CN102647250A - A Cooperative Communication Method Based on Clustered Sphere Decoding in Virtual MIMO - Google Patents

Abstract

The invention discloses a cooperative communication method based on clustering sphere decoding in virtual MIMO (Multiple-Input Multiple-Output). The cooperative communication method comprises the following steps of: at first, clustering sensor nodes in a wireless sensing network by using a load balanced type responsive distributed clustering algorithm (RDCA); secondly, selecting a cooperative node (CN) of each cluster head after clustering; thirdly, transferring a cluster head node (CH) to the cooperative node (CN) after data combination; and finally, commonly sending data to a convergence point, and using a sphere decoding algorithm at a receiving end, so that the transmission efficiency of the wireless sensing network is improved by compromising complexity and error rate property. With the adoption of the method, the advantages of space-time coding are expanded into a plurality of clusters, and the cooperative control among the clusters in a traditional scheme is avoided. Meanwhile, the improved sphere decoding strategy is used at the receiving end, so that the transmission efficiency of the wireless sensing network is improved by improving a throughput rate of the entire wireless sensing network system.

Description

A Cooperative Communication Method Based on Clustered Sphere Decoding in Virtual MIMO

technical field

The present invention relates to the field of wireless cooperative communication, especially for wireless sensor networks, and more specifically, relates to a cooperative communication method based on clustering spherical decoding in virtual MIMO (Multiple-Input Multiple-Out-put) .

Background technique

Wireless sensor network integrates multiple technologies such as sensor technology, wireless communication technology, and embedded technology, and can be widely used in military, medical, business, and family fields. In recent years, it has attracted extensive attention from academia and industry. The wireless sensor network needs to improve the signal-to-noise ratio, reduce the bit error rate, extend the communication distance, expand the network range, reduce the number of redundant retransmissions and relay routes, and indirectly increase the channel at a relatively small transmission power. data throughput and improve transmission efficiency. Virtual MIMO technology, as one of the technologies to effectively improve the transmission efficiency of wireless sensor networks, makes full use of the characteristics of high node density of sensor networks, and allows multiple single-antenna nodes to perform cooperative information processing and transmission to form a virtual MIMO antenna array. Reliable communication in a multipath fading environment with less total energy consumption, thereby prolonging network life, improving data reliability, reducing bit error rate, and improving the throughput of the entire transmission system. At present, the sensor nodes in the wireless sensor network will be clustered according to a specific algorithm, but the network scale will be reduced. When encountering multiple clusters and needing to send data to the data center (AP) at the same time, it will not only affect the real-time performance of the transmission system, but also generate access conflicts. In order to prevent access conflicts, the traditional solution is to coordinate each cluster and introduce a priority control strategy, which reduces the transmission efficiency of the wireless sensor network system. In addition, in the traditional scheme, a low-complexity linear receiver is first used for the VBLAST system. The decoding process is equivalent to using a linear filter to separate the transmitted data streams, and then makes a separate judgment for each data stream. The linear decoding algorithm is: zero-forcing detection (ZF) algorithm and minimum mean square error (MMSE) algorithm. Among them, the zero-forcing detection algorithm ignores the influence of noise correlation and reduces the bit error rate (BER) performance (BER: bit error ratio) is an indicator to measure the accuracy of data transmission within a specified time. Bit error rate = bit errors in transmission / total number of codes transmitted * 100%.) Although the MMSE algorithm has a good compromise in reducing noise interference and weakening adjacent signal interference, the bit error rate BER performance has not improved. Moreover, the two above-mentioned zero-forcing detection (ZF) linear decoding algorithms and the minimum mean square error (MMSE) linear decoding algorithm do not consider the special structure of the multi-cluster channel matrix, and need to calculate the inverse matrix of the multi-dimensional matrix, and its decoding complexity is high. The processing complexity is still relatively large, and the transmission efficiency will still be reduced.

