CN103220757B - A kind of optimum relay selection method based on two-way auction model - Google Patents

Abstract

The invention discloses a kind of optimum relay selection method based on two-way auction model, first auctioneer notifies that auction starts, each node determines respective bid or asked price according to own resource situation, again according to actual conditions definition both parties energy efficiency function to each other, set up complete weight pre-matching bigraph (bipartite graph); Maximum weight algorithms is adopted to obtain internodal maximum power efficiency matching relationship; Finally delete virtual pair relationhip according to actual relationship, obtain final successfully transaction.The present invention adopts two-way auction model to realize optimum relay selection method, and the method, for the optimum trunk node selection of edge customer, obtains higher-energy efficiency, expanding communication scope.Effectively reduce system and node energy consumption, adopt maximum weight algorithms auxiliary system to choose optimum energy efficiency coupling combination, delete void and join the matching relationship that finally struck a bargain.Effectively reduce individual consumer and even whole system energy ezpenditure, improve network performance.

Description

Optimal relay selection method based on two-way auction model

Technical Field

The invention relates to a cooperative communication relay node selection method in a wireless communication network, in particular to an optimal relay selection method based on a two-way auction model.

Background

Wireless communication is one of the most active research hotspots in today's communication field, and the fading characteristics of wireless channels are the main reasons that hinder the increase of channel capacity and the improvement of service quality. The space diversity technology is a simpler and effective method for inhibiting channel fading, and the multiple-input multiple-output (MIMO) technology greatly improves the reliability of transmission and has double functions of space diversity and space multiplexing. The cooperative communication technology is a virtual MMO technology, which can effectively improve the transmission rate and reliability of the wireless communication network and enlarge the transmission distance and coverage of the wireless network.

In a wireless communication network, some edge users exist, the quality of a communication link between the edge users and a base station is poor, and even the minimum communication signal-to-noise ratio cannot be achieved, namely, a direct communication link does not exist. The cooperative communication technology can solve communication troubles for the users, but the users serving as relay nodes are limited by own resources (such as residual energy, transmission power and the like), are not willing to forward information for other nodes, and even source nodes needing relaying also want to acquire communication with minimum cost, and the selfish nature causes the cooperative mechanism not to work normally and effectively, thereby causing user loss, and blindly cooperating to degrade network performance. Therefore, the problem of selecting a suitable relay node for an edge user and considering node selfish in cooperative communication is a problem which needs to be solved urgently. In addition, how to obtain higher system capacity under the condition of minimum overall system overhead is also important to improve the overall system performance.

At present, the existing research mainly provides an incentive strategy of corresponding reward for the nodes to solve the selfishness problem of the nodes. The auction model can effectively solve this problem and can select a more appropriate relay node for a source node that needs relay service.

Both academia and industry are very concerned with the superiority of cooperative communication. In recent years, relay scheduling algorithms dedicated to cooperative communication have been increasingly studied, with no auction theory-based strategy. The auction of relay services between a source node and a plurality of relay nodes (see the documents: Beibei Wang, Zhu Han, and K.J. ray Liu, "distributed Relay Selection and Power Control for Multi user Cooperative Communication network Using Stacking Game," IEEE Transactions on Mobile Computing, Vol.8, No.7, pp.975-990, July 2009) is also a unidirectional auction mode. In the mode, no consideration is given to that any node in the cooperative communication network can be used as a source node and the source node has competition when selecting a relay, so that the two-way auction is more suitable for a cooperative communication mechanism. TASC (see Dejun Yang, X.F., and GuiliangXue, "Truthful Automation for collaborative Communications," ACM Mobile hoc2011, Paris, France, May2011) is a relay selection strategy based on a two-way Auction model. The main idea is to use honest bids and asking prices among nodes for relay service to conduct auction activities. The relay arrangement is completed by arranging the quotation of the source node from high to low and arranging the ask price of the relay node from low to high, but the method sacrifices partial nodes to ensure an honest auction mechanism and cannot obtain better system energy efficiency.

Disclosure of Invention

In view of the above, the technical problem to be solved by the present invention is to provide an optimal relay selection method for a node-excited game strategy based on a bi-directional auction model, which determines a pricing function by comprehensively considering the self-state and the income of the node, thereby reasonably exciting the cooperation among the nodes, effectively reducing the energy consumption of the system and the nodes, improving the system performance of the network, and obtaining higher system capacity.

The purpose of the invention is realized as follows:

the invention provides an optimal relay selection method based on a two-way auction model, which comprises the following steps:

s1: establishing a timely auction model according to the network topology, and informing an auctioneer of the start of auction;

the timely auction model comprises buyers, sellers and auctioneers, wherein the buyers are edge users needing relaying, the sellers are nodes providing relaying services, and the auctioneers are base stations;

s2: the buyer proposes the quote function according to the following formulaSimultaneously, the seller puts forward an ask function according to the following formula ${\mathrm{Ask}}_{R_{\left(n\right)}D};$

