| CPC H04B 17/373 (2015.01) [H04L 25/0222 (2013.01); H04L 25/023 (2013.01)] | 20 Claims |
|
1. A wireless communication method, comprising:
estimating, based on channel quality information for a first communication channel during a first time interval, a predicted quality of a second communication channel during a second time interval that is a latency interval after the first time interval,
wherein the estimating includes 1) determining a prediction filter for estimating the predicted quality, and 2) estimating the predicted quality by applying the prediction filter to the channel quality information,
wherein the determining the prediction filter includes: 1) generating one or more pairs of channel quality information vectors representing channel quality measurements for the first communication channel and/or the second communication channel using a training step; 2) determining a maximum likelihood cross-covariance matrix for a matrix whose entries correspond to the one or more pairs of channel quality information vectors; and 3) determining the prediction filter from the maximum likelihood cross-covariance matrix,
wherein the one or more pairs of channel quality information vectors are represented as: Θ1=[Vt1|Vt2| . . . | VtK], Θ2=[Vt1+Δt|Vt2+Δt| . . . | Vtk+Δt], where Vx represents an N×1 vector of channel quality measurements at time x, and wherein Δt corresponds to the latency interval; and wherein the matrix is represented as:
 and wherein the maximum likelihood cross-covariance matrix is determined by maximizing a probability:
 where R is the maximum likelihood cross-covariance matrix represented as R=
 wherein the prediction filter C corresponds to: C=R21·R11−1, wherein R11, R12, R21 and R22 are Toeplitz matrices; and
using the predicted quality for processing transmissions on the second communication channel during the second time interval, wherein the second communication channel is in a reverse direction of the first communication channel.
|