WO2012103716A1

WO2012103716A1 - Detecting method and system for multiple-input multiple-output system

Info

Publication number: WO2012103716A1
Application number: PCT/CN2011/076483
Authority: WO
Inventors: 李景玉; 高振兴; 刘华斌; 江长国
Original assignee: 华为技术有限公司
Priority date: 2011-06-28
Filing date: 2011-06-28
Publication date: 2012-08-09
Also published as: CN102246452A; CN102246452B

Abstract

Disclosed are a detecting method and system for a multiple-input multiple-output system. The method includes: generating a processed received signal according to a received signal and a receiving channel matrix, and determining a first stage node according to the processed received signal; and repeatedly performing the process as follows until the stage number i equals the number of transmitting antennas: obtaining the ranking sequence of the reliability of at least two nodes of a i ^th stage according to the nodes of a (i - 1)^th stage; calculating the measurement values of the survival paths corresponding to each node of the i ^th stage, according to the ranking sequence of the reliability of the at least two nodes of the i ^th stage; repeatedly selecting a group of nodes of the i ^th stage in the ascending order of the measurement values of the survival paths until the number of the selected nodes of the i ^th stage equals the number of the survival paths of the i ^th stage; and outputting the survival paths and the measurement values respectively corresponding to the selected nodes of the i ^th stage for decoding transmitted signals.

Description

Detection method and system for multiple input multiple output system

The present invention relates to the field of communications technologies, and in particular, to a method and system for detecting a multiple input multiple output system.

Background technique

In a multiple-input multiple-output (MIMO) system, the model of the received signal can be expressed as: Y=HS+N, where Y is the received signal, S is the transmit antenna, and H is the channel response. , N is noise. For example, referring to FIG. 1A, in the MIMO system, there are T transmitting antennas at the transmitting end and R receiving antennas at the receiving end, and the transmission signal sequence of the T transmitting antennas can be expressed as S = [ _SI , S ₂ ,... _St ] ^t , the signal vector of the corresponding RX wife antenna can be expressed as γ = [ _Υι , _{Υ 2} ,... _Υκ ] ^τ , zero mean value on the receiving antenna, white Gaussian noise with variance _σ ² It can be expressed as N = [ _ni , n ₂ ,... _nR ] ^T , and H is specifically a channel matrix of R x T dimension; where _Si is the signal transmitted by the ith transmit antenna, and _yi is the ith root The signal received by the receiving antenna.

The purpose of the receiver detector of the multiple input multiple output system is to recover the transmitted symbol 8 from the received vector Y. In order to fully obtain the receive diversity gain, the Maximum Likelihood (MLD) algorithm is usually used for detection to achieve decoding; MLD refers to all possibilities of traversing S, and finds the smallest |Y - HS|| ² S. However, since the MLD algorithm needs to traverse all possible transmission vectors during detection, the calculation amount is large and the implementation is complicated. In order to effectively reduce the amount of computation without loss of performance, the prior art proposes to combine QR decomposition (a matrix decomposition) of the channel matrix with the M algorithm (a node search method of the ML). The method is used for detection. This method is called QRM (ie QR Decomposition and M-algorithm) algorithm. In this method, it is first necessary to determine the ordering of the child nodes, and then select the child nodes one by one according to the sorting. And output the corresponding survival path.

In the research and practice of the prior art, the inventors of the present invention have found that the existing QRM algorithm needs to retain more performance for each stage in order to achieve better performance when the modulation order or the number of transmitting antennas is large. The nodes that cause the node selection are still complex.

Summary of the invention

Embodiments of the present invention provide a detection method and a detection system for a multiple input multiple output system, which can be Under the premise of achieving better detection performance, the node is selected.

A method for detecting a multiple input multiple output system, comprising:

Generating a processed received signal according to the received signal and the received channel matrix, and determining a first level node based on the processed received signal;

Repeat the following process until the number i reaches the number of transmit antennas, and i is a positive integer greater than 1: Obtain the reliability order of at least two i-th nodes according to the i-1th node; according to at least two i-th nodes The reliability rankings calculate the metric values of the respective survival paths corresponding to the respective i-th nodes; repeatedly performing the metric values according to the respective survival paths to select a group of i-th nodes from the smallest to the largest, until the selected The number of i-level nodes is equal to the number of survival paths of the i-th level;

The survival path and the metric value for decoding the transmitted signal corresponding to the selected i-th node are respectively output.

A detection system comprising:

a first processing module, configured to generate a processed received signal according to the received signal and the received channel matrix, and determine a first level node based on the processed received signal;

The second processing module is configured to repeatedly perform the following process until the number of stages i reaches the number of transmitting antennas, and i is a positive integer greater than 1:

Acquiring the reliability ranking of at least two i-th nodes according to the i-1th node; calculating the metric values of the respective survival paths corresponding to the i-th nodes according to the reliability ranking of the at least two i-th nodes; Selecting a group of i-th nodes according to the metric values of the respective survival paths, in descending order, until the number of selected i-th nodes is equal to the number of survival paths of the i-th level; output and selected i-th nodes Corresponding to the survival path and metric value for decoding the transmitted signal.

In the embodiment of the present invention, a group of i-th sub-nodes are selected at a time according to the metric value from small to large, so that one sub-node can be selected at a time, and the better detection performance can be reduced compared with the prior art. The number of times the node is selected, that is, the node selection.

