CN104363619A - System and method for load balance of mobile communication network - Google Patents

Abstract

The invention relates to a system and a method for load balance of a mobile communication network. The method includes: monitoring loads of cells in the network; when the load of a certain cell is larger than a trigger threshold, selecting a load balance area, marking a switchable terminal, marking a switching group, generating an adjacent map according to adjacency relation of the cells, and setting a peak weight and a side weight of the adjacent map according to terminal loads; setting an equation by taking an adjacent map partition mode as an independent variable, taking minimization of weight difference of all parts of the partitioned adjacent map as a primary target and taking minimization of quantity of switching execution terminals as a secondary target; working out an optimal solution of the equation; obtaining user terminals, needed to be switched to the adjacent cell, in each cell in the balance area according to the optimal solution and an initial solution of the equation. Due to the fact that the target of the set equation is equivalent to minimization of load difference of the cells, the acquired optimal solution enables highest load balance of selected areas, network load balance solution is optimized, and failure in treatment of gathering of a great quantity of high-load cells is avoided.

Description

A kind of load-balancing method of mobile radio communication and system

Technical field

The present invention relates to communication technical field, in particular to a kind of load-balancing method of mobile radio communication and a kind of SiteServer LBS of mobile radio communication.

Background technology

Along with user is to the sharply increase of mobile access demand, there is the feature of Access Network community dense distribution in LTE mobile communication system, while improve unit are user capacity, also improves the signal strength signal intensity of unit are.But being limited to physical resource, the open ended number of users in community is still limited.In addition due to city mobile subscriber's dense distribution, current LTE system there will be the situation of multiple neighbor cell high capacity simultaneously.

Load-balancing methods more of the prior art often only consider the ordinary circumstance of single or a small amount of base station high capacity, and ignore LTE wireless access network new feature as above, when there is the high capacity simultaneously of a large amount of neighbor cell, the user being arranged in the high capacity community at center, owing to temporarily not having suitable switching target, causes algorithm to lose efficacy.

In addition, mobility load balance has two main targets: reduce the load of high capacity community and improve the equilibrium degree of each cell load in network.And load-balancing methods more of the prior art only load to certain threshold for target with what reduce high capacity community, although the number of cells had influence in implementation is less, mechanism is simple, but to improve the load balancing degrees of each community in network for target in current theoretical research, the load of high capacity community must be reduced, but the whole network load balancing degrees to be promoted due to the method, portfolio effect is good but relate to all communities in network due to needs, and execution efficiency is lower.And owing to only having high capacity community to be the short slab affecting user's access performance in network, load-balancing method relates to low loaded cell in a large number can't have obvious lifting to network performance.

Summary of the invention

Technical problem to be solved by this invention is, the problem of load balancing integrated solution how monitoring network region built, meet the overall load requirement of network, and according to community and terminal set up adjacent map solve how by terminal distribution to community, the settling mode of optimized network load balancing, and the situation avoiding that method lost efficacy when processing a large amount of high capacity community and assembling.

For this purpose, the present invention proposes a kind of load-balancing method of mobile radio communication, comprising: S1, the load of multiple community in monitoring objective network, when the load monitoring at least one community is greater than the first load threshold, enter step S2, S2, community load being greater than the second load threshold is labeled as Hot Spot, the community adjacent with described Hot Spot is labeled as peripheral cell, described Hot Spot and described peripheral cell form selection area, changeable terminal is marked in described selection area, in the neighbor cell of community, changeable terminal place, neighbor cell maximum for changeable terminal received signals intensity is set to Target cell corresponding to terminal changeable with this, changeable terminal is divided group according to Target cell, first community will be in and the changeable endpoint groups being Target cell with the second community, group is switched with being in the second community and merging into one with the changeable endpoint groups that the first community is Target cell, mark described switching group, S3, switching group information according to cell neighboring relations each in described selection area and mark generates cell adjacency figure, wherein, each summit in described adjacent map represents community, limit between two summits represents that corresponding Liang Ge community has neighbouring relations, and the switching group between corresponding Liang Ge community representated by two summit, every bar limit, the weight on limit is the quantity of changeable terminal in corresponding switching group, and the weight on summit is switching the terminal quantity outside group in respective cell, , S4, remembers that the set on described adjacent map summit is V, and the set on limit is E, then there is N in described adjacent map _vindividual point and N _ebar limit, limit E _iweights be m _i, some V _iweights be m _j, according to N _eindividual variable x _idetermine N _ethe slit mode on bar limit, wherein, i ∈ 1,2,3 ..., x _ne, limit E _i=(V _ij1, V _ij2), j ₁and j ₂for limit E _ithe mark of two end points, wherein, j ₁<j ₂, and j ₁, j ₂∈ 1,2,3 ..., N _e, if x _irepresent the weights identifying the limit that less point is got in two end points of back, a cut-out limit, if vectorial X=(x ₁, x ₂, x ₃..., x _ne),