<math> <mfenced open='{' close=''> <mtable> <mtr> <mtd> <msub> <mi>Bid</mi> <mrow> <msub> <mi>S</mi> <mrow> <mo>(</mo> <mi>m</mi> <mo>)</mo> </mrow> </msub> <msub> <mi>R</mi> <mrow> <mo>(</mo> <mi>n</mi> <mo>)</mo> </mrow> </msub> </mrow> </msub> <mo>=</mo> <msub> <mi>V</mi> <mrow> <msub> <mi>S</mi> <mrow> <mo>(</mo> <mi>m</mi> <mo>)</mo> </mrow> </msub> <msub> <mi>R</mi> <mrow> <mo>(</mo> <mi>n</mi> <mo>)</mo> </mrow> </msub> </mrow> </msub> <mrow> <mo>(</mo> <mn>1</mn> <mo>-</mo> <msub> <mi>&lambda;</mi> <mi>i</mi> </msub> <mo>)</mo> </mrow> <mo>,</mo> </mtd> <mtd> <msub> <mi>&lambda;</mi> <mi>i</mi> </msub> <mo>&Element;</mo> <mo>[</mo> <mrow> <mn>0,1</mn> <mo>)</mo> </mrow> </mtd> </mtr> <mtr> <mtd> <msub> <mi>Ask</mi> <mrow> <msub> <mi>R</mi> <mrow> <mo>(</mo> <mi>n</mi> <mo>)</mo> </mrow> </msub> <mi>D</mi> </mrow> </msub> <mo>=</mo> <msub> <mi>Q</mi> <mrow> <msub> <mi>R</mi> <mrow> <mo>(</mo> <mi>n</mi> <mo>)</mo> </mrow> </msub> <mi>D</mi> </mrow> </msub> <mrow> <mo>(</mo> <mn>1</mn> <mo>+</mo> <msub> <mi>&xi;</mi> <mi>j</mi> </msub> <mo>)</mo> </mrow> <mo>,</mo> </mtd> <mtd> <msub> <mi>&xi;</mi> <mi>j</mi> </msub> <mo>&Element;</mo> <mo>[</mo> <mrow> <mn>0,1</mn> <mo>]</mo> </mrow> </mtd> </mtr> </mtable> </mfenced> </math>

Wherein,bidding for a buyer represents a true valuation of the service of the relay node n by the source node m,representing the true valuation of the relay node n to the destination node for the seller ask price, λ_iElevation and xi of buyer_jFor the seller elevation, the buyer elevation and the buyer elevation are [0,1 ]]A random value in between;

s3: judging whether the quotation function and the ask function meet the following formula:

<math> <mrow> <msub> <mi>Bid</mi> <mrow> <msub> <mi>S</mi> <mrow> <mo>(</mo> <mi>m</mi> <mo>)</mo> </mrow> </msub> <msub> <mi>R</mi> <mrow> <mo>(</mo> <mi>n</mi> <mo>)</mo> </mrow> </msub> </mrow> </msub> <mo>&GreaterEqual;</mo> <msub> <mi>Ask</mi> <mrow> <msub> <mi>R</mi> <mrow> <mo>(</mo> <mi>n</mi> <mo>)</mo> </mrow> </msub> <mi>D</mi> </mrow> </msub> <mo>,</mo> </mrow> </math> if not, returning to step S1;

s4: if yes, indicating that the transaction is present, and constructing a complete weight pre-matching bipartite graph under the two-way auction modeWhen the source node is a buyer m, an adjacent edge exists between the relay node and a seller n;

s5: pre-matching a bipartite graph according to the complete weight, and obtaining a maximum energy efficiency matching relation between nodes by adopting a maximum weight matching algorithm in a graph theory;

s6: judging whether a virtual matching transaction relationship exists in the maximum energy efficiency matching relationship, if not, entering the step S8;

wherein, the virtual allocation trading relation is the trading relation when the quotation function is smaller than the ask price function;

s7: if yes, deleting the virtual matching transaction relationship according to the actual relationship;

s8: acquiring the transaction price of the buyer and the seller;

s9: the buyer pays payment and obtains communication opportunity, and the seller provides relay service to obtain income and complete transaction.

Further, the weight value of the complete weight pre-matching bipartite graph constructed in the step S4 in the two-way auction mode is performed according to the following formula:

${\mathrm{EE}}_{S_{\left(m\right)}{R}_{\left(n\right)}}=\frac{C_{S_{\left(m\right)}{R}_{\left(n\right)}}}{P_{S_{\left(m\right)}{R}_{\left(n\right)}}+P_{R_{\left(n\right)}D}},$

wherein,the function of the energy efficiency is expressed,for a capacity of a certain pair of matching relationships,the representation represents the transmission power of the nth relay node, namely the seller to the base station;represents the transmission power representing the transmission power from the source node, i.e. buyer m, to the relay node, i.e. seller n; the SNR representation represents the signal-to-noise ratio; sigma²Representing a gaussian white noise power;the expression represents the distance from the nth relay node, namely the seller to the base station, and k is a path fading factor;representing the distance from the source node, i.e. buyer m, to the relay node, i.e. seller n.

Further, the transaction price of the buyer and the seller in the step S8 is calculated by the following formula:

$F_{S_{\left(m\right)}{R}_{\left(n\right)}}=\frac{({\mathrm{Bid}}_{s_{\left(m\right)}{R}_{\left(n\right)}}+{\mathrm{Ask}}_{R_{\left(n\right)}D})}{2}$

wherein,representing the buyer paying a virtual reward to the seller.

Further, the buyer proposes a quotation function in the step S2Simultaneously, the seller proposes an ask functionIs determined by evaluating own resources, including remaining energy and transmission power.

