Estimating signals in sub-sampled systems

A method of operating a base station in a radio frequency communication network, comprising:

• receiving pilot signals on multiple subchannels of a resource block;
• computing channel estimation matrices for subchannels of a resource block on which pilot signals are received;
• determining a relationship between the computed channel estimation matrices for the subchannels on which the pilot signals are received and a subchannel number;
• selecting an anchor subchannel for the resource block where a desired error in the accuracy of a precoding matrix for the subchannel is minimized;
• computing weights for each subchannel of the resource block using the computed relationship and a distance that the subchannel is away from the selected anchor subchannel;
• computing channel estimation matrices for each subchannel in the resource block by weighting the channel matrices for the subchannels on which pilot signals are received with the weights that are computed for each of the subchannels;
• and
• computing precoding matrices for each subchannel in the resource block from the channel estimation matrices computed for the corresponding subchannel.

A base station configured for use in a radio frequency communication network, comprising:

• an antenna array;
• a memory for storing instructions that are executable by a processor;
• a processor configured to execute the instructions to determine precoding matrices for subchannels of a resource block by:
• receiving pilot signals on a number of subchannels of a resource block;
• computing channel estimation matrices for subchannels of a resource block on which pilot signals are received;
• determining a relationship between the computed channel estimation matrices for the subchannels on which the pilot signals are received and a subchannel number;
• selecting an anchor subchannel for the resource block where a desired accuracy of a precoding matrix for the subchannel is highest for the resource block;
• computing weights for each subchannel of the resource block using the linear regressor and a distance that the subchannel is away from the selected anchor subchannel;
• computing channel estimation matrices for each subchannel in the resource block by weighting the channel matrices for the subchannels on which pilot signals are received with the weights that are computed for the corresponding subchannel;
• and
• computing precoding matrices for each subchannel in the resource block from the channel estimation matrices determined for the corresponding subchannel.

A non-transitory computer-readable storage medium storing instructions configured to cause a computer system to perform a method, the instructions comprising:

• instructions for receiving a plurality of multipath channel estimates from a plurality of user devices, the plurality of multipath channel estimates within a plurality of frequency channel resource blocks, the frequency channel resource blocks between boundaries of a plurality of considered frequencies;
• averaging the multipath channel estimates falling within each of the instructions for resource blocks to compute average channel estimates for each of the blocks;
• determining frequency offsets for each of the average estimates in the instructions for resource blocks;
• and
• generating a plurality of interpolated channel matrices using the average instructions for estimates offset by the frequency offsets.

A base station, comprising:

• a plurality of antennas;
• at least one processor;
• at least one memory comprising instructions configured to cause the at least one processor to perform a method comprising:
• receiving a plurality of multipath channel estimates from a plurality of user devices, the plurality of multipath channel estimates within a plurality of frequency channel resource blocks, the frequency channel resource blocks between boundaries of a plurality of considered frequencies;
• averaging the multipath channel estimates falling within each of the resource blocks to compute average channel estimates for each of the blocks;
• computing frequency offsets for each of the average estimates in the resource blocks;
• and
• generating a plurality of interpolated channel matrices using the average estimates offset by the frequency offsets.