$\mathrm{Fitness}\left(X\right)=\underset{j\&Element;V}{\mathrm{\&Sigma;}}{(M_j-\frac{\underset{j\&Element;V}{\mathrm{\&Sigma;}}{M}_j}{\left|V\right|})}^2+\mathrm{\&alpha;}\underset{i\&Element;E}{\mathrm{\&Sigma;}}|x_i-x_{0i}|,$

Wherein, the required terminal number switched in feasible solution during for getting minimum value to Fitness (X), α is 10 ^-3positive number, the initial segmentation state of adjacent map is X=(x ₀₁, x ₀₂, x ₀₃..., x _0Ne); S5, the feasible solution according to getting minimum value to Fitness (X) is split described adjacent map, according to each point after segmentation and the weights on residual limit that are connected to each point, the changeable terminal in described objective network is divided to corresponding community.

Preferably, described step S2 comprises: received signal strength in described selection area is less than the terminal of predetermined power, or the terminal label that signal to noise ratio is less than default signal to noise ratio is changeable terminal.

Preferably, described step S5 comprises: calculate feasible solution when getting minimum value to Fitness (X) according to quantum cluster ion algorithm.

Preferably, also comprise: S6, according to the quantity changeable terminal in described objective network being divided to terminal in rear community, and the Reference Signal Received Power that each terminal is uploaded, upgrade the bigoted parameter of individual character of community.

The present invention also proposes a kind of SiteServer LBS of mobile radio communication, comprising: monitoring means, for the load of community multiple in monitoring objective network, indexing unit, when the load that described monitoring means monitors at least one community is greater than the first load threshold, community load being greater than the second load threshold is labeled as Hot Spot, the community adjacent with described Hot Spot is labeled as peripheral cell, described Hot Spot and described peripheral cell form selection area, changeable terminal is marked in described selection area, in the neighbor cell of community, changeable terminal place, neighbor cell maximum for changeable terminal received signals intensity is set to Target cell corresponding to terminal changeable with this, changeable terminal is divided group according to Target cell, first community will be in and the changeable endpoint groups being Target cell with the second community, group is switched with being in the second community and merging into one with the changeable endpoint groups that the first community is Target cell, mark described switching group, adjacent map generation unit, cell adjacency figure is generated for the switching group information according to cell neighboring relations each in described selection area and mark, wherein, each summit in described adjacent map represents community, limit between two summits represents that corresponding Liang Ge community has neighbouring relations, and the switching group between corresponding Liang Ge community representated by two summit, every bar limit, the weight on limit is the quantity of changeable terminal in corresponding switching group, and the weight on summit is switching the terminal quantity outside group in respective cell, computing unit, remembers that the set on described adjacent map summit is V, and the set on limit is E, then there is N in described adjacent map _vindividual point and N _ebar limit, limit E _iweights be m _i, some V _iweights be m _j, according to N _eindividual variable x _idetermine N _ethe slit mode on bar limit, wherein, i ∈ 1,2,3 ..., x _ne, limit E _i=(V _ij1, V _ij2), j ₁and j ₂for limit E _ithe mark of two end points, wherein, j ₁<j ₂, and j ₁, j ₂∈ 1,2,3 ..., N _e, if x _irepresent the weights identifying the limit that less point is got in two end points of back, a cut-out limit, if vectorial X=(x ₁, x ₂, x ₃..., x _ne),

