Fast self-adaptive cellular-network-free clustering method

Through the fast adaptive clustering method based on leakage antennas, dynamically forming and merging AP clusters, the problem of dense user cluster overload and time-consuming access processes in traditional methods is solved, and the scalability and high transmission of cellular networks are achieved rate.

CN119012212APending Publication Date: 2024-11-22FUDAN UNIVERSITY
View PDF 0 Cites 2 Cited by

Patent Information

Application Number
CN202410985949.5
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-23
Publication Date
2024-11-22

AI Technical Summary

Technical Problem

The traditional cellular terahertz network clustering method leads to overload problems in dense user clusters, and the traditional phased array antenna access process takes a long time, hindering the deployment and scalability of large-scale cellular networks.

Method used

A fast adaptive cellular network clustering method based on leakage antenna is adopted, and the affinity propagation algorithm and the radiation characteristics of the leakage antenna are used to dynamically form AP clusters, merge AP clusters to improve spectrum efficiency, and avoid interference through frequency division multiplexing.

Benefits of technology

It realizes scalability of cellular-free networks, reduces system overhead, improves transmission rates, supports the deployment of large-scale cellular-free terahertz networks and reliable and efficient services.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119012212A_ABST
    Figure CN119012212A_ABST
Patent Text Reader

Abstract

The invention belongs to the technical field of large-scale cellular-free networks, and particularly relates to a fast self-adaptive cellular-free network clustering method based on leaky-wave antennas. The method comprises the following steps: firstly, constructing a cellular-free terahertz communication network based on a leaky-wave antenna, and giving a signal-to-interference noise ratio of a downlink of a user when any precoding scheme is provided; the self-adaptive non-cellular network clustering method comprises the following steps: forming an initialized AP cluster by using an affinity propagation algorithm; combining the AP clusters which do not provide the signal transmission service for the user and the AP clusters which provide the signal transmission service for the user in the clusters according to a single AP spectrum efficiency maximization criterion; and when the spectrum efficiency of each AP in the cluster is improved after the two AP clusters are combined, the two AP clusters are further combined. On the basis, a frequency division multiplexing method is provided to eliminate AP inter-cluster interference. According to the method, a higher transmission rate can be realized, and the user service quality and the resource utilization rate of a cellular-free network are improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention belongs to the technical field of large-scale non-cellular network, and in particular relates to a fast adaptive non-cellular network clustering method. Background Art

[0002] 6G cellular-free network architecture may evolve according to different levels of cooperation, and access point (AP) clusters play a vital role in practical deployment. In large-scale cellular-free terahertz downlink networks with densely distributed APs, AP clusters with inter-CPU coordination are essential for scalability, that is, AP clusters composed of multiple APs coordinate to serve multiple users, and each cluster is controlled by a single CPU. In addition, AP clusters will help reduce the signal processing complexity under large-scale cellular-free networks, because the large-scale cellular-free network signal processing of each cluster is in a lower-dimensional space, so it is necessary to build an AP cluster solution.

[0003] Regarding the research on clustering methods for cellular-free terahertz networks, the traditional network-centric clustering is used, where the radius of a base station cluster is calculated as the minimum and maximum distance between base stations in the same cluster, and each user is associated with a base station by following the optimal channel rule. However, such a design forms a fixed AP cluster and may cause overload problems in some clusters with dense users, which is not scalable. Therefore, we need to develop a scalable large-scale cellular-free network AP clustering method.

[0004] Regarding the selection of terahertz antennas, the traditional phased array antenna access process and beam search take a long time, which hinders the deployment of large-scale non-cellular terahertz networks. Leaky wave antennas are low-cost, easy-to-produce traveling wave antennas with the characteristics of space-frequency coupling, narrow transmission beams, and high gain, making them a better choice to support the deployment of large-scale non-cellular terahertz networks. Summary of the invention

[0005] The object of the present invention is to provide a cellular-free terahertz network clustering method based on leaky wave antennas to reduce system overhead, increase transmission rate and enhance scalability.

