Symbol de-mapping methods in multiple-input multiple-output systems
View Patent ↗In a multiple-input multiple-output (MIMO) system, multiple receive antennas produce a received signal vector, Y, which includes an element for each of the receive antennas. In an embodiment of a de-mapping method performed within a MIMO receiver, a quadrature phase shift keying (QPSK) search is performed within a search space that includes the full constellation of symbol points. Based on the results of the QPSK search, the search space is reduced to fewer than all of the quadrants, and the received signal vector data is scaled and transformed to the reduced search space. A lower-level QPSK search is performed, and the process is repeated until the modulation order is reduced to a QPSK constellation. Hard or soft decisions corresponding to the search results may then be passed to a decoder.
1. A method for demapping symbols comprising:
performing a first elemental search over a highest-order elementary modulation on a received signal vector that includes multiple elements, wherein the first elemental search is performed within a first search space and produces an identified vector of elementary modulation symbols;
transforming the received signal vector to a new origin that corresponds to the identified vector, resulting in a transformed, received signal vector;
performing a subsequent elemental search on the transformed, received signal vector, wherein the subsequent elemental search is performed within a reduced search space defined by the identified vector, and wherein the subsequent elemental search produces a next identified vector of elementary modulation symbols; and
producing demapped bit values that correspond to a next identified vector of elementary modulation symbols of a lowest-level search.
2. The method of claim 1 , wherein the received signal vector is modulated using quadrature amplitude modulation, and quadrature phase shift keying is an elementary modulation.
3. The method of claim 1 , wherein the received signal vector is modulated using pulse amplitude modulation, and binary phase shift keying is an elementary modulation.
4. A method for demapping symbols comprising:
performing a first quadrature phase shift keying (QPSK) search on a received signal vector that includes multiple elements, wherein the first QPSK search is performed within a first search space and produces an identified QPSK vector;
transforming the received signal vector to a new origin that corresponds to the identified QPSK vector, resulting in a transformed, received signal vector;
performing a subsequent QPSK search on the transformed, received signal vector, wherein the subsequent QPSK search is performed within a reduced search space defined by the identified QPSK vector, and wherein the subsequent QPSK search produces a next identified QPSK vector; and
producing search results that include de-mapped bit values corresponding to a QPSK vector identified as a result of a lowest-level search.
5. The method of claim 4 , further comprising:
producing the received signal vector, wherein each of the multiple elements corresponds to a signal received by one of multiple receive antennas of a multiple-input multiple-output receive antenna array.
6. The method of claim 4 , further comprising:
scaling the transformed, received signal vector, prior to performing the subsequent QPSK search.
7. The method of claim 4 , further comprising:
until the subsequent QPSK search results in a next identified QPSK vector that corresponds to a constellation point,
repeating transforming the transformed, received signal vector; and
repeating performing the subsequent QPSK search.
8. The method of claim 4 , further comprising:
incorporating a tree-searching algorithm into either or both the first QPSK search and the subsequent QPSK search to produce multiple identified QPSK vectors that are used to define the reduced search space.
9. The method of claim 8 , wherein incorporating the tree-searching algorithm comprises:
incorporating a tree search into a QPSK search in which a number of the identified QPSK vectors with smallest Euclidian distance values are included in the reduced search space for a subsequent iteration.
10. The method of claim 8 , wherein incorporating the tree-searching algorithm comprises:
incorporating a tree search into a QPSK search in which a number of the identified QPSK vectors with Euclidian distance values that fall with a threshold of a best of the identified QPSK vectors are included in the reduced search space for a subsequent iteration.
11. The method of claim 4 , further comprising:
producing search results that include at least one soft decision for use by a decoder.
12. The method of claim 11 , wherein producing the search results comprises:
producing the at least one soft decision as a set of log-likelihood ratios or approximations of log-likelihood ratios.
13. A method comprising:
performing a first quadrature phase shift keying (QPSK) search on a received signal vector, Y, which includes multiple elements, wherein the first QPSK search is performed within a first search space and produces an identified QPSK vector; and
until a reduced search space corresponds to a QPSK constellation,
canceling higher-order interference based on the identified QPSK vector and scaling the multiple elements within the received signal vector according to
Y
~
k
=
1
2
(
Y
~
k
-
1
-
x
^
k
-
1
)
,
where {tilde over (Y)} k is a scaled version of the received signal vector at search level k, and {circumflex over (X)} k is a QPSK vector at search level k, and
performing a level-k QPSK search according to
x
^
k
=
arg
min
QPSK
vectors
x
Y
~
k
-
Hx
2
,
where is a channel transfer matrix, and x is a transmit signal vector.