DRAWINGS

In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the embodiments or the description of the prior art will be briefly described below. Obviously, the drawings in the following description It is only some embodiments of the present invention, and those skilled in the art can obtain other drawings according to these drawings without any creative work. Figure la is a schematic diagram of a model of a multi-antenna system;

FIG. 1b is a flowchart of a method for detecting a multiple input multiple output system according to an embodiment of the present invention; FIG. 1c is a schematic diagram of partitioning of a preset two-dimensional space;

2a is a flowchart of a method for detecting a multiple input multiple output system according to an embodiment of the present invention; FIG. 2b is a flow chart of node selection in a detection method of a multiple input multiple output system according to an embodiment of the present invention;

2c is a schematic diagram of node selection in an embodiment of the present invention;

2d is another schematic diagram of node selection in an embodiment of the present invention;

2e is another schematic diagram of node selection in the embodiment of the present invention;

2f is another schematic diagram of node selection in an embodiment of the present invention;

3 is a schematic diagram of mapping in a preset two-dimensional space;

4a is a schematic structural diagram of a detection system according to an embodiment of the present invention;

FIG. 4b is another schematic structural diagram of a detection system according to an embodiment of the present invention.

Detailed ways

BRIEF DESCRIPTION OF THE DRAWINGS The technical solutions in the embodiments of the present invention will be described in detail below with reference to the accompanying drawings. All other embodiments obtained by a person skilled in the art based on the embodiments of the present invention without creative efforts are within the scope of the present invention.

Embodiments of the present invention provide a method, apparatus, and system for detecting a multiple input multiple output system. The following is a detailed description. Embodiment 1

A method for detecting a multiple input multiple output system, comprising: generating a processed received signal according to a received signal and a received channel matrix, and determining a first level node based on the processed received signal, and repeatedly performing the following process until the number of stages i The number of transmit antennas is reached, and i is a positive integer greater than one:

Acquiring the reliability ranking of at least two i-th nodes according to the i-1th node; calculating the metric values of the respective survival paths corresponding to the i-th nodes according to the reliability ranking of the at least two i-th nodes; Selecting a group of i-th nodes according to the metric values of the respective survival paths, in descending order, until the number of selected i-th nodes is equal to the number of survival paths of the i-th level;

Outputting a survival path and metric for decoding the transmitted signal corresponding to the selected i-th node respectively Value.

The survival path and the corresponding metric value are used to further calculate the transmission symbol S. For the specific calculation method, refer to the prior art, which is not described in this embodiment. In the process of solving the signal transmitted by the MIMO system, each node corresponds to a solution of the transmitted signal, that is, the possible value of the transmitted signal.

Referring to Figure lb, the specific process can be as follows:

101. Generate a processed received signal according to the received signal and the received channel matrix, and determine a first level node based on the processed received signal;

The specific received signal according to the received signal and the received channel matrix may be as follows:

The receiving channel matrix is QR decomposed to obtain a 酉 matrix Q and an upper triangular matrix R; the conjugate matrix of the 酉 matrix Q is multiplied with the received signal to obtain a processed received signal.

The method may further include: before generating the processed received signal according to the received signal and the received channel matrix (ie, step 101),

Obtain the received signal, the receive channel matrix, the number of transmit antennas, and the number of survival paths at each level.

For example, the details can be as follows:

(1) Obtaining the received signal Y, the receiving channel matrix Η, the number of transmitting antennas, and the number of survival paths at each level;

(2) Decomposing the receiving channel matrix H to obtain a 酉 matrix Q and an upper triangular matrix R: H=QR;

(3) multiplying the conjugate matrix β ^{Η of the 酉} matrix Q by the received signal Y to obtain a vector for convenience of description, which is referred to as a processed received signal in the embodiment of the present invention.

Ϋ = Q ^H HS + Q ^H N = Q ^H QRS + Q ^H N = RS + N;

Among them, two Q ^H N , statistical characteristics and noise N-like.

102. Repeat the following process, the direct number i (and i indicates the number of stages) reaches the number of transmitting antennas, where i is a positive integer greater than 1:

Obtaining, according to the i-1th node, a reliability ranking of the plurality of (ie, at least two) i-th nodes; calculating, according to the reliability ranking of the plurality of i-th nodes, each survival path corresponding to each i-th node Metric value; repeatedly performed according to the calculated metrics of each survival path to select from small to large Selecting a group of i-th nodes until the number of selected i-th nodes is equal to the number of surviving paths of the i-th level; and outputting the survival path and the metric for decoding the transmitted signals respectively corresponding to the selected i-th nodes . For example, the details can be as follows:

(1) Acquiring the reliability order of the plurality of i-th nodes according to the i-1th node, as follows: determining the node estimation values of the plurality of i-th nodes according to the processed received signals, and determining the plurality of i-th nodes The node estimation value is in the preset two-dimensional space, and the constellation point order is obtained according to the node estimation values of the plurality of i-th nodes in the preset two-dimensional space, and the reliability of the i-th node is determined according to the obtained constellation point ordering. Sexual ordering. The two-dimensional space can be set according to the node distribution of the entire network.

For example, if in step 101, the processed received signal is obtained.

Ϋ = Q ^H HS + Q ^H N = Q ^H QRS + Q ^H N = RS + N ; Then the processed received signal Ϋ can be regarded as the received signal, and the matrix R is regarded as the channel, starting from the last line of R , MLD detection algorithm for selecting points by level. as follows:

= Q ^H HS + Q ^H N = Q ^H QRS + Q ^H N = RS + NJ ^ , get the following formula:

The detection process starts from the beginning, until _Sl . Among them, the first level, _Sl is the T level. Set the number of paths to survive at each level in QRM and calculate the Euclidean distance for each level. The method for determining the order of the child nodes according to the parent node may be as follows:

Suppose the parent node is i-1 level and the child node is i level. If you want to determine the reliability order of the i-th child nodes under each i-1 level parent node. You can get:

N

^N +l― ^j ^r N-i+l,j ^S j

^ j=N-i+2 ·

^S N- 1 = ,

rN-i+l, N-i+l

Where – _i+2 ,..., the i-1 level of this path, ... the second level, the node estimate of level 1. Calculate the corresponding constellation point order according to the position of ^ _—i+1 (ie, the symbol estimation of the i-th level) in the two-dimensional space. Since the position of the constellation point in the two-dimensional space is relatively fixed, it is equivalent to the coordinate point, so Root According to the sorting of constellation points, combined with the position of other nodes in the two-dimensional space, the reliability ranking of the nodes can be introduced.

