Equalization in a wireless communication system
The present subject matter relates to a method comprising: receiving a radio frequency signal, the radio frequency signal comprising components, wherein each component is associated with a time unit and a frequency unit; selecting components of the radio frequency signal, the selected components comprising reference data; computing, for the selected components, equalization coefficients, referred to as initial equalization coefficients; deriving the equalization coefficients for non-reference data components of the radio frequency signal by interpolating, along at least one of time domain or frequency domain, the initial equalization coefficients.
1. An apparatus for a wireless communication system, the apparatus comprising at least one processor; and at least one memory storing instructions that, when executed by the at least one processor, cause the apparatus at least to:
receive a radio frequency signal, the radio frequency signal comprising components, wherein each component of the components is associated with a time unit and a frequency unit;
select components of the radio frequency signal, the selected components comprising reference data;
receive channel information in accordance with a first granularity of the time units and frequency units associated with the selected components,
resample the channel information for obtaining channel information in accordance with a second granularity that is coarser than the first granularity;
compute, for the selected components, equalization coefficients, referred to as initial equalization coefficients, wherein the computation of the initial equalization coefficients is performed using the resampled channel information; and
derive the equalization coefficients for non-reference data components of the radio frequency signal by interpolating the initial equalization coefficients along at least one of time domain or frequency domain.
2. The apparatus of claim 1 , wherein the initial equalization coefficients are computed per distinct pair of time unit and frequency unit of the selected components using channel information for the pair of time unit and frequency unit, wherein the channel information comprises a channel estimate.
3. The apparatus of claim 1 , wherein the instructions, when executed by the at least one processor, further cause the apparatus to perform interpolation comprising at least:
selecting at least one domain of: time domain or frequency domain; and
performing the interpolation along the selected domain.
4. The apparatus of claim 1 , wherein the instructions, when executed by the at least one processor, further cause the apparatus to perform interpolation along the time domain and the frequency domain by interpolating along the time domain before the interpolation along the frequency domain.
5. The apparatus of claim 1 , wherein the instructions, when executed by the at least one processor, further cause the apparatus to perform the computation of the initial equalization coefficients and the interpolation concurrently.
6. The apparatus of claim 1 , wherein the instructions, when executed by the at least one processor, further cause the apparatus to perform the computation of the initial equalization coefficients before the interpolation starts.
7. The apparatus of claim 1 , the apparatus comprising: a first hardware accelerator that is configured to compute the initial equalization coefficients and a second hardware accelerator that is configured to:
receive the initial equalization coefficients through a hardware interface between the first and second hardware accelerators, and
perform the interpolation.
8. The apparatus of claim 1 , the selected components represent Demodulation Reference Signal (DM-RS) occasions respectively.
9. The apparatus of claim 1 , wherein the instructions, when executed by the at least one processor, further cause the apparatus to suppress interference from the received non-reference data components by weighting the data in the non-reference data components using the derived equalization coefficients.
10. The apparatus of claim 1 , the apparatus comprising multiple receiving antennas, wherein each component of the radio frequency signal is a vector of symbols received by the receiving antennas respectively.
11. The apparatus of claim 1 , the apparatus comprising multiple receiving antennas, wherein the equalization coefficients are interference rejection combining (IRC) weights, wherein the initial IRC weights for a specific time-frequency block associated with the selected components are provided as a matrix W IRC,d (Original)=(H H R −1 H) −1 H H R −1 , where H is a channel estimate N RX ×N L matrix and R is an interference covariance N RX ×N RX matrix for the specific time-frequency block, where N RX is the number of receiving antennas and N L is the number of transmitted layers.
12. A method comprising:
receiving a radio frequency signal, the radio frequency signal comprising components, wherein each component is associated with a time unit and a frequency unit;
selecting components of the radio frequency signal, the selected components comprising reference data;
receiving channel information in accordance with a first granularity of the time units and frequency units associated with the selected components,
resampling the channel information for obtaining channel information in accordance with a second granularity that is coarser than the first granularity;
computing, for the selected components, equalization coefficients, referred to as initial equalization coefficients, wherein the computation of the initial equalization coefficients is performed using the resampled channel information; and
deriving the equalization coefficients for non-reference data components of the radio frequency signal by interpolating, along at least one of time domain or frequency domain, the initial equalization coefficients.
13. The method of claim 12 , wherein the radio frequency signal comprises a set of components that comprise reference data, the method further comprising:
receiving channel estimates for the set of components, wherein the selected components is a subset of the set of components, and wherein the computation of the initial equalization coefficients is performed using the received channel estimates which are associated with the selected components.
14. The method of claim 12 , the interpolation comprising:
selecting at least one domain of: time domain or frequency domain; and
performing the interpolation along the selected domain.
15. The method of claim 12 , wherein the interpolation is performed along the time domain and the frequency domain by interpolating along the time domain before the interpolation along the frequency domain starts.
16. The method of claim 12 , wherein the computation of the initial equalization coefficients and the interpolation are performed concurrently.
17. The method of claim 12 , wherein the computation of the initial equalization coefficients is performed before the interpolation starts.
18. The method of claim 12 , further comprising suppressing interference from the received non-reference data components by weighting the data in the non-reference data components using the derived equalization coefficients.