CN102170700A - Cellular ad hoc network radio resource optimization and allocation method - Google Patents

Abstract

The invention relates to a cellular ad hoc network radio resource optimization and allocation method which comprises the following steps: for a given system model and parameters, obtaining a performance objective function which can be realized under a cooperative relay protocol and can select a monotonic increasing function reflecting the generation speed of application layer data required by users; giving the following constraint conditions required to be met by realizing performance indexes: different QoS (quality of service) requirements of end-users, a flow conservation principle, maximum and minimum transmission power requirements and attainable speed requirements; carrying out Lagrangian dual decomposition according to the objective function and the constraint conditions, respectively decomposing to a physical layer, a network layer and an application layer of OSI (open system interconnection), and respectively carrying out the optimized solution on each layer by using a sub-gradient mathematical method; and after respectively solving the optimization sub-problems of the physical layer, the network layer and the application layer, obtaining the optimal flow distribution, relay selection, transmission power and data generation speed distribution of each sub-carrier. The method provided by the invention ensures better stability and reality, and is suitable for the next generation of cellular ad hoc cooperative networks.

Description

A kind of honeycomb ad hoc network radio resources optimized distribution method

Technical field

The invention belongs to the wireless network communication technique field, be specifically related to a kind of honeycomb ad hoc network radio resources optimized distribution method.

Background technology

Next generation wireless network need provide the multimedia service and the data service of higher rate, and for example IEEE 802.11a/g requires the data rate of 54Mbps, and obtaining higher data speed with lower cost is the key problem of wireless communication field.In addition, another key element of following cordless communication network is in the ability that does not have to keep under the situation of intrinsic network configuration communication, and existing cellular network is owing to be subjected to the restriction of self character, promoting throughput and reducing the power loss only way is exactly the coverage that reduces each base station, thereby has reduced the radius of sub-district and the number of users of sub-district.But this way need provide bigger infrastructure expense, more will dispose a large amount of base stations.Obviously, this is not that we want.Ad hoc network is because the flexible nature of himself receives increasing concern in recent years; it is under the situation without any existing network infrastructure or centralized management; cooperation by one group of mobile node dynamically forms the casual network structure; the link of network internal is dynamic, usually can disconnect because of moving of node.Ad hoc has become an important field of research, this emerging technological expansion portable access, and make the communication under the emergency case become possibility.

In recent years, a large amount of scientific research institutions and manufacturer consider ad hoc network is combined with cellular system, form a kind of system of complementation, existing cellular system will rely on centralized control and management, and the standard of mobile radio system of future generation will make great efforts to develop towards the direction of ad hoc.And the complementary function that ad hoc and Cellular Networks are had makes the two the possibility that is combined into, and the characteristics of ad hoc network can solve the above-mentioned difficult point problem in the existing cellular network system effectively.Honeycomb ad hoc network is a kind of network schemer of mobile cellular and ad hoc coexistence, is one of principal mode of NGBW communication system.

Resource allocation plays crucial effects for performance and each functional module of wireless network, different with the Resource Allocation Formula of existing 2.5G, 3G communication network, honeycomb ad hoc network needs the access at random of multiple business such as control more effectively is real-time, non real-time, reduces the influence that the channel quality changeability is brought.Collaboration diversity was subjected to industry in recent years and paid close attention to widely as a kind of wireless channel multipath fading and method that promotes network coverage of resisting effectively.A plurality of independently terminal equipments form a virtual aerial array by cooperating mutually, under the situation that does not increase equipment complexity, have promoted the overall performance of system.Therefore, the method for salary distribution of the choice criteria of collaboration relay node, relaying strategy, user's resources such as qos requirement, channel, data flow and power all affects the arrangement of antenna and the systematic function of communication network.