Further, the maximum weight matching algorithm in step S5 is a matching operation for maximizing a total weight based on a complete weight pre-matching bipartite graph and an edge weight, and specifically includes the following steps:

s51: constructing a weight matrix according to the complete weight pre-matching bipartite graph to form a complete weight bipartite graph G (V)₁,V₂) (ii) a Wherein, G (V)₁,V₂) The representation contains a set of vertices V₁And V₂The bipartite graph of (1); v₁Represents a collection of source nodes, i.e., buyers; v₂Representing a set of relay nodes, i.e. sellers;

s52: to the full weighted bipartite graph G (V)₁,V₂) And executing the Hungarian algorithm, and modifying the feasible top mark to find the match with the maximum weight value, thereby finally obtaining the maximum weight matching relation.

Further, the weight matrix is a square matrix, virtual nodes are arranged in the square matrix, the virtual nodes are source nodes or relay nodes, and the corresponding weight values of the virtual nodes and other nodes are 0.

Further, the feasible superscript is performed according to the following formula:

<math> <mfenced open='{' close=''> <mtable> <mtr> <mtd> <mi>l</mi> <mrow> <mo>(</mo> <mi>x</mi> <mo>)</mo> </mrow> <mo>=</mo> <munder> <mi>max</mi> <mrow> <mi>y</mi> <mo>&Element;</mo> <msub> <mi>V</mi> <mn>2</mn> </msub> </mrow> </munder> </mtd> <mtd> <mi>&omega;</mi> <mrow> <mo>(</mo> <mi>x</mi> <mo>,</mo> <mi>y</mi> <mo>)</mo> </mrow> <mo>&Element;</mo> <msub> <mi>V</mi> <mn>1</mn> </msub> </mtd> </mtr> <mtr> <mtd> <mi>l</mi> <mrow> <mo>(</mo> <mi>y</mi> <mo>)</mo> </mrow> <mo>=</mo> <mn>0</mn> </mtd> <mtd> <mi>y</mi> <mo>&Element;</mo> <msub> <mi>V</mi> <mn>2</mn> </msub> </mtd> </mtr> </mtable> </mfenced> </math>

wherein, ω represents the weight value of each edge, and l (y) represents the top mark of the relay node, i.e. the seller; l (x) indicates the topmark of the source node, i.e. buyer; y represents a certain relay node, namely a seller; x represents a certain source node, i.e. a buyer.

Further, the method for modifying the superscript in step S5 is performed according to the following formula:

wherein,representing a source node, namely a buyer set, and T representing a relay node, namely a seller set, defining the nodes asT=H∩V₂H is a bipartite graph G (V)₁,V₂) A vertex set of a staggered subgraph with a middle root at a certain node; l' (u) is the modified topmark; ω (x, y) represents the weight value between the source node, i.e., buyer x, and the relay node, i.e., seller y; alpha is alpha_lRepresents a topmark adjustment factor; u represents a certain node, namely a buyer or a seller; l (u) represents the corresponding bid of a node, i.e., buyer or seller.

The invention has the advantages that: the invention adopts a bidirectional auction model to realize the optimal relay selection method, and the method aims at the optimal relay node selection of the edge user to obtain higher energy efficiency and expand the communication range. The node is comprehensively considered to evaluate according to the resource condition of the node, the selfish psychology of pursuing the benefit maximization of the node is met by introducing the elevation, the state (residual energy and transmission power) of the node and the income (a seller obtains virtual payment paid by a buyer after auction is finished and the buyer obtains relay service) are comprehensively considered from the system perspective to determine a pricing function and construct a price asking function, so that the cooperation among the nodes is reasonably stimulated, the energy consumption of the system and the nodes is effectively reduced, the maximum weight matching algorithm is adopted to assist the system to select the optimal energy efficiency matching combination, and the virtual matching is deleted by combining with the auction condition of the actual condition to obtain the final transaction matching relationship. The energy consumption of single users and even the whole system is effectively reduced, and the network performance is improved.

Drawings

In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention will be further described in detail with reference to the accompanying drawings, in which:

FIG. 1 is a diagram of a system model for edge users in the present invention;

FIG. 2 is a flow chart of an algorithm based on a two-way auction model in the present invention;

FIG. 3 is a two-way auction mechanism of the present invention;

FIG. 4 is a quote and ask function of the present invention;

FIG. 5 is a complete weight pre-match bipartite graph for a bi-directional auction in accordance with the present invention;

FIG. 6 is a schematic diagram of a possible virtual match, i.e., a non-existing transaction, in a concurrent invention.

Detailed Description

The preferred embodiments of the present invention will be described in detail below with reference to the accompanying drawings; it should be understood that the preferred embodiments are illustrative of the invention only and are not limiting upon the scope of the invention.

Example 1

Fig. 1 is a diagram of a system model for an edge user in the present invention, fig. 2 is an algorithm flowchart based on a two-way auction model in the present invention, fig. 3 is a two-way auction mechanism in the present invention, fig. 4 is a function of bid price and bid price in the present invention, fig. 5 is a two-part diagram of complete weight pre-matching in a two-way auction in the present invention, fig. 6 is a schematic diagram of a transaction in which virtual matches possibly exist or do not exist in the present invention, as shown in the following figures: the invention provides an optimal relay selection method based on a two-way auction model, which comprises the following steps:

s1: establishing a timely auction model according to the network topology, and informing an auctioneer of the start of auction; after the last auction is completed, the auctioneer begins broadcasting a notification to all free nodes to prepare for the round of auction activity.

The timely auction model comprises buyers, sellers and auctioneers, wherein the buyers are edge users needing relaying, the sellers are nodes providing relaying services, and the auctioneers are base stations;

as shown in fig. 1, the edge user is located at the edge of the cell, and cannot directly communicate with the base station, and can only obtain communication through the relay of other users. The communication resources of the nodes are limited in the communication, e.g. the minimum SNR threshold SNR of the communication is limited_THOr maximum transmit power, etc. Not all nodes can communicate directly.

