Many-to-one relay selection method for distributed network

Abstract

The invention discloses a many-to-one relay selection method oriented to a distributed network, relates to the technical field of distributed network communication, studies the fairness optimization problem of a global transmission rate in a wireless network with limited relay nodes, considers the performance of source nodes of the whole network in a resource allocation process, and considers the overall performance of relays at the same time, thereby improving the resource allocation efficiency. Therefore, the average performance of the relay network is optimized, and the stable result of the whole network is realized. The method is characterized in that both a source node and a relay node have own decision-making capabilities, the source node selects a relay link for efficient transmission to meet own transmission rate requirements, and the relay node knows transmission rate information of the source node connected with the relay node and decides whether to accept a request of the source node or not according to a selection standard of the relay node. And the transmission rates of all source nodes are maximized by optimizing the relay selection strategy of the source nodes, so that the optimal relay distribution result of the whole network is realized.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of distributed network communication, in particular to a multi-to-one relay selection method for distributed network. BACKGROUND

[0002] With the rapid development of wireless communication technology, distributed network has been widely used in the fields of Internet of Things, Internet of Vehicles, emergency communication and the like due to its self-organization, high expansibility and invulnerability. However, the characteristics such as dynamic node, resource limitation and channel time variation result in the challenges of reduced transmission reliability, limited coverage range and unbalanced energy consumption. Relay communication technology can effectively expand the network coverage range, enhance the reliability of the link and improve the system capacity through the cooperation and forwarding among nodes, and has become a key supporting technology for improving the performance of distributed network. In the distributed network, the relay node can effectively overcome the channel defects such as path loss and multipath fading by forwarding the signal of the source node, and improve the overall capacity and user service quality of the network. In addition, the traffic scheduling of the relay node can balance the network load, avoid congestion in hot spot areas and improve the spectrum utilization.

[0003] Although the relay communication technology has significant advantages in the communication of distributed network, its performance is largely dependent on the reasonable allocation of relay resources. In the dynamic and complex distributed network, the fairness optimization of relay allocation is particularly critical. On the one hand, the energy difference of nodes, channel time variation and dynamic change of topology may cause some nodes to be in a high load state for a long time, which aggravates the unbalanced energy consumption of the network. On the other hand, when multiple users compete for the same relay node, if only the instantaneous channel quality or local information is used as the basis, it is easy to cause unbalanced resource allocation and affect the long-term stability of the network. There are numerous nodes in the network, and reasonable allocation of relay resources can greatly increase the transmission capacity and efficiency of the network.

[0004] Therefore, from the perspective of optimizing the overall transmission rate of the network, how to solve the problem of relay node allocation in wireless network to achieve global optimization of relay distribution has not yet been related. SUMMARY

[0005] Therefore, the present application provides a multi-to-one relay selection method for distributed network, which can study the fairness optimization problem of global transmission rate in the limited wireless network of relay nodes, consider the performance of the source nodes in the whole network and the overall performance among relays in the resource allocation process, so that the relay network reaches the optimization of average performance and realizes the stable result of the whole network.

[0006] To achieve the above purpose, the technical scheme of the present application includes the following steps: Step 1, constructing a relay network system model of multi-source nodes to one destination node, including a source node set , a relay node set , destination node set .

[0007] Step 2, all source nodes in the network build relay node preference list; each source node ranks relay nodes based on transmission rate, and the source node applies for connection according to the ranking of relay nodes in the preference list, and the relay node accepts all connection applications and feeds back the actual transmission rate that the source node can get, and ranks the source nodes by transmission rate to form a source node list.

[0008] Step 3, all source nodes compare the current transmission rate with the transmission rate that can be achieved in direct transmission mode; if the transmission rate that can be achieved by the current connection is lower than that in direct transmission mode, the source node gives up the current relay connection and tries to select the next relay node in the preference list.

[0009] Each source node that does not establish a connection with a relay node updates the preference list until only the direct transmission mode can be selected.

[0010] Step 4, the relay node randomly selects a source node connected to itself , the source node finds a better relay node ; if the connection between the source node and the relay node can improve the total transmission rate of and two relay nodes, the relay node accepts the application of the source node ; at the same time, the relay node deletes the source node from the source node list connected to itself.

