COMMUNICATION DEVICE AND METHOD FOR EFFICIENTLY RECEIVING MIMO SIGNALS

Abstract

Communication device adapted for receiving a MIMO signal is provided. The device comprises a first detector adapted to perform a first symbol detection on the MIMO signal using a first detection method, a detection error determination unit adapted to determine a first detection error of the first symbol detection, a detection error judging unit adapted to determine if the first detection error is above or below a detection threshold, and a second detector, adapted to perform a second symbol detection on the MIMO signal using a second detection method, if the detection error judging unit has determined that the first detection error is above the detection threshold. The communication device is adapted to use results of the symbol detection as final symbol detection results, if the detection error judging unit has determined that the first detection error is below the detection threshold.

Claims

1. A communication device adapted for receiving a multiple-input and multiple-output (MIMO) signal, comprising: a first detector adapted to perform a first symbol detection on the MIMO signal using a first detection method, a detection error determination component adapted to determine a first detection error of the first symbol detection, a detection error judging component adapted to determine if the first detection error is below or above a detection threshold, and a second detector, adapted to perform a second symbol detection on the MIMO signal using a second detection method, when the detection error judging component has determined that the first detection error is above the detection threshold, wherein the communication device is adapted to use results of the first symbol detection as final symbol detection results, when the detection error judging component has determined that the first detection error is below the detection threshold.

2. The communication device according to claim 1, wherein the detection error determination component is adapted to determine a second detection error of the second symbol detection, wherein the detection error judging component is adapted to determine if the second detection error is below or above the detection threshold, and wherein the communication device is adapted to use results of the second symbol detection as final symbol detection results, when the detection error judging component has determined that the second detection error is below the detection threshold.

3. The communication device according to claim 2, wherein the second symbol detection is an iterative symbol detection method, wherein the detection error determination component is adapted to determine the second detection error after each iteration, wherein the detection error judging component is adapted to determine if the second detection error is below or above the detection threshold after each iteration, wherein the second detector is adapted to perform a further iteration, when the detection error judging component has determined that the second detection error is above the detection threshold, and wherein the communication device is adapted to use results of the second symbol detection as final symbol detection results, when the detection error judging component has determined that the second detection error is below the detection threshold.

4. The communication device according to claim 1, wherein the first detection method comprises a lower detection accuracy and a lower computational complexity than the second detection method.

5. The communication device according to claim 1, wherein the first detection method is a minimum mean square error (MMSE) detection method or a zero-forcing (ZF)method, and wherein the second detection method is a successive interference cancellation (SIC) method or a sphere decoder (SD) method or a maximum likelihood (ML) method.

6. The communication device according to claim 2, wherein the communication device comprises a third detector adapted to perform a third symbol detection on the MIMO signal using a third detection method, when the detection error judging component has determined that the second detection error is above the detection threshold, and wherein the communication device is adapted to use results of the third symbol detection independent of a third detection error of the third symbol detection.

7. The communication device according to claim 6, wherein the first detection method comprises a lower detection accuracy and a lower computational complexity than the second detection method, and wherein the second detection method comprises a lower detection accuracy and a lower computational complexity than the third detection method.

8. The communication device according to claim 7, wherein the first detection method is a minimum mean square error (MMSE) detection method or a zero-forcing (ZF) method, wherein the second detection method is asuccessive interference cancellation (SIC) method or a sphere decoder (SD) method, and wherein the third detection method is a maximum likelihood (ML) method.

9. The communication device according to claim 1, wherein the communication device comprises a detection threshold determining component adapted to determine the detection threshold adaptively.

10. The communication device according to claim 1, wherein the detection threshold determining component is adapted to determine the detection threshold adaptively dependent upon a signal-to-noise-ratio of the MIMO signal, and/or a battery level of a battery of the communication device, and/or a temperature of the communication device, and/or an availability of computational resources, and/or an accuracy of available channel state information of the MIMO signal.

11. The communication device according to claim 1, wherein the detection threshold determining component is adapted to determine the detection threshold as a higher value for a lower signal-to-noise-ratio, a lower battery level of the battery of the communication device, a higher temperature of the communication device, a lower availability of computational resources, and a lower accuracy of available channel state information of the MIMO signal, and wherein the detection threshold determining component is adapted to determine the detection threshold as a lower value for a higher signal-to-noise-ratio, a higher battery level of the battery of the communication device, a lower temperature of the communication device, a higher availability of computational resources, and a higher accuracy of the available channel state information of the MIMO signal.