[0006] The fast adaptive non-cellular network clustering method provided by the present invention firstly constructs a non-cellular terahertz network based on leaky wave antennas, wherein the signal-to-interference-noise ratio of the user downlink is calculated when any precoding scheme is adopted.

[0007] In a non-cellular terahertz network, APs are uniformly deployed at a high density. Each AP is equipped with a multi-surface leaky-wave antenna with 360-degree coverage as a transmitting antenna, and the user end is equipped with an omnidirectional antenna as a receiving antenna. The effective antenna gain of the leaky-wave antenna is:

[0008]

[0009] Where ξ is the radiation efficiency factor, L is the antenna aperture length, ρ is the attenuation coefficient, f is the center frequency of the subchannel, θ is the line-of-sight propagation angle in the transmission direction, c is the speed of light in vacuum, and f co is the cut-off frequency of the leaky-wave antenna.

[0010] There are a total of K users in the network. For each user, the number of APs that can provide services is recorded as M. The downlink channel matrix is:

[0011]

[0012] Where H = [h1…h k …h K ] T is the conventional channel matrix with omnidirectional antennas, h k =[h 1k ,…,h Mk ] T is the channel vector between the kth user and the AP, θ km is the line-of-sight propagation angle between the mth AP and the kth user.

[0013] For any precoding matrix F, the kth column vector of the column-normalized precoding matrix is The signal-to-interference-noise ratio of the kth user downlink is:

[0014]

[0015] Among them, q k is the power spectrum density of the data signal transmitted to the kth user, δ 2 is the power spectral density of the noise.

[0016] The fast adaptive non-cellular network clustering method based on leaky wave antennas has the following specific steps:

[0017] (1) Use the affinity propagation algorithm (Frey BJ, Dueck D. Clustering by passing messages between data points [J]. Science, 2007, 315 (5814): 972-976) to form an initialized AP cluster. This is an effective method to adaptively find network-centric clusters without initializing the number of AP clusters. In the information transmission process of the affinity propagation algorithm, the negative Euclidean distance between the mth AP and its sample m′th AP is used as the similarity S (m, m′) between them, and the median of the similarity is selected as the reference. Assuming that all APs have the same chance of becoming the center of an AP cluster, a network-centric AP cluster is obtained. In order to determine the association between each user and the AP cluster, the present invention adopts a fast initial access method based on the radiation characteristics of leaky wave antennas (Lin Z, Wang L, Tan B, et al. Spatial-spectral terahertz networks[J]. IEEE Transactions on Wireless Communications, 2021, 21(6): 3881-3892), according to the line-of-sight propagation angle θ in the transmission direction. mk (θ mk is the line-of-sight propagation angle between the mth AP and the kth user), the maximum radiation level of the leaky wave antenna can be achieved by using the following frequencies:

[0018]

[0019] In this way, the kth user and the mth AP at the frequency Downlink received signal strength for:

[0020]

[0021] Each user chooses to associate with the desired AP that provides the maximum received signal strength: That is, the mth * APs, and then each user is served by the AP cluster that includes its desired AP.

[0022] (2) Based on the obtained AP cluster Some AP clusters do not provide signal transmission services to any users. This type of AP cluster is defined as Remove those AP clusters that do not provide signal transmission services to any users and The remaining AP clusters that provide signal transmission services to users are defined as An AP cluster that does not provide signal transmission services to any user Should be connected to the AP cluster that provides signal transmission services to users Merge, select an AP cluster that does not provide signal transmission services to any user Then select it and AP cluster Merge to maximize the spectrum efficiency of each AP in the merged AP cluster:

[0023]

[0024] in, represents the merged AP cluster, represents the number of APs in the AP cluster after merging, γ k represents the signal-to-interference-to-noise ratio of the kth user’s downlink. AP cluster that provides signal transmission services to users After the merger, the merged AP cluster is obtained

[0025] (3) Based on the obtained AP cluster If the spectrum efficiency of each AP in the cluster is improved after two AP clusters are merged, the two AP clusters should be further merged:

[0026]

[0027] in, represents the merged AP cluster, Indicates the number of APs in the merged AP cluster. The AP clusters whose spectrum efficiency of each AP in the merged cluster is further improved are further merged to obtain the final merged AP cluster.