[0011] Step 5, if in the selection process, the relay selection strategy of all source nodes cannot be changed, it means that the network has reached a stable solution, and the output is the final distributed many-to-one selection result; otherwise, steps 3-5 are repeated.

[0012] Further, in step 1, a relay network system model of multiple source nodes to one destination node is constructed, specifically: In the relay network, the source node applies for forwarding to the relay node to achieve cooperative spatial diversity, the relay node regularly broadcasts its own position information, and forwards the received signal to the destination node; the source node and the relay node select appropriate items from the opposite set according to their own selection criteria.

[0013] There are N source nodes, M relay nodes and K destination nodes; among them, the source node set is: , the relay node set is: and destination nodes ; one source node is allowed to connect to at most one relay node, while one relay node is allowed to connect to multiple source nodes.

[0014] For a source-relay node pair, the selection criterion for each source node is to find a relay node that can provide the maximum transmission rate to fulfill its own transmission rate requirement; define the relayed node served source node s transmission rate as:

[0015] relay node s transmission rate as:

[0016] where is an indicator function, when source node successfully connects to relay node , , otherwise equals 0.

[0017] Further, in step 2, when multiple source nodes including source node select the same relay node , source node will share the relay link resource with other source nodes.

[0018] Source node constructs a preference list according to transmission rate ; where is the real transmission rate that source node can get.

[0019] s preference list is: ; this preference relationship represents that source node s best choice is , if refuses s request, will apply to the next relay node , and so on. indicates that no relay node is selected.

[0020] The relay node accepts all the applied source nodes according to the first-come-first-served principle, allocates time resources evenly, and rejects other source nodes; the rejected source nodes receive feedback and find and apply to the next relay connection; similarly, the relay node feeds back the real transmission rate that the received source nodes can get to the source nodes and ranks the source nodes according to the transmission rate to form a source node list.

[0021] Further, For source nodes The achievable real transmission rate:

[0022] wherein, denotes the number of source nodes accessing the relay node; denotes the transmission rate transmitted by, forwarded by and received by; denotes the signal-to-noise ratio from the transmission of a source node to the reception at the relay node; denotes the signal-to-noise ratio from the transmission of a source node to the reception at the destination node; denotes the signal-to-noise ratio from the transmission of a relay node to the reception at the destination node; denotes the channel bandwidth; wherein denotes the transmission to the destination node with the assistance of a relay and denotes the direct communication of a source node with a destination node , i.e. direct mode.

[0023] Further, step 3, in particular: The source node updates the preference list, in particular for each source node when the transmission rate with the selected relay node is smaller than the transmission rate in direct mode, i.e. the current relay node is abandoned and the preference list is changed; until a relay node is found, which results in a transmission rate through the relay node which is greater than or equal to the transmission rate in direct mode, i.e. , .

[0024] Each source node without a connection to a relay node will update the preference list until only the direct mode is left. Source nodes which do not meet the requirements for changing the relay node remain connected to the original relay node.

[0025] Further, step 4, for all source nodes connected to a relay node one of the source nodes is randomly selected to suggest it to find a better relay selection; when the source node finds a new relay node​ When applying, it will carry the previously connected relay node The transmission rate of the source node allows the relay node to compare the transmission rates and make a choice; if the source node Connecting to a relay node When the source node The transmission efficiency is greater than that of the source node Connecting to a relay node The transmission efficiency when , then the source node To relay node Make a connection request; if this option allows the relay node With relay nodes The total transmission rate Increase, then the relay node Accept source node Otherwise, the application will be rejected, that is, the source node Will not abandon the selected relay node ; Source node Before connecting to another relay node After that, the previous relay node Remove it from the list.

