A d2d communication multicast method based on equal-angle clustering and adaptive bandwidth allocation

Abstract

The application discloses a D2D communication multicast method based on equal-angle clustering and adaptive bandwidth allocation. The terminal of a D2D multicast communication system under a cellular network comprises a D2D transmitting terminal (cluster head) and a D2D receiving terminal, wherein the D2D transmitting terminal comprises S D2D transmitting users and the D2D receiving terminal comprises R D2D receiving users. The method comprises the following steps: solving the communication distance and the number of clusters under the base station which make the system time delay minimum according to a brute force search algorithm; clustering the users through the equal-angle clustering algorithm of the application and selecting the optimal cluster head according to the distance, and the cluster head obtains data from the base station; and the cluster head distributes the data to the users in the cluster which can establish D2D reliable communication through the D2D multicast mode, wherein the cluster head performs bandwidth allocation according to the optimal solution obtained through the proportional bandwidth allocation algorithm of the application. The application can efficiently utilize the spectrum resource of the cellular network and effectively reduce the transmission time delay of the multicast communication.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of D2D multicast communication, in particular to a D2D communication multicast method based on equal-angle clustering and adaptive bandwidth allocation. BACKGROUND

[0002] In the traditional cellular network, the direct communication between mobile devices is not allowed, even if the transmitting terminal and the receiving terminal are very close, the control signaling and data need to be transferred through the core network. D2D communication technology can effectively reduce the burden of the base station, avoid congestion, and reduce the transmission power of the terminal device and the transmission delay. As a relay, D2D communication technology can greatly increase the coverage of the cellular network and effectively improve the system capacity.

[0003] However, due to the difference in channel state between users, the transmission rate of users is greatly different, which makes it difficult for the multicast transmitting terminal, such as the base station, to transmit data at a rate suitable for all receiving terminals. In most cases, the base station will select the multicast transmission rate according to the user with the worst channel condition to ensure that each receiving terminal can successfully receive. Therefore, when most (but not all) receiving terminals are in good channel conditions and can transmit at high rates, one or a few receiving terminals with poor channel conditions may become the bottleneck of the throughput and other performances of multicast communication. Therefore, the D2D multicast technology combines the traditional cellular network multicast technology with D2D communication, and transmits the multicast data correctly received by the receiving terminal with good channel conditions to the receiving terminal with relatively poor channel conditions through the D2D link. In this way, even if a few receiving terminals may have very poor channel conditions, the BS can still multicast at a high data rate.

[0004] Therefore, we need to design how to efficiently form a specific multicast cluster and how to select the best cluster head user for each multicast cluster. In the existing D2D multicast scheme under the cellular network, most of them do not consider the coupling problem of clustering, cluster head selection and bandwidth resource allocation when D2D forwarding. In view of these problems, the present application proposes a D2D communication multicast method based on equal-angle clustering and adaptive bandwidth allocation, which efficiently utilizes the spectrum resources of the cellular network and reduces the transmission delay. SUMMARY

[0005] The purpose of the present application is to propose a D2D communication multicast method based on equal-angle clustering and adaptive bandwidth allocation for how to efficiently form a specific multicast cluster and how to select the best cluster head user for each multicast cluster, and under the constraint of bandwidth resources, allocate bandwidth to different clusters to make the delay of D2D multicast forwarding as short as possible.

[0006] Technical scheme: In order to achieve the above-mentioned purpose of the application, the present application adopts the following technical scheme:

[0007] A D2D communication multicast method based on equal-angle clustering and adaptive bandwidth allocation, characterized in that it comprises the following steps:

[0008] (1) initializing system parameters, including cell radius R, base station transmission power P BS , D2D transmission terminal power P T , multicast transmission packet size D, thermal noise power n0, system transmission bandwidth B, path loss system α, terminal number N, and terminal location information, including step length j and δ in proportional bandwidth algorithm;

[0009] (2) calculating the number of clusters S, base station communication distance d BS，Rs , and retransmission communication distance d Rs，Zr that minimize the total delay T of the communication system according to the brute force search algorithm;

[0010] (3) clustering according to the equal-angle clustering algorithm proposed in the application, and selecting the optimal cluster head according to the distance;

[0011] (4) after obtaining the D2D user attachment relationship, finding the worst user of the D2D communication link in all clusters, and calculating the channel threshold C th according to the user, and allocating bandwidth to different clusters according to the proportional bandwidth allocation algorithm proposed in the application.

