CN103415040B - Wireless Heterogeneous Networks multi-node optimal relay alliance cooperation motivational techniques - Google Patents

Abstract

The present invention proposes Wireless Heterogeneous Networks multi-node optimal relay alliance cooperation motivational techniques, first described method constructs alliance's earnings pattern in Wireless Heterogeneous Networks, and take into full account that the channel circumstance of reality sets up cost function, by complete mathematical derivation solving-optimizing model, it is obtained in that the theoretical optimal solution of Optimized model, then pass through and optimize the relay alliance that alliance's acquisition is best, thus effectively excitation node participates in cooperation.The heterogeneous network multi-node optimal relay alliance cooperation incentive mechanism mode that the present invention produces is very simple and is easily achieved, and has good application prospect.

Description

Wireless heterogeneous network multi-node optimal relay alliance cooperation excitation method

Technical Field

The invention belongs to the technical field of wireless communication, and particularly relates to a cooperative excitation method of a wireless heterogeneous network multi-node optimal relay alliance.

Background

In recent years, wireless mobile communication technology has been developed, and systems such as WLAN, UMTS and WiMAX are introduced globally and various existing second-generation mobile communication networks continue to operate, which brings coexistence of multiple types of communication networks, i.e., a Wireless Heterogeneous Network (WHN) wireless heterogeneous network. The WHN is an aggregate formed by mutually fusing multiple technologies, multiple networks and multiple services, can greatly improve the performance of a single network, and creates conditions for introducing new services while supporting the traditional services; it can also provide higher data transmission rate, wider signal coverage and support higher rate mobility for future mobile communication systems. In a typical cellular and Ad-hoc heterogeneous convergence network, the bottleneck effect of a cellular network base station can be relieved, the traffic flow can be balanced, the space multiplexing rate of the network can be improved and the capacity of the network can be correspondingly improved through the self-organization and the multi-hop relay capacity of the mobile Ad-hoc network. However, the existence of some selfish nodes in the network can also affect the performance of the whole network by avoiding the consumption of own resources to refuse to participate in relay forwarding. Of course, the relay alliance incentive as a new research technology has many problems to be solved, and the core difficulty is as follows: how to select a federation of nodes to participate in a collaboration and how to perform collaboration within the federation. Therefore, the research on the cooperation incentive of the node optimal relay alliance in the heterogeneous wireless convergence network is widely concerned.

At present, for the research of a node cooperation excitation mechanism in an Ad-hoc network with the worst security in a heterogeneous wireless network, the method mainly comprises an excitation mechanism based on a reputation value, an excitation scheme selected by a mechanism and an analysis method based on a game theory. The above excitation mechanism has the following disadvantages: (1) only considering the case of single or two relay nodes, the single or two node alliance is too simple and the consideration of the cooperation and mutual influence among the relay nodes is insufficient; (2) the incentive scheme is complex to implement, for example, the maintenance and propagation mechanism of the reputation value is complex and unreliable, which causes the problem of inconsistency of the reputation value; (3) most studies only prove the existence of nash equilibrium based on the game theory incentive mechanism, but do not propose a specific cooperation promotion scheme.

Disclosure of Invention

In order to overcome the defects of the prior art and particularly aim at the cooperative excitation problem of the multi-node relay alliance in the wireless heterogeneous network, the invention provides a cooperative excitation method of the multi-node optimal relay alliance in the wireless heterogeneous network. The method takes a wireless heterogeneous network as a model, analyzes the node profit in the network, and performs joint optimization on the utility profit of the multi-node relay alliance and the alliance forwarding cost function.