S2: the buyer proposes the quote function according to the following formulaSimultaneously, the seller puts forward an ask function according to the following formula ${\mathrm{Ask}}_{R_{\left(n\right)}D};$

<math> <mfenced open='{' close=''> <mtable> <mtr> <mtd> <msub> <mi>Bid</mi> <mrow> <msub> <mi>S</mi> <mrow> <mo>(</mo> <mi>m</mi> <mo>)</mo> </mrow> </msub> <msub> <mi>R</mi> <mrow> <mo>(</mo> <mi>n</mi> <mo>)</mo> </mrow> </msub> </mrow> </msub> <mo>=</mo> <msub> <mi>V</mi> <mrow> <msub> <mi>S</mi> <mrow> <mo>(</mo> <mi>m</mi> <mo>)</mo> </mrow> </msub> <msub> <mi>R</mi> <mrow> <mo>(</mo> <mi>n</mi> <mo>)</mo> </mrow> </msub> </mrow> </msub> <mrow> <mo>(</mo> <mn>1</mn> <mo>-</mo> <msub> <mi>&lambda;</mi> <mi>i</mi> </msub> <mo>)</mo> </mrow> <mo>,</mo> </mtd> <mtd> <msub> <mi>&lambda;</mi> <mi>i</mi> </msub> <mo>&Element;</mo> <mo>[</mo> <mrow> <mn>0,1</mn> <mo>)</mo> </mrow> </mtd> </mtr> <mtr> <mtd> <msub> <mi>Ask</mi> <mrow> <msub> <mi>R</mi> <mrow> <mo>(</mo> <mi>n</mi> <mo>)</mo> </mrow> </msub> <mi>D</mi> </mrow> </msub> <mo>=</mo> <msub> <mi>Q</mi> <mrow> <msub> <mi>R</mi> <mrow> <mo>(</mo> <mi>n</mi> <mo>)</mo> </mrow> </msub> <mi>D</mi> </mrow> </msub> <mrow> <mo>(</mo> <mn>1</mn> <mo>+</mo> <msub> <mi>&xi;</mi> <mi>j</mi> </msub> <mo>)</mo> </mrow> <mo>,</mo> </mtd> <mtd> <msub> <mi>&xi;</mi> <mi>j</mi> </msub> <mo>&Element;</mo> <mo>[</mo> <mrow> <mn>0,1</mn> <mo>]</mo> </mrow> </mtd> </mtr> </mtable> </mfenced> </math>

Wherein,bidding for a buyer represents a true valuation of the service of the relay node n by the source node m,representing the true valuation of the relay node n to the destination node for the seller ask price, λ_iElevation and xi of buyer_jIs the seller elevation. The real estimation is the estimation price of the buyer and the seller to the own resources, such as the estimation price of the buyer and the seller to the own residual energy, the estimation price of the own transmission power, and the like, taking the transmission power estimation as an example: in order to optimize energy efficiency, the nodes transmit data with the lowest transmission power, and therefore the data are used as the real evaluation. Since both buyer and seller are mobile devices with limited power, the system and method are applicable to mobile devices with limited powerIs satisfied with the constraint that: p_min≤P≤P_max. The node takes the minimum transmitting power capable of completing communication as the real evaluation value thereof, and sets the minimum SNR threshold capable of communicating as SNR_THThen said real valuation can be expressed asWherein sigma²Is Gaussian white noise power, and G is the path loss expressed as G ═ d^-kD is the distance between two nodes, and k is the path fading factor. Due to the randomness of the topology, the difference in distance between nodes may cause the true valuation to be larger than P_maxWhen it is defined as P ═ P_maxThe constraint is still satisfied. Due to non-negativity of the buyer's offer, the buyer's elevation is a random value between [0,1 ], and the seller's elevation is [0,1 ]]Random value to satisfy the self-privacy psychology of the buyer and the seller;

the buyer proposes the quotation function in said step S2Simultaneously, the seller proposes an ask functionIs determined by evaluating own resources, including remaining energy and transmission power.

S3: judging whether the quotation function and the ask function meet the following formula:

<math> <mrow> <msub> <mi>Bid</mi> <mrow> <msub> <mi>S</mi> <mrow> <mo>(</mo> <mi>m</mi> <mo>)</mo> </mrow> </msub> <msub> <mi>R</mi> <mrow> <mo>(</mo> <mi>n</mi> <mo>)</mo> </mrow> </msub> </mrow> </msub> <mo>&GreaterEqual;</mo> <msub> <mi>Ask</mi> <mrow> <msub> <mi>R</mi> <mrow> <mo>(</mo> <mi>n</mi> <mo>)</mo> </mrow> </msub> <mi>D</mi> </mrow> </msub> <mo>,</mo> </mrow> </math> if not, returning to step S1;

s4: if yes, indicating that the transaction is present, and constructing a complete weight pre-matching bipartite graph under the two-way auction modeWhen the source node is a buyer m, an adjacent edge exists between the relay node and a seller n; the weight value of the complete weight pre-matching bipartite graph constructed in the step S4 in the bi-directional auction mode is performed according to the following formula:

${\mathrm{EE}}_{S_{\left(m\right)}{R}_{\left(n\right)}}=\frac{C_{S_{\left(m\right)}{R}_{\left(n\right)}}}{P_{S_{\left(m\right)}{R}_{\left(n\right)}}+P_{R_{\left(n\right)}D}},$

wherein,the function of the energy efficiency is expressed,the capacity of a certain pair of matching relations, omega represents the weight value of each edge; the representation represents the transmission power of the nth relay node, namely the seller to the base station;represents the transmission power representing the transmission power from the source node, i.e. buyer m, to the relay node, i.e. seller n; the SNR representation represents the signal-to-noise ratio; sigma²Representing a gaussian white noise power;the expression represents the distance from the nth relay node, namely the seller to the base station, and k is a path fading factor;to representThe distance from the source node, i.e., buyer m, to the relay node, i.e., seller n.