In order to improve the detection efficiency, the constellation points obtained by the above calculation may be sorted and stored, for example, may be recorded in the reliability ranking table, so that the reliability of the pre-stored by searching for the bit position of the preset two-dimensional space according to the node estimation value is obtained. Sort the table, you can get the corresponding constellation point sort.

For example, referring to FIG. 1c, taking the constellation point number as 16QAM as an example, first, the space can be divided into several regions according to the constellation points, and one region is called a block, for example, the two-dimensional space is divided into 128 regions, and the two-dimensional space includes 128 blocks. The intersection between the block and the block is the constellation point (see the black point in Figure lc), which has a fixed position in the two-dimensional space, and the application of the constellation point facilitates positioning of the node. For each block, a corresponding reliability ordering table can be stored. In this case, as long as it is determined which block the node is located, the reliability ranking table corresponding to the block can be searched to obtain the constellation point order corresponding to the block where the node is located, thereby Get the reliability order of the child nodes under the node.

Of course, in addition to the above described detection system can store the reliability ranking table of all blocks, the detection system can also store only the reliability ranking table of the partial blocks. as follows:

(A) Stores the reliability ranking table for all blocks

If the reliability ranking table of all the blocks is stored, the constellation point order of the block where the node is located may be obtained by directly searching the reliability ranking table corresponding to the block where the node is located according to the node position; that is, according to the node estimation value of the i-th node The position of the preset two-dimensional space to obtain the corresponding constellation points can be specifically:

The block where the i-th node is located is determined according to the node estimation value of the i-th node, and the constellation point corresponding to the block where the sub-node is located is obtained by searching the reliability ranking table corresponding to the block where the i-th node is located.

(B) Reliability ranking table for storing partial blocks

If only the reliability ordering table of the partial block is stored, after determining the block where the stator node is located according to the estimated value of the node of the i-th node, the following processing may be performed as follows:

The first case: if the block where the i-th node is located has a corresponding reliability ranking table, the constellation points corresponding to the block where the i-th node is located are obtained by searching the reliability ranking table corresponding to the block where the i-th node is located;

The second case: if the block where the i-th node is located does not have a corresponding reliability ranking table, the node estimation value of the i-th node is mapped to the block with the corresponding reliability ranking table, and the mapping is searched by The reliability ranking table corresponding to the block in which the point is located obtains the constellation point order corresponding to the block where the mapping point is located, and the constellation points corresponding to the block in which the mapping point is located are inversely mapped, and the constellation points corresponding to the block where the i-th level node is located are obtained.

(2) Calculate the metric value of the survival path corresponding to each i-th node according to the reliability ranking of the i-th node. For a specific method for calculating the metric value, refer to the prior art, and details are not described herein again.

(3) Selecting a group of i-th nodes at a time according to the metric value from small to large, until the number of selected i-th nodes is equal to the number of survival paths of the i-th level; for example, the details may be as follows:

Selecting a group of i-th sub-nodes according to the metric value from small to large, determining whether the number of selected i-th nodes is equal to the number of survival paths of the i-th level, and if so, performing corresponding output according to the selected i-th node output a step of decoding a survival path of the transmitted signal and a corresponding metric value (ie, step (4)); if not, updating the ordering of the metric values of the remaining path of the remaining i-th node, and returning execution according to the metric value The step of selecting a set of i-th sub-nodes in a large order (ie, step (3)).

For example, a group of 3 nodes can be set. If the number of survival paths of the i-th level is 12, three nodes can be selected at a time according to the metric value from small to large. Since 3 is less than 12, the remaining i-th is updated. Sorting the metrics of the survival path corresponding to the level node, selecting 3 nodes at a time according to the metric value from small to large, and then determining whether the number of selected i-th nodes has reached 12, since 3 + 3 = 6 If it is less than 12, it is necessary to continue to select nodes from the remaining i-th nodes, and so on, until step (4) is reached until the number of selected i-th sub-nodes reaches the number of survival paths of the i-th level.

The size of the group (that is, the number of nodes included in a group) can be set according to the needs of the actual application; it should be noted that if the group is larger, the selection of the node is more single, but its performance will be somewhat Reduced, so, when setting the size of the group, it needs to be considered comprehensively according to the needs of the actual application.

(4) outputting a corresponding survival path for decoding the transmitted signal and a corresponding metric value according to the selected i-th node.

Thereafter, the selection of the node enters the i + 1 level, and the node selection method is the same as the above node selection method. After the execution of the i + 1 level, the i i + 2 level is entered, and so on, until all the nodes are selected. .

As can be seen from the above, in this embodiment, a group of i-th nodes are selected at a time according to the metric value from small to large, so that a better detection can be achieved in comparison with the prior art that one sub-node is selected at a time. Under the premise of performance, reduce the number of node selections, that is, the node selection. The method provided in this embodiment selects a predetermined number of nodes to achieve near maximum likelihood performance when traversing all possible transmission vectors when detecting in the prior art. According to the method described in the first embodiment, the details will be further exemplified in the second and third embodiments. Embodiment 2

As shown in FIG. 3, in the embodiment, the preset two-dimensional space is a two-dimensional space with a constellation point number of 16QAM, and the two-dimensional space is divided into 128 regions, and each region is called a block, corresponding to each one. The blocks are stored with a reliability ranking table, that is, in this embodiment, 128 reliability ranking tables are stored. Each reliability ranking table correspondingly records the corresponding star point order when the node is located in the area. That is to say, as long as it is determined which block the node is located, the constellation point ordering can be obtained by finding the reliability ranking table of the block, thereby obtaining the reliability ranking of the node.