$\mathrm{Fitness}\left(X\right)=\underset{j\&Element;V}{\mathrm{\&Sigma;}}{(M_j-\frac{\underset{j\&Element;V}{\mathrm{\&Sigma;}}{M}_j}{\left|V\right|})}^2+\mathrm{\&alpha;}\underset{i\&Element;E}{\mathrm{\&Sigma;}}|x_i-x_{0i}|,$

Wherein, the required terminal number switched in feasible solution during for getting minimum value to Fitness (X), α is 10 ^-3positive number, the initial segmentation state of adjacent map is X=(x ₀₁, x ₀₂, x ₀₃..., x _0Ne), calculate the feasible solution of Fitness (X) being got to minimum value; Allocation units, for splitting described adjacent map according to described feasible solution, are divided to corresponding community according to each point after segmentation and the weights on residual limit that are connected to each point by the changeable terminal in described objective network.

Preferably, described indexing unit is used for received signal strength in described selection area to be less than the terminal of predetermined power, or the terminal label that signal to noise ratio is less than default signal to noise ratio is changeable terminal.

Preferably, described computing unit calculates feasible solution when getting minimum value to Fitness (X) according to quantum cluster ion algorithm.

Preferably, also comprise: updating block, for basis, the changeable terminal in described objective network is divided to the quantity of terminal in rear community, and the Reference Signal Received Power that each terminal is uploaded, upgrade the bigoted parameter of individual character of community.

Pass through technique scheme, whether changeable according to terminal (user) and switch target BS, terminal is divided into groups, and build the cell adjacency figure that band has a weight and limit weight, on the basis of this adjacent map, structure adjacent map cutting problems, each part of the adjacent map obtained after to every bar limit cutting (summit and several sections of residual limits be connected on this summit) is made to have weight close as far as possible, namely make the load of terminal be assigned to as far as possible equably in each community, thus ensure that the overall load in network area is minimum.

Accompanying drawing explanation

Can understanding the features and advantages of the present invention clearly by reference to accompanying drawing, accompanying drawing is schematic and should not be construed as and carry out any restriction to the present invention, in the accompanying drawings:

Fig. 1 shows the schematic flow diagram of the load-balancing method of mobile radio communication according to an embodiment of the invention;

Fig. 2 A to Fig. 2 C shows the schematic diagram building adjacent map according to an embodiment of the invention;

Fig. 3 A to Fig. 3 C shows the effect schematic diagram of load balancing according to an embodiment of the invention;

Fig. 3 D shows the effect schematic diagram of the whole network load balancing in prior art;

Fig. 4 A shows the difference schematic diagram carrying out the area load after the area load after load balancing and prior art the whole network load balancing according to one embodiment of the invention;

Fig. 4 B shows area ratio cumulative distribution schematic diagram corresponding to zones of different load difference that two kinds of equalization schemes cause;

After Fig. 5 shows and carries out load balancing according to an embodiment of the invention, cells block rate adds up schematic diagram;

Fig. 6 shows the schematic block diagram of the SiteServer LBS of mobile radio communication according to an embodiment of the invention.

Embodiment

Can more clearly understand above-mentioned purpose of the present invention, feature and advantage, below in conjunction with the drawings and specific embodiments, the present invention is further described in detail.It should be noted that, when not conflicting, the feature in the embodiment of the application and embodiment can combine mutually.

Set forth a lot of detail in the following description so that fully understand the present invention; but; the present invention can also adopt other to be different from other modes described here and implement, and therefore, protection scope of the present invention is not by the restriction of following public specific embodiment.