[0028] Compared with the existing methods, the fast adaptive non-cellular network clustering method proposed in the present invention has the following advantages: it does not need to give the number of final clusters in advance, avoiding the problem of fixed AP cluster number, and fully considering the nature of user access link in the AP clustering process, which is scalable. The leaky wave antenna used is a low-cost, easy-to-produce traveling wave antenna with the characteristics of space-frequency coupling, narrow transmission beam and high gain, which is a better choice to support large-scale non-cellular terahertz network deployment. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 The present invention is a flow chart of a method for clustering a cellular-free terahertz communication network based on leaky wave antennas.

[0030] Figure 2 This is a structural model diagram of the cellular-free terahertz communication network clustering method based on leaky wave antennas of the present invention.

[0031] Figure 3 and Figure 4 It is a schematic diagram of the performance effect of the cellular-free terahertz communication network clustering method based on leaky wave antennas of the present invention. DETAILED DESCRIPTION

[0032] In order to make the purpose and technical solution of the present invention clearer and easier to understand, the present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0033] The cellular-free terahertz network constructed by the present invention has a structure as follows: Figure 2 shown.

[0034] In a non-cellular terahertz network, APs are uniformly deployed at a high density. Each AP is equipped with a multi-surface leaky-wave antenna with 360-degree coverage as a transmitting antenna, and the user end is equipped with an omnidirectional antenna as a receiving antenna. The effective antenna gain of the leaky-wave antenna is:

[0035]

[0036] Where ξ is the radiation efficiency factor, L is the antenna aperture length, ρ is the attenuation coefficient, f is the center frequency of the subchannel, θ is the line-of-sight propagation angle in the transmission direction, c is the speed of light in vacuum, and f co is the cutoff frequency of the leaky antenna. There are a total of K users in the network. For each user, the number of APs that can provide services is recorded as M. The channel matrix of the downlink is:

[0037]

[0038] Where H = [h1…h k …h K ] T is the traditional channel matrix based on omnidirectional antennas, h k =[h 1k ,…,h Mk ] T is the channel vector between the kth user and the AP, θ km is the line-of-sight propagation angle between the mth AP and the kth user.

[0039] For any precoding matrix F, the kth column vector of the column-normalized precoding matrix is The signal-to-interference-noise ratio of the kth user downlink is:

[0040]

[0041] Among them, q k is the power spectrum density of the data signal transmitted to the kth user, δ 2 is the power spectral density of the noise.

[0042] The hierarchical clustering method proposed in the present invention comprises the following specific steps:

[0043] (1) The affinity propagation algorithm is used to form an initial AP cluster. During the information transmission process of the affinity propagation algorithm, the negative Euclidean distance between the mth AP and its sample m′th AP is used as the similarity S(m, m′) between them, and the median of the similarity is selected as the reference degree. It is assumed that all APs have the same chance to become the center of the AP cluster, thereby obtaining a network-centered AP cluster. In order to determine the association between each user and the AP cluster, a fast initial access method based on the radiation characteristics of leaky wave antennas is adopted. mk , the maximum radiation level of the leaky wave antenna is achieved by using the following frequencies:

[0044]

[0045] In this way, the kth user and the mth AP at the frequency The downlink received signal strength is for:

[0046]

[0047] Each user chooses to associate with the desired AP that provides the maximum received signal strength: That is, the mth * APs, and then each user is served by the AP cluster that includes its desired AP.