Contents of the invention

The purpose of the present invention is to address the above technical problems and propose a cooperative communication method based on clustered spherical decoding in virtual MIMO. This method extends the advantages of space-time coding to multiple clusters and avoids the coordination between clusters in traditional schemes. At the same time, an improved spherical decoding strategy is adopted at the receiving end, which improves the throughput of the entire wireless sensor network system and thus improves the transmission efficiency of the wireless sensor network.

In order to achieve the above-mentioned purpose of the invention, the idea of the present invention is: firstly, the sensor nodes in the wireless sensor network are clustered by using load-balanced responsive distributed clustering algorithm (RDCA), and then the cooperative nodes of each cluster head are selected after clustering. (CN), forward the fused data of the cluster head node (CH) to the cooperative node (CN), and finally jointly send the data to the aggregation point, and adopt the spherical decoding algorithm at the receiving end to achieve a compromise between complexity and bit error rate performance to improve wireless Transmission Efficiency of Sensor Networks.

According to above-mentioned inventive design, the present invention adopts following technical scheme:

A cooperative communication method based on clustering spherical decoding in virtual MIMO of the present invention, the specific steps of the method are as follows:

(1), in the virtual MIMO domain, the cluster head node (CH) selects several cooperative nodes (CN) from the acquisition nodes (ED) in the cluster;

，另一种为，2种状态轮流出现，在整个通信传输过程中时长满足

； (2), the communication transmission in the local cluster, the duration used is two states, one duration is

, another for , the two states appear in turn, and the duration of the entire communication transmission process satisfies

;

(3) The long-distance communication transmission between the coordination node (CN) and the data center (AP) takes two states, one of which is , another for , the two states appear in turn, and the duration of the entire communication transmission process satisfies ;

(4), define the threshold value that is less than the individual power supply value of the terminal collection node (ED) in wireless communication to be exhausted, and judge whether the individual power supply value of each terminal collection node is greater than the threshold value of power exhaustion in turn, if the terminal collection node If the value of the individual power supply is equal to or greater than the threshold of power depletion, go to step (1), and the terminal acquisition node participates in the next round of communication transmission, otherwise the terminal acquisition does not participate in the next round of communication transmission, and the communication transmission ends.

The cluster head node (CH) in the virtual MIMO domain described in the above steps (1) selects several cooperative nodes (CN) from the collection node (ED) in the cluster, and its specific steps are as follows:

(1-1), in the virtual MIMO domain, according to the size of the remaining energy of the wireless sensor network node power supply battery of all acquisition nodes (ED), select the cluster head node (CH) according to the responsive distributed clustering algorithm (RDCA); The cluster head collection nodes select their own cluster heads according to the cost function based on the closest distance with load balancing performance;

(1-2) After the clustering is formed, the cluster head node (CH) allocates time slots to the acquisition nodes (ED) in each cluster by means of time division multiple access (TDMA) to reduce transmission interference and facilitate data transmission within the cluster;

(1-3), the cluster head node (CH) will select several cooperative nodes (CN) from the collection nodes in the cluster.

The local intra-cluster communication transmission described in the above step (2), its specific steps are as follows:

(2-1), the terminal acquisition node (ED) sends the data to the cluster head node (CH) in turn according to the TDMA time slot allocated by the step (1-3);

(2-2), the cluster head node (CH) aggregates all the collected data together with its own data;

(2-3), then the aggregated data streams are serially converted into Nt sub-data streams and broadcast to the corresponding collaborative nodes (CN);

The long-distance communication transmission between the coordination node (CN) and the data center (AP) described in the above step (3), its specific steps are as follows:

(3-1), the coordinating nodes (CN) of each cluster perform space-time block coding (STBC) encoding on the data;

(3-2), the STBC data stream of each cluster is transmitted to the data center (AP) in the traditional VBLAST mode;

(3-3) The cooperative VBLAST antenna node at the receiving end samples and quantizes the received data, and the processed data stream is sent to the data center (AP) for decoding and recovery, and the improved spherical decoding (SD) is adopted at the receiving end The algorithm decodes the received signal;

The processed data stream described in the above steps (3-3) is sent to the data center (AP) for decoding recovery, and the improved spherical decoding (SD) algorithm is used at the receiving end to decode the received signal, which is as follows:

中搜索以ρ为半径的超球内的点，离球心最近的格点即为Y的译码，解码时以分簇为单位进行分组，然后对每个分块分别进行球形解码， The basic idea of the improved spherical decoding algorithm is to accept the signal vector Y as the center of the sphere, and set

Search for the points in the hypersphere with ρ as the radius, and the grid point closest to the center of the sphere is the decoding of Y. When decoding, it is grouped in units of clusters, and then each block is decoded separately.