As shown in Figure 1, the load-balancing method of mobile radio communication according to an embodiment of the invention, comprising: S1, and the load of multiple community in monitoring objective network, when the load monitoring at least one community is greater than the first load threshold, enters step S2, S2, community load being greater than the second load threshold is labeled as Hot Spot, the community adjacent with described Hot Spot is labeled as peripheral cell, described Hot Spot and described peripheral cell form selection area, changeable terminal is marked in described selection area, in the neighbor cell of community, changeable terminal place, neighbor cell maximum for changeable terminal received signals intensity is set to Target cell corresponding to terminal changeable with this, changeable terminal is divided group according to Target cell, first community will be in and the changeable endpoint groups being Target cell with the second community, group is switched with being in the second community and merging into one with the changeable endpoint groups that the first community is Target cell, mark described switching group, and the signal strength signal intensity of the terminal in this group according to the less numbering end points respective cell received is sorted, S3, switching group information according to cell neighboring relations each in described selection area and mark generates cell adjacency figure, wherein, each summit in described adjacent map represents community, limit between two summits represents that corresponding Liang Ge community has neighbouring relations, and the switching group between corresponding Liang Ge community representated by two summit, every bar limit, the weight on limit is the quantity of changeable terminal in corresponding switching group, and the weight on summit is switching the terminal quantity outside group in respective cell, S4, the set on note adjacent map summit is V, and the set on limit is E, then there is N in adjacent map _vindividual point and N _ebar limit, limit E _iweights be m _i, some V _iweights be m _j, according to N _eindividual variable x _idetermine N _ethe slit mode on bar limit, wherein, i ∈ 1,2,3 ..., x _ne, limit E _i=(V _ij1, V _ij2), j ₁and j ₂for limit E _ithe mark of two end points, wherein, j ₁<j ₂, and j ₁, j ₂∈ 1,2,3 ..., N _e, if x _irepresent the weights identifying the limit that less point is got in two end points of back, a cut-out limit, if vectorial X=(x ₁, x ₂, x ₃..., x _ne),

$\mathrm{Fitness}\left(X\right)=\underset{j\&Element;V}{\mathrm{\&Sigma;}}{(M_j-\frac{\underset{j\&Element;V}{\mathrm{\&Sigma;}}{M}_j}{\left|V\right|})}^2+\mathrm{\&alpha;}\underset{i\&Element;E}{\mathrm{\&Sigma;}}|x_i-x_{0i}|,$

Wherein, the required terminal number switched in feasible solution during for getting minimum value to Fitness (X), α is 10 ^-3positive number, the initial segmentation state (i.e. equation initial solution) of adjacent map is X=(x ₀₁, x ₀₂, x ₀₃..., x _0Ne); S5, feasible solution (i.e. equation optimal solution) according to getting minimum value to Fitness (X) is split adjacent map, according to each point after segmentation and the weights on residual limit that are connected to each point, the changeable terminal in objective network is divided to corresponding community.

The condition that said method triggers is that then informing network activates load-balancing function when whether certain cell load in monitoring network has exceeded the first load threshold, otherwise continues monitoring cell load.

When each base station activation load-balancing function of network advertisement, the second load threshold is notified in the lump each community in network, in network, be designated as Hot Spot more than the community of the second load threshold, form set C _h, the neighboring station outside Hot Spot is the lower base station of load, is designated as peripheral cell, forms set C _p.Due to the randomness of location of mobile users and business shape body, Hot Spot in a network position is random, so region may have separate multiple subregions to form, can search these subregions in the following manner.

Due to the existence of X2 interface in LTE, Hot Spot can obtain its information of adjacent cells accurately, this Hot Spot and its all neighbor cell form a S set, if another set of S and Hot Spot and all neighbor cells formation thereof has common factor, then union is asked to merge these two set, continuous merging, until remaining set is not all occured simultaneously, remaining each set constitutes each independently subregion, and the quantity of remainder set is exactly the separate subregion quantity of selected areas in network.

After being divided into different group to the changeable terminal in each community according to Target cell, there is certain in two adjacent sectors to group, the changeable terminal in one of them group is with the community at another group place for Target cell, and vice versa.The group with above-mentioned feature exists in pairs, and has following feature: or must will access in Liang Ge community after the terminate load equilibrium in group switches.So two groupings can merge, and can sort according to the less community acknowledge(ment) signal intensity of numbering.

