Multiple-input and multiple-output (MIMO) detection in wireless communications

A method comprising:

• receiving a reference signal;
• determining a noise estimation of the reference signal;
• determining a distance metric and a log-likelihood ratio (LLR) of the determined noise estimation, wherein determining the distance metric and the LLR comprises:
• decomposing the noise estimation based on Cholesky decomposition techniques;
• and
• applying sphere decoding techniques;
• and
• determining a final LLR based on the distance metric and the LLR of the noise estimation,
• wherein determining the final LLR further comprises:
• combining the LLR of the noise estimation with an LLR of another noise estimation;
• or
• selecting the LLR of the noise estimation based on the distance metric of the noise estimation;
• wherein the reference signal is a first reference signal, the noise estimation is a first noise estimation, the distance metric is a first distance metric, and the LLR is a first LLR, the method further comprising:
• receiving a second reference signal;
• determining a second noise estimation of the second reference signal;
• determining a second distance metric and a second (LLR) of the second noise estimation;
• and
• determining a combined noise estimate, wherein determining the combined noise estimate comprises:
• combining the first LLR and the second LLR;
• or
• determining that the second distance metric is nearer to a threshold value than the first distance metric;
• and determining the combined noise estimate according to the second noise estimation.

A receiver comprising:

• a receiver operable to receive a plurality of reference signals;
• and
• a processor operable to:
• for a given reference signal of the plurality of reference signals:
• determining a noise estimation;
• and
• determine a distance metric and a log-likelihood ratio (LLR) of the noise estimation;
• and
• determine a final LLR based on of the distance metric and the LLR of each noise estimation;
• wherein the processor is further operable to:
• combine the LLRs of at least two noise estimations;
• or
• select the LLR of a particular noise estimation based on the distance metric of the particular noise estimation;
• wherein for a first reference signal of the plurality of reference signal, the noise estimation is a first noise estimation, the distance metric is a first distance metric, and the LLR is a first LLR;
• for a second reference signal of the plurality of reference signals, the noise estimation is a second noise estimation, the distance metric is a second distance metric, and the LLR is a second LLR;
• wherein the processor is further operable to:
• determine a combined noise estimate, wherein determining of the combined noise estimate comprises:
• combining the first LLR and the second LLR;
• or
• determining that the second distance metric is nearer to a threshold value than the first distance metric;
• and determining the combined noise estimate according to the second noise estimation.