In order to solve the technical problems, the invention adopts the following technical scheme:

the method adopts a multi-node relay alliance model to obtain the number of the optimal nodes of the multi-node relay alliance; comprises the following steps:

step A, setting a relay alliance set asWherein tau is_kRepresents an arbitrary multi-node federation,is a natural number; the maximum total profit is realized by maximizing the alliance utility profit and minimizing the alliance node cost function, and then the relay alliance model is as follows:

$\max z=\max (a_{{\mathrm{\τ}}_k}-c_{{\mathrm{\τ}}_k})$

$s.t.a_{{\mathrm{\τ}}_k}=\frac{n(n-1)}{2}(2p-2c+{2m}_r-{2m}_p)$

p>m_p

m_r>c

$c_{{\mathrm{\τ}}_k}=\underset{i=1}{\overset{n}{\mathrm{\Σ}}}({\mathrm{\λ}}_i{p}_i+{({\mathrm{\γ}}_{\mathrm{tar}}-{\mathrm{\γ}}_i)}^2)$

${\mathrm{\γ}}_i=\frac{h_i{p}_i}{N_{r}W}$

0≤p_i≤p_m

wherein,

c is resource cost consumed by forwarding the packet by the node, p is the income of the node when the packet of the node is successfully forwarded by another node, m_rFor forwarding the reward, m_pIs the forwarding price;

z represents the total revenue of the federation,representing a node federation utility revenue function,representing a relay alliance cost function;

n represents the number of nodes in any alliance, and n is more than or equal to 2;

γ_taris a target signal-to-interference ratio, gamma_iRepresenting the signal-to-interference ratio SNR, h of the signal received by node i_iDenotes the link gain, λ, between node i and other nodes in the federation_iProportional to the link gain, i.e. λ_i=kh_iK is a constant greater than zero, i is a natural number less than or equal to n;

p_irepresenting the transmission power, p, of any node i_mRepresents the maximum value of the transmitting power of any node, W represents the bandwidth of the signal transmitted by each node, N_rRespectively representing the power spectrum density of channel additive Gaussian noise of a receiving end;

step B, as alliance utility revenue functionAnd when the maximum value is taken, performing power control on the nodes in the alliance, so that the cost function of the nodes takes the minimum value, and simplifying the relay alliance model in the step A into:

$\{p_1^{*},p_2^{*},\·\·\·p_n^{*}\}=\underset{n}{\arg }\min f(p_1,p_2,\·\·\·p_n)$

$=\underset{n}{\arg }\min \left(c_{{\mathrm{\τ}}_k}\right)$

$s.t.c_{{\mathrm{\τ}}_k}=\underset{i=1}{\overset{n}{\mathrm{\Σ}}}({\mathrm{\λ}}_i{p}_i+{({\mathrm{\γ}}_{\mathrm{tar}}-{\mathrm{\γ}}_i)}^2)$

wherein, f (p)₁,p₂,…p_n) Representing a cost function of the relay alliance forwarding packets;respectively representing the optimal power of nodes in the alliance, and arg represents a variable meeting the function maximum value;

calculating an optimized target value of the optimized model;

step C, comparing the total earnings of the alliances formed by different numbers of nodes in the relay alliance set M, namely obtaining the number n of the optimal nodes of the relay alliance^*：

$n^{*}=\underset{\mathrm{\τ}\∈M,n}{\arg }\max z^{*}$

Wherein z is^*=maxz。

The invention has the beneficial effects that: the invention provides a wireless heterogeneous network multi-node relay cooperation excitation method which includes the steps of firstly, constructing an alliance profit model in a wireless heterogeneous network, fully considering the actual channel environment to establish a cost function, solving an optimization model through complete mathematical derivation, obtaining a theoretical optimal solution of the optimization model, and then obtaining an optimal relay alliance through the optimization alliance, so that nodes are effectively excited to participate in cooperation. The heterogeneous network multi-node optimal relay alliance cooperation incentive mechanism generated by the invention is very simple in mode and easy to realize, and has good application prospect.

Drawings

Fig. 1 is a flowchart of an optimal relay cooperation excitation method for a wireless heterogeneous network.

Detailed Description

The following describes the wireless heterogeneous network multi-node relay cooperation excitation method in detail with reference to the accompanying drawings.