[0026] Beneficial effects: (1) The present invention provides a many-to-one relay selection method for distributed networks, which is characterized in that: the source node and the relay node have their own decision-making capabilities. The source node selects a relay link with efficient transmission to meet its own transmission rate requirements. The relay node knows the transmission rate information of the source node connected to it and decides whether to accept the source node's request based on its selection criteria. By optimizing the relay selection strategy of the source node, the transmission rate of all source nodes is maximized, thereby achieving the optimal relay allocation result for the entire network. It can be seen that the present invention studies the fairness optimization problem of the global transmission rate in a wireless network with limited relay nodes. By considering the performance of the source nodes of the entire network in the resource allocation process and the overall performance between relays, the relay network achieves the optimization of the average performance and achieves a stable result for the entire network.

[0027] (2) The present invention improves the fairness of the global transmission rate of a wireless network with limited relay nodes. By effectively sharing the relays, the source nodes promote all source nodes to achieve a balance in their transmission rates, thereby achieving a balanced distribution of resources across the entire network.

[0028] (3) The present invention takes into account the performance of source nodes in the entire network during resource allocation, and improves the performance of all source nodes as much as possible. In addition, during the relay allocation process, it recommends that source nodes with poor performance find better relay nodes for connection, thereby solving the fairness problem of relay allocation in the relay network and achieving global optimization of relay allocation. BRIEF DESCRIPTION OF DRAWINGS

[0029] Figure 1 is a relay network system model of multiple source nodes to one destination node; Figure 2 is a flow chart of a many-to-one relay selection algorithm for a distributed network. DETAILED DESCRIPTION

[0030] The application will be described in detail below with reference to the accompanying drawings and embodiments.

[0031] A many-to-one relay selection method for a distributed network is proposed by the application, and a flow thereof is shown in Figure 2 The fairness of global transmission rate in a wireless network with limited relay nodes is researched, and the average performance of the relay network is optimized by considering the performance of all source nodes and the overall performance between relays in the resource allocation process, so as to achieve stable results in the whole network. The algorithm has the following characteristics: the source nodes and the relay nodes have their own decision-making ability, the source nodes select efficient relay links to meet their own transmission rate requirements, the relay nodes know the transmission rate information of the source nodes connected to them, and decide whether to accept the request of the source nodes according to the selection criteria. The transmission rate of all source nodes is maximized by optimizing the relay selection strategy of the source nodes, so as to achieve the optimal relay allocation result in the whole network.

[0032] The technical scheme of the application comprises the following steps: Step 1, constructing a relay network system model of multiple source nodes to one destination node, including a source node set , a relay node set , and a destination node set . The source nodes make forwarding applications to the relay nodes to achieve cooperative spatial diversity, the relay nodes periodically broadcast their own position information, and forward the received signals to the destination node. The source nodes and the relay nodes select appropriate items in the opposite set according to their respective selection criteria.

[0033] Step 2, all source nodes in the network construct a relay node preference list. Each source node arranges the relay nodes in order of preference based on the transmission rate, the source nodes make connection applications according to the relay node ranking in the preference list, the relay nodes accept all connection applications and feed back the actual transmission rate that the source nodes can get, and the source nodes are arranged in order of transmission rate to form a source node list.

[0034] Step 3, the source nodes update the preference list. All source nodes compare the current transmission rate with the transmission rate that can be achieved in the direct transmission mode. If the transmission rate obtained by the current connection is lower than the transmission rate that can be achieved in the direct transmission mode, i.e. , the source nodes The current relay connection will be abandoned and the next relay node in the preference list will be selected. Each source node that has not established a connection with a relay node will update the preference list until only direct transmission mode can be selected.

[0035] Step 4, relay node Randomly select a source node that has established a connection with itself Ask it to find a better relay node If the source node The connection with the relay node Can improve The total transmission rate of the two relay nodes, the relay node Will accept the application of the source node At the same time, the source node Will be deleted from the source node list connected by the relay node .

[0036] Step 5, if the relay selection strategy of all source nodes cannot be changed in the selection process, it means that the network has reached a stable solution, and the output is the final distributed many-to-one selection result; otherwise, steps 3-5 are repeated.

[0037] In the embodiment of the application, for step 1, a relay network system model of multiple source nodes to a destination node is constructed, as shown in Figure 1 In the relay network, there are N source nodes, M relay nodes and K destination nodes. The source node set is: , the relay node set is: , and the destination node . A source node is allowed to be connected to at most one relay node, and a relay node can be connected to multiple source nodes at the same time.