[0012] 2. In the embodiment, the function T calculation formula in step (2) is:

[0013]

[0014] wherein the retransmission communication distance d Rs，Zr can be represented by the base station communication distance d BS，Rs , that is wherein represents the included angle between adjacent cluster heads,

[0015] 3. In the embodiment, the users are clustered according to the following steps in step (3):

[0016] (3.1) let the initial value of s be 1, obtain the positions of all users, and the horizontal and vertical coordinates of user i are represented as x_user i and y_user i,

[0017] (3.2) calculate the included angle of user i

[0018] (3.3) calculate the user node and the base station communication radius d BS，RsThe distance between the nodes is found, and the node with the minimum value and its angle are recorded as Rs and w respectively, and the node is taken as the first cluster head;

[0019] (3.4) Find the angle between the user nodes that satisfies The nodes are denoted as Zr, and they are attached to the cluster head, b Rs，Zr =1, otherwise b Rs，Zr =0;

[0020] (3.5) Let s = s + 1, and according to the angle w of the first cluster head, we can get the horizontal and vertical coordinates of the sth theoretical cluster head point. The calculation formulas are:

[0021] (3.6) Calculate the distance between the user node and the theoretical cluster head, find the node with the minimum value and its angle, record them as Rs and w respectively, take this node as the sth cluster head, and get the attachment relationship b according to step (3.4) Rs，Zr ;

[0022] (3.7) Repeat steps (3.5)-(3.6) until s=S.

[0023] 4. In the implementation scheme, in step (4), bandwidth is allocated to the cluster head according to the following steps:

[0024] (4.1) According to the formula Get the channel threshold C th The value range is 0<C th ≤C m ;

[0025] (4.2) According to Shannon's formula and channel threshold C th The ratio of the bandwidth of the worst node in each cluster’s D2D link to the total bandwidth B is calculated using the bisection method and is denoted as λ. s , and according to Get μ;

[0026] (4.3) If μ>1, according to C th =C th -j·|μ-1|·B adjusts the channel threshold C th Otherwise, according to C th =C th +j·|μ-1|·B adjusts the channel threshold C th The value of

[0027] (4.4) Repeat steps (4.2)-(4.3) until |1-μ|<δ;

[0028] (4.5) If μ>1, find λ sfind the maximum value in the set and reassign the value plus mu-1; otherwise find the minimum value in the set and reassign the value plus 1-mu. s find the maximum value in the set and reassign the value plus mu-1; otherwise find the minimum value in the set and reassign the value plus 1-mu.

[0029] Beneficial effects: compared with the prior art, the clustering and relay selection method for D2D multicast communication of the cellular network of the application can adaptively select the number of cluster heads in multicast retransmission and the distance from the cluster head to the base station, and at the same time, an optimal bandwidth resource allocation mechanism is given. The results show that the application can more evenly distribute users in each cluster, thereby effectively reducing the system delay, improving the edge user experience and system performance. BRIEF DESCRIPTION OF DRAWINGS

[0030] Figure 1 Method flowchart implemented by the application;

[0031] Figure 2 Cellular D2D multicast communication system model diagram of the method of the application;

[0032] Figure 3 Principle diagram of the clustering algorithm in the application;

[0033] Figure 4 Delay variation with cell user number comparison diagram of the algorithm in the application and other algorithms;

[0034] Figure 5 Delay variation with cell radius comparison diagram of the algorithm in the application;

[0035] Figure 6 Delay variation with transmission power comparison diagram of the algorithm in the application and other algorithms. DETAILED DESCRIPTION

[0036] The application will be further illustrated below in conjunction with the drawings and specific embodiments, and it should be understood that these embodiments are only used to illustrate the application and not to limit the scope of the application, and after reading the application, those skilled in the art can modify various equivalent forms of the application, which all fall within the scope defined by the claims attached to the application.

[0037] The scene setting and parameter setting will be analyzed in detail below, and the application is not limited to the specific scene and parameter setting in the embodiments.