14. The method of claim 13 , further comprising:
incorporating a tree-searching algorithm into either or both the first QPSK search and the level-k QPSK search to produce multiple identified QPSK vectors that are used to define the reduced search space.
15. The method of claim 13 , further comprising:
producing search results that include at least one soft decision for use by a decoder.
16. The method of claim 15 , wherein producing the search results comprises:
producing the at least one soft decision as a set of log-likelihood ratios or approximations of log-likelihood ratios.
17. The method of claim 13 , further comprising:
producing search results that include de-mapped bit values corresponding to a QPSK vector identified as a result of a lowest-level search.
18. A computer-readable medium having computer program instructions stored thereon which, when executed within a multiple-input multiple-output device, results in:
performing a first quadrature phase shift keying (QPSK) search on a received signal vector that includes multiple elements, wherein the first QPSK search is performed within a first search space and produces an identified QPSK vector;
transforming the received signal vector to a new origin that corresponds to the identified QPSK vector, resulting in a transformed, received signal vector; and
performing a subsequent QPSK search on the transformed, received signal vector, wherein the subsequent QPSK search is performed within a reduced search space defined by the identified QPSK vector, and wherein the subsequent QPSK search produces a next identified QPSK vector; and
producing search results that include de-mapped bit values corresponding to a QPSK vector identified as a result of a lowest-level search.
19. The computer-readable medium of claim 18 , wherein execution of the instructions further results in:
incorporating a tree-searching algorithm into either or both the first QPSK search and the subsequent QPSK search to produce multiple identified QPSK vectors that are used to define the reduced search space.
20. The computer-readable medium of claim 18 , wherein execution of the instructions further results in:
producing search results that include at least one soft decision for use by a decoder.
21. An apparatus comprising:
multiple receive antennas operable to receive multiple received signals; and
a symbol-processing element, operable to
perform a first quadrature phase shift keying (QPSK) search on a received signal vector that includes multiple elements corresponding to the multiple received signals, wherein the first QPSK search is performed within a first search space and produces an identified QPSK vector;
transform the received signal vector to a new origin that corresponds to the identified QPSK vector, resulting in a transformed, received signal vector; and
perform a subsequent QPSK search on the transformed, received signal vector, wherein the subsequent QPSK search is performed within a reduced search space defined by the identified QPSK vector, and wherein the subsequent QPSK search produces a next identified QPSK vector.
22. The apparatus of claim 21 , wherein the symbol-processing element is further operable to:
incorporate a tree-searching algorithm into either or both the first QPSK search and the subsequent QPSK search to produce multiple identified QPSK vectors that are used to define the reduced search space.
23. The apparatus of claim 21 , wherein the symbol-processing element is further operable to:
produce search results that include at least one soft decision for use by a decoder.
24. The apparatus of claim 21 , wherein the symbol-processing element is further operable to:
produce search results that include de-mapped bit values corresponding to a QPSK vector identified as a result of a lowest-level search.
25. A multiple-input multiple-output communication device, comprising:
multiple receive antennas operable to receive multiple received signals; and
a symbol-processing element, operable to
perform a first quadrature phase shift keying (QPSK) search on a received signal vector that includes multiple elements corresponding to the multiple received signals, wherein the first QPSK search is performed within a first search space and produces an identified QPSK vector;
transform the received signal vector to a new origin that corresponds to the identified QPSK vector, resulting in a transformed, received signal vector; and
perform a subsequent QPSK search on the transformed, received signal vector, wherein the subsequent QPSK search is performed within a reduced search space defined by the identified QPSK vector, and wherein the subsequent QPSK search produces a next identified QPSK vector.
26. The multiple-input multiple-output communication device of claim 25 , wherein the symbol-processing element is further operable to:
incorporate a tree-searching algorithm into either or both the first QPSK search and the subsequent QPSK search to produce multiple identified QPSK vectors that are used to define the reduced search space.
27. The multiple-input multiple-output communication device of claim 25 , wherein the symbol-processing element is further operable to:
produce search results that include at least one soft decision for use by a decoder.
28. The multiple-input multiple-output communication device of claim 25 , wherein the symbol-processing element is further operable to:
produce search results that include de-mapped bit values corresponding to a QPSK vector identified as a result of a lowest-level search.