[0038] For a source-relay node pair, the selection criterion of each source node is to find a relay node that can provide the maximum transmission rate, so as to realize the transmission rate requirement of each source node. The transmission rate of the source node Served by the relay node Is defined as:

[0039] A relay node knows the transmission rate of the source node connected to it, and is more inclined to serve a transmission link with a high transmission rate, so as to maximize its own transmission rate. The transmission rate of the relay node Can be written as:

[0040] Wherein,​ is the indicator function, when the source node Successfully connected to the relay node hour, , otherwise it is equal to 0.

[0041] In the embodiment of the present invention, for step 2, when the source node is included Multiple source nodes select the same relay node , source node The relay link resources will be shared with other source nodes. The actual transmission rate that can be obtained for:

[0042] in, Indicates access to a relay node The number of source nodes; Indicated by emission, Forward and Received transmission rate; It represents the signal-to-noise ratio transmitted from the source node to the relay node; It represents the signal-to-noise ratio transmitted from the source node to the destination node; It represents the signal-to-noise ratio transmitted from the relay node to the destination node; represents the channel bandwidth; express In the relay With the help of ,and Represents the source node With the destination node Communicate directly, that is, direct transmission mode.

[0043] Source Node According to the transmission rate , build a preference list. Example Preference list: This preference relationship represents the source node The best choice is ,if Rejected Request, Will be the next relay node Submit an application, and so on.

[0044] Relay nodes accept all requesting source nodes on a first-come, first-served basis, evenly distribute time resources, and reject other source nodes. Rejected source nodes receive feedback and re-apply for the next relay connection. Similarly, relay nodes provide feedback to the receiving source nodes regarding their actual available transmission rate and rank the source nodes by transmission rate to form a source node list.

[0045] In the embodiment of the present invention, for step 3, the source node updates the preference list. , when selecting relay nodes The transmission rate is lower than the direct transmission rate, that is, hour, Abandon the current relay node , change the preference list; until the relay node is found , so that through the relay node The transmission rate is greater than or equal to the transmission rate of the direct transmission mode, that is, , .

[0046] Each source node that has not established a connection with a relay node will update its preference list until only direct transmission mode can be selected. Source nodes that do not meet the requirements for changing relay nodes will remain connected to the original relay node. Indicates that no relay node is selected.

[0047] In the embodiment of the present invention, for step 4, for connecting to the relay node All source nodes of Randomly select one of the source nodes It is recommended to find a better relay option. To the new relay node When applying, it will carry the previously connected relay node The transmission rate of the source node allows the relay node to compare the transmission rates and make a choice. Connecting to a relay node When the source node The transmission efficiency is greater than that of the source node Connecting to a relay node The transmission efficiency when , then the source node To relay node Make a connection request. If this option allows the relay node With relay nodes The total transmission rate Increase, then the relay node Accept source node Otherwise, the application will be rejected, that is, the source node Not to abandon the selected relay node . Source node After determining to connect another relay node The previous relay node Will be removed from the list.

[0048] In the embodiment of the present application, for step 5, if no source node can change the current selection in the process of selecting the relay node, it indicates that the whole network has reached a stable solution, and the output is the final distributed many-to-one selection result.

[0049] To sum up, the above is only a preferred embodiment of the present application, and is not used to limit the protection scope of the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A many-to-one relay selection method for distributed networks, characterized in that: The method comprises the following steps: Step 1, constructing a relay network system model of a plurality of source nodes to a destination node, including a source node set , a relay node set , a destination node set ; Step 2, all source nodes in the network construct a relay node preference list; Each source node ranks the relay nodes according to the transmission rate, and the source node applies for connection according to the order of the relay nodes in the preference list, the relay node accepts all connection applications and feeds back the actual transmission rate that the source node can obtain, and ranks the source nodes according to the transmission rate to form a source node list; Step 3, all source nodes compare the current transmission rate with the transmission rate that can be achieved in the direct transmission mode; if the transmission rate that can be obtained by the current connection is lower than that in the direct transmission mode, the source node gives up the current relay connection and tries to select the next relay node in the preference list; Each source node that has not established a connection with a relay node updates the preference list until only the direct transmission mode can be selected; Step 4, relay node Randomly select a source node to establish connection with itself , source node Find a better relay node ; if the connection between the source node and the relay node can improve and the total transmission rate of the two relay nodes, the relay node accepts the application of the source node ; at the same time, the source node will be deleted from the source node list of the relay node connected Step 5, if the relay selection strategy of all source nodes cannot be changed in the selection process, it is indicated that the network has reached a stable solution, and the output is the final distributed many-to-one selection result; otherwise, steps 3-5 are cycled.