12. The communication device according to claim 9, wherein the detection threshold determining component is adapted to determine the detection threshold through the following:
t=N.sub.Rx*.sup.2+ wherein t is the detection threshold, wherein N.sub.Rx is a number of receiver antennas receiving the MIMO signal, wherein .sup.2 is the variance of the noise of the transmission channel of the MIMO signal, wherein is a detection tolerance parameter adjusting the detection threshold.

13. The communication device according to claim 1, wherein the detection error determination component is adapted to determine the detection errors through the following:
d=H*xy.sub.2.sup.2 wherein d is a scalar representing a sufficient statistic for the received vector y. wherein H is an effective channel matrix of a transmission channel of the MIMO signal, where x is a vector of the detection results.

14. A reception method for receiving a multiple-input and multiple-output (MIMO) signal, comprising: performing a first symbol detection on the MIMO signal using a first detection method, determining a first detection error of the first symbol detection, determining if the first detection error is below or above a detection threshold, performing a second symbol detection on the MIMO signal using a second detection method, when it has been determined that the first detection error is above the detection threshold, using results of the first symbol detection as final symbol detection results, when it has been determined that the first detection error is below the detection threshold.

15. A non-transitory computer readable medium, comprising processor-executable instructions stored thereon, which when executed by a hardware processor cause the processor to implement operations including: performing a first symbol detection on the MIMO signal using a first detection method, determining a first detection error of the first symbol detection, determining if the first detection error is below or above a detection threshold, performing a second symbol detection on the MIMO signal using a second detection method, when it has been determined that the first detection error is above the detection threshold, using results of the first symbol detection as final symbol detection results, when it has been determined that the first detection error is below the detection threshold.

16. The method according to claim 14, further comprising: determining a second detection error of the second symbol detection, determining if the second detection error is below or above the detection threshold, and using results of the second symbol detection as final symbol detection results, when it has determined that the second detection error is below the detection threshold.

17. The method according to claim 16, wherein the second symbol detection is an iterative symbol detection method, and the method further comprises: determining the second detection error after each iteration, determining if the second detection error is below or above the detection threshold after each iteration, performing a further iteration, when it has determined that the second detection error is above the detection threshold, and using results of the second symbol detection as final symbol detection results, when it has determined that the second detection error is below the detection threshold.

18. The method according to claim 14, wherein the first detection method comprises a lower detection accuracy and a lower computational complexity than the second detection method.

19. The method according to claim 14, wherein the first detection method is a minimum mean square error (MMSE) detection method or a zero-forcing (ZF) method, and wherein the second detection method is a successive interference cancellation (SIC) method or a sphere decoder (SD) method or a maximum likelihood (ML) method.

20. The computer readable medium according to claim 15, wherein the operations further include: determining a second detection error of the second symbol detection, determining if the second detection error is below or above the detection threshold, and using results of the second symbol detection as final symbol detection results, when it has determined that the second detection error is below the detection threshold.

Description

BRIEF DESCRIPTION OF DRAWINGS

[0045] The present application is in the following explained in detail in relation to embodiments of the application in reference to the enclosed drawings, in which:

[0046] FIG. 1 shows an overview of a MIMO communication system;

[0047] FIG. 2 shows a simplified block diagram of the proposed approach;

[0048] FIG. 3 shows an embodiment of the proposed approach in pseudo-code;

[0049] FIG. 4 shows a further block diagram of an embodiment of the proposed approach;

[0050] FIG. 5 shows an embodiment of the first aspect of the application in a block diagram;

[0051] FIG. 6 shows an embodiment of the second aspect of the application in a flow diagram;

[0052] FIG. 7 shows achievable results by use of the application;

[0053] FIG. 8 shows achievable results by use of the application.

DESCRIPTION OF EMBODIMENTS

[0054] The general setup of a MIMO communication system and the underlying problem has been discussed along FIG. 1. With use of FIGS. 2-4, the application is described in a general manner. Along FIG. 5, an embodiment of the first aspect of the application and its function is described in detail. Along FIG. 6, an embodiment of the second aspect of the application and its function is described in detail. Along FIG. 7 and FIG. 8, the benefits of the application are shown. Similar entities and reference numbers in different figures have been partially omitted.

[0055] The basic idea of the proposed approach is shown in FIG. 2. A low complexity first detection method also referred to as educated guess is employed to a received signal Y in a first step 20. In a second step 21, it is determined if the achieved accuracy is sufficiently good. If so, the first detection result is output in a third step 22. In case it is not sufficiently good, in a fourth step 23, the first result is at least partially refined then tested again. This refinement typically uses a second more complex, but also more accurate detection method.

[0056] Here, the value E is a tolerance threshold that tunes between the correctness of the estimation and the complexity, that is, =0 the best available estimation is performed, while for = only the low complexity guess is performed.