For the community of seven in Fig. 2 A, in community 1, changeable terminal can be divided into 6 groups, numbering is respectively { 1.2,1.3,1.4,1.5,1.6,1.7}, the changeable terminal in community 2 is divided into 3 groups (neglect one of them group, the target BS of the user's optimum signal in this group is outside community 1 ~ 7), and numbering is respectively { 2.1,2.7,2.3}, for the changeable terminal in the group 1.2 and 2.1 in community 1 and 2, corresponding Target cell is respectively community 1 and community 2, thus can merge into one and switch group, as shown in Figure 2 B.Switch group owing to existing between community 1 and 2, can determine that two minizones exist neighbouring relations and have limit E thus in adjacent map ₁, the community at Liang Ge group place is converted to limit E respectively ₁two end points V ₁and V ₂all groupings are changed in the manner described above and then merges, adjacent map as that shown in fig. 2 c can be obtained, its angular vertex weight is the terminal quantity switched in this respective cell outside group, limit weight is the terminal quantity in switching group corresponding to limit, and namely the larger represented number of users of weight is more.

Just in Access Network, a unit load is produced when a terminal sends service request, then cell load can represent by the quantity of any active ues terminal of community accommodation, thus problem of load balancing can be converted into how cutting adjacent map, make the weight of each part after cutting (summit and several the residual limits be connected with this summit) closest, what also just make user load distribute is more balanced.

Preferably, step S2 comprises: terminal received signal strength in selection area being less than predetermined power, or the terminal label that signal to noise ratio is less than default signal to noise ratio is changeable terminal.

Mark can carry out the user of load balancing switching according to the RSRP (Reference Signal Received Power) of reporting of user in community or SINR (signal and interference-to-noise ratio), and these users are divided into groups according to the handover-target base station of signal the best.

Can assess the Received signal strength from current service cell and neighbor cell, the acknowledge(ment) signal of current service cell is changeable terminal lower than the terminal label of predetermined power.Because after user is switched to Target cell, the signal of former Serving cell becomes the interference signal of Target cell, the user that acknowledge(ment) signal intensity is high in former Serving cell, in Target cell, its service quality cannot meet the minimum requirements of proper communication, thus need such power threshold to determine changeable terminal.In order to ensure there is best signal strength signal intensity after user switches, terminal can be set and can only be switched to the strongest neighbor cell of acknowledge(ment) signal, after such load balancing switches, can only accessing at Current Serving BTS of changeable terminal, or selected target BS.

Preferably, step S5 comprises: calculate feasible solution when getting minimum value to Fitness (X) according to quantum cluster ion algorithm.

Preferably, also comprise: S6, according to the quantity changeable terminal in objective network being divided to terminal in rear community, and the Reference Signal Received Power that each terminal is uploaded, upgrade the bigoted parameter of individual character of community.

Because the user in adjacent map representated by limit sorts compared with the acknowledge(ment) signal of small-cell according to numbering, and the element x in adjacent map cutting scheme X _i, represent E _ishould to numbering less cell allocation x _iindividual user, to another cell allocation m _i-x _iindividual user.Thereafter according to user's distribution condition, and the RSRP information of reporting of user, upgrade community individual character offset parameter CIO.

E _icorresponding Liang Ge community is designated as j ₁and j ₂, and j ₂corresponding cell number is less than community j ₁if, j ₂the number of terminals x distributed _i<x _0i, then this x _ithere is the terminal k of minimum received signal strength, RSRP in individual terminal _{j1, k}be more than or equal to RSRP _{j2, k}, and CIO _{j1, j2}=RSRP _{j1, k}-RSRP _{j2, k}, CIO _{j1, j2}=-CIO _{j2, j1}; If x _0i<x _i, then community j ₂the number of terminals m distributed _i-x _i<m _i-x _0i, this m _i-x _ithere is the terminal k of minimum acknowledge(ment) signal intensity, RSRP in individual terminal _{j1, k}be less than or equal to RSRP _{j2, k}, then CIO _{j1, j2}=RSRP _{j2, k}-RSRP _{j1, k}, CIO _{j1, j2}=-CIO _{j2, j1}.

Particularly, suppose each community of this LTE Access Network, due to the restriction of physical resource, guarantee QoS of customer, can hold at most difficult 100 load units, other relevant parameters of network are as shown in table 1 below.Terminal in network is divided into background terminal and hot terminal, and wherein background terminal is uniformly distributed according in network range, hot terminal with ad-hoc location in network for average, with 2 times of minizone distances for variance normal distribution.