[0038] 1. Classification and number of mobile terminals

[0039] In the D2D multicast communication system under the cellular network, the base station sends broadcast data packets to users in the cell. Due to the different channel conditions from the base station to the user nodes, the nodes with good channel conditions can receive the data packets in a timely and correct manner, while other users with poor channel conditions need to forward them through D2D communication. The multicast receiving end set with low link quality will be served by the multicast receiving end with high link quality through the D2D mode. Figure 2 As shown in Figure 1, multicast is divided into two transmission phases. The first phase is multicast transmission from the base station to the cluster head, and the second phase is forwarding from the cluster head to other nodes. Assuming that the users in the cell are uniformly distributed and the total number of users is N, multiple D2D multicast groups (D2DMs) can be formed. Each D2DM consists of a D2D forwarder and multiple D2D receivers. The number of D2D transmitting users is denoted as S, and the number of D2D receiving users is denoted as R. S + R = N, so the cluster head and D2D receiving user sets can be denoted as: A = R1, R2, ..., R s , Z=Z1,Z2,...,Z R During D2D communication transmission, each cluster uses orthogonal frequency division multiplexing to seek the best clustering and relay selection method to complete the multicast task with the lowest possible delay.

[0040] 2. Establishment of channel model

[0041] The transmitting terminal uses a fixed transmission power, denoted as P T Assume that the channel model between the transmitting terminal and the receiving terminal is a free space attenuation model, that is, P r / P t =1 / r α , where P r is the power received by the receiving terminal, P T is the transmit power of the transmitting terminal, r is the distance between terminals, and α is the path loss factor.

[0042] 3. System latency

[0043] In the first phase of D2D multicast communication, the latency of the base station downlink transmission phase can be expressed as follows:

[0044]

[0045] Among them, P BS is the base station transmission power, d BS，Rs is the distance from the base station to the sth cluster head Rs, the noise power spectral density n0, B is the channel bandwidth, and D is the packet size of multicast transmission.

[0046] In the base station to cluster head transmission stage, the base station can use all the spectrum bandwidth resources B, and in the transmission process of cluster head to the rest of the nodes, the bandwidth B will be divided for all cluster heads to use. The time delay of the second stage is that all cluster heads complete the forwarding to all other nodes in their clusters, which is expressed as follows:

[0047]

[0048] Where, P T is the data transmission power of the node, d Rs，Zr is the distance from the s-th cluster head Rs to the r-th D2D receiving terminal user Rs, λ Rs is the proportion of the total channel bandwidth of the s-th multicast cluster, and the binary symbol b Rs，Zr is used to represent the attachment relationship between the s-th cluster head Rs and the r-th D2D receiving terminal user Rs.

[0049] Finally, the total system time delay is recorded as:

[0050]

[0051]

[0052]

[0053]

[0054]

[0055] Based on the above theoretical basis, the clustering and relay selection method of the D2D multicast communication of the cellular network is designed.

[0056] As Figure 1 shown, the D2D communication multicast method based on equal-angle clustering and adaptive bandwidth allocation disclosed by the embodiment of the application comprises the following steps:

[0057] (1) Initialize system parameters, which include cell radius R, base station transmission power P BS , D2D transmitting terminal power P T , multicast transmission data packet size D, thermal noise power n0, system transmission bandwidth B, path loss system α, terminal number N and terminal position information, including step length j and δ in the proportional bandwidth algorithm.

[0058] (2) According to the brute force search algorithm, the cluster number S, the communication distance d BS，Rs from the base station and the retransmission communication distance d Rs，Zr that make the total time delay T of the communication system minimum are calculated, and the function T calculation formula is as follows:

[0059]

[0060] wherein the retransmission communication distance d Rs，Zr may be expressed as BS，Rs , that is wherein represents the included angle between adjacent cluster heads,

[0061] (3) The equal-angle clustering algorithm is used for clustering, and the optimal cluster head is selected according to the distance. The equal-angle clustering algorithm includes the following steps:

[0062] 1) Let the initial value of s be 1, and obtain the positions of all users. The horizontal and vertical coordinates of the user i are represented as x_user i and y_user i respectively,