Assuming Nc=3, Nt=4, Nr=3, divide the 3 clusters into two groups, the first group corresponds to two clusters, and the second group corresponds to the remaining cluster, which is expressed as:

(1)

，

为

向量，

，

为

向量，

为

的子向量， in

,

for

vector,

,

for

vector,

for

subvector of

的维度为

， 

 = 4  则利用球形解码SD算法求解第二分组

，得：  set up

has a dimension of

,

= 4, use the spherical decoding SD algorithm to solve the second group

,have to:

                    (2) 

(2)

为第二分组解向量格点集合，为信号矢量第二分组向量，为上三角矩阵R的第二行第二列分量，

为第二分组向量， in,

Solve the set of vector lattice points for the second group, is the second grouping vector of signal vectors, is the second row and second column component of the upper triangular matrix R,

is the second grouping vector,

，

的表达式为：  Then use the spherical decoding (SD) algorithm to solve the first group , has a dimension of

,

The expression is:

(3)

为第二分组解向量格点集合，

为信号矢量第二分组向量，

为上三角矩阵R的第i行第j列分量，为第二分组向量, 

为第二分组解向量格点集合。 in,

Solve the set of vector lattice points for the second group,

is the second grouping vector of signal vectors,

is the i-th row and j-th column component of the upper triangular matrix R, is the second grouping vector,

Solve the set of vector lattice points for the second group.

Compared with the prior art, the advantage of the cooperative communication method based on clustered spherical decoding in virtual MIMO of the present invention is that the method adopts STBC encoding for local communication within a cluster, and adopts VBLAST mode for long-distance communication between clusters, achieving Diversity and multiplexing effects, with the advantages of STBC and VBLAST at the same time, reducing the overhead of clustering; the receiver uses the spherical decoding algorithm (SD) to decompose the transmission signal into multiple groups, weakening the impact of error propagation effects on the transmission system , and the larger the number of groups, the smaller the decoding complexity, which can make the data center achieve a balance between the algorithm complexity and the bit error rate performance, extend the advantages of space-time coding to multiple clusters, and avoid the traditional solution between clusters Coordinated control, while greatly improving the throughput rate of the entire transmission system, thereby improving the transmission efficiency of the wireless sensor network.

Description of drawings

Fig. 1 is three sub-clusters in the communication method of the present invention, collection node (ED), data center (AP) in the cluster, each sub-cluster has a channel structure diagram of a cluster head node and four cooperative nodes;

FIG. 2 is a flow chart of a cooperative communication method based on clustered sphere decoding in virtual MIMO according to the present invention;

Fig. 3 is the flowchart of step (1) in Fig. 2;

Fig. 4 is the flowchart of step (2) among Fig. 2;

Fig. 5 is the flowchart of step (3) among Fig. 2;

Fig. 6 is the BER performance comparison chart (in the figure, the vertical axis is the bit error rate, and the horizontal axis represents the average bit energy/white noise power spectral density) of different decoding algorithms in the present invention under a single clustering;

Fig. 7 is single cluster and multi-cluster BER performance contrast figure among the present invention (vertical axis represents bit error rate, and horizontal axis represents average bit energy/white noise power spectral density);

Fig. 8 is a comparison diagram of BER performance between packet spherical decoding and traditional spherical decoding in the present invention (the vertical axis represents bit error rate, and the horizontal axis represents average bit energy/white noise power spectral density);

Fig. 9 is a comparison diagram of computational complexity between grouped spherical decoding and traditional spherical decoding in the present invention (the vertical axis is the simulation time, and the horizontal axis is the average bit energy/white noise power spectral density). the

Detailed ways

The present invention will be further described in detail below in conjunction with the accompanying drawings and embodiments.