[0057] In one embodiment, depicted in FIG. 4, it is proposed to use ordered MMSE-SIC as a non-linear algorithm and MMSE as a low complexity educated guess. Therefore, MMSE is the first detection method, while ordered MMSE-SIC is the second detection method. This choice is motivated by the following reasons: [0058] Linear MMSE is one of the best known linear detection available and its complexity is of the same order of the one of less performing algorithm such a ZF; [0059] Linear MMSE is a step of MMSE-SIC, hence the low complexity guess comes for free in computational terms; [0060] SIC takes care of high Eb/N0 terms first, and linear MMSE is optimal at low Eb/N0, thus there is a reasonable hope that many useless operation will be avoided; [0061] SIC complexity does not depend on the constellation size.

[0062] In FIG. 3, a pseudo-code implementing this embodiment is shown.

In a first step 30, a linear MMSE equalized version of the transmit vector ({tilde over (x)}) is computed. Notice that this step is necessary also in standard MMSE-SIC, hence it does not increase the complexity of this algorithm. In a second step 31, this vector is passed through a step function (hard decoding) to obtain an educated guess X. The educated guess transmission is simulated (Hx) and the distance between this vector and the actually received vector is determined d=Hxy.sub.2.sup.2 in a third step 32.

[0063] This distance d is compared to a detection threshold dN.sub.Rx.sup.2+ to decide if the guess is sufficiently accurate or not. This test takes the name of likelihood test. Notice that, if the likelihood test is performed with =0 then it theoretically guarantees that the educated guess is the best possible guess (i.e., if dN.sub.Rx.sup.2 then the educated guess corresponds to the ML equalization). Henceforth, values of >0 will decrease the performance in terms of BER, but they will decrease the amount of complexity of the algorithm.

[0064] If the test is negative, it means that, in the alphabet, there exists one element that has a higher probability of being the better one than the educated guess. In this case, we proceed with one step of SIC in a further step 38. This means that among the non-decoded element of the vector x the one with the highest SINR is selected. The interference created by the already decoded element is now subtracted and a new element is therefore decoded. Henceforth, this newly obtained vector is treated as the new educated guess. In a further step 37, it is determined, if all elements of the vector have already been decoded. If this is the case, in a step 36, a hard decoding is performed and the result is output in a step 35.

[0065] Since SIC decodes the elements with the highest SINR first there exists a non-negligible probability that the likelihood test d(N.sub.Rx.sup.2+) is passed after a few rounds of SIC. This means that the same performance of SIC can be obtained with much less complexity.

[0066] A simple SIC is just one possible embodiment. A multi-branch SIC would further increase the performance, at the cost, however, of complexity.

[0067] In FIG. 5, an embodiment of the inventive communication device 10 is shown. Here, only the components relevant for the application, especially all aspects of the symbol detection, which correspond to the MIMO equalizer 5 of FIG. 1, are shown. For reasons of conciseness, other components of the communication device 10 are omitted here.

[0068] The communication device 10 comprises a first detector 11, and error determining unit 12, an error judging unit 13, a second detector 14, a control unit 15 and a detection threshold determining unit 16. All units 11, 12, 13, 14 and 16 are connected to the control unit 15. Moreover, the first detector 11 and the second detector 14 are connected to the error determining unit 12. The error determining unit 12 is furthermore connected to the error judging unit 13.

[0069] After receiving a MIMO signal, the first detector 11 performs a first symbol detection on the MIMO signal using a first detection method. The detection error determining unit 12 determines a first detection error of the first symbol detection, as shown above. The error judging unit then determines if the first detection error is below or above a detection threshold, as also explained above. If the first detection error is below the detection threshold, the results of the first detector 11 are used as final detection results. In case the first detection error is above the detection threshold, the second detector 14 performs a second symbol detection on the MIMO signal using a second detection method. The results of this second detection can be directly used as the output symbols.

[0070] Alternatively, the second symbol detection method can be an iterative method, as shown above. In this case, the iterative method is performed until the error judging unit 13 determines that the second detection error is below the detection threshold. In this case, after each iteration, the error determining unit 12 and the error judging unit 13 perform their functions.

[0071] Moreover, in an alternative embodiment, a third detector can be present. The third detector then is connected to the error determining unit 12 and to the control unit 15. The third detector is then configured to perform a third symbol detection, if the second detection error is above the detection threshold.

[0072] The first detection method used by the first detector 11 has a lower computational complexity and accuracy than the second detection method employed by the second detector 14. In case of a third detector been present, the computational complexity and accuracy of the third detection method employed by the third detector is higher than the accuracy and computational complexity of the first and second detection methods used by the first and second detectors 11, 14.