[0048] (2) Based on the obtained AP cluster Some AP clusters do not provide signal transmission services to any users. This type of AP cluster is defined as Remove those AP clusters that do not provide signal transmission services to any users and The remaining AP clusters that provide signal transmission services to users are defined as An AP cluster that does not provide signal transmission services to any user Should be connected to the AP cluster that provides signal transmission services to users Merge, select an AP cluster that does not provide signal transmission services to any user Then select it and AP cluster Merge to maximize the spectrum efficiency of each AP in the merged AP cluster:

[0049]

[0050] in, represents the merged AP cluster, represents the number of APs in the AP cluster after merging, γ k represents the signal-to-interference-to-noise ratio of the kth user’s downlink. AP cluster that provides signal transmission services to users After the merger, the merged AP cluster is obtained

[0051] (3) Based on the obtained AP cluster If the spectrum efficiency of each AP in the cluster is improved after two AP clusters are merged, the two AP clusters should be further merged:

[0052]

[0053] in, represents the merged AP cluster, Indicates the number of APs in the merged AP cluster. The AP clusters whose spectrum efficiency of each AP in the merged cluster is further improved are further merged to obtain the final merged AP cluster.

[0054] After the clustering of the non-cellular terahertz network is completed, the interference between AP clusters is avoided by using frequency division multiplexing. Each AP cluster provides the same sub-channel with the same center frequency; different AP clusters provide different sub-channels with different center frequencies. After clustering is achieved, the total transmission rate of all users is:

[0055]

[0056] For each AP cluster obtained, the average transmission rate of users in the cluster should be greater than the minimum average transmission rate. This ensures reliable and efficient service quality of the non-cellular network. Therefore, the non-cellular terahertz network obtained after clustering needs to meet the following conditions 1:

[0057]

[0058] in, Indicates an AP cluster The number of users in .

[0059] This algorithm allocates candidate subchannels to AP clusters according to the following two situations. The specific steps are as follows:

[0060] (1) For each AP cluster, condition 1 should be met first. AP clusters that do not meet condition 1 will be given priority to occupy candidate sub-channels. Among these AP clusters that do not meet condition 1, candidate sub-channels are preferentially allocated to the AP cluster that can maximize the total transmission rate of all users in the network.

[0061] (2) After all AP clusters meet condition 1, the remaining unassigned candidate subchannels will be used to improve the total transmission rate of the non-cellular terahertz network. Each unassigned candidate subchannel is assigned to the AP cluster that can maximize the total transmission rate of all users in the network.

[0062] In this embodiment, all APs comply with equal power allocation to provide communication services to users. The radiation efficiency factor ξ is 1, the leaky wave antenna aperture length L is 0.15m, the attenuation coefficient ρ is 130rad / m, and the total transmission power of the downlink is The noise power spectral density is δ 2 is -168dBm / Hz, the total number of subchannels is 30, all APs and users are independently and evenly distributed in an area of ​​200m×200m, the absolute antenna elevation angle difference between APs and users is evenly distributed within 7-10m, and the propagation angle θ between the mth AP and the kth user mk Evenly distributed in The free space path loss model is used as the channel model, and the zero-forcing precoding scheme is adopted. The K-means partition clustering scheme with a predetermined number of clusters is used for comparison, and the predetermined number of clusters K is 2 and 3.

[0063] Specific implementation effects such as Figure 3 and Figure 4 As shown, the sub-channel frequencies are all between 200 GHz and 300 GHz. Figure 3 Shows the total bandwidth B total When the bandwidth is 10 GHz and the number of users is 10, the total transmission rate obtained by the two clustering methods changes with the increase in the number of APs. It can be seen that the hierarchical clustering scheme we proposed achieves the maximum transmission rate. Compared with the K-means partition clustering scheme with a predetermined number of clusters, the proposed scheme has more significant advantages when more APs are deployed. Figure 4 Shows the total bandwidth B total When the bandwidth is 10 GHz and the number of APs is 64, the total transmission rate obtained by the two clustering methods changes with the increase in the number of users. The proposed hierarchical clustering scheme still achieves the maximum transmission rate, and it can be clearly seen that different clustering schemes will have a huge impact on the total transmission rate. However, the proposed hierarchical clustering scheme has obvious advantages over the K-means partition clustering scheme with a pre-given number of clusters.

[0064] In summary, the fast adaptive non-cellular network clustering method of the present invention can effectively reduce system complexity and significantly improve system throughput.