For the convenience of simulation, as shown in Figure 1, a cooperative communication method based on clustered spherical decoding in virtual MIMO of the present invention makes the following assumptions: for the communication of acquisition nodes (EDs) within a cluster, the channel model is additive Gaussian white Noise; for the long-distance communication from the cooperative node (CN) to the data center (AP), the channel model is a Rayleigh channel with square rate fading; all nodes are isomorphic and have the same power supply battery energy; all nodes become cluster heads Nodes (CH) have equal communication opportunities; the transmission model adopts a single-hop structure; all nodes are synchronized.

As shown in Figure 2-5, a kind of cooperative communication method based on clustering spherical decoding in virtual MIMO of the present invention, this method comprises the following steps:

(1), in the virtual MIMO domain, the cluster head node (CH) selects several cooperative nodes (CN) from the collection node (ED) in the cluster, as shown in Figure 3, the specific steps are as follows:

(1-1) In the virtual MIMO domain, select the cluster head node (CH) according to the Responsive Distributed Clustering Algorithm (RDCA) according to the remaining energy of the wireless sensor network node power supply battery of all acquisition nodes (ED); The non-cluster head collection nodes select their respective cluster heads according to the cost function based on the closest distance with load balancing performance;

(1-2) After the clustering is formed, the cluster head node (CH) uses time division multiple access (TDMA) to allocate time slots to each collection node (ED) in the cluster to reduce transmission interference and facilitate data transmission within the cluster;

(1-3), the cluster head node (CH) will select several cooperative nodes (CN) from the collection nodes (ED) in the cluster;

(2), communication transmission in the local cluster, as shown in Figure 4, its specific steps are as follows:

(2-1), terminal acquisition node (ED) sends data to this cluster head node (CH) in turn according to the TDMA time slot that step (1-3) distributes, enters sleep state afterwards, saves battery energy;

(2-2), the cluster head node (CH) aggregates all the collected data together with its own data;

(2-3), then the aggregated data streams are serially converted into Nt sub-data streams and sent to corresponding cooperative nodes (CN) by broadcast;

(3), the long-distance communication transmission between the coordination node (CN) and the data center (AP), as shown in Figure 5, the specific steps are as follows:

(3-1), the coordinating nodes (CN) of each cluster perform space-time block coding (STBC) encoding on the data to counteract the spatial correlation between adjacent antennas in the cluster, obtain diversity gain, and reduce the error of the entire transmission system. code rate;

(3-2), the STBC data flow of each cluster is transmitted to the data center (AP) in a traditional VBLAST mode, assuming that all cluster head nodes (CH) are synchronized to obtain multiplexing gain;

(3-3) The cooperative VBLAST antenna node at the receiving end samples and quantizes the received data, and the processed data stream is sent to the data center (AP) for decoding and recovery, and the improved spherical decoding (SD) is adopted at the receiving end The algorithm decodes the received signal;

(4), define the threshold value that is less than the individual power supply value of the terminal collection node (ED) in wireless communication to be exhausted, and judge whether the individual power supply value of each terminal collection node is greater than the threshold value of power exhaustion in turn, if the terminal collection node If the value of the individual power supply is equal to or greater than the threshold of power depletion, go to step (1), and the terminal acquisition node participates in the next round of communication transmission, otherwise the terminal acquisition does not participate in the next round of communication transmission, and the communication transmission ends.

The processed data stream described in the above steps (3-3) is sent to the data center (AP) for decoding recovery, and the decoding recovery at the receiving end adopts an improved spherical decoding (SD) algorithm, which is as follows:

中搜索以ρ为半径的超球内的点，离球心最近的格点即为Y的译码，解码时以分簇为单位进行分组，然后对每个分块分别进行球形解码， The basic idea of the improved spherical decoding algorithm is to accept the signal vector Y as the center of the sphere, and set

Search for the points in the hypersphere with ρ as the radius, and the grid point closest to the center of the sphere is the decoding of Y. When decoding, it is grouped in units of clusters, and then each block is decoded separately.