The details will be described below by way of example.

Scenario: Referring to FIG. 1A, in the MIMO system, there are T transmitting antennas at the transmitting end and R receiving antennas at the receiving end, and the transmission signal sequence of the T transmitting antennas can be expressed as S = [ _Sl , s ₂ ,...s _T ] ^T , the signal vector of the corresponding RX 1 receiving antenna can be expressed as 丫 ^ ^ ! ^, noise can be expressed as Ν

And Η is specifically a R x T-dimensional receiving channel matrix; where _Si is the signal transmitted by the ith transmitting antenna, and _yi is the signal received by the ith receiving antenna, the model of the received signal can be expressed as: Y=HS+N.

Then, as shown in FIG. 2a, the adaptive node selection method may be specifically as follows:

201. The receiving end acquires the received signal Y, the receiving channel matrix Η, the number of transmitting antennas, and the number of the storage paths at each level;

202. The receiving end performs QR decomposition on the receiving channel matrix , to obtain a 酉 matrix Q and an upper triangular matrix R: H=QR.

203. The receiving end multiplies the conjugate matrix β ^{Η of the 酉} matrix Q by the received signal Y to obtain the processed received signal.

Y = Q ^H HS + Q ^H N = Q ^H QRS + Q ^H N = RS + N ;

Among them, two Q ^H N , statistical characteristics and noise N-like. 204. The receiving end adaptively selects the i-th node according to the processed received signal, and selects Mi survival paths, where i=l, 2...T, that is, the initial value of i is incremented to T. For example, referring to Figure 2b, the details can be as follows:

Step 1: After obtaining the metric values of the My survival paths of the i-1th node, obtain the constellation points corresponding to the i-th sub-nodes with the i-1th node as the parent node under each survival path. Sort. The number of survival paths of the i-th level is M _; The details can be as follows:

The receiving end determines the node estimation value of the i-th sub-node under the i-1th parent node according to the processed received signal ¥, and determines the position of the node estimation value of the i-th sub-node in the preset two-dimensional space, according to The node estimation value of the i-th child node acquires the corresponding constellation point order in the position of the preset two-dimensional space. For example, you can: 3⁄4 down:

^ Y = Q ^H HS + Q ^H N = Q ^H QRS + Q ^H N = RS + NJ ^ , get the following formula:

Wherein, _1⁄2 — _i+2 , . . . , _1⁄2 are the i-1th level of the path corresponding to the processed received signal ¥, ... the second stage, the first stage node estimated value.

If it is necessary to determine the position of the node estimation value of the i-th sub-node under the i-1th parent node in the preset two-dimensional space, then ₊₁ can be _mapped into the preset two-dimensional space, for example, see the figure. Lc, if _{; +1} is located in the block "1" (that is, the block numbered "1" in the figure lc), then the corresponding constellation points can be sorted by finding the reliability ranking table corresponding to the block "1".

Step 2. Determine the reliability ranking of the i-th sub-node according to the constellation point order obtained in step 1.

Step 3: Calculate the metric value of the survival path corresponding to each i-th sub-node (ie, the path between the i-th parent node and the i-th-th sub-node) according to the reliability ranking of the i-th sub-node To get a metric.

Step 4: Select a group of i-th child nodes at a time according to the metric value from small to large, and output a corresponding survival path. Among them, a group of at least two, the size of the group (that is, the number of nodes included in a group) can be set according to the needs of the actual application. For example, among the obtained metrics, the smallest metric value may be selected to select the corresponding i-th child node. Then, the metric value of the survival path corresponding to the second sub-nodes of the parent node is calculated, and the corresponding i-th child node is output; wherein, N^ Mr l , when the number of remaining child nodes under the parent node is less than , actually output the remaining child nodes.

Step 5: Determine whether the number of selected i-th child nodes is equal to the number of survival paths of the i-th level, and if yes, that indicates that a survival path has been selected, and then perform step 7 and enter the i+1th level; if not, It means that the survival path has not been selected yet, and then step 6 is performed.

Step 6. Re-update the metrics under the survival path according to the survival path selected in step 4 and the reliability order of the i-th sub-nodes, and then return to step 4.

Step 7. Output the survival path and the corresponding metric according to the selected i-th sub-node.

The process of selecting a survival path is described in detail below with reference to an example.

Scenario: The tree diagram of the i-1th and ith levels is shown in Figure 2c, where the nodes are represented by circles and the lines between the nodes represent path metrics. The initial known information is the three nodes with the i-1 level labeled 1, 2, and 3 and the corresponding metric values. The three nodes are the parent nodes, and each parent node has 8 child nodes. The labels are respectively 1, 2, 3, 4, 5, 6, 7, and 8. In addition, it is also known that the number of survival paths of the i-th level is M _i4 = 3 , the number of survival paths of the i-th level ^^ , and Ν^3. According to the methods in steps 1 and 2, the reliability of the child nodes under each parent node from d to large is ranked as follows:

Parent node 1: child node 1, child node 2, child node 3, child node 4, child node 5, child node 6, child node 7, child node 8;

Parent node 2: child node 1, child node 2, child node 3, child node 4, child node 5, child node 6, child node 7, child node 8;

Parent node 3: child node 1, child node 2, child node 3, child node 4, child node 5, child node 6, child node 7, child node 8;

It should be noted that only the ordering of each child node can be known at this time, and the specific metric value cannot be known. Therefore, in step 3, the metrics of the corresponding survival paths of each parent node are calculated according to the reliability ranking of each parent node of the i-1th level and the i-th child node of each parent node.