Parameter name (unit)	Value
		Bandwidth (MHz)	20
Minizone distance (m)	500
		Path loss (dB)	L(d)＝34+40log(d)
Shadow fading	With δ s=8dB is the logarithm normal distribution of variance
		Through-put power (dBm)	46
Thermal noise power (dBm)	-104

Load balancing triggers thresholding	0.9
		Load balancing regional choice thresholding	0.6

Table 1

When in network, certain cell load exceedes certain threshold, network enters load balancing state, and by regional choice load-threshold, selects Hot Spot, and in conjunction with the information of adjacent cells of community, determines equalization region.As shown in Fig. 4 .b, shade mark community is Hot Spot, and the community of shadow-free mark is peripheral cell, and the Hot Spot in figure and peripheral cell form two independently load balancing subregions altogether, can build cell adjacency figure and adjacent map cutting problems respectively.

Wherein M _jthe weight of each part after expression cutting.Utilize quanta particle swarm optimization to solve this problem, N=40 particle upgrades particle position vector X according to following formula _i=(x _i1, x _i2, x _i3..., x _iD), then have

$x_{\mathrm{id}}=\left\{\begin{array}{cc}\left(\right|p_d+\mathrm{\&beta;}|{\mathrm{Mbest}}_d-x_{\mathrm{id}}|\ln \left(u\right)|+m_d)\mathrm{mod}{m}_d,& \mathrm{\&mu;}>0.5\\ \left(\right|p_d-\mathrm{\&beta;}|{\mathrm{Mbest}}_d-x_{\mathrm{id}}|\ln \left(u\right)|+m_d)\mathrm{mod}{m}_d,& \mathrm{\&mu;}\&le;0.5\end{array}\right.$

Wherein, p _dparticle local optimum position p _idwith global optimum position p _gdstochastic averagina, i.e. p _d=rp _id+ (1-r) p _gd, μ and r is random number between [0,1], and β is regulating parameter, and getting 0.9, Mbest is the average of all particle optimal solutions, namely

Load balancing effect, as shown in Figure 4, the area that in network, load gap is no more than 3% accounts for 98.79% of the gross area, load balance ability is excellent, Fig. 6 represents the impact of load balancing on network cell blocking rate, the load-balancing method of the mobile radio communication of known selection area effectively can reduce the blocking rate of network, and excellent effect.

As shown in Figure 6, the SiteServer LBS 20 of mobile radio communication comprises according to an embodiment of the invention: monitoring means 21, for the load of community multiple in monitoring objective network, indexing unit 22, when the load that monitoring means 21 monitors at least one community is greater than the first load threshold, community load being greater than the second load threshold is labeled as Hot Spot, the community adjacent with Hot Spot is labeled as peripheral cell, described Hot Spot and described peripheral cell form selection area, changeable terminal is marked in described selection area, in the neighbor cell of community, changeable terminal place, neighbor cell maximum for changeable terminal received signals intensity is set to Target cell corresponding to terminal changeable with this, changeable terminal is divided group according to Target cell, first community will be in and the changeable endpoint groups being Target cell with the second community, group is switched with being in the second community and merging into one with the changeable endpoint groups that the first community is Target cell, mark described switching group, adjacent map generation unit 23, group information for marking according to syntople and the indexing unit of selection area small area generates adjacent map, wherein, each summit of adjacent map represents a community, the limit connecting two summits represents that corresponding Liang Ge community has neighbouring relations, the limit of adjacent map is according to two summits corresponding switching group respectively, in adjacent map, the weight definition on limit is the number of changeable terminal in the switching group of its correspondence, and vertex weights is defined as in community the terminal number switched outside group, computing unit 24, the set on note adjacent map summit is V, and the set on limit is E, then there is N in adjacent map _vindividual point and N _ebar limit, limit E _iweights be m _i, some V _iweights be m _j, according to N _eindividual variable x _idetermine N _ethe slit mode on bar limit, wherein, i ∈ 1,2,3 ..., x _ne, limit E _i=(V _ij1, V _ij2), j ₁and j ₂for limit E _ithe mark of two end points, wherein, j ₁<j ₂, and j ₁, j ₂∈ 1,2,3 ..., N _e, if x _ito represent in two end points of back, a cut-out limit and identify that less point gets