Assuming Nc=3, Nt=4, Nr=3, divide the 3 clusters into two groups, the first group corresponds to two clusters, the second group corresponds to the remaining another cluster, the expression is :

(1)

为向量，，

为

向量，为的子向量， in ,

for vector, ,

for

vector, for subvector of

的维度为

， 

 = 4  则利用球形解码SD算法求解第二分组

，得：  set up

has a dimension of

,

= 4, use the spherical decoding SD algorithm to solve the second group

,have to:

                           (2)

(2)

为第二分组解向量格点集合，

为信号矢量第二分组向量，

为上三角矩阵R的第二行第二列分量，

为第二分组向量， in,

Solve the set of vector lattice points for the second group,

is the second grouping vector of signal vectors,

is the second row and second column component of the upper triangular matrix R,

is the second grouping vector,

，

的维度为

，

的表达式为：  Then use the spherical decoding SD algorithm to solve the first group

,

has a dimension of

,

The expression is:

(3)

为第二分组解向量格点集合，

为信号矢量第二分组向量，为上三角矩阵R的第i行第j列分量，

为第二分组向量, 

为第二分组解向量格点集合。 in,

Solve the set of vector lattice points for the second group,

is the second grouping vector of signal vectors, is the i-th row and j-th column component of the upper triangular matrix R,

is the second grouping vector,

Solve the set of vector lattice points for the second group.

In order to prove the superiority of a cooperative communication method based on clustered spherical decoding in virtual MIMO of the present invention, the following simulations are performed to compare the bit error rate (BER) performance of different algorithms under single clustering, single cluster and multi-cluster BER performance, packet sphere decoding and traditional sphere decoding BER performance, packet sphere decoding and traditional sphere decoding computational complexity, the comparison results are as follows:

As shown in Figure 6, in the figure, the vertical axis represents the bit error rate, and the horizontal axis represents the average bit energy/white noise power spectral density. ), the dashed line with a triangle is the bit error rate performance curve using the minimum mean square error algorithm (MMSE), and the dashed line with a diamond is the improved spherical decoding algorithm (SD) of the present invention Bit error rate performance curve. It can be seen from the comparison results in the figure that the bit error rate can be reduced by adopting the decoding method of the present invention, and the system performance can be improved.

As shown in Figure 7, in the figure, the vertical axis represents the bit error rate, and the horizontal axis represents the average bit energy/white noise power spectral density. The four dashed lines are: the dashed line with black dots is a single cluster with 4 nodes Adopt the bit error rate performance curve of STBC scheme below, the dashed line with triangle, the dashed line with cross and the dashed line with circle are 3 sub-clusters 4 cooperative nodes, adopt the STBC proposed by the present invention for multi-cluster transmission Three clustering BER performance curves of the combined scheme with VLBAST. It can be seen from the comparison results in the figure that the system multiplexing gain is obviously improved by adopting the method of the present invention, and the bit error rate of the system is reduced.

As shown in Figure 8, in the figure, the vertical axis indicates that the coordinate is the bit error rate, and the horizontal axis indicates the average bit energy/white noise power spectral density. The bit error rate performance curve of the algorithm (that is, the improved spherical decoding algorithm), and the curve with triangles is the bit error rate performance curve of the traditional spherical decoding algorithm. It can be seen from the comparison results in the figure that the decoding method of the present invention has a lower BER performance than the traditional decoding method, but the reduction is limited.

As shown in Figure 9, in the figure, the vertical axis is the simulation time, and the horizontal axis is the average bit energy/white noise power spectral density. The simulation time curve of the improved spherical decoding algorithm), and the curve with triangles is the simulation time curve of the traditional spherical decoding algorithm. It can be seen from the comparison results in the figure that the decoding method of the present invention greatly reduces the simulation time compared with the traditional decoding method under the same performance. Combining Fig. 8 and Fig. 9 and adopting the method of the present invention, although a certain bit error performance is sacrificed, the complexity of the algorithm can be greatly reduced.