2. The many-to-one relay selection method for a distributed network of claim 1, wherein, In step 1, a relay network system model of multiple source nodes to one destination node is constructed, specifically: In the relay network, the source node applies for forwarding to the relay node to achieve spatial diversity, the relay node regularly broadcasts its own position information, and forwards the received signal to the destination node; the source node and the relay node select appropriate items in the opposite set according to their own selection criteria; There are N source nodes, M relay nodes and K destination nodes; wherein the source node set is: , the relay node set is: , and the destination node ; A source node is allowed to be connected with at most one relay node, and a relay node is simultaneously connected with multiple source nodes; For each source-relay node pair, the selection criterion for each source node is to find a relay node that can provide the maximum transmission rate to fulfill its own transmission rate requirement; the transmission rate of the relayed node served by the source node is defined as: Relay node The transmission rate is: wherein, is an indicator function that equals 1 when the source node successfully connects to the relay node , , otherwise equals 0.

3. The many-to-one relay selection method for a distributed network of claim 1, wherein, In step 2, when multiple source nodes including the source node select the same relay node , the source node will share the relay link resource with other source nodes; source node according to the transmission rate , a preference list is constructed; wherein the real transmission rate available to the source node ; the preference list of ; this preference relation represents the source node the best choice is if rejected request, will make an application to the next relay node , and so on; denotes the unselected relay node; The relay node accepts all application source nodes according to the first-come-first-served principle, evenly allocates time resources, and rejects other source nodes; the rejected source node receives feedback, re-finds and applies for the next relay connection; similarly, the relay node feeds back the actual transmission rate that the received source node can obtain, and ranks the source nodes according to the transmission rate to form a source node list.

4. The many-to-one relay selection method for a distributed network of claim 3, wherein, The As the source node Real transmission rate that can be obtained: in, Indicates access to a relay node The number of source nodes; Indicated by emission, Forward and Received transmission rate; It represents the signal-to-noise ratio transmitted from the source node to the relay node; It represents the signal-to-noise ratio transmitted from the source node to the destination node; It represents the signal-to-noise ratio transmitted from the relay node to the destination node; represents the channel bandwidth; express In the relay With the help of ,and Represents the source node With the destination node Communicate directly, that is, direct transmission mode.

5. A many-to-one relay selection method for a distributed network as claimed in claim 3 or 4, characterized in that, In step 3, specifically: The source node updates the preference list in such a way that for each source node , the transmission rate when selecting a relay node is less than the transmission rate in direct mode, i.e. , , the current relay node is abandoned and the preference list is changed; until a relay node is found such that the transmission rate when going through the relay node is greater than or equal to the transmission rate in direct mode, i.e. , . Each source node that has not established a connection with a relay node updates the preference list until only the direct transmission mode can be selected; the source node that does not meet the requirement of replacing the relay node maintains the original relay node.

6. The many-to-one relay selection method for distributed network of claim 1, wherein, The step 4, for all source nodes connected to a relay node , randomly select one of them suggest it to find a better relay selection; when the source node approaches a new relay node , it will carry the transmission rate of the previous connected relay node , so that the relay node can compare the transmission rates and make a selection; If the source node Connecting to a relay node When the source node The transmission efficiency is greater than that of the source node Connecting to a relay node The transmission efficiency when , then the source node To relay node Make a connection request; if this option allows the relay node With relay nodes The total transmission rate Increase, then the relay node Accept source node Otherwise, the application will be rejected, that is, the source node Will not abandon the selected relay node ; Source node Before connecting to another relay node After that, the previous relay node Remove it from the list.

Images