CN106911443B - Pilot tone optimum design method in compressed sensing based M2M communication system - Google Patents

Abstract

The present invention provides a pilot optimization design method in an M2M communication system based on compressed sensing. The method includes: determining the initial pilot sequence p _k of the user nodes according to the constellation point set of the modulation scheme adopted by the M2M communication system, and combining the initial pilot sequences of all user nodes into an initial pilot matrix P; The frequency matrix P is transformed from the complex number domain to the real number domain to obtain the matrix Q, and the optimization algorithm based on SVD decomposition is used to optimize the matrix Q. The present invention converts the initial pilot matrix to the real number domain, and then optimizes it using an optimization algorithm based on SVD decomposition. Under the conditions of different signal-to-noise ratios, the optimized pilot sequence and random pilot Compared with the matrix, the bit error rate is significantly reduced; or, in the case of the same bit error rate, the length of the required pilot sequence is shorter, so that the accuracy of multi-user access detection and joint channel estimation in the M2M communication system can be improved. .

Description

Pilot optimization design method in M2M communication system based on compressive sensing

technical field

The present invention relates to the technical field of M2M (Machine to Machine, machine to machine), and in particular to a method for optimal design of pilot frequency in M2M communication based on compressed sensing.

Background technique

M2M communication is the main manifestation of today's Internet of Things. With the explosive development of information technology, people are no longer limited to using machines to complete daily work. We hope that machines and equipment can pass without human intervention. Network interconnection to communicate and complete corresponding tasks. For example, the temperature and humidity sensors in the farmland collect data on the temperature and humidity of the land, and return the data to the server in the management center, and the server analyzes the data, and manages the farmland by adjusting the temperature and humidity controller ecological environment. This process does not require the participation of human activities, and the entire task is completed through mutual communication between devices.

Now LTE-Advanced has been used in M2M communication, but unlike the voice call system, user nodes in the M2M communication system usually have the characteristics of low activity rate and low data rate. Compared with the traditional communication system, the number of user nodes in the M2M communication system usually communicates at the same time is not large, and the data packets transmitted are relatively small, so the M2M communication system is a sparse communication system.

CS (Compressive Sensing, Compressed Sensing) is an emerging theory in recent years. Its core is to project a sparse or compressible high-dimensional signal to a low-latitude space through a specific matrix transformation. When performing signal reconstruction , using the sparse signal or the sparsity of the compressed signal to reconstruct the original signal using a linear or nonlinear restoration algorithm. In the channel estimation of the pilot method in the M2M communication system, due to the low activity of the user nodes, the channel impulse response corresponding to the inactive user node is regarded as zero value, and the channel impulse response of the active user node is regarded as non-zero value, that is, user access has sparse characteristics, so multi-user access detection and channel state information estimation can be performed through the corresponding compressed sensing signal reconstruction algorithm.

At present, the existing multi-user access detection technology needs to assign mutually orthogonal pilot sequences to each user, while non-orthogonal pilot sequences are used in multi-user access and joint channel estimation based on compressed sensing, so it can Save pilot resources. Currently, commonly used compressed sensing signal recovery algorithms include Matching Pursuit (MP), Orthogonal Matching Pursuit (OMP), Compressive Sampling Matching Pursuit (CoSaMP), etc.

In multi-user access and joint channel estimation based on compressed sensing, the pilot sequences of multi-users are usually randomly generated. However, the performance of multi-user access detection and joint channel estimation using randomly generated pilot sequences is not optimal. Therefore, it is urgent to design a new optimal design method of pilot sequence, and the pilot sequence generated by this method can further improve the accuracy of multi-user access detection and joint channel estimation.

Contents of the invention

Embodiments of the present invention provide a compressive sensing-based pilot optimization design method in an M2M communication system, so as to improve the accuracy of multi-user access detection and channel joint estimation in the M2M communication system.

In order to achieve the above object, the present invention adopts the following technical solutions.