The weights on limit, if vectorial X=(x ₁, x ₂, x ₃..., x _ne),

$\mathrm{Fitness}\left(X\right)=\underset{j\&Element;V}{\mathrm{\&Sigma;}}{(M_j-\frac{\underset{j\&Element;V}{\mathrm{\&Sigma;}}{M}_j}{\left|V\right|})}^2+\mathrm{\&alpha;}\underset{i\&Element;E}{\mathrm{\&Sigma;}}|x_i-x_{0i}|,$

Wherein, the required terminal number switched in feasible solution during for getting minimum value to Fitness (X), α is 10 ^-3positive number, the initial segmentation state of adjacent map is X=(x ₀₁, x ₀₂, x ₀₃..., x _0Ne), calculate the feasible solution of Fitness (X) being got to minimum value; Allocation units 25, for splitting adjacent map according to feasible solution, are divided to corresponding community according to each point after segmentation and the weights on residual limit that are connected to each point by the changeable terminal in objective network.

Preferably, indexing unit 22 is for being less than the terminal of predetermined power by received signal strength in selection area, or the terminal label that signal to noise ratio is less than default signal to noise ratio is changeable terminal.

Preferably, computing unit 24 calculates feasible solution when getting minimum value to Fitness (X) according to quantum cluster ion algorithm.

Preferably, also comprise: updating block 26, for basis, the changeable terminal in objective network is divided to the quantity of terminal in rear community, and the Reference Signal Received Power that each terminal is uploaded, upgrade the bigoted parameter of individual character of community.

According to the embodiment of the present invention, additionally provide a kind of non-volatile machine readable media, store the program product for load balancing.

According to the embodiment of the present invention, additionally provide a kind of machine readable program, program makes machine perform the load-balancing method of mobile radio communication arbitrary in as above technical scheme.

According to the embodiment of the present invention, additionally provide a kind of storage medium storing machine readable program, wherein, machine readable program makes machine perform the load-balancing method of mobile radio communication arbitrary in as above technical scheme.

In the present invention, term " first ", " second " only for describing object, and can not be interpreted as instruction or hint relative importance.Term " multiple " refers to two or more, unless otherwise clear and definite restriction.

The foregoing is only the preferred embodiments of the present invention, be not limited to the present invention, for a person skilled in the art, the present invention can have various modifications and variations.Within the spirit and principles in the present invention all, any amendment done, equivalent replacement, improvement etc., all should be included within protection scope of the present invention.

Claims (8)

1. a load-balancing method for mobile radio communication, is characterized in that, comprising:

S1, the load of multiple community in monitoring objective network, when the load monitoring at least one community is greater than the first load threshold, enters step S2;

S2, community load being greater than the second load threshold is labeled as Hot Spot, the community adjacent with described Hot Spot is labeled as peripheral cell, described Hot Spot and described peripheral cell form selection area, changeable terminal is marked in described selection area, in the neighbor cell of community, changeable terminal place, neighbor cell maximum for changeable terminal received signals intensity is set to Target cell corresponding to terminal changeable with this, changeable terminal is divided group according to Target cell, first community will be in and the changeable endpoint groups being Target cell with the second community, group is switched with being in the second community and merging into one with the changeable endpoint groups that the first community is Target cell, mark described switching group,

S3, switching group information according to cell neighboring relations each in described selection area and mark generates cell adjacency figure, wherein, each summit in described adjacent map represents community, limit between two summits represents that corresponding Liang Ge community has neighbouring relations, and the switching group between corresponding Liang Ge community representated by two summit, every bar limit, the weight on limit is the quantity of changeable terminal in corresponding switching group, and the weight on summit is switching the terminal quantity outside group in respective cell;