S5: pre-matching a bipartite graph according to the complete weight, and obtaining a maximum energy efficiency matching relation between nodes by adopting a maximum weight matching algorithm in a graph theory; the maximum weight matching algorithm is a matching operation for maximizing the total weight based on the complete weight pre-matching bipartite graph and the side weight thereof of S4. The edge weight, i.e. the weight value on each adjacent edge, is defined as the corresponding energy efficiency function

Firstly, a weight matrix is constructed according to a complete weight pre-matching bipartite graph, the weight matrix is a square matrix, virtual nodes are introduced for constructing the square matrix, the virtual nodes can be source nodes (buyers) or relay nodes (sellers), the corresponding weight values of the virtual nodes and other nodes are 0, and a complete weighting bipartite graph G (V) can be constructed₁,V₂). Executing a Hungarian algorithm on the completely weighted bipartite graph, and continuously modifying the feasible topmarks to find out the matching with the maximum weight value, so that the perfect matching relation of the maximum weight can be obtained finally; the feasible superscript is carried out according to the following formula:

<math> <mfenced open='{' close=''> <mtable> <mtr> <mtd> <mi>l</mi> <mrow> <mo>(</mo> <mi>x</mi> <mo>)</mo> </mrow> <mo>=</mo> <munder> <mi>max</mi> <mrow> <mi>y</mi> <mo>&Element;</mo> <msub> <mi>V</mi> <mn>2</mn> </msub> </mrow> </munder> </mtd> <mtd> <mi>&omega;</mi> <mrow> <mo>(</mo> <mi>x</mi> <mo>,</mo> <mi>y</mi> <mo>)</mo> </mrow> <mo>&Element;</mo> <msub> <mi>V</mi> <mn>1</mn> </msub> </mtd> </mtr> <mtr> <mtd> <mi>l</mi> <mrow> <mo>(</mo> <mi>y</mi> <mo>)</mo> </mrow> <mo>=</mo> <mn>0</mn> </mtd> <mtd> <mi>y</mi> <mo>&Element;</mo> <msub> <mi>V</mi> <mn>2</mn> </msub> </mtd> </mtr> </mtable> </mfenced> </math>

the weight matrix is a square matrix, virtual nodes are arranged in the square matrix, the virtual nodes are source nodes or relay nodes, and the weight values of the virtual nodes and other nodes are 0.

The method for modifying the topmark can be carried out according to the following formula:

whereinT=H∩V₂Wherein H is a bipartite graph G (V)₁,V₂) The vertex set of the staggered subgraph with the root at a certain node. l' (u) is the modified topmark;representing a source node, namely a buyer set, and T representing a relay node, namely a seller set, defining the nodes asT=H∩V₂H is a bipartite graph G (V)₁,V₂) Middle root is atA set of vertices of a staggered subgraph of a node; l' (u) is the modified topmark; ω (x, y) represents the weight value between the source node, i.e., buyer x, and the relay node, i.e., seller y; alpha is alpha_lRepresents a topmark adjustment factor; u represents a certain node, namely a buyer or a seller; l (u) represents the corresponding bid of a node, i.e., buyer or seller.

S6: judging whether a virtual matching transaction relationship exists in the maximum energy efficiency matching relationship, if not, entering the step S8;

wherein, the virtual matching transaction relationship is the transaction relationship when the matched relationship edge weight is 0;

s7: if yes, deleting the virtual matching transaction relationship according to the actual relationship;

s8: acquiring the transaction price of the buyer and the seller;

the transaction price of the buyer and the seller in the step S8 is calculated by the following formula:

$F_{S_{\left(m\right)}{R}_{\left(n\right)}}=\frac{({\mathrm{Bid}}_{s_{\left(m\right)}{R}_{\left(n\right)}}+{\mathrm{Ask}}_{R_{\left(n\right)}D})}{2}$

wherein,representing the buyer paying a virtual reward to the seller.

S9: the buyer pays payment and obtains communication opportunity, and the seller provides relay service to obtain income and complete transaction.

Example 2

The embodiment 2 specifically describes the principle and implementation process of the optimal relay selection method based on the two-way auction model:

as shown in fig. 2, which is an algorithm flowchart based on the bi-directional auction model in the present invention, the specific implementation steps are as follows:

1) after the last auction is completed, the auctioneer begins broadcasting a notification to all free nodes to prepare for the round of auction activity. As shown in fig. 3. The method comprises the steps that m source nodes and n relay nodes are provided, the source nodes are buyers, the relay nodes are sellers, after the notification is received, the buyers submit own quotes and asking prices in the relay auction activity to auctioneers at the same time, and the bids or asking prices are generated through a bid or asking price function respectively. The auctioneer begins the auction after receiving all of the bids and asks.