After calculating the metric value of the corresponding survival path under each parent node, step 4 is performed, that is, the i-1th node corresponding to the minimum metric value is selected one by one according to the metric value calculated in step 3.

For example, as shown in steps 1 and 2, under parent node 1, the minimum metric lifetime is 1 →1 ; Under parent node 2, the survival path of the smallest metric is 2→1; under parent node 3, the survival path of the smallest metric is 3→1 (see the thick black line in Figure 2c), after After the calculation in step 3, it is known that among these metrics, the metric value of the survival path 2→1 is the smallest, then the survival path 2→1 can be selected (ie, the child node 1 under the parent node 2 is selected), and the survival path is output. 2→1 (see the black dotted line in Figure 2c). Then calculate the minimum metrics of the parent node 2 except 2→1. Because of NF3, the three survival paths except 2→1 are selected according to the ranking of the metrics from small to large, ie 2→2, 2→3 And 2→4 (that is, select child node 2 under parent node 2, child node 3 under parent node 2, and child node 4 under parent node 2), and output these three survival paths. Then, step 5 is executed. Since the selected survival path is 4 and the number of survival paths has not yet reached 16 (the number of survival paths of the i-th level is ^=16), the minimum metric value under the parent node 2 needs to be updated according to the reliability ranking table. , that is, perform step 6. Since 2→1, 2→2, 2→3, and 2→4 under parent node 2 have been selected as the survival path, according to the reliability ranking table, the minimum metric under parent node 2 is 2→5. Then calculate the metric value of 2→5 and update it to the minimum metric value under the parent node 2, and then return to perform the operation of selecting the smallest metric value, that is, perform step 4 as follows:

See Figure 2d. At this time, under parent node 1, the survival path of the smallest metric is 1→1; under parent node 2, the survival path of the smallest metric is 2→5; under parent node 3, the minimum metric is The survival path is 3→1 (see the black thick line in Figure 2d); among these metrics, the survival path 3→1 has the smallest metric, so you can choose the survival path 3→1 (ie select the parent). The child node 1) under node 3, and outputs the survival path 3→1 (see the black dotted line in Figure 2d). Then calculate the minimum metrics of the parent node 3 except 3→1. Because Ν^3, the three survival paths except 3→1 are selected according to the ranking of the metrics from small to large, namely 3→2, 3 →3 and 3→4 (that is, select child node 2 under parent node 3, child node 3 under parent node 3, and child node 4 under parent node 3), and output these three survival paths. Then, in step 5, since the selected survival path is 8 and the number has not yet reached 16, the minimum metric value under the parent node 3 needs to be updated according to the reliability ranking table, that is, step 6 is performed. Since 3→1, 3→2, 3→3, and 3→4 under the parent node 3 have been selected as the survival path, according to the reliability ranking table, the minimum metric under the parent node 3 is 3→5. Then calculate the metric value of 3→5 and update it to the minimum metric value under the parent node 3, and then return to the operation of selecting the smallest metric value, that is, perform step 4, :3⁄4 port:

See Figure 2e. At this time, under parent node 1, the survival path of the smallest metric is 1→1; under parent node 2, the survival path of the smallest metric is 2→5; under parent node 3, the minimum metric is Survival path 3→5 (see the black thick line in Figure 2e); among these metrics, the metric of the survival path 2→5 is the smallest, so you can choose the survival path 2→5 (ie select the parent node 2) The child node 5) and output the survival path 2 → 5 (see the black dotted line in Figure 2e). Then calculate the minimum metrics of the parent node 2 except 2→1, 2→2, 2→3, 2→4, 2→5. Since Ν,=3, the metrics are selected from small to large. 3 survival paths except 2→1, →, 2→3, 2→4, 2→5, ie 2→6, 2→7 and 2→8 (ie select child node 6 under parent node 2, parent The child node 7 under node 2 and the child node 8 under parent node 2), and output these three survival paths. Then step 5 is executed. Since the selected survival path is 12 and the number has not yet reached 16, the selection needs to be continued. Step 4 is performed as follows:

See Figure 2f. At this time, under parent node 1, the minimum metric has a survival path of 1→1; under parent node 3, the minimum metric has a survival path of 3→5 (see the black thick line in Figure 2f). Part)) Among these metrics, the metric of the survival path 3→5 is the smallest, so you can select the survival path 3→5 (ie select the child node 5 under the parent node 3) and output the survival path 3→5 ( See the black dotted line in Figure 2f). Then calculate the minimum metric value of the parent node 3 except 3→1, 3→2, 3→3, 3→4, 3→5, and select 除^3 according to the metric value from small to large. 3 survival paths other than 3→1, 3→2, 3→3, 3→4, 3→5, ie 3→6, 3→7 and 3→8 (ie select child node 6 under parent node 3, The child node 7 under the parent node 3 and the child node 8 under the parent node 3), and outputs the three survival paths. Then, step 5 is executed. Since the selected survival path has reached 16 at this time, the node of this level is selected, and step 205 can be performed.

It should be noted that the number of parent nodes, the number of child nodes, and the number of survival paths at each level and the specific values may be set according to actual application requirements, and are not limited to the values listed in this example. It should be understood that In other application scenarios, the implementation method is the same.

205. The receiving end determines whether the last level is reached. If the last level is reached, step 206 is performed. If the last level has not been reached, the value of i is incremented by one, and the node of the i+1th level is adaptively selected. , that is, return to step 204. For example, it can be judged whether i is equal to T, and if so, it indicates that the last level has been reached, and if not, it indicates that the last level has not yet been reached.