A method for optimally designing pilots in an M2M communication system based on compressed sensing, further comprising:

Determine the initial pilot sequence p _k of the user nodes according to the constellation point set of the modulation scheme adopted by the M2M communication system, and combine the initial pilot sequences of all user nodes into an initial pilot matrix P;

Transforming the initial pilot matrix P from the complex domain to the real domain to obtain a matrix Q, and optimizing the matrix Q using an optimization algorithm based on SVD decomposition to obtain an optimized pilot matrix P _opti ;

The described initial pilot matrix P is transformed from complex number domain to real number domain to obtain matrix Q, including:

Define the function f( ) that matrix is converted into real number domain from complex number domain, utilize described function f( ) to carry out the conversion from complex number domain to real number domain to described initial pilot matrix P, obtain 2N * 2K dimension matrix Q;

where P _r represents the real part of matrix P, and P _i represents the imaginary part of matrix P;

The matrix Q is optimized using an optimization algorithm based on SVD decomposition, and the optimized pilot matrix P _opli is obtained, including:

① Perform column normalization on the 2N×2K dimensional matrix Q to obtain in, are respectively obtained by dividing each element of the corresponding column in the 2N×2K dimensional matrix Q by the square root of the sum of the squares of all elements of the column to which the element belongs;

②The initial set Ω is an empty set; the vector w with an initial length of K, the jth element of the vector w is The sum of squared correlations with the other columns plus The sum of squared correlations with other columns; the initial length is the array Γ of K, and the elements of the array Γ are the serial numbers of the elements of the vector w from large to small, so that j=Γ(1), where Γ(1) is the array The first element; where, The N is to randomly select N elements repeatedly;

③ Add element j to set Ω, use 2N×(2K-2) dimensional matrix express remove and After the remaining matrix, by pairing Carrying out SVD decomposition into the transposed matrix of the product of matrix U, matrix S and unitary matrix V, we get Then select the last column V _end in the unitary matrix V to update Let ρ denote the average energy of the set of modulation symbols, then update for

④Update the value of the real part of the pilot vector p _j as The first N elements of the update pilot vector p _j imaginary part value is The last N elements of ; update each element of p _j as the constellation point closest to the element, update according to steps ① and ②

⑤ Define a vector of length K The jth element of the vector is The sum of squared correlations with the other columns plus Correlation sum of squares with other columns; defines an array of length K Array elements in turn are vectors The sequence number of elements arranged from largest to smallest;

6. Make t=1; Wherein, t is the number of iterations;

⑦if Let j=t, let Skip to step ④; if and Let j=t, let Skip to step ④; if And Γ(t)∈Ω, let t=t+1;

⑧If t≤K, repeat step ⑦; if t>K, stop the iteration to obtain the optimal pilot matrix P _opli .

Further, according to the constellation point set of the modulation scheme adopted by the M2M communication system, the initial pilot sequence p _k of the user node is determined, and the initial pilot sequences of all user nodes are combined into an initial pilot matrix P, including :

Determine the constellation point set Λ corresponding to the modulation scheme adopted by the M2M communication system, and randomly select N elements from the constellation point set Λ to form the initial pilot sequence p _k ∈Λ ^N of the k-th user, and sequentially generate The initial pilot sequences of all user nodes, combine the initial pilot sequences of all user nodes into an initial pilot matrix P=[p ₁ ,p ₂ ,...,p _K ]∈ΛN ^×K , where K is the total number of users.

Further, the method also includes:

The optimized pilot matrix P _opti is used for multi-user access detection and channel joint estimation in the M2M communication system, and the channel estimation result is used to recover the information sent by the users.

Further, using the optimized pilot matrix P _opti to perform detection and joint channel estimation on multi-user access in the M2M communication system includes:

The channel impulse response corresponding to user node i is Where L _h represents the total number of tap delays of the discrete channel, and the observation result of the pilot sequence received at the receiving end is:

in Indicates the optimized pilot of user node i, * indicates convolution, n indicates additive Gaussian white

noise;

According to the matrix convolution transformation, the formula (1) is transformed to get

in representation vector convolution moment

array. Therefore, it can be further obtained from formula (2)

y _p =A _p h+n (6)

in represents a collection of pilot convolution matrices, while Represents the set of channel impulse responses of all user nodes, where K is the total number of users;

The sparse vector h is solved using the compressive sensing reconstruction signal algorithm.