2) As shown in fig. 4, the bid and ask functions of the buyer and seller in the two-way auction campaign are shown, and both buyer and seller can obtain their own bid and ask functions through their own resources (for example: remaining energy, transmit power, etc.) to determine that it can provide a price code. We define offer and ask points for buyers and sellersIs otherwise provided withAndrespectively representing the true valuation of the service of the relay node n by the source node m and the true valuation of the service from the relay node n to the destination node. Since each node has selfishness, each node wants to obtain the maximum benefit by using the minimum expense, then the node serving as the source node of the buyer can perform certain price reduction on the basis of real valuation, and similarly, the node serving as the relay node of the seller can perform certain price raising on the basis of real valuation, so that the concept of elevation is introduced, and the quotation and the asking price of the buyer and the seller are defined asAndλ_iand xi_jIs taken as elevation, the elevation is [0,1 ]]By taking random values, the value can be obtained

<math> <mfenced open='{' close=''> <mtable> <mtr> <mtd> <msub> <mi>Bid</mi> <mrow> <msub> <mi>S</mi> <mrow> <mo>(</mo> <mi>m</mi> <mo>)</mo> </mrow> </msub> <msub> <mi>R</mi> <mrow> <mo>(</mo> <mi>n</mi> <mo>)</mo> </mrow> </msub> </mrow> </msub> <mo>=</mo> <msub> <mi>V</mi> <mrow> <msub> <mi>S</mi> <mrow> <mo>(</mo> <mi>m</mi> <mo>)</mo> </mrow> </msub> <msub> <mi>R</mi> <mrow> <mo>(</mo> <mi>n</mi> <mo>)</mo> </mrow> </msub> </mrow> </msub> <mrow> <mo>(</mo> <mn>1</mn> <mo>-</mo> <msub> <mi>&lambda;</mi> <mi>i</mi> </msub> <mo>)</mo> </mrow> <mo>,</mo> </mtd> <mtd> <msub> <mi>&lambda;</mi> <mi>i</mi> </msub> <mo>&Element;</mo> <mo>[</mo> <mn>0,1</mn> <mo>]</mo> </mtd> </mtr> <mtr> <mtd> <msub> <mi>Ask</mi> <mrow> <msub> <mi>R</mi> <mrow> <mo>(</mo> <mi>n</mi> <mo>)</mo> </mrow> </msub> <mi>D</mi> </mrow> </msub> <mo>=</mo> <msub> <mi>Q</mi> <mrow> <msub> <mi>R</mi> <mrow> <mo>(</mo> <mi>n</mi> <mo>)</mo> </mrow> </msub> <mi>D</mi> </mrow> </msub> <mrow> <mo>(</mo> <mn>1</mn> <mo>+</mo> <msub> <mi>&xi;</mi> <mi>j</mi> </msub> <mo>)</mo> </mrow> <mo>,</mo> </mtd> <mtd> <msub> <mi>&xi;</mi> <mi>j</mi> </msub> <mo>&Element;</mo> <mo>[</mo> <mrow> <mn>0,1</mn> <mo>]</mo> </mrow> </mtd> </mtr> </mtable> </mfenced> </math>

3) After comprehensively considering the quotes and ask prices of all buyers and sellers, the relationship existsIn this case, only the seller with the lowest price need to be selected, however, from the perspective of the whole system, it is expected that the most transactions can be achieved, and therefore, a complete weight pre-matching bipartite graph under a bidirectional auction mode is constructed next.

4) As shown in fig. 5, the full weight pre-matching bipartite graph in the bi-directional auction mode of the present invention is characterized. The solid lines in the figure represent the pairings that can be traded for any node, i.e. areThe dotted line in the figure indicates that the transaction pair does not exist, i.e. isTaking the source node 2 as an example, the adjacent edges of the source node 2 in FIG. 5 are shown by solid lines satisfyingAnd the dotted line representsThe edge of (2). Therefore, based on the transaction pre-matching relationship between all source nodes and the relay nodes, we define the weight value of each edge, and we use omega to represent the weight value. To maximize energy efficiency with minimal overhead in the overall system, we define an energy efficiency function

${\mathrm{EE}}_{S_{\left(m\right)}{R}_{\left(n\right)}}=\frac{C_{S_{\left(m\right)}{R}_{\left(n\right)}}}{P_{S_{\left(m\right)}{R}_{\left(n\right)}}+P_{R_{\left(n\right)}D}}$

Wherein,is the capacity of a certain pair of possible matching relationships. Here, we define the edge weight value of each possible transaction as its corresponding energy efficiency function, and the impossible transactions we define the edge weight value as 0, as shown by the dotted line in the figure. Taking the source node 2 as an example, the solid line indicates that the edge weight is non-zero, and the dotted line indicates that the edge weight is 0.

5) After a complete weight pre-matching bipartite graph is established, a maximum weight matching algorithm in a graph theory is adopted to complete final matching relation determination, and a perfect matching relation can be obtained after the algorithm is executed. A planned transaction, i.e. an energy efficient optimal relay schedule, can be obtained.

As shown in fig. 6, since there may be a transaction in which a matching relationship of a weight value with an edge weight of 0 exists, that is, there is no possibility, the execution algorithm deletes the transaction that does not exist, and a final transaction, that is, a relay pairing relationship is obtained. Completing the auction process, and defining the transaction price as:

$F_{S_{\left(m\right)}{R}_{\left(n\right)}}=\frac{({\mathrm{Bid}}_{s_{\left(m\right)}{R}_{\left(n\right)}}+{\mathrm{Ask}}_{R_{\left(n\right)}D})}{2}$

and finally, the buyer pays corresponding payment to the corresponding seller.

