Method of decoding a spatially multiplexed signal and its corresponding receiver

1. A method of decoding a received signal, the method comprising:

receiving a signal with a receiver via at least one of first and second receive antennas, the signal being a function at least of a noise term matrix, of a channel matrix, and of a first and second symbols and belonging to at least a signal constellation, said channel matrix comprising a first and a second columns, said first column comprising components representing a propagation channel response between a first transmit antenna and at least the first and a second receive antennas, said second column comprising components representing the propagation channel response between a second transmit antenna and at least the first and second receive antennas,

selecting with the receiver a set of possible values of the first symbol belonging to the signal constellation;

for each selected value of the first symbol performing the steps of:

estimating with the receiver the value of the second symbol using the value of the first symbol, to generate an estimated value of the second symbol; and

calculating with the receiver an Euclidean distance between the received signal and a noiseless signal defined as the product of the matrix H by the vector formed of the first symbol with said selected value and by the second symbol with said estimated value;

selecting with the receiver the minimal Euclidean distance among the Euclidean distances respectively calculated for the different selected values belonging to the set or a subset of said set; and

selecting with the receiver decoded first and second symbols corresponding to the selected minimum Euclidean distance.

2. Method according to claim 1 , further comprising, before performing the step of estimating the value of the second symbol, at least the steps of:

calculating with the receiver a first quantity equal to the complex conjugate of the second column multiplied by the received signal;

calculating with the receiver a second quantity equal to the complex conjugate of the second column multiplied by the first column; and

calculating with the receiver a third quantity equal to the complex conjugate of the second column multiplied by said second column.

3. Method according to claim 1 , wherein said step of estimating the value of the second symbol comprises at least the steps of:

generating with the receiver an intermediate signal representative of the received signal in which the contribution of the first symbol is subtracted; and

taking with the receiver a decision on the value of the second symbol according to the intermediate signal.

4. Method according to claim 3 , wherein said decision on the value of the second symbol is taken by sending the intermediate signal to a threshold detector Q, which generates the estimated value of the second symbol according to the intermediate signal.

5. Method according to claim 3 , wherein said decision on the value of the second symbol is taken by using a look-up table to find the estimated value of the second symbol.

6. Method according to claim 1 , wherein said set includes all the values of the first symbol belonging to the signal constellation.

7. Method according to claim 1 , wherein said set is selected inside a sphere belonging to the signal constellation, centered on the received signal and the radius of which is equal to a predefined value.

8. Method according to claim 7 , wherein said sphere is such that |s 1 −ε l | 2 ≦ρ 2 /|l 22 | 2 ,l 22 being the component of the last line and column of an upper triangular matrix obtained by performing at least a QR decomposition of the channel matrix, and ε 1 being a component of a vector equal to the inverse of the channel matrix multiplied by the received signal.

9. Method according to claim 1 , wherein the first symbol is defined by a plurality of symbolic bits, each symbolic bit being designated by its rank and being equal to 0 or 1,

and in that the step of selecting the minimal Euclidean distance is performed for a plurality of subsets, each subset including all the possible values of the first symbol in which the symbolic bit of a predetermined rank has a predetermined value.

10. Method according to claim 1 , further comprising the steps of:

calculating with the receiver a first soft symbol

s

1

soft

=

c

1

H

⁢

y

-

c

1

H

⁢

c

2

⁢

s

2

ML

c

1

H

⁢

c

1

,

c

1

H

being the complex conjugate of the first column, and s 2 ML being the value of the second symbol corresponding to said estimated value; and

calculating with the receiver a second soft symbol

s

2

soft

=

c

2

H

⁢

y

-

c

2

H

⁢

c

1

⁢

s

1

ML

c

2

H

⁢

c

2

,

c

2

H

being the complex conjugate of the second column, and s 1 ML being the value of the first symbol corresponding to said estimated value.

11. A receiver implementing at least said method according to claim 1 .

12. Method according to claim 2 , wherein said step of estimating the value of the second symbol comprises at least the steps of:

generating with the receiver an intermediate signal representative of the received signal in which the contribution of the first symbol is subtracted; and

taking with the receiver a decision on the value of the second symbol according to the intermediate signal.

13. Method according to claim 2 , wherein said set includes all the values of the first symbol belonging to the signal constellation.

14. Method according to claim 3 , wherein said set includes all the values of the first symbol belonging to the signal constellation.

15. Method according to claim 4 , wherein said set includes all the values of the first symbol belonging to the signal constellation.

16. Method according to claim 5 , wherein said set includes all the values of the first symbol belonging to the signal constellation.

17. Method according to claim 2 , wherein said set is selected inside a sphere belonging to the signal constellation, centered on the received signal and the radius of which is equal to a predefined value.

18. Method according to claim 3 , wherein said set is selected inside a sphere belonging to the signal constellation, centered on the received signal and the radius of which is equal to a predefined value.

19. Method according to claim 4 , wherein said set is selected inside a sphere belonging to the signal constellation, centered on the received signal and the radius of which is equal to a predefined value.

20. Method according to claim 5 , wherein said set is selected inside a sphere belonging to the signal constellation, centered on the received signal and the radius of which is equal to a predefined value.