In realizing process of the present invention, the inventor finds that there is following shortcoming in existing wireless network cooperation resource allocation techniques:

Existing wireless cooperation Resource Allocation Formula mainly concentrates on centralized beehive network system or distributed ad hoc network system.Two kinds of channel condition information differences that system can use when resource allocation, therefore the strategy that adopts is also different, and be not suitable for the centralized multi-hop resource allocation of honeycomb ad hoc network system, the existing allocative decision of industry is not considered the effect of relevant mutual information between the network different levels mostly based on traditional hierarchical network yet; In addition, for the multi-carrier OFDM A that can receive and dispatch simultaneously (Orthogonal Frequency Division Multiple Access) network, the various radio resource allocation strategy that the terminal use is local to transmit data, cooperate other phone user and ad hoc user transmit data all can affect the performance gain of cooperation; Add the different requirements of different user, thereby affect the overall performance of system, need more optimal Resource Allocation Formula QoS.

Summary of the invention

The objective of the invention is to have overcome the deficiencies in the prior art, can stride the method that layer resource optimization distributes according to the good and bad situation of the finiteness of user QoS demand, resource and channel conditions for honeycomb ad hoc network provides a kind of, this method obtains the maximum performance gain of collaboration communication technology under the situation of upgrading hardware not.

For solving the problems of the technologies described above, technical scheme of the present invention is:

A kind of honeycomb ad hoc network radio resources optimized distribution method, the method comprising the steps of:

(1) for given system model and parameter, obtain the performance objective function that can realize under the cooperating relay agreement, this target function selection can reflect the monotonic increasing function of the application layer data generating rate of user's request;

(2) provide the required satisfied constraints of following realization performance index: the qos requirement that the terminal use is different, stream conserva-tion principle, maximum and minimum transmit power require and the achievable rate requirement;

(3) carry out the Lagrange duality decomposition according to target function and constraints, decompose physical layer, network layer and the application layer of OSI respectively, the mathematical method of employing subgradient is carried out optimization respectively to each layer and is found the solution;

(4) after solving the optimization subproblem of physical layer, network layer and application layer respectively, the optimization flow distribution, relay selection, transmitting power and the data generation rate that obtain each subcarrier distribute.

The present invention with respect to the beneficial effect of prior art is:

The present invention is according to different user's requests, channel condition is selected collaboration relay node, coordination strategy, and the cooperation resource allocation methods proposed on this basis, compare with power distribution method with relay selection commonly used at present, because (multidimensional refers to flow rate in the present invention and distributes in the resource allocation of use multidimensional, relay selection, transmit power allocations, data generation rate distribution etc.), it carries out relative complex, but hardware does not need upgrading, convenient, expense is little, the resource allocation methods that proposes compared with prior art, under identical resource consumption situation, promote the overall utility of system, more adapted to the channel status that changes.So the present invention has better robustness and actuality, more be applicable to honeycomb ad hoc collaborative network of future generation.

Description of drawings

Fig. 1 is the architectural schematic of a specific embodiment of honeycomb ad hoc network of the present invention;

Fig. 2 is a honeycomb ad hoc network radio resources optimized distribution method flow diagram of the present invention.

Embodiment

Present embodiment is implemented under the prerequisite of technical solution of the present invention; provided a concrete network model configuration (see figure 1); and provided detailed execution mode and concrete operating process, but protection scope of the present invention is not limited to following network model and embodiment.

The embodiment of the invention is based upon on MMR (Mobile Multihop Relay) the honeycomb ad hoc network integration model, and the relay station RS in the model has played the part of two roles, and for base station BS, it is a mobile station MS; For travelling carriage, it is the relay station of base station.Whole network is a tree structure, and each bar relay data path all should terminate in the base station.The principle of model is to utilize many reliable links of two-forty to replace the low rate unreliable link, thereby it is multiplexing with the empty frequency of dividing to improve spectrum utilization.In addition, and cooperation transmission strategy employing decoding re-transmission of the present invention (Decode-and-Forward, DF) mode, but do not influence a characteristic of stock of the present invention.

A kind of honeycomb ad hoc network radio resources optimized distribution method that the embodiment of the invention provided as shown in Figure 2, may further comprise the steps:

Step s301, set up the basic model of cooperation communication system.

Concrete, being set in the cellular cell, a base station comprises the existence of a plurality of mobile phone users on every side.These terminal uses adopt the mode of OFDMA to send data to the base station, and some user may be through can arriving the base station once going on foot, and some user may need just can arrive the base station through multi-hop ad hoc network.Can assist to transmit data by the cooperating relay station between the different user.Each user has an antenna of supporting omnidirectional.