206. Output Μ _τ survival paths of the last stage and corresponding metric values.

As can be seen from the above, in this embodiment, a group of i-th sub-nodes are selected at a time according to the metric value from small to large, so that a better check can be achieved in comparison with the prior art in selecting one sub-node at a time. Under the premise of measuring performance, reducing the number of node selections, that is, the node selection. Embodiment 3

In the second embodiment, the constellation point ordering of all areas is stored. Unlike the second embodiment, in the present embodiment, only the constellation point ordering of the partial areas is stored.

Example: 3⁄4 port, specifically: 3⁄4 mouth:

As shown in Figure lc, due to the symmetry of the constellation point coordinates, only the constellation points of the partial regions can be stored. If the node estimation value is not stored in the region where the reliability is sorted, the node estimation value can be mapped to the storage reliability. Within the region of the sexual ordering, and looking up the table to get the constellation point sorting, and then the resulting sorting is inversely mapped to obtain the sorting corresponding to the block where the original node estimated value is located.

In the following, the reliability ranking table corresponding to the block in the first quadrant area is stored as an example for description. Referring to Figure 3, where the constellation point is 16QAM, the first quadrant is divided into 32 blocks, as follows:

For example, the node estimate A is in the second quadrant, and the real part is taken to the first quadrant after taking the absolute value, as shown in A. A, falls in block 1, and the table is sorted as "a, b, c ...". Then find the constellation points of each constellation point about the imaginary axis, such as "a2, b2, c2", where a2 and a are symmetric about the imaginary axis, b2 and b are symmetric about the imaginary axis, and so on. In this way, the constellation points corresponding to the node estimate A are sorted, and the reliability order of the child nodes under the node is obtained.

For another example, the node estimated value B is in the third quadrant, and the real imaginary part is mapped to the first quadrant after taking the absolute value, as shown in FIG. A, falls in block 1, and the table is sorted as "a, b, c ...". Then find the constellation points of each constellation point that are symmetric about the origin, such as "a3, b3, c3, ...", so that the constellation points corresponding to the node estimation values are obtained, thereby obtaining the sub-nodes of the node. Reliability sorting.

For another example, the node estimated value C is in the fourth quadrant, and the imaginary part is mapped to the first quadrant after taking the absolute value, as shown in FIG. Α, falling in block 1, the lookup table gets the constellation points corresponding to block 1 sorted as "a, b, c ...". Then find the constellation points for each constellation point about the real axis, such as "a4, b4, c4". In this way, the constellation points corresponding to the estimated value C are obtained, so that the reliability ordering of the child nodes under the node is obtained.

Among them, the constellation points are mostly Gray code, and the symmetric constellation points about the origin of the real axis of the virtual axis can be obtained only by bit inversion at certain positions. For example, refer to Table 1. The mapping relationship of 16QAM modulation in LTE can be as follows:

Where "b(i), b(i + 1), b(i + 2), b(i + 3)" means that 4 bits can be used to represent the coordinates of a constellation point, ie "b(i) The columns in which b(i + 1) , b(i + 2), b(i + 3)" are located, such as "0000", "0001", "0010", etc. all refer to the constellation point coordinates. While I represents the real part of the constellation point, Q represents the imaginary part of the constellation point, that is, the two-dimensional coordinates of the constellation point.

Simply invert b ( i ) to get the point where the constellation point is symmetric about the imaginary axis. Simply invert b ( i+1 ) to get the point where the constellation point is symmetric about the real axis, and b ( i ) b ( The i+1) inversion can get the symmetry point about the origin.

Further, the respective regions of the division of the first quadrant are symmetric with respect to the 45-degree diagonal, and it is also possible to store only the region below or above the diagonal, further reducing the amount of storage.

The other implementations of the present embodiment are the same as those of the second embodiment except for the storage of the reliability ranking table. For details, refer to the second embodiment, and details are not described herein again. As can be seen from the above, in this embodiment, a group of i-th sub-nodes are selected at a time according to the metric value from small to large, so that it is possible to achieve better detection performance than selecting one sub-node at a time in the prior art. In the following, the number of times of node selection is reduced, that is, the node selection is simplified. In addition, since the reliability ranking table of the partial area can be stored only in this embodiment, the storage amount can be reduced, and the storage resource can be saved. Embodiment 4

In order to better implement the above method, the embodiment of the present invention further provides a detection system, as shown in FIG. 4a, the detection system includes a first processing module 401 and a second processing module 402;

a first processing module 401, configured to generate a processed received signal according to the received signal and the received channel matrix, and determine a first level node based on the processed received signal;

The second processing module 402 is configured to repeatedly perform the following process until the number of stages i reaches the number of transmitting antennas, and i is a positive integer greater than one:

As shown in FIG. 4b, the second processing module 402 can include a sort acquisition unit 4021, a calculation unit 4022, a selection unit 4023, and an output unit 4024. The acquisition unit 401, the calculation unit 402, the selection unit 403, and the output unit 404 may each be a processor unit.

a sorting obtaining unit 4021, configured to acquire, according to the (i-1)th node, a reliability ranking of the plurality of (ie, at least two) i-th nodes, where i is a positive integer and is less than or equal to the number of transmitting antennas;

The calculating unit 4022 is configured to calculate, according to the reliability ranking of the plurality of (ie, at least two) i-th nodes acquired by the order obtaining unit 4021, the metric values of the respective survival paths corresponding to the respective i-th nodes; For the specific method of the value, refer to the prior art, and details are not described herein again.

The selecting unit 4023 is configured to repeatedly perform a set of (ith select one set) i-th nodes in a descending order of the metric values of the respective survival paths obtained by the calculating unit 4022, until the number of the selected i-th nodes is equal to the first Number of survival paths of level i;

The output unit 4024 is configured to output corresponding information according to the ith level node selected by the selecting unit 4023. The decoding of the survival path of the transmitted signal and the corresponding metric value.