From the technical solutions provided by the above-mentioned embodiments of the present invention, it can be seen that in the embodiments of the present invention, after converting the initial pilot matrix to the real number domain, and then using an optimization algorithm based on SVD decomposition for optimization, under different SNR conditions In this case, compared with the random pilot matrix, the bit error rate of the optimized pilot sequence optimized by the pilot sequence optimization design method of the embodiment of the present invention is significantly reduced; The length is shorter, so that the accuracy of multi-user access detection and joint channel estimation in the M2M communication system can be improved.

Additional aspects and advantages of the invention will be set forth in part in the description which follows, and will become apparent from the description, or may be learned by practice of the invention.

Description of drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following will briefly introduce the accompanying drawings that need to be used in the description of the embodiments. Obviously, the accompanying drawings in the following description are only some embodiments of the present invention. For Those of ordinary skill in the art can also obtain other drawings based on these drawings without any creative effort.

FIG. 1 is a schematic diagram of a sparse M2M communication system provided by an embodiment of the present invention;

Fig. 2 is a kind of flow chart that carries out pilot optimization design to matrix Q to obtain optimal pilot matrix P _opti provided by the embodiment of the present invention;

FIG. 3 is a schematic diagram of a principle of pilot sequence detection provided by an embodiment of the present invention;

Fig. 4 is a schematic diagram of a comparison of SNR and BER provided by an embodiment of the present invention (Pa=0.03, PL ₌ 48);

FIG. 5 is a schematic diagram of a comparison between a pilot length and a bit error rate provided by an embodiment of the present invention (SNR=8, O _a =0.04);

FIG. 6 is a schematic diagram of a comparison between pilot length and bit error rate provided by an embodiment of the present invention (SNR=16, P _a =0.04).

Detailed ways

Embodiments of the present invention are described in detail below, examples of which are shown in the drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below by referring to the figures are exemplary only for explaining the present invention and should not be construed as limiting the present invention.

Those skilled in the art will understand that unless otherwise stated, the singular forms "a", "an", "said" and "the" used herein may also include plural forms. It should be further understood that the word "comprising" used in the description of the present invention refers to the presence of said features, integers, steps, operations, elements and/or components, but does not exclude the presence or addition of one or more other features, Integers, steps, operations, elements, components, and/or groups thereof. It will be understood that when an element is referred to as being "connected" or "coupled" to another element, it can be directly connected or coupled to the other element or intervening elements may also be present. Additionally, "connected" or "coupled" as used herein may include wirelessly connected or coupled. As used herein, the term "and/or" includes any and all combinations of one or more of the associated listed items.

Those skilled in the art can understand that, unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. It should also be understood that terms such as those defined in commonly used dictionaries should be understood to have a meaning consistent with the meaning in the context of the prior art, and will not be interpreted in an idealized or overly formal sense unless defined as herein explain.

In order to facilitate the understanding of the embodiments of the present invention, several specific embodiments will be taken as examples for further explanation below in conjunction with the accompanying drawings, and each embodiment does not constitute a limitation to the embodiments of the present invention.

The embodiment of the present invention proposes a pilot sequence optimization design method for multi-user access detection and channel joint estimation applied in an M2M communication system, which can reduce the overhead of pilot resources and improve multi-user access detection and channel joint estimation. Estimated precision.

A schematic diagram of a sparse M2M communication system provided by an embodiment of the present invention is shown in FIG. 1. There are K user nodes in the sparse M2M communication system, and only A users in the system need to send data to the base station at the same time. , which is the user’s active probability Active users send their respective pilots, and the base station performs multi-user access detection and joint channel estimation through compressed sensing algorithms. Then the base station uses the estimated channel state information to estimate the data transmitted by the user subsequently.

The present invention provides a pilot optimization design method for multi-user access detection and channel joint estimation in an M2M communication system based on compressed sensing, which includes the following steps:

Step 1. Determine the initial pilot sequence p _k of user k according to the modulation scheme adopted by the system.

