Scalable massive MIMO
View Patent ↗Techniques are disclosed relating to massive MIMO communications. In some embodiments, a base station is configured to dynamically adjust the number of processing elements used for MIMO signal estimation (e.g., the number of MIMO RX chains used for parallel processing). In some embodiments, the number of processing elements may be based on the number of antennas currently being used, the number of spatial streams, interconnect throughput thresholds, sampling rate, etc. In some embodiments, the base station includes configurable MIMO cores configured to dynamically switch between MIMO signal estimation techniques, e.g., on a per-symbol basis. In some embodiments, the base station includes configurable linear decoders configured to separately multiply input matrices and combine or refrain from combining the results based on the number of antennas and/or processing elements currently in use.
1. An apparatus, comprising:
a plurality of antennas;
a plurality of processing elements coupled to the plurality of antennas and configured to perform processing for received wireless communications via the plurality of antennas;
one or more interconnects configured to couple ones of the plurality of processing elements with other ones of the plurality of processing elements;
wherein the apparatus is configured to combine signals received by multiple antennas of the plurality of antennas;
wherein, for at least a subset of the processing elements, each processing element is configured to operate on a different portion of the combined signal in parallel, wherein each portion includes signals from multiple antennas of the plurality of antennas; and
wherein the apparatus is configured to operate in:
a first mode in which a first number of the plurality of processing elements process received wireless communications from a first number of the plurality of antennas; and
a second mode in which a second number of the plurality of processing elements that is smaller than the first number of processing elements process received wireless communications from a second number of the plurality of antennas that is smaller than the first number of antennas.
2. The apparatus of claim 1 , wherein the apparatus is configured to determine the second number of processing elements for use in the second mode based on at least:
the value of the second number of antennas;
a sampling rate for signals from the second number of antennas;
an amount of data per sample for signals from the second number of antennas; and
one or more threshold throughput values for the one or more interconnects.
3. The apparatus of claim 2 , wherein the apparatus is configured to determine the second number of processing elements based on a product of: the value of the second number of antennas, the sampling rate, and the amount of data per sample, divided by one of the one or more threshold throughput values.
4. The apparatus of claim 1 , wherein the apparatus is configured to determine the second number of processing elements for use in the second mode based on at least:
the value of the second number of antennas;
a number of separate spatial streams used for communication via the second number of antennas;
one or more threshold throughput values for the one or more interconnects;
an amount of data per sample for a generated matrix; and
an output rate for the separate spatial streams after processing based on the generated matrix.
5. The apparatus of claim 4 , wherein the apparatus is configured to determine the second number of processing elements based on a product of: the value of the second number of antennas, the number of separate spatial streams, the output rate, and the amount of data per sample, divided by one of the one or more threshold throughput values.
6. The apparatus of claim 1 ,
wherein a portion of the plurality of processing elements are linear decoders;
wherein, in the first mode, a particular linear decoder of the plurality of processing elements includes multiplication circuitry configured to process a single input matrix using; and
wherein, in the second mode, the multiplication circuitry is configured to process a plurality of input matrices.
7. The apparatus of claim 6 , further comprising:
adder tree circuitry configured to add multiple smaller matrix multiplication results to process the single input matrix in the first mode and bypass addition of smaller matrix multiplication results in the second mode.
8. The apparatus of claim 1 , wherein the apparatus is configured to process a first symbol of a received wireless signal according to a first signal estimation technique and a second, immediately subsequent symbol of the received wireless signal according to a second, different signal estimation technique.
9. The apparatus of claim 8 , wherein the first and second techniques are different signal estimation techniques selected from the group consisting of: minimum mean square error (MMSE), zero forcing (ZF), and maximal-ratio combining (MRC).
10. A method, comprising:
selecting a number of antennas from among a plurality of available antennas for use in wireless communications in a multiple-input multiple-output (MIMO) system;
determining a number of processing elements for processing received signals from the selected number of antennas, wherein for at least a subset of the processing elements, each processing element is configured to operate on a different portion of combined signal from the selected antennas in parallel, wherein each portion includes signals from multiple antennas; and
processing, by the determined number of processing elements, incoming wireless communications;
wherein the determining is performed based on:
the number of antennas selected; and
one or more threshold throughput values for one or more interconnects between the processing elements.
11. The method of claim 10 , wherein the determining is further performed based on:
a number of separate spatial streams used for communication via the number of antennas selected.
12. The method of claim 11 , wherein the determining is based on a product of: the value of the number of antennas selected, the number of separate spatial streams, an output rate, and an amount of data per sample, divided by one of the one or more threshold throughput values.
13. The method of claim 10 , wherein the determining is based on a product of: the value of the number of antennas selected, a sampling rate for signals from the selected number of antennas, and an amount of data per sample, divided by one of the one or more threshold throughput values.
14. The method of claim 10 , further comprising:
selecting a processing mode for one of the number of processing elements, wherein the selecting is performed from among a first mode in which multiplication circuitry in the processing element is configured to process a single input matrix and a second mode in which the multiplication circuitry is configured to process a plurality of input matrices.
15. The method of claim 14 , further comprising adding results of smaller matrix multiplications to generate a multiplication result for the single input matrix in the first mode.
16. The method of claim 10 , further comprising:
increasing the number of processing elements in response to a change in operating conditions.
17. The method of claim 10 , further comprising:
processing different symbols of an incoming wireless signal according to different signal estimation techniques.
18. The method of claim 17 , wherein the different signal estimation techniques include at least one of: minimum mean square error (MMSE), zero forcing (ZF), or maximal-ratio combining (MRC).
19. The method of claim 17 , further comprising:
bypassing orthogonalization circuitry used for one or more other signal estimation techniques when processing the incoming wireless communications according to a maximal-ratio combining signal estimate technique.
20. The method of claim 17 , further comprising:
asserting a pre-determined scaling factor when processing the incoming wireless communications according to a zero forcing signal estimate technique.