Typical Dual base station Wireless communication topology model, wherein base stations BS₁And BS₂In the independent coverage area, a cellular communication mode is adopted, and in the overlapping coverage area, relay nodes connected by Ad-hoc exist, so when a base station BS₁Nodes in independent coverage area to access BS₂Base station, which must be forwarded through relay nodes in the overlap area, relay alliance setWherein tau is_kRepresenting an arbitrary two-node federation, is a natural number and the number of nodes in each federation is represented by the variable n. The method can effectively relieve the bottleneck effect of the cellular network base station, balance the traffic flow and improve the space reuse rate of the network, and is also relativeThe communication quality of the mobile network should be improved.

Considering the actual network environment, some nodes in the network may adopt a cooperative denial strategy to save their own resources, but the amount of virtual currency of each node is constant, and in order to ensure that each node can obtain the services provided by the network, it must be satisfied that the amount of virtual currency of itself is greater than zero, and it is assumed that the node accepts the number of packets n_rNumber of discarded packets n_dAnd node forwarding the number n of packets it generates itself_gIt must satisfy:

(n_r-n_d)m_r+V-n_gm_p≧ 0 (1) where V represents the virtual currency amount of the node.

We now assume that all participants reach a gaming agreement through a trusted third party, maximizing the revenue per node by forming a gaming league. There are four optional policy sets { (C) between any two participants_i,I_j),(C_i,C_j),(I_i,C_j),(I_i,I_j) And the gains under different strategies are respectively:

U(C_i,C_j)=(p-c-m_p+m_r,p-c-m_p+m_r)（2）

U(I_i,I_j)=(0,0)（3）

$U(C_i,I_j)=\left\{\begin{array}{cc}(-c+m_r,p-m_p)& \mathrm{if}{V}_j>0\\ \left(\mathrm{0,0}\right)& \mathrm{if}{V}_j<0\end{array}\right.---\left(4\right)$

$U(I_i,C_j)=\left\{\begin{array}{cc}(p-m_p,-c+m_r)& \mathrm{if}{V}_i>0\\ \left(\mathrm{0,0}\right)& \mathrm{if}{V}_i<0\end{array}\right.---\left(5\right)$

wherein i and j respectively represent relay participants, the resource cost consumed by the node for forwarding the packet is c, if the packet of one node is successfully forwarded by another node, the profit of the node is p, and the forwarding reward in the transmission transaction is m_rForward price of m_pOne node can buy resources to forward packets with the won forward reward, and we assume m_r，m_pThe same scaling mechanism is used for c, p. If node i forwards the packet for node j, and node j also forwards the packet for node i, then each node receives a certain forwarding reward m_rThe packet is forwarded to obtain the corresponding profit p, but the node has to have a certain forwarding cost c and a forwarding price m paid to the cooperative node for successfully forwarding the packet_pSimilarly, when both nodes i and j adopt the strategy of rejecting to forward the packet, each node has no payment and can not obtain any compensation, but when one of the two nodes adopts cooperation and the other node rejects cooperation, the node i forwards the packet of the node j but does not forward the packet of the node i, so that the benefit of the node i is m_r-c; at the same time, since the packet of node j is forwarded, it should pay a certain currency, V_i,V_jRespectively represent the currency amounts of nodes i, j, when V_i<0 or V_j<At 0, there is no interaction between the nodes.

In a multi-node league gaming process, a group of interests, made up of a plurality of participants, is typically represented in aggregate form. Based on three nodes i, j, k as an example, the federation that can be selected by any node i includes { i }, { i, j }, { i, k }, { i, j, k }, and nodes j, k in the same way can form four different federation sets respectively.

When a node exists in the alliance, the alliance income is a Nash equilibrium solution in a non-cooperative game according to the formulas (2) to (5); when two nodes exist in the alliance, the two nodes adopt a cooperation strategy meeting the pareto optimal, but other participants adopt a Nash equilibrium strategy; when three nodes exist in the alliance, all the nodes in the system participate in cooperation, the alliance income is the maximum, and every two nodes adopt cooperation strategies. Therefore, when there are n nodes in the union, any two nodes cooperate with each other, and the number of cooperation combinations isThe maximum benefit of the cooperation set formed by any two nodes is 2p-2c +2m_r-2m_pTo sum up, the maximum profit of the n-node federation is:

$a_{{\mathrm{\&tau;}}_k}=\frac{n(n-1)}{2}(2p-2c+2m_r-2m_p)---\left(6\right)$

in a typical league game, to obtain the maximum benefit of the league, the league is playedThe maximum profit of the alliance is strived to always minimize alliance consumption. For a single node, the higher the signal-to-interference ratio, the better the service quality and transmission efficiency, but this will increase the battery consumption and the interference to other nodes in the federation, so we establish a cost function defining the node based on the node transmission power and the signal-to-interference ratio, since the signal-to-interference ratio γ is_iIs a function related to factors such as transmission power, link gain, interference power and the like, and a non-linear cost function which can properly embody fairness and is relatively effective is provided as follows:

$c_{{\mathrm{\&tau;}}_k}=\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}({\mathrm{\&lambda;}}_i{p}_i+{({\mathrm{\&gamma;}}_{\mathrm{tar}}-{\mathrm{\&gamma;}}_i)}^2)---\left(7\right)$

${\mathrm{\&gamma;}}_i=\frac{h_i{p}_i}{N_{r}W}---\left(8\right)$

0≤p_i≤p_m（9）

where n represents the number of nodes in any federation, γ_tarIs a target signal-to-interference ratio, gamma_iRepresents the signal-to-interference ratio (SNR), λ, of node i_iProportional to the link gain, i.e. λ_i=kh_iK is a normal number, h_iRepresenting the link gain, p, between node i and other nodes in the federation_iRepresenting the transmission power, p, of node i_mRepresenting the maximum value of the transmission power of the node, W representing the bandwidth of the signal transmitted by each node, N_rRespectively, the channel additive gaussian noise power spectral densities of the receiving ends.

Therefore, in the wireless heterogeneous network multi-node relay alliance cooperative communication, for the maximum total income, the maximum alliance utility income is realized by minimizing an node cost function, and therefore, the following relay alliance optimization model is established:

$\max z=\max (a_{{\mathrm{\&tau;}}_k}-c_{{\mathrm{\&tau;}}_k})$

$s.t.a_{{\mathrm{\&tau;}}_k}=\frac{n(n-1)}{2}(2p-2c+{2m}_r-{2m}_p)$

p>m_p

m_r>c

$c_{{\mathrm{\&tau;}}_k}=\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}({\mathrm{\&lambda;}}_i{p}_i+{({\mathrm{\&gamma;}}_{\mathrm{tar}}-{\mathrm{\&gamma;}}_i)}^2)$

${\mathrm{\&gamma;}}_i=\frac{h_i{p}_i}{N_{r}W}$

0≤p_i≤p_m(10)

by optimizing the target, we can see that, in order to find the optimal number of nodes in the federation, we need to optimize the power of the nodes in the fixed federation. Therefore, when the number n of nodes in the alliance is any constant, the utility gain functionWhen the maximum value is taken, power optimization of nodes in the alliance needs to be carried out, so that the cost function of the nodes takes the minimum value, and the alliance optimization model in the step A is simplified into the following steps:

$\{p_1^{*},p_2^{*},\&CenterDot;\&CenterDot;\&CenterDot;p_n^{*}\}=\underset{n}{\arg }\min f(p_1,p_2,\&CenterDot;\&CenterDot;\&CenterDot;p_n)$

$=\underset{n}{\arg }\min \left(c_{{\mathrm{\&tau;}}_k}\right)$

$s.t.c_{{\mathrm{\&tau;}}_k}=\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}({\mathrm{\&lambda;}}_i{p}_i+{({\mathrm{\&gamma;}}_{\mathrm{tar}}-{\mathrm{\&gamma;}}_i)}^2)$

${\mathrm{\&gamma;}}_i=\frac{h_i{p}_i}{N_{r}W}$

0≤p_i≤p_m(11)

wherein, f (p)₁,p₂,…p_n) Representing a cost function of the relay alliance forwarding packets;respectively representing the optimal power of nodes in the alliance, and arg represents a variable meeting the function maximum value;

how to solve the optimal power of each node in the fixed alliance then becomes the core of the problem. By the pair formula f (p)₁,p₂,…p_n) The partial derivatives for each node transmit power are calculated and made zero:

$\frac{\&PartialD;f(p_1,p_2,\&CenterDot;\&CenterDot;\&CenterDot;p_n)}{\&PartialD;p_1}={\mathrm{\&lambda;}}_1-2({\mathrm{\&gamma;}}_{\mathrm{tar}}-{\mathrm{\&gamma;}}_1)\frac{h_1}{N_{r}W}=0---\left(12\right)$

similarly, a partial derivative formula of the n transmitting power of the alliance node can be obtained:

$\frac{\&PartialD;f(p_1,p_2,\&CenterDot;\&CenterDot;\&CenterDot;p_n)}{\&PartialD;p_n}={\mathrm{\&lambda;}}_n-2({\mathrm{\&gamma;}}_{\mathrm{tar}}-{\mathrm{\&gamma;}}_n)\frac{h_n}{N_{r}W}=0---\left(13\right)$

the optimal power of the nodes in the consolidated-available federation can be expressed as:

$p_1^{*}=\frac{N_{r}W}{h_1}{\mathrm{\&gamma;}}_{\mathrm{tar}}-\frac{{\mathrm{\&lambda;}}_1{N}_r^{2}W}{2h_1}$

$p_n^{*}=\frac{N_{r}W}{h_n}{\mathrm{\&gamma;}}_{\mathrm{tar}}-\frac{{\mathrm{\&lambda;}}_n{N}_r^{2}W}{2h_n}---\left(14\right)$

the formula shows that:

$\min f(p_1,p_2,\&CenterDot;\&CenterDot;\&CenterDot;p_n)=f(\frac{N_{r}W}{h_1}{\mathrm{\&gamma;}}_{\mathrm{tar}}-\frac{{\mathrm{\&lambda;}}_1{N}_r^{2}W}{2h_1},\&CenterDot;\&CenterDot;\&CenterDot;,\frac{N_{r}W}{h_n}{\mathrm{\&gamma;}}_{\mathrm{tar}}-\frac{{\mathrm{\&lambda;}}_n{N}_r^{2}W}{2h_n})---\left(15\right)$

the optimal total profit z under the fixed alliance condition can be obtained from the formula (6) and the formula (16)^*：

$z^{*}=\frac{n(n-1)}{2}(2p-2c+2m_r-2m_p)-f(\frac{N_{r}W}{h_1}{\mathrm{\&gamma;}}_{\mathrm{tar}}-\frac{{\mathrm{\&lambda;}}_1{N}_r^{2}W}{2h_1},\&CenterDot;\&CenterDot;\&CenterDot;,\frac{N_{r}W}{h_n}{\mathrm{\&gamma;}}_{\mathrm{tar}}-\frac{{\mathrm{\&lambda;}}_n{N}_r^{2}W}{2h_n})---\left(16\right)$

Comparing the total profit of alliances formed by different numbers of nodes in the relay alliance set M, so as to obtain the optimal number n of nodes in the multi-node relay alliance^*：：

$n^{*}=\underset{\mathrm{\&tau;}\&Element;M,n}{\arg }\max z^{*}---\left(17\right)$

Wherein z is^*=maxz。

For describing the optimal relay cooperation excitation method in the wireless heterogeneous network proposed by the present invention in more detail, a flowchart of the optimal relay cooperation excitation method in the wireless heterogeneous network shown in fig. 1 is illustrated as follows:

the first step is as follows: establishing network topology, initializing network environment (such as any node i in relay alliance, candidate alliance set)Signal bandwidth W, maximum value of transmitted power p_mForward cost c, forward reward m_rForward revenue p, forward price m_p）。

The second step is that: the node obtains various channel state information through the environment perception technology: channel gain h_iChannel noise power spectral density N_i。

The third step: according to a formula (6), firstly, calculating the maximum utility benefit of any alliance;

the fourth step: calculating the optimal power under the alliance according to the formula (15); the maximum total alliance yield under this power control condition is then calculated according to equation (16).