Step s302, for any mobile phone users, obtain the performance objective function under the cooperating relay agreement.

Concrete, at first set up the network performance objective function, this function is made as the non-linear increasing function of data generation rate, and concrete method to set up can be referring to execution mode; To setting up this application layer parameter of data generation rate of using in the performance objective functional procedure, set up its equation according to the stream conserva-tion principle.

Step s303, draw the constraints that realizes that target function need satisfy.

Concrete, the constraints of setting allowable resource; According to different terminals user's different demands, obtain the constraints of different data generation rates; According to the channel condition between the different user node, (referring to channel gain here) obtains the standard that relay station is selected.

Step s304, decompose, the subproblem after decomposing is optimized processing respectively according to target function and the constraints Lagrange duality of advancing.

Concrete, according to target function and constraints, carry out lagrangian optimization; The majorized function that obtains is carried out antithesis to be handled; The dual function that obtains is decomposed according to the OSI different levels, a complicated problems is decomposed on the different sublayers solve respectively; In the process that subproblem is optimized, can find the solution Lagrange multiplier, according to the convergence multiplier that obtains, be updated in the optimization solution of subproblem then, obtain best resource allocation parameters with the method for subgradient.

Step s305, be optimized resource allocation at each word carrier wave.

Concrete, top step only is that the resource of different user is optimized, and this step is at each subcarrier, and each user's total resources are assigned to respectively on the different subcarriers.

The present invention is further detailed explanation by embodiment below in conjunction with accompanying drawing.

That provide as Fig. 1 is the basic model figure of honeycomb ad hoc network cooperation communication system of the present invention, a base station BS is arranged in the network cell, M mobile subscriber MS arranged on every side, reticulate topology distribution around BS, distributed areas are 200m*200m, and mobile subscriber's set is expressed as

System adopts the OFDMA up link, N subcarrier, and the subcarrier set is expressed as The system link set is

Each mobile phone users can realize sending simultaneously own data and other user's relay transmission of cooperating in different subcarriers.

If P is (i, j, n _Ij), x (i, j, n _Ij) represent respectively mobile subscriber i to user j at subcarrier

In transmitted power and average Business Stream speed, S (i) expression mobile subscriber i generates the Mean Speed of data, O (i), I (i) represents that respectively the adfluxion that flows out with inflow user i closes.According to NUM (Network Utility Maximization) principle, the present invention selects terminal mobile user data generating rate utility function U _iThe summation of (S (i)) is as target function, and the selection of utility function is the monotonic increasing function of user data generating rate according to application layer user's demand, and it can reflect the network user's satisfaction, promptly

S(i)≥R(i)，?

0≤x(i，j，n _ij)≤r(i，j，n _ij)，?

Wherein R (i) is the minimum-rate requirement of terminal use i, n _Ij, n _JiBe respectively user i, the selected subcarrier of j, constraint formulations (3) has guaranteed different application layer user's different QoS requirements; Formula (2) is according to the stream conserva-tion principle, does not produce at certain node under the situation of data flow, flows into all data flow and all data flow conservations that flow out this node of this node; Constraint (4) is to guarantee that streaming rate can not surpass the flank speed that channel capacity may reach on every link; Formula (5) has retrained each terminal mobile subscriber's transmitted power less than maximum.By finding the solution the optimal user transmitted power P of above-mentioned target function ^*(i, j, n _Ij), flow rate x ^*(i, j, n _Ij), application layer user data generating rate S ^*(i), subcarrier n _Ij ^*And relaying i ^*, trunk subscriber i ^*At subcarrier

Interior transmitted power

Obtain optimum resource distribution mode.