[0073] In case of using a first and second detector 11, 14 as shown in FIG. 5, the first detection method is a minimum means square error detection method or a zero forcing method. The second detection method is a successive interference cancellation method or a sphere decoder or a maximum likelihood method. In case of using additionally a third detector, the second detection method is a successive interference cancellation method or a sphere decoder method and the third detection method is a maximum likelihood method.

[0074] Advantageously, the detection threshold is determined adaptively by the detection threshold determining unit 16. The detection threshold determination unit 16 though is an optional component. The detection threshold determining unit 16 determines the detection threshold adaptively dependent upon a signal-to-noise-ratio of the MIMO signal, and/or a battery level of a battery of the communication device and/or a temperature of the communication device, and/or an availability of computational resources, and/or an accuracy of available channel state information of the MIMO signal.

[0075] Especially, the detection threshold is determined as a higher value for a lower signal-to-noise-ratio, a lower battery level of the battery of the communication device, a higher temperature of the communication device, a lower availability of the computational resources, and a lower accuracy of the available channel state information of the MIMO signal. The detection threshold determining unit 16 determines the detection threshold as a lower value for a higher signal-to-noise-ratio, a higher battery level of the battery of the communication device, a lower temperature of the communication device, a higher availability of computational resources and a higher accuracy of the available channel state information of the MIMO signal.

[0076] In FIG. 6, a flow diagram of an embodiment of the second aspect of the application is shown. In a first step 100, a MIMO signal is received. In a second step 101 a first symbol detection is performed on the MIMO signal. In a third step 102, a first detection error of the first symbol detection is determined. In a fourth step 103, it is determined if the first detection error is below or above the detection threshold. In case it is below, in a fifth step 104 the results of the first symbol detection are used as final symbol detection results. In case the detection error is above the detection threshold, in a sixth step 105, a second symbol detection is performed on the MIMO signal. In a seventh step 106, the results of the symbol detection are then used as final symbol detection results.

[0077] Alternatively, the second symbol detection can be performed as an iterative method. In this case, after each step of the iterative method, again the detection error is determined and compared to the detection threshold. In case the detection error is below the threshold, the results are output as final detection results. In case they are above the threshold, a further iteration is performed.

[0078] Alternatively, a third symbol detection can be performed after the second symbol detection, and after it has been determined that the second detection error is above the detection threshold. The results of the third symbol detection are then used as final detection results no matter of the achieved detection error.

[0079] In the following, some advantages of the application are described.

[0080] If =0, then the proposed approach has an equalization capability non-inferior to the one of MMSE-SIC. Its complexity is also strictly inferior to the one of MMSE-SIC and it is independent from the constellation size.

[0081] In order to showcase the performance of the algorithm, a first simulation assessing the equalization ability is performed against MMSE-SIC. The results of this simulation are depicted in FIG. 7. In accordance with the theory, the performance of the proposed algorithm is not inferior to the one of MMSE-SIC, and is sensibly better than the one of linear MMSE.

[0082] In order to assess the complexity gain, a simulation with M=16 and N.sub.t=N.sub.Rx=4 is performed. The complexity is evaluated in terms of percentage of MMSE-SIC operation, in other words, the complexity of MMSE-SIC is equal to 1. It can be noticed how in at low Eb/N0 the educated guess is almost always ML, and thus no more SIC operation are needed. When the Eb/N0 increases, more SIC operations are needed to improve the quality of the educated guess, until a certain regime is reached. This proves that the proposed algorithm can yield a gain in complexity that is of around the 90% at low Eb/N0 and of the 60% at high Eb/N0. This is depicted in FIG. 8.

[0083] The application is not limited to the examples and especially not to a specific number of antennas or detection methods. The application discussed above can also be applied to many MIMO communication schemes. The characteristics of the exemplary embodiments can be used in any combination.

[0084] The application has been described in conjunction with various embodiments herein. However, other variations to the disclosed embodiments can be understood and effected by those skilled in the art in practicing the claimed application, from a study of the drawings, the disclosure and the appended claims. In the claims, the word comprising does not exclude other elements or steps and the indefinite article a or an does not exclude a plurality. A single processor or other unit may fulfill the functions of several items recited in the claims. The mere fact that certain measures are recited in usually different dependent claims does not indicate that a combination of these measures cannot be used to advantage. A computer program may be stored/distributed on a suitable medium, such as an optical storage medium or a solid-state medium supplied together with or as part of other hardware, but may also be distributed in other forms, such as via the internet or other wired or wireless communication systems.