S4, remembers that the set on described adjacent map summit is V, and the set on limit is E, then there is N in described adjacent map _vindividual point and N _ebar limit, limit E _iweights be m _i, some V _iweights be m _j, according to N _eindividual variable x _idetermine N _ethe slit mode on bar limit, wherein, i ∈ 1,2,3 ..., x _ne, limit E _i=(V _ij1, V _ij2), j ₁and j ₂for limit E _ithe mark of two end points, wherein, j ₁<j ₂, and j ₁, j ₂∈ 1,2,3 ..., N _e, if x _irepresent the weights identifying the limit that less point is got in two end points of back, a cut-out limit, if vectorial X=(x ₁, x ₂, x ₃..., x _ne),

$\mathrm{Fitness}\left(X\right)=\underset{j\&Element;V}{\mathrm{\&Sigma;}}{(M_j-\frac{\underset{j\&Element;V}{\mathrm{\&Sigma;}}{M}_j}{\left|V\right|})}^2+\mathrm{\&alpha;}\underset{i\&Element;E}{\mathrm{\&Sigma;}}|x_i-x_{0i}|,$

Wherein, $M_j=\underset{i\&Element;E}{\mathrm{\&Sigma;}}(m_j+m_{\mathrm{ij}}),\underset{i\&Element;E}{\mathrm{\&Sigma;}}|x_i-x_{0i}|$ The required terminal number switched in feasible solution during for getting minimum value to Fitness (X), α is 10 ^-3positive number, the initial segmentation state of adjacent map is X=(x ₀₁, x ₀₂, x ₀₃..., x _0Ne);

S5, the feasible solution according to getting minimum value to Fitness (X) is split described adjacent map, according to each point after segmentation and the weights on residual limit that are connected to each point, the changeable terminal in described objective network is divided to corresponding community.

2. the load-balancing method of mobile radio communication according to claim 1, it is characterized in that, described step S2 comprises: received signal strength in described selection area is less than the terminal of predetermined power, or the terminal label that signal to noise ratio is less than default signal to noise ratio is changeable terminal.

3. the load-balancing method of mobile radio communication according to claim 1, it is characterized in that, described step S5 comprises: calculate feasible solution when getting minimum value to Fitness (X) according to quantum cluster ion algorithm.

4. the load-balancing method of mobile radio communication according to any one of claims 1 to 3, it is characterized in that, also comprise: S6, according to the quantity changeable terminal in described objective network being divided to terminal in rear community, and the Reference Signal Received Power that each terminal is uploaded, upgrade the bigoted parameter of individual character of community.

5. a SiteServer LBS for mobile radio communication, is characterized in that, comprising:

Monitoring means, for the load of community multiple in monitoring objective network;

Indexing unit, when the load that described monitoring means monitors at least one community is greater than the first load threshold, community load being greater than the second load threshold is labeled as Hot Spot, the community adjacent with described Hot Spot is labeled as peripheral cell, described Hot Spot and described peripheral cell form selection area, changeable terminal is marked in described selection area, in the neighbor cell of community, changeable terminal place, neighbor cell maximum for changeable terminal received signals intensity is set to Target cell corresponding to terminal changeable with this, changeable terminal is divided group according to Target cell, first community will be in and the changeable endpoint groups being Target cell with the second community, group is switched with being in the second community and merging into one with the changeable endpoint groups that the first community is Target cell, mark described switching group,

Adjacent map generation unit, cell adjacency figure is generated for the switching group information according to cell neighboring relations each in described selection area and mark, wherein, each summit in described adjacent map represents community, limit between two summits represents that corresponding Liang Ge community has neighbouring relations, and the switching group between corresponding Liang Ge community representated by two summit, every bar limit, the weight on limit is the quantity of changeable terminal in corresponding switching group, and the weight on summit is switching the terminal quantity outside group in respective cell;

Computing unit, remembers that the set on described adjacent map summit is V, and the set on limit is E, then there is N in described adjacent map _vindividual point and N _ebar limit, limit E _iweights be m _i, some V _iweights be m _j, according to N _eindividual variable x _idetermine N _ethe slit mode on bar limit, wherein, i ∈ 1,2,3 ..., x _ne, limit E _i=(V _ij1, V _ij2), j ₁and j ₂for limit E _ithe mark of two end points, wherein, j