Next, under above-mentioned constraints (2), (3), (4), (5), the optimization aim function is carried out Lagrangian conversion

Wherein λ (i), μ (i), ρ (i, j), τ (i) is Lagrange multiplier, λ, μ, ρ, τ are the Lagrange multiplier vector, S, x, P are respectively packet generating rate vector, link flow velocity vectors, transmitted power vector.Therefore, the definition dual function is

$G(\mathrm{\λ},\mathrm{\μ},\mathrm{\ρ},\mathrm{\τ})=\underset{S,x,P}{\max }L(S,x,P,\mathrm{\λ},\mathrm{\μ},\mathrm{\ρ},\mathrm{\τ})---\left(7\right)$

The optimization problem of formula (1) can be equivalent to

maximize?G(λ，μ，ρ，τ) (8)

According to duality theory,

The optimization problem of formula (8) can be decomposed into the application layer subproblem

And network layer subproblem

And physical layer subproblem

ε x wherein ²(i, j, n _Ij) be the auxiliary function that adds in order to find the solution subproblem (11), when ε is tending towards infinitesimal, ε x ²(i, j, n _Ij) be tending towards 0.

Subproblem (12) is the function of user's transmitted power, solves easily:

(A) user i is correctly decoded relay station, and relay station is when the channel condition of user j is better than channel condition between user i, the j, promptly

Wherein DS (i) representative can be correctly decoded terminal use i, and channel condition satisfies the relay station set of above-mentioned requirements.Then achievable rate is

Wherein P (i ', j, n _{Ij '}') be that the source sends the relay station node i j ' of user i at subcarrier n _{Ij '}' interior transmitted power, h (i, j, n _Ij), | h (i ', j, n _{Ij '}') | ²Be node i and j, node i ' and j between at subcarrier n _Ij, n _{Ij '}' interior channel gain, N ₀W is the mean square deviation power of Gaussian noise, and Γ is the capacity gap, r ₀(i, i ', n _Ij) be that via node i ' is at subcarrier n _IjThe interior end-to-end spectrum efficiency that can be correctly decoded the source node i data,

For being assigned with the subcarrier of use,

Representative is gathered DS (i), sets of sub-channels the user

Middle selection P (i ', j, n _{Ij '}') | h (i ', j, n _{Ij '}') | ²I ' the node of value maximum is as the via node i of user i ^*, subcarrier n _{Ij '}' as i ^*The subcarrier of selecting

For saving power, via node i ^*, subcarrier

Selection can be according to following standard:

Subproblem (12) solves easily by differentiate so

$P^{*}(i,j,n_{\mathrm{ij}})=\frac{\frac{\mathrm{\ρ}(i,j){\left|h(i,j,n_{\mathrm{ij}})\right|}^2}{2\ln 2\mathrm{\τ}\left(i\right)}-\mathrm{\Γ}{N}_{0}W-P^{*}(i^{*},j,{n_{{\mathrm{ij}}^{*}}}^{*}){\left|h(i^{*},j,{n_{{\mathrm{ij}}^{*}}}^{*})\right|}^2}{{\left|h(i,j,n_{\mathrm{ij}})\right|}^2}---\left(16\right)$

It should be noted that the subcarrier n of user i _Ij ^*It also is basis

Principle choose.Be that following formula can be rewritten as

$P^{*}(i,j,{n_{\mathrm{ij}}}^{*})={\left[\frac{\frac{\mathrm{\ρ}(i,j){\left|h(i,j,{n_{\mathrm{ij}}}^{*})\right|}^2}{2\ln 2\mathrm{\τ}\left(i\right)}-\mathrm{\Γ}{N}_{0}W-P^{*}(i^{*},j,{n_{{\mathrm{ij}}^{*}}}^{*}){\left|h(i^{*},j,{n_{{\mathrm{ij}}^{*}}}^{*})\right|}^2}{{\left|h(i,j,{n_{\mathrm{ij}}}^{*})\right|}^2}\right]}^{+}---\left(17\right)$

Wherein [x] ⁺=max{0, x}.

(B) user i does not have the relay station of being correctly decoded, or relay station is when the channel condition of user j is inferior to channel condition between user i, the j

$P^{*}(i,j,{n_{\mathrm{ij}}}^{*})={\left[\frac{\frac{\mathrm{\ρ}(i,j){\left|h(i,j,{n_{\mathrm{ij}}}^{*})\right|}^2}{2\ln 2\mathrm{\τ}\left(i\right)}-\mathrm{\Γ}{N}_{0}W}{2{\left|h(i,j,{n_{\mathrm{ij}}}^{*})\right|}^2}\right]}^{+}---\left(18\right)$

Do not need the cooperation of relay nodes this moment.

