plurality of receive signals.