1. Determine the constellation point set Λ corresponding to the modulation scheme adopted by the system. For example, the Λ of the BPSK (Binary PhaseShift Keying) modulation scheme is [+1, -1], and the QPSK (Quadrature Phase ShiftKeying, Λ of the quadrature phase shift keying) modulation scheme is [1+i, 1-i, -1+i, -1-i], or other modulation schemes can be used.

2. N elements can be repeatedly randomly selected from the constellation point set Λ to form the initial pilot sequence p _k ∈Λ ^N of the kth user, and the initial pilot sequences of all user nodes are sequentially generated.

3. Combine the initial pilot sequences of all user nodes into an initial pilot matrix P=[p ₁ , p ₂ , . . . , p _K ]∈ΛN ^×K , where K is the total number of users.

Step 2. Transform the initial pilot matrix P from the complex domain to the real domain to obtain the matrix Q. The matrix Q is optimized by using an optimization algorithm based on SVD decomposition, and an optimized pilot matrix P _opli is obtained.

Since the elements in the constellation point set Λ are complex numbers, the elements in the matrix P are also complex numbers. Therefore, the present invention defines a function f(·) that converts the matrix from the complex number domain to the real number domain, then we have Where P _r represents the real part of matrix P, and P _i represents the imaginary part of matrix P. Using the pilot sequence optimization method based on Singular value decomposition (SVD) proposed by the present invention, the matrix Q is optimized to obtain the optimal pilot matrix P _opti .

Fig. 2 is a kind of flow chart that matrix Q is carried out pilot optimization design to obtain optimum pilot matrix _Popli that the embodiment of the present invention provides, and specific processing process comprises:

① Transform the initial pilot matrix P from the complex domain to the real domain to obtain a 2N×2K dimensional matrix

② Perform column normalization on the 2N×2K dimensional matrix Q to obtain in, are respectively obtained by dividing each element of the corresponding column in the 2N×2K dimensional matrix Q by the root sum of the squares of all elements of the column to which the element belongs;

③The initial set Ω is an empty set; the vector w with an initial length of K, the jth element of the vector w is The sum of squared correlations with the other columns plus The sum of squared correlations with other columns; an array Γ with an initial length of K, and the elements of the array Γ are the sequence numbers of the elements of the vector w arranged from large to small. Let j=Γ(1), where Γ(1) is the first element of the array.

④ Add element j to set Ω, use 2N×(2K-2) dimensional matrix express remove and After the remaining matrix, by pairing SVD is decomposed into the transposed matrix of the product of matrix U, matrix S and unitary matrix V, and we get Then select the last column V _end in the unitary matrix V to update Let ρ denote the average energy of the set of modulation symbols, then update for

⑤Update the value of the real part of the pilot vector p _j as The first N elements of the update pilot vector p _j imaginary part value is The last N elements of ; update each element of p _j as the constellation point closest to the element, update according to steps ① and ②

⑥ Define a vector with length K The jth element of the vector is The sum of squared correlations with the other columns plus Correlation sum of squares with other columns; defines an array of length K Array elements in turn are vectors The sequence number of elements arranged from largest to smallest.

⑦Let t=1, where t is the number of iterations.

⑧ if Let j=t, let Skip to step ④; if and Let j=t, let Skip to step ④; if And Γ(t)∈Ω, let t=t+1;

⑨If t≤K, repeat step ⑧; if t>K, then stop the iteration to obtain the optimal pilot matrix P _opti .

Step 3: Use the optimized pilot matrix P _opti to detect multi-user access in the M2M communication system and jointly estimate the channel, and use the channel estimation result to recover the information sent by the users.

A schematic diagram of the principle of pilot sequence detection provided by an embodiment of the present invention is shown in Figure 3, including the following processing procedures;

① The channel impulse response corresponding to user node i is Among them, L _h represents the total number of tap delays of discrete channels. Then the observation result of the pilot sequence received at the receiver is

in Indicates the optimized pilot of user node i, * indicates convolution, and n indicates additive white Gaussian noise;

②According to the matrix convolution transformation, the formula (4) is transformed to get

in representation vector convolution moment

array. Therefore, it can be further obtained from formula (5)

y _p =A _p h+n (9)

in Represents a set of pilot convolution matrices, where K is the total number of users, and Represents the set of channel impulse responses of all user nodes.