The fifth step: when tau is_k∈ M, comparing the profits of different relay alliances, and obtaining the optimal number n of the relay alliance nodes according to the formula (17)^*. Therefore, the optimum number of nodes n under optimum power control^*The formed union can obtain the maximum total yield, namely, the maximum total yield is the optimal solution of the optimization model (formula (10)).

Other advantages and modifications will readily occur to those skilled in the art, based upon the above description. Therefore, the present invention is not limited to the above specific examples, and a detailed and exemplary description of one aspect of the present invention will be given by way of example only. Those skilled in the art should appreciate that they can readily use the disclosed conception and specific embodiments as a basis for designing or modifying other structures for carrying out the same purposes of the present invention without departing from the spirit and scope of the invention as defined by the appended claims and their equivalents.

Claims (1)

1. The method is characterized in that a multi-node relay alliance model is adopted to obtain the number of the optimal nodes of the multi-node relay alliance; comprises the following steps:

step A, setting a relay alliance set asWherein tau is_kRepresents an arbitrary multi-node federation, is a natural number; the maximum total profit is realized by maximizing the alliance utility profit and minimizing the alliance node cost function, and then the relay alliance model is as follows:

$\max z=max(a_{{\mathrm{\&tau;}}_k}-c_{{\mathrm{\&tau;}}_k})$

$a_{{\mathrm{\&tau;}}_k}=\frac{n(n-1)}{2}(2p-2c+2m_r-2m_p)$

p>m_p

m_r>c

$c_{{\mathrm{\&tau;}}_k}=\underset{i=1}{\overset{n}{\&Sigma;}}({\mathrm{\&lambda;}}_i{p}_i+{\left({\mathrm{\&gamma;}}_{tar}-{\mathrm{\&gamma;}}_i\right)}^2)$

${\mathrm{\&gamma;}}_i=\frac{h_i{p}_i}{N_{r}W}$

0≤p_i≤p_m

wherein,

c is resource cost consumed by forwarding the packet by the node, p is the income of the node when the packet of the node is successfully forwarded by another node, m_rFor forwarding the reward, m_pIs the forwarding price;

z represents the total revenue of the federation,representing a node federation utility revenue function,representing a relay alliance cost function;

n represents the number of nodes in any alliance, and n is more than 2;

γ_taris a target signal-to-interference ratio, gamma_iRepresenting the signal-to-interference ratio SNR, h of the signal received by node i_iDenotes the link gain, λ, between node i and other nodes in the federation_iProportional to the link gain, i.e. λ_i＝kh_iK is a constant greater than zero, i is a natural number less than or equal to n;

p_irepresenting the transmission power, p, of any node i_mRepresents the maximum value of the transmitting power of any node, W represents the bandwidth of the signal transmitted by each node, N_rRespectively representing the power spectrum density of channel additive Gaussian noise of a receiving end;

step B, as alliance utility revenue functionAnd when the maximum value is taken, performing power control on the nodes in the alliance, so that the cost function of the nodes takes the minimum value, and simplifying the relay alliance model in the step A into:

$\begin{array}{c}\{p_1^{*},p_2^{*},...p_n^{*}\}=\underset{n}{\arg }\min f(p_1,p_2,...p_n)\\ =\underset{n}{\arg }\min \left(c_{{\mathrm{\&tau;}}_k}\right)\end{array}$

$c_{{\mathrm{\&tau;}}_k}=\underset{i=1}{\overset{n}{\&Sigma;}}({\mathrm{\&lambda;}}_i{p}_i+{\left({\mathrm{\&gamma;}}_{tar}-{\mathrm{\&gamma;}}_i\right)}^2)$

wherein, f (p)₁,p₂,…p_n) Representing a cost function of the relay alliance forwarding packets;respectively representing the optimal power of nodes in the alliance, and arg represents a variable meeting the function maximum value;

calculating an optimized target value of the model;

step C, comparing the total earnings of the alliances formed by different numbers of nodes in the relay alliance set M, namely obtaining the number n of the optimal nodes of the relay alliance^*：

$n^{*}=\underset{\mathrm{\&tau;}\&Element;M,n}{\arg }max{z}^{*}$

Wherein z is^*＝maxz。