Claims (4)

1. A cooperative communication method based on clustering sphere decoding in virtual MIMO is characterized in that a sensor node in a wireless sensor network is clustered by adopting a load balancing response type distributed clustering algorithm (RDCA), then a Cooperative Node (CN) of each cluster head is selected after clustering, data fused by the cluster head nodes (CH) are forwarded to the Cooperative Node (CN), and finally the data are jointly sent to a gathering point, a sphere decoding algorithm is adopted at a receiving end, the compromise between complexity and error rate performance is achieved, and the transmission efficiency of the wireless sensor network is improved, and the method specifically comprises the following steps:

(1) selecting a plurality of Cooperative Nodes (CN) from cluster acquisition nodes (ED) by cluster head nodes (CH) in the virtual MIMO domain;

(2) local intra-cluster communication transmission, the used time length is 2 states, one time length is

The other is

2 states appear in turn, and the duration in the whole communication transmission process is satisfied

；

(3) The time length for the remote communication transmission of the Cooperative Node (CN) and the data center (AP) is 2 states, one time length isThe other is

2 states appear in turn, and the duration in the whole communication transmission process is satisfied

；

(4) Defining a threshold value which is smaller than the independent power supply value of a terminal acquisition node (ED) in wireless communication and is used up for electric quantity, sequentially judging whether the independent power supply value of each terminal acquisition node is larger than the threshold value used up for electric quantity, if the independent power supply value of the terminal acquisition node is equal to or larger than the threshold value used up for electric quantity, turning to the step (1), enabling the terminal acquisition node to participate in next round of communication transmission, otherwise, enabling the terminal acquisition node not to participate in next round of communication transmission, and finishing the communication transmission.

2. The cooperative communication method based on cluster sphere decoding in virtual MIMO according to claim 1, wherein said step (1) selects a plurality of Cooperative Nodes (CN) from the cluster acquisition nodes (ED) in the virtual MIMO domain by the cluster head node (CH), and comprises the following steps:

(1-1) selecting a cluster head node (CH) in a virtual MIMO domain according to a response type distribution clustering algorithm (RDCA) according to the residual energy of a power supply battery of wireless sensor network nodes of all acquisition nodes (ED); the non-cluster-head acquisition nodes select respective cluster heads according to a cost function with load balancing performance and based on the shortest distance (closest);

(1-2) after clustering is formed, the cluster head node (CH) allocates time slots to each cluster acquisition node (ED) in a Time Division Multiple Access (TDMA) mode so as to reduce transmission interference and facilitate cluster data transmission;

(1-3) the cluster head node (CH) selects a plurality of Cooperative Nodes (CN) from the cluster collection nodes.

3. The cooperative communication method based on cluster sphere decoding in virtual MIMO according to claim 2, wherein the local intra-cluster communication transmission in the step (2) comprises the following specific steps:

(2-1) the terminal acquisition node (ED) sends data to the cluster head node (CH) in turn according to the TDMA time slots distributed in the step (1-3);

(2-2) the cluster head node (CH) converges all the collected data together with the data of the cluster head node;

and (2-3) converting the aggregated data stream into Nt sub-data streams in a serial-parallel mode, and broadcasting and transmitting the Nt sub-data streams to the corresponding Cooperative Nodes (CN).

4. The cooperative communication method based on cluster sphere decoding in virtual MIMO as claimed in claim 3, wherein said step (3) of remote communication transmission between the Cooperative Node (CN) and the data center (AP) comprises the following steps:

(3-1) the Cooperative Nodes (CN) of each cluster respectively carry out space-time block coding (STBC) coding on the data;

(3-2) transmitting the STBC data stream of each cluster to a data center (AP) by adopting a traditional VBLAST mode;

and (3-3) sampling and quantizing the received data by a cooperative VBLAS antenna node of the receiving end, sending the processed data stream to a data center (AP) for decoding and recovering, and decoding the received signal by adopting a Spherical Decoding (SD) algorithm at the receiving end.

Images