③Because the channel information of inactive users is regarded as zero elements, and the user's activity rate P _a <<1, h is a sparse vector, so formula (6) can use compressive sensing to reconstruct the signal algorithm, such as group orthogonality Matching pursuit algorithm (Group Orthogonal Matching Pursuit, GOMP) or group least absolute shrinkage and selection algorithm (Groupleast absolute shrinkage and selection operator, Group Lasso) solves this formula.

The pilot sequence optimization algorithm based on SVD decomposition proposed by the present invention is in the process of each round of iterative solution. The result obtained is not a pilot sequence, but a real number vector that does not meet the constraints of the constellation point set of the modulation scheme. The vector should be converted first to the complex number field, and then select the closest constellation point as the element of the new pilot sequence according to the distance.

In summary, the embodiment of the present invention proposes a design method for optimizing a multi-user pilot sequence composed of constellation point elements. It can be seen from FIG. 4 that under different SNR conditions, after Compared with the pilot sequence optimized by the pilot sequence optimization design method of the present invention and the random pilot matrix, the bit error rate is obviously reduced. As can be seen from Fig. 5 and Fig. 6, in the situation of low signal-to-noise ratio (8dB) and high signal-to-noise ratio (16dB), after using the pilot sequence optimal design method of the present invention, in the same situation of pilot length The lower bit error rate is significantly reduced, or the length of the required pilot sequence is shorter when the bit error rate is the same. Therefore, the accuracy of multi-user access detection and channel joint estimation in the M2M communication system can be improved.

Usually, the pilot sequence used in channel estimation is a randomly generated sequence. The present invention combines compressed sensing technology and considers the elements of the modulated signal set of the pilot sequence, and the randomly generated sequence in the M2M communication system The pilot matrix composed of pilot sequences is optimized based on SVD decomposition, so that the optimized pilot matrix can improve the accuracy of multi-user access detection and joint channel estimation in different situations.

Those skilled in the art can understand that the accompanying drawing is only a schematic diagram of an embodiment, and the modules or processes in the accompanying drawing are not necessarily necessary for implementing the present invention.

It can be seen from the above description of the implementation manners that those skilled in the art can clearly understand that the present invention can be implemented by means of software plus a necessary general hardware platform. Based on this understanding, the essence of the technical solution of the present invention or the part that contributes to the prior art can be embodied in the form of software products, and the computer software products can be stored in storage media, such as ROM/RAM, disk , CD, etc., including several instructions to make a computer device (which may be a personal computer, server, or network device, etc.) execute the methods described in various embodiments or some parts of the embodiments of the present invention.

Each embodiment in this specification is described in a progressive manner, the same and similar parts of each embodiment can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the device or system embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and for relevant parts, refer to part of the description of the method embodiments. The device and system embodiments described above are only illustrative, and the units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, It can be located in one place, or it can be distributed to multiple network elements. Part or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. It can be understood and implemented by those skilled in the art without creative effort.

The above is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed in the present invention can easily think of changes or Replacement should be covered within the protection scope of the present invention. Therefore, the protection scope of the present invention should be determined by the protection scope of the claims.

Claims (4)

1. A pilot optimization design method in an M2M communication system based on compressed sensing is characterized by comprising the following steps:

determining an initial pilot sequence p of a user node according to a constellation point set of a modulation scheme adopted by an M2M communication system_kCombining the initial pilot sequences of all user nodes into an initial pilot matrix P;

performing complex number domain to real number domain conversion on the initial pilot frequency matrix P to obtain a matrix Q, and optimizing the matrix Q by adopting an optimization algorithm based on SVD decomposition to obtain an optimized pilot frequency matrix P_opti；