Subproblem (10), (11) separate S ^*(i), x ^*(i, j, n _Ij) can obtain by differentiation:

$S^{*}\left(i\right)={\left[U_i^{\′-1}(\mathrm{\μ}\left(i\right)-\mathrm{\λ}\left(i\right))\right]}^{+}---\left(19\right)$

$x^{*}(i,j,{n_{\mathrm{ij}}}^{*})={\left[\frac{\mathrm{\μ}\left(i\right)-\mathrm{\μ}\left(j\right)-\mathrm{\ρ}(i,j)}{2\mathrm{\ϵ}}\right]}^{+}---\left(20\right)$

Wherein

It is function U _iThe inverse function of derivative (x).

Lagrange multiplier λ in the following formula (i), μ (i), ρ (i, j), τ (i) can find the solution subprogram 1 by the method for following subgradient:

(1) initialization λ ⁽⁰⁾(i), μ ⁽⁰⁾(i), ρ ⁽⁰⁾(i, j), τ ⁽⁰⁾(i);

(2) known λ ^(t)(i), μ ^(t)(i), ρ ^(t)(i, j), τ ^(t)(i), try to achieve r according to (14), (15), (17), (18), (19), (20) ^*(i, j, n _Ij ^*), S ^*(i), x ^*(i, j, n _Ij ^*), P ^*(i, j, n _Ij ^*),

(3), upgrade Lagrange multiplier according to iterative formula

λ ^(t+1)(i)＝λ ^(t+1)(i)+β ^(t)(S ^*(i)-R _i) (21)

${\mathrm{\μ}}^{(t+1)}\left(i\right)={\mathrm{\μ}}^{\left(t\right)}\left(i\right)+{\mathrm{\β}}^{\left(t\right)}((\underset{j\∈O\left(i\right)}{\mathrm{\Σ}}{x}^{*}(i,j,{n_{\mathrm{ij}}}^{*})-\underset{j\∈I\left(i\right)}{\mathrm{\Σ}}{x}^{*}(j,i,{n_{\mathrm{ji}}}^{*})-S^{*}\left(i\right))---\left(22\right)$

ρ ^(t+1)(i，j)＝ρ ^(t)(i，j)+β ^(t)(r ^*(i，j，n _ij ^*)-x ^*(i，j，n _ij ^*)) (23)

${\mathrm{\τ}}^{(t+1)}\left(i\right)={\mathrm{\τ}}^{\left(t\right)}\left(i\right)+{\mathrm{\β}}^{\left(t\right)}(P_{\max }\left(i\right)-\underset{j\∈O\left(i\right)}{\mathrm{\Σ}}{P}^{*}(i,j,{n_{\mathrm{ij}}}^{*}))---\left(24\right)$

(4) return (2), till iteration convergence.

So far, each layer resource optimization of having finished honeycomb ad hoc network cooperation communication system distributes, and finished the resource allocation to relay station simultaneously.

Claims (2)

1. honeycomb ad hoc network radio resources optimized distribution method, the method comprising the steps of:

(1) for given system model and parameter, obtain the performance objective function that can realize under the cooperating relay agreement, this target function selection can reflect the monotonic increasing function of the application layer data generating rate of user's request;

(2) provide the required satisfied constraints of following realization performance index: the qos requirement that the terminal use is different, stream conserva-tion principle, maximum and minimum transmit power require and the achievable rate requirement;

(3) carry out the Lagrange duality decomposition according to target function and constraints, decompose physical layer, network layer and the application layer of OSI respectively, the mathematical method of employing subgradient is carried out optimization respectively to each layer and is found the solution;

(4) after solving the optimization subproblem of physical layer, network layer and application layer respectively, the optimization flow distribution, relay selection, transmitting power and the data generation rate that obtain each subcarrier distribute.

2. honeycomb ad hoc network radio resources optimized distribution method is characterized in that described monotonic increasing function is:

The constraints of formula (1) is:

S(i)≥R(i)，

0≤x(i，j，n _ij)≤r(i，j，n _ij)，

In formula (1)～(5), R (i) is the minimum-rate requirement of terminal use i, n _Ij, n _JiBe respectively user i, the selected subcarrier of j.

Images