The selecting unit 4023 is specifically configured to select a group of i-th nodes according to the metric value from small to large, and determine that the number of selected i-th nodes is smaller than the number of survival paths of the i-th level, and update the remaining i-th nodes. Sorting the metric values of the corresponding survival paths, and returning to perform the steps of selecting a group of i-th nodes in descending order of the metric values;

The output unit 4024 is specifically configured to: when the selecting unit 4023 determines that the number of selected i-th nodes is equal to the number of survival paths of the i-th stage, output the survival path and the corresponding metric value according to the i-th node selected by the selecting unit.

As shown in FIG. 4b, the first processing module 401 may include a decomposing unit 4012 and an operation unit 4013. Further, the first processing module 401 may further include an obtaining unit 4011; any one of the obtaining unit 4011, the decomposing unit 4012, and the operation unit 4013. Can also be a processor unit;

The obtaining unit 4011 is configured to acquire a received signal, a receiving channel matrix, a number of transmitting antennas, and a number of survival paths of each level;

Decomposing unit 4012, configured to perform fast response code QR decomposition on the receiving channel matrix to obtain a matrix Q and an upper triangular matrix R;

The operation unit 4013 is configured to multiply the conjugate matrix of the matrix Q obtained by the decomposition unit 4012 by the received signal to obtain the processed received signal;

The order obtaining unit 4021 is configured to determine, according to the processed received signal, a plurality of (ie, at least two) node estimation values of the i-th node, and determine a position of the node estimation value of the i-th node in a preset two-dimensional space, Obtaining corresponding constellation point order according to the position estimation values of the plurality of (ie, at least two) i-th nodes in the preset two-dimensional space, and determining a plurality (ie, at least two) according to the acquired constellation point ordering The reliability ordering of the subnodes.

For example, the details can be as follows:

The obtaining unit 4011 obtains the received signal Y, the received channel matrix Η, the number of transmitting antennas Τ, and the number of survival paths of each level, and then the decomposition unit 4012 decomposes the received channel matrix H to obtain the 酉 matrix Q and the upper triangular matrix R: H=QR; thereafter, the arithmetic unit 4013 multiplies the conjugate matrix β ^{Η of the} matrix Q by the received signal , to obtain the processed received signal.

Y = Q ^H HS + Q ^H N = Q ^H QRS + Q ^H N = RS + N ; Among them, two Q ^H N , statistical characteristics and noise Ν. After the processed received signal is obtained, the order obtaining unit 4021 may determine the node estimated value of the i-th child node under the i-1th parent node according to the processed received signal, and determine the node of the i-th child node. The estimated value is in a preset two-dimensional space, and then the corresponding constellation points are sorted according to the position estimation value of the i-th sub-node in the preset two-dimensional space, thereby further determining the reliability order of the i-th sub-node, For details, refer to the previous embodiment, and details are not described herein again.

In order to improve the detection efficiency, the constellation points obtained by the above calculation may be sorted and stored, for example, may be recorded in the reliability ranking table, so that the reliability of the pre-stored by searching for the bit position of the preset two-dimensional space according to the node estimation value is obtained. Sort the table, you can get the corresponding constellation point sort. For example, the details can be as follows:

The preset two-dimensional space is divided into several regions according to a preset strategy, and each region is called a block, and each block has a corresponding reliability ranking table, wherein the reliability ranking table records the constellation point order corresponding to the block. ;

Then, the order obtaining unit 4021 is configured to determine, according to the node estimation value of the i-th node, the block where the i-th node is located, and obtain the block where the i-th node is located by searching the reliability ranking table corresponding to the block where the i-th node is located. The corresponding constellation points are sorted.

The preset two-dimensional space is divided into a plurality of regions according to a preset strategy, and each region is referred to as a block, and the partial blocks have a corresponding reliability ranking table, wherein the reliability ranking table records the constellation point order corresponding to the block;

Then, the order obtaining unit 4021 is specifically configured to determine, according to the node estimation value of the i-th node, the block where the i-th node is located; if the block where the i-th node is located has a corresponding reliability ranking table, search for the i-th level The reliability ranking table corresponding to the block where the node is located, obtains the constellation point order corresponding to the block where the i-th level node is located; if the block where the i-th level node is located does not have the corresponding reliability ranking table, the node estimated value of the i-th level node is mapped. In the block with the corresponding reliability ranking table, by searching the reliability ranking table corresponding to the block where the mapping point is located, the constellation point corresponding to the block where the mapping point is located is obtained, and the constellation points corresponding to the block in which the mapping point is located are inversely mapped. , get the constellation point order corresponding to the block where the i-th level node is located. For the method of the mapping and the demapping, refer to the foregoing method embodiments, and details are not described herein. In the specific implementation, each of the foregoing units may be implemented as an independent entity, or may be implemented in any combination, and may be implemented as the same entity or a plurality of entities. For the implementation of the foregoing various units, refer to the foregoing embodiments, and details are not described herein.

As can be seen from the above, the selecting unit 403 of the detecting system of the embodiment can select a group of the i-th sub-nodes at a time according to the metric value from the smallest to the largest, so that one sub-node can be selected at a time compared to the prior art. Under the premise of achieving better detection performance, the number of node selections is reduced, that is, the node selection is completed. In addition, the detection system can store the reliability ranking table of all blocks, and can also utilize the symmetric relationship to store only part of the area. The reliability ranking table, and then using the mapping and demapping methods, the nodes in all regions can obtain the reliability ranking of the corresponding constellation points. Since this embodiment can store only the reliability ranking table of the partial regions, it can be reduced. Storage capacity, saving storage resources.