[0063] 2) Calculate the included angle of the user i

[0064] 3) Calculate the distance between the user node and the communication radius d BS，Rs issued by the base station, find the minimum value corresponding to the node and its angle, and record them as Rs and w respectively. The node is taken as the first cluster head;

[0065] 4) Find the node whose included angle satisfies , and record them as Zr. Attach them to the cluster head, and b Rs，Zr = 1, otherwise b Rs，Zr = 0;

[0066] 5) Let s = s + 1, and obtain the horizontal and vertical coordinates of the s-th theoretical cluster head according to the included angle w of the first cluster head. The calculation formulas are respectively

[0067] 6) Calculate the distance between the user node and the theoretical cluster head, find the minimum value corresponding to the node and its angle, and record them as Rs and w respectively. The node is taken as the s-th cluster head, and the attachment relationship b Rs，Zr is obtained according to step (3.4);

[0068] 7) Repeat steps 5) to 6) until s = S.

[0069] (4) After obtaining the attachment relationship of the D2D user, find the worst user in the D2D communication link in all clusters, and calculate the channel threshold C th according to the user. Different clusters are given bandwidth allocation according to the proportional bandwidth allocation algorithm proposed in the application. The proportional bandwidth allocation algorithm includes the following steps:

[0070] 1) Obtain the value range of the channel threshold C th according to the formula 0 < Cth ≤C m ;

[0071] 2) Order According to Shannon's formula and channel threshold C th The ratio of the bandwidth of the worst node in each cluster’s D2D link to the total bandwidth B is calculated using the bisection method and is denoted as λ. s , and according to Get μ;

[0072] 3) If μ>1, according to C th =C th -j·|μ-1|·B adjusts the channel threshold C th Otherwise, according to C th =C th +j·|μ-1|·B adjusts the channel threshold C th The value of

[0073] 4) Repeat steps 2)-3) until |1-μ|<δ;

[0074] 5) If μ>1, find λ s The maximum value among them and re-assign it by adding μ-1; otherwise find λ s The minimum value in is obtained and the value is re-assigned by adding 1-μ.

[0075] like Figure 3 As shown in the figure, the present invention takes 5 clusters as an example. The hexagon in the figure is the D2D transmitter. The three sides of the five triangles of the same size and shape represent the transmission communication radius from the base station to the D2D transmitter, the communication radius within the cluster and the cell radius. BS，Rs The blue dotted circle with a radius of is the transmission communication range from the base station to the D2D transmitter, and the cluster heads are distributed on the blue dotted circle. The angle between two adjacent cluster heads and the base station is 72 degrees. The angle between adjacent cluster heads is equal and is represented by express, In the figure, d Rs，Zr The red dotted circle with a radius of represents the transmission communication range from the red cluster head to other nodes in its cluster.

[0076] To verify the feasibility of the clustering and relay selection method for D2D multicast communication in cellular networks proposed in this paper, MATLAB simulation experiments were conducted on three performance indicators: the number of cell users, cell radius, and transmit power. The clustering method was compared with two classic clustering algorithms, K-means and AP; and the bandwidth allocation method was compared with random bandwidth allocation and equal bandwidth allocation.

[0077] Figure 4is a comparison chart of time delay of the algorithm in the application and other algorithms varying with the number of users in a cell. To verify the advantage of the method of the application over the prior art, the application sets the following simulation parameters: the transmission power is 28dBm, and the cell radius is 400m. With the increase of the number of D2D users, the time delay under the K-means algorithm presents a downward trend. This is because the more the users, the more uniform the clustering effect of K-means; while the AP algorithm presents an upward trend, because the number of clusters generated after automatic clustering is large and uneven, and part of the cluster heads are closer to the cell edge, resulting in longer transmission time delay. Compared with the other two algorithms, the algorithm of the application is less affected by the number of users in a cell under the given cell radius, transmission power and other parameters, because the clustering is analyzed based on angle in the application.