Aiming at the problem of difficult communication of edge users, the invention provides an optimal relay selection strategy by combining a two-way auction model, considers the selfness of the nodes, reasonably defines the quotation and ask price functions of buyers and sellers, gives an incentive strategy of virtual reward and encourages the nodes to actively participate in cooperation. Meanwhile, the energy efficiency of the whole system is considered, and the energy consumption in the network is reduced, the capacity is improved, and the communication range is widened.

The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention, and it is apparent that those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if such modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include such modifications and variations.

Claims (5)

1. An optimal relay selection method based on a two-way auction model is characterized in that: the method comprises the following steps:

s1: establishing a timely auction model according to the network topology, and informing an auctioneer of the start of auction;

the timely auction model comprises buyers, sellers and auctioneers, wherein the buyers are edge users needing relaying, the sellers are nodes providing relaying services, and the auctioneers are base stations;

s2: the buyer proposes the quote function according to the following formulaSimultaneously, the seller puts forward an ask function according to the following formula

<math> <mrow> <mfenced open='{' close=''> <mtable> <mtr> <mtd> <msub> <mi>Bid</mi> <mrow> <msub> <mi>S</mi> <mrow> <mo>(</mo> <mi>m</mi> <mo>)</mo> </mrow> </msub> <msub> <mi>R</mi> <mrow> <mo>(</mo> <mi>n</mi> <mo>)</mo> </mrow> </msub> </mrow> </msub> <mo>=</mo> <msub> <mi>V</mi> <mrow> <msub> <mi>S</mi> <mrow> <mo>(</mo> <mi>m</mi> <mo>)</mo> </mrow> </msub> <msub> <mi>R</mi> <mrow> <mo>(</mo> <mi>n</mi> <mo>)</mo> </mrow> </msub> </mrow> </msub> <mrow> <mo>(</mo> <mn>1</mn> <mo>-</mo> <msub> <mi>&lambda;</mi> <mi>i</mi> </msub> <mo>)</mo> </mrow> <mo>,</mo> </mtd> <mtd> <msub> <mi>&lambda;</mi> <mi>i</mi> </msub> <mo>&Element;</mo> <mo>[</mo> <mn>0,1</mn> <mo>)</mo> </mtd> </mtr> <mtr> <mtd> <msub> <mi>Ask</mi> <mrow> <msub> <mi>R</mi> <mrow> <mo>(</mo> <mi>n</mi> <mo>)</mo> </mrow> </msub> <mi>D</mi> </mrow> </msub> <mo>=</mo> <msub> <mi>Q</mi> <mrow> <msub> <mi>R</mi> <mrow> <mo>(</mo> <mi>n</mi> <mo>)</mo> </mrow> </msub> <mi>D</mi> </mrow> </msub> <mrow> <mo>(</mo> <mn>1</mn> <mo>+</mo> <msub> <mi>&xi;</mi> <mi>j</mi> </msub> <mo>)</mo> </mrow> <mo>,</mo> </mtd> <mtd> <msub> <mi>&xi;</mi> <mi>j</mi> </msub> <mo>&Element;</mo> <mo>[</mo> <mn>0,1</mn> <mo>]</mo> </mtd> </mtr> </mtable> </mfenced> <mo>;</mo> </mrow> </math>

Wherein,bidding for a buyer represents a true valuation of the service of the relay node n by the source node m,representing the true valuation of the relay node n to the destination node for the seller ask price, λ_iElevation and xi of buyer_jThe seller is marked;

s3: judging whether the quotation function and the ask function meet the following formula:

if not, returning to step S1;

s4: if yes, indicating that the transaction is present, and constructing a complete weight pre-matching bipartite graph under the two-way auction modeWhen the source node is a buyer m, an adjacent edge exists between the relay node and a seller n;

s5: pre-matching a bipartite graph according to the complete weight, and obtaining a maximum energy efficiency matching relation between nodes by adopting a maximum weight matching algorithm in a graph theory;

s6: judging whether a virtual matching transaction relationship exists in the maximum energy efficiency matching relationship, if not, entering the step S8;

wherein, the virtual matching transaction relationship is the transaction relationship when the matched relationship edge weight is 0;

s7: if yes, deleting the virtual matching transaction relationship according to the actual relationship;

s8: acquiring the transaction price of the buyer and the seller;

s9: the buyer pays a reward and obtains a communication opportunity, and the seller provides a relay service to obtain a profit and complete a transaction;

the weight value of the complete weight pre-matching bipartite graph constructed in the step S4 in the bi-directional auction mode is performed according to the following formula:

${\mathrm{EE}}_{S_{\left(m\right)}{R}_{\left(n\right)}}=\frac{C_{S_{\left(m\right)}{R}_{\left(n\right)}}}{P_{S_{\left(m\right)}{R}_{\left(n\right)}}+P_{R_{\left(n\right)}D}},$