Claims

1. A fast adaptive non-cellular network clustering method, characterized in that: Firstly, a cell-free terahertz communication network based on leaky wave antennas is constructed, where the signal-to-interference-noise ratio of the user downlink is studied when any precoding scheme is adopted; In a non-cellular terahertz communication network, APs are uniformly deployed with high density. Each AP is equipped with a multi-surface leaky-wave antenna with 360-degree coverage as a transmitting antenna, and the user end is equipped with an omnidirectional antenna as a receiving antenna. The effective antenna gain of the leaky-wave antenna is: Where ξ is the radiation efficiency factor, L is the antenna aperture length, ρ is the attenuation coefficient, f is the center frequency of the subchannel, θ is the line-of-sight propagation angle in the transmission direction, c is the speed of light in vacuum, and f co is the cutoff frequency of the leaky-wave antenna; There are a total of K users in the network. For each user, the number of APs that can provide services is recorded as M. The downlink channel matrix is: Where H = [h1…h k …h K ] T is the conventional channel matrix with omnidirectional antennas, h k =[h 1k ,…,h Mk ] T is the channel vector between the kth user and the AP, θ km is the line-of-sight propagation angle between the mth AP and the kth user; For any precoding matrix F, the kth column vector of the column-normalized precoding matrix is The signal-to-interference-noise ratio of the kth user downlink is: Among them, q k is the power spectrum density of the data signal transmitted to the kth user, δ 2 is the power spectral density of the noise; The specific steps of the fast adaptive non-cellular network clustering method based on leaky wave antenna are as follows: (1) The affinity propagation algorithm is used to form an initial AP cluster. During the information transmission process of the affinity propagation algorithm, the negative Euclidean distance between the mth AP and its sample m′th AP is used as the similarity S(m, m′) between them, and the median of the similarity is selected as the reference degree. It is assumed that all APs have the same chance to become the center of the AP cluster, thereby obtaining the network-centered AP cluster {C z }, in order to determine the association between each user and the AP cluster, a fast initial access method based on the radiation characteristics of the leaky wave antenna is adopted, according to the line-of-sight propagation angle θ in the transmission direction mk , i.e., the line-of-sight propagation angle between the mth AP and the kth user, by using the following frequencies to achieve the maximum radiation level of the leaky wave antenna: In this way, the kth user and the mth AP at the frequency Downlink received signal strength Each user chooses to associate with the desired AP that provides the maximum received signal strength: That is, the m*th AP, then each user is served by the AP cluster that includes its desired AP; (2) The obtained AP cluster {C z }, some AP clusters do not provide signal transmission services to any users. This type of AP cluster is defined as Remove the AP clusters that do not provide signal transmission services to any user, and replace the original AP clusters {C z The remaining AP clusters that provide signal transmission services to users are defined as {C z′ }; AP cluster that does not provide signal transmission service to any user The AP cluster that provides signal transmission services to users z′ Merge, select an AP cluster that does not provide signal transmission services to any user Then, combine it with the AP cluster {C z′ }Merge to maximize the spectrum efficiency of each AP in the merged AP cluster: in, represents the merged AP cluster, represents the number of APs in the AP cluster after merging, γ k represents the signal-to-interference-to-noise ratio of the kth user's downlink; the AP cluster that does not provide signal transmission services to any user The AP cluster that provides signal transmission services to users z′ After the merger, the merged AP cluster is obtained. (3) Obtained AP cluster In the example, if the spectrum efficiency of each AP in the cluster is improved after two AP clusters are merged, the two AP clusters are further merged: in, represents the merged AP cluster, Indicates the number of APs in the merged AP cluster; further merge the AP clusters whose spectrum efficiency of each AP in the merged cluster is further improved to obtain the final merged AP cluster

Citation Information

Cited By

  • Terahertz cellular-free receiver extrapolation method based on multilateral Schrodinger bridge

    CN122554013A

  • A terahertz cell-free receiver extrapolation method based on multi-edge trace schrodinger bridge

    CN122554013B