[0078] Figure 5 is a comparison chart of time delay of the algorithm in the application varying with the cell radius. To verify the feasibility of the method of the application, the application sets the following simulation parameters: the transmission power is 28dBm, and the number of users in a cell is 200. Compared with multicast without relay, the multicast service after introducing D2D communication has a significant decrease in time delay. Because the bandwidth is allocated randomly without considering the channel quality of users in a cluster, the time delay is longer than that of the average bandwidth scheme and the proportional bandwidth scheme, and the user with the worst channel quality in a certain cluster is allocated the smallest bandwidth, resulting in a time delay close to that of no relay transmission. Compared with the other three algorithms, the algorithm of the application can obtain the lowest time delay.

[0079] Figure 6 is a comparison chart of time delay of the algorithm in the application and other algorithms varying with the transmission power. To verify the advantage of the method of the application over the prior art, the application sets the following simulation parameters: the number of users in a cell is 200. With the increase of the transmission power, the transmission time delay of different algorithms presents a significant downward trend. Because the number of clusters generated by the AP algorithm is large and the cluster centers are distributed randomly, when the transmission power is higher than 28dBm, the time delay of the AP algorithm is larger than that without relay. Compared with the other algorithms, the algorithm of the application can obtain the lowest time delay.

Claims

1. A D2D communication multicast method based on equi-angle clustering and adaptive bandwidth allocation, characterized in that: The method comprises the following steps: (1) Initialize system parameters, including cell radius R, base station transmission power P BS , the power P of the D2D transmitting terminal T , the packet size D of multicast transmission, the thermal noise power n0, the system transmission bandwidth B, the path loss system α, the number of terminals N, and the location information of the terminals, including the step size j and δ in the proportional bandwidth algorithm; (2) According to the brute force search algorithm, calculate the number of clusters S and the angle between adjacent cluster heads that minimize the total delay T of the communication system. The base station sends the communication distance d BS，Rs And the retransmission communication distance d Rs，Zr ; (3) Clustering is performed according to the equal-angle clustering algorithm, and the optimal cluster head is selected based on the distance, specifically including: (3.1) Let the initial value of s be 1, and obtain the positions of all users. The horizontal and vertical coordinates of user i are expressed as (3.2) Calculate the angle of user i (3.3) Calculate the communication radius d between the user node and the base station BS，Rs The distance between the nodes is found, and the node with the minimum value and its angle are recorded as Rs and ω respectively, and the node is used as the first cluster head; (3.4) Find the angle between the user nodes that satisfies The node is recorded as Zr, and Zr is attached to the cluster head. Rs，Zr =1, otherwise b Rs，Zr =0; (3.5) Let s = s + 1, and according to the angle ω of the first cluster head, we can get the horizontal and vertical coordinates of the sth theoretical cluster head. The calculation formulas are (3.6) Calculate the distance between the user node and the theoretical cluster head, find the node with the minimum value and its angle, record them as Rs and ω respectively, take this node as the sth cluster head, and get the attachment relationship b according to step (3.4) Rs，Zr ; (3.7) Repeat steps (3.5)-(3.6) until s=S; (4) After obtaining the D2D user attachment relationship, find the user with the worst D2D communication link in all clusters and calculate the channel threshold C based on this user. th , bandwidth is allocated to different clusters according to the proportional bandwidth allocation algorithm, specifically including: (4.1) According to the formula Get the channel threshold C th The value range is 0<C th ≤C m , A represents the cluster head, Z represents the set of D2D receiving end users; (4.2) According to Shannon's formula and channel threshold C th The ratio of the bandwidth of the worst node in each cluster’s D2D link to the total bandwidth B is calculated using the bisection method and is denoted as λ. s , and according to Get μ; (4.3) If μ>1, according to C th =C th -j·|μ-1|·B adjusts the channel threshold C th Otherwise, according to C th =C th +j·|μ-1|·B adjusts the channel threshold C th The value of (4.4) Repeat steps (4.2)-(4.3) until |1-μ|<δ; (4.5) If μ>1, find λ s The maximum value among them and re-assign it by adding μ-1; otherwise find λ s The minimum value in is re-assigned by adding 1-μ to the minimum value.

2. A D2D communication multicast method based on equi-angle clustering and adaptive bandwidth allocation according to claim 1, characterized in that: The calculation formula of function T in step (2) is: Among them, the retransmission communication distance d Rs，Zr Use the base station to send the communication distance d BS，Rs To express, that is

Images