wherein,the function of the energy efficiency is expressed,for a capacity of a certain pair of matching relationships, <math> <mrow> <msub> <mi>P</mi> <mrow> <msub> <mi>S</mi> <mrow> <mo>(</mo> <mi>m</mi> <mo>)</mo> </mrow> </msub> <msub> <mi>R</mi> <mrow> <mo>(</mo> <mi>n</mi> <mo>)</mo> </mrow> </msub> </mrow> </msub> <mo>=</mo> <mfrac> <mrow> <mi>SNR</mi> <msup> <mi>&sigma;</mi> <mn>2</mn> </msup> </mrow> <msubsup> <mi>d</mi> <mrow> <msub> <mi>S</mi> <mrow> <mo>(</mo> <mi>m</mi> <mo>)</mo> </mrow> </msub> <msub> <mi>R</mi> <mrow> <mo>(</mo> <mi>n</mi> <mo>)</mo> </mrow> </msub> </mrow> <mrow> <mo>-</mo> <mi>k</mi> </mrow> </msubsup> </mfrac> <mo>,</mo> <msub> <mi>P</mi> <mrow> <msub> <mi>R</mi> <mrow> <mo>(</mo> <mi>n</mi> <mo>)</mo> </mrow> </msub> <mi>D</mi> </mrow> </msub> <mo>=</mo> <mfrac> <mrow> <mi>SNR</mi> <msup> <mi>&sigma;</mi> <mn>2</mn> </msup> </mrow> <msubsup> <mi>d</mi> <mrow> <msub> <mi>R</mi> <mrow> <mo>(</mo> <mi>n</mi> <mo>)</mo> </mrow> </msub> <mi>D</mi> </mrow> <mrow> <mo>-</mo> <mi>k</mi> </mrow> </msubsup> </mfrac> <mo>;</mo> </mrow> </math> the representation represents the transmission power of the nth relay node, namely the seller to the base station;represents the transmission power representing the transmission power from the source node, i.e. buyer m, to the relay node, i.e. seller n; the SNR representation represents the signal-to-noise ratio; sigma²Representing a gaussian white noise power;the expression represents the distance from the nth relay node, namely the seller to the base station, and k is a path fading factor;representing the distance from the source node, i.e. buyer m, to the relay node, i.e. seller n;

the transaction price of the buyer and the seller in the step S8 is calculated by the following formula:

$F_{S_{\left(m\right)}{R}_{\left(n\right)}}=\frac{({\mathrm{Bid}}_{S_{\left(m\right)}{R}_{\left(n\right)}}+{\mathrm{Ask}}_{R_{\left(n\right)}D})}{2},$

wherein,representing the buyer paying a virtual reward to the seller;

the buyer proposes the quotation function in said step S2Simultaneously, the seller proposes an ask functionIs determined by evaluating own resources, including remaining energy and transmission power.

2. The optimal relay selection method based on the bi-directional auction model of claim 1, wherein: the maximum weight matching algorithm in step S5 is a matching operation for maximizing a total weight based on a complete weight pre-matching bipartite graph and a side weight, and specifically includes the following steps:

s51: constructing a weight matrix according to the complete weight pre-matching bipartite graph to form a complete weight bipartite graph G (V)₁,V₂) (ii) a Wherein, G (V)₁,V₂) The representation contains a set of vertices V₁And V₂The bipartite graph of (1); v₁Represents a collection of source nodes, i.e., buyers; v₂Representing a set of relay nodes, i.e. sellers;

s52: to the full weighted bipartite graph G (V)₁,V₂) And executing the Hungarian algorithm, and modifying the feasible top mark to find the match with the maximum weight value, thereby finally obtaining the maximum weight matching relation.

3. The optimal relay selection method based on the bi-directional auction model of claim 2, wherein: the weight matrix is a square matrix, virtual nodes are arranged in the square matrix, the virtual nodes are source nodes or relay nodes, and the weight values of the virtual nodes and other nodes are 0 correspondingly.

4. The optimal relay selection method based on the bi-directional auction model of claim 2, wherein: the feasible superscript is carried out according to the following formula:

<math> <mfenced open='{' close=''> <mtable> <mtr> <mtd> <mi>l</mi> <mrow> <mo>(</mo> <mi>x</mi> <mo>)</mo> </mrow> <mo>=</mo> <munder> <mi>max</mi> <mrow> <mi>y</mi> <mo>&Element;</mo> <msub> <mi>V</mi> <mn>2</mn> </msub> </mrow> </munder> </mtd> <mtd> <mi>&omega;</mi> <mrow> <mo>(</mo> <mi>x</mi> <mo>,</mo> <mi>y</mi> <mo>)</mo> </mrow> <mo>&Element;</mo> <msub> <mi>V</mi> <mn>1</mn> </msub> </mtd> </mtr> <mtr> <mtd> <mi>l</mi> <mrow> <mo>(</mo> <mi>y</mi> <mo>)</mo> </mrow> <mo>=</mo> <mn>0</mn> </mtd> <mtd> <mi>y</mi> <mo>&Element;</mo> <msub> <mi>V</mi> <mn>2</mn> </msub> </mtd> </mtr> </mtable> </mfenced> </math>

omega represents the weight value of each edge, and l (y) represents the top mark of the relay node, namely the seller; l (x) indicates the topmark of the source node, i.e. buyer; y represents a certain relay node, namely a seller; x represents a certain source node, i.e. a buyer.

5. The optimal relay selection method based on the bi-directional auction model of claim 2, wherein: the modification method of the topmark in the step S5 is performed according to the following formula:

wherein,representing a source node, namely a buyer set, and T representing a relay node, namely a seller set, defining the nodes asT＝H∩V₂H is a bipartite graph G (V)₁,V₂) A vertex set of a staggered subgraph with a middle root at a certain node; l' (u) is the modified topmark; ω (x, y) represents the weight value between the source node, i.e., buyer x, and the relay node, i.e., seller y; alpha is alpha_lRepresents a topmark adjustment factor; u represents a certain node, namely a buyer or a seller; l (u) represents the corresponding topmark of a certain node, i.e. buyer or seller.