The detection system mentioned in the embodiment of the present invention may be located in a terminal device or other type of communication device, and used to obtain a plurality of surviving paths and corresponding metric values, if the modulation order or the number of transmitting antennas is large, To achieve near maximum likelihood performance, the system in this embodiment can cause multiple nodes to be reserved for each stage for detecting the transmitted signal S. The device may be in the form of a mobile phone, a laptop computer, or a tablet computer. The specific embodiment of the device is not limited. A person skilled in the art may understand that all or part of the various steps of the foregoing embodiments may be completed by a program instructing related hardware. The program may be stored in a computer readable storage medium, and the storage medium may include: Read only memory (ROM, Read Only Memory), random access memory (RAM), disk or optical disk.

The above is a detailed description of a QRM method and system provided by the embodiments of the present invention. The present invention is only used to help understand the method and core idea of the present invention. Meanwhile, for those skilled in the art, according to the idea of the present invention, The details of the present invention and the scope of the application are subject to change. The contents of the present specification are not to be construed as limiting the invention.

Claims

Rights request

A method for detecting a multiple input multiple output system, comprising:

The method according to claim 1, wherein the obtaining the reliability ranking of the at least two i-th nodes according to the i-1th-level node comprises:

Determining, according to the processed received signal, node estimation values of at least two i-th nodes, determining location of the node estimation values of the at least two i-th nodes in a preset two-dimensional space, according to at least two i-th nodes The node estimation value obtains the constellation point order in the preset two-dimensional space, and determines the reliability order of the at least two i-th nodes according to the obtained constellation point order.

The method according to claim 2, wherein the obtaining the constellation point order according to the node estimation value of the at least two i-th nodes in the preset two-dimensional space comprises:

The preset two-dimensional space is divided into blocks according to a preset policy, and each block has a reliability ranking table, and the reliability ranking table records the constellation point order corresponding to the block;

Determining, according to the node estimation value of the i-th node, the block where the i-th node is located, and obtaining the constellation point order corresponding to the block where the i-th node is located by searching the reliability ranking table corresponding to the block where the i-th node is located.

The preset two-dimensional space is divided into blocks according to a preset strategy, and some blocks have a reliability ranking table, and the reliability ranking table records the constellation point order corresponding to the block;

Determining, according to the node estimation value of the i-th node, a block in which the i-th node is located; If the block where the i-th node is located has a reliability ranking table, the constellation point corresponding to the block where the i-th node is located is obtained by searching the reliability ranking table corresponding to the block where the i-th node is located;

If the block where the i-th node is located does not have a reliability ranking table, the node estimation value of the i-th node is mapped to the block with the reliability ranking table to obtain the mapping point, and the reliability ranking table corresponding to the block where the mapping point is located is searched. The constellation points corresponding to the blocks in which the mapping points are located are sorted, and the constellation points corresponding to the blocks in which the mapping points are located are inversely mapped, and the constellation points corresponding to the blocks in which the i-th level nodes are located are obtained.

The method according to any one of claims 1 to 4, wherein the receiving the received signal according to the received signal and the received channel matrix comprises:

Performing QR decomposition on the received channel matrix to obtain a matrix Q and an upper triangular matrix R;

The conjugate matrix of the matrix Q is multiplied by the received signal to obtain a processed received signal.

6. A detection system, comprising:

7. The detection system of claim 6, the second processing module comprising:

a sorting obtaining unit, configured to obtain reliability rankings of at least two level i nodes according to the i-1th level node;

a calculation unit, configured to calculate a metric value of each survival path corresponding to each i-th node according to a reliability ranking of at least two i-th nodes acquired by the sort acquisition unit;

a selecting unit, configured to repeatedly perform the metric values of the respective survival paths obtained according to the calculating unit to select a group of i-th nodes from small to large, until the number of selected i-th nodes is equal to the survival of the i-th level Number of paths;

An output unit, configured to output, respectively, corresponding to the selected i-th node for decoding the sent signal The survival path and the corresponding metric.

8. The detection system of claim 7 wherein:

The ordering obtaining unit is configured to determine node estimation values of at least two i-th nodes according to the processed received signals, and determine a position of the node estimation values of the at least two i-th nodes in a preset two-dimensional space, Obtain constellation point order according to the node estimation values of the at least two i-th nodes in a preset two-dimensional space, and determine a reliability order of the at least two i-th nodes according to the obtained constellation point order.

The detection system according to claim 7 or 8, wherein the preset two-dimensional space is divided into blocks according to a preset policy, each block has a reliability ranking table, and the reliability ranking table records Sorting the constellation points corresponding to this block;

And the ordering obtaining unit is configured to determine, according to the node estimation value of the i-th level node, the block where the i-th level node is located, and obtain the i-th level by searching a reliability ranking table corresponding to the block where the i-th level node is located The constellation points corresponding to the block in which the node is located are sorted.

The detection system according to claim 7 or 8, wherein the preset two-dimensional space is divided into blocks according to a preset policy, and some blocks have a reliability ranking table, and the reliability ranking table records Sort the constellation points corresponding to the block;

And the sequence obtaining unit is configured to determine, according to the node estimation value of the i-th node, a block where the i-th node is located; if the block where the i-th node is located has a reliability ranking table, search for a block corresponding to the block where the i-th node is located The reliability ranking table obtains the constellation point order corresponding to the block where the i-th node is located; if the block where the i-th node is located does not have the reliability ranking table, the node estimation value of the i-th node is mapped to the reliability ranking table. A mapping point is obtained in the block, and the constellation points corresponding to the block in which the mapping point is located are obtained by finding the reliability ranking table corresponding to the block in which the mapping point is located, and the constellation points corresponding to the block in which the mapping point is located are inversely mapped to obtain the i-th level. The constellation points corresponding to the block in which the node is located are sorted.