The converting the initial pilot matrix P from a complex number domain to a real number domain to obtain a matrix Q includes:

defining a function f (-) for converting a matrix from a complex number domain to a real number domain, and performing complex number domain to real number domain conversion on the initial pilot frequency matrix P by using the function f (-) to obtain a 2 Nx 2K dimensional matrix Q;

wherein P is_rRepresenting the real part of the matrix P, P_iRepresents the imaginary part of the matrix P;

the matrix Q is optimized by adopting an optimization algorithm based on SVD decomposition to obtain an optimized pilot matrix P_optiThe method comprises the following steps:

firstly, carrying out column normalization on the 2 Nx 2K dimensional matrix Q to obtainWherein,the square of each element in the corresponding column in the 2 Nx 2K dimensional matrix Q is divided by the square sum of all elements in the column to which the element belongs;

the initial set omega is an empty set; a vector w of initial length K, the jth element of the vector w beingSum of squares of correlation with other columns plusThe sum of squared correlation with other columns; the array Γ is the initial length K, the elements of the array Γ are sequentially the serial numbers of the elements of the vector w arranged from large to small, and j is equal to Γ (1), wherein Γ (1) is the 1 st element of the array; wherein,n is repeatedly and randomly selecting N elements;

③ adding element j into set omega, using 2 Nx (2K-2) dimensional matrixTo representIn (1) removingAndthe matrix left over later, by pairThe SVD is decomposed into a transposed matrix of the product of the matrix U, the matrix S and the unitary matrix V to obtainThe last column V in the unitary matrix V is selected_endUpdateLet ρ represent the average energy of the modulation symbol set, then updateIs composed of

Updating the pilot frequency vector p_jThe value of the real part isFirst N elements of (d), update pilot vector p_jThe value of the imaginary part isThe last N elements of (1); updating p_jEach element of (1) is a constellation point nearest to the element, and is updated according to the steps of (i) and (ii)

Defining vector with length of KThe jth element of the vector isSum of squares of correlation with other columns plusThe sum of squared correlation with other columns; defining an array of length KThe array elements being in turn vectorsThe sequence numbers of the elements arranged from large to small;

sixthly, making t equal to 1; wherein t is the number of iterations;

seventhly, ifLet j equal t, letJumping to the step (IV); if it is notAnd isLet j equal t, letJumping to the step (IV); if it is notAnd Γ (t) ∈ Ω, let t ═ t + 1;

eighthly, if t is not larger than K, repeating the step (seventhly); if t is more than K, stopping iteration to obtain the optimal pilot frequency matrix P_opti。

2. The method of claim 1, wherein the initial pilot sequence p of the user node is determined according to a set of constellation points of a modulation scheme adopted by the M2M communication system_kCombining the initial pilot sequences of all user nodes into an initial pilot matrix P, including:

determining a constellation point set Lambda corresponding to a modulation scheme adopted by an M2M communication system, and repeatedly and randomly selecting N elements from the constellation point set Lambda to form an initial pilot sequence p of a kth user_k∈Λ^NSequentially generating initial pilot sequences of all user nodes, and combining the initial pilot sequences of all user nodes into an initial pilot matrix P ═ P₁，p₂，...，P_k]∈Λ^N×KAnd K is the total number of users.

3. The method of claim 1, further comprising:

using the optimized pilot matrix P_optiAnd detecting multi-user access in the M2M communication system, performing channel joint estimation, and recovering information sent by a user by using a channel estimation result.

4. The method of claim 3 wherein the optimized pilot matrix P is utilized_optiDetecting and channel joint estimation for multi-user access in an M2M communication system, comprising:

the channel impulse response corresponding to the user node i isWherein L is_hThe total number of tap time delays of the discrete channel is represented, and the observation result of the pilot frequency sequence received at the receiving end is as follows:

juquePilot frequency of the user node i after representing optimization represents convolution, and n represents additive white Gaussian noise;

transforming the formula (1) according to a matrix convolution transformation

WhereinRepresenting a vectorThe convolution matrix of (a); thus, the compound of formula (2) can further obtain

y_p＝A_ph+n (3)

WhereinRepresents a set of pilot convolution matrices, K being the total number of users, anda set of channel impulse responses representing all user nodes;

and solving the sparse vector h by using a compressed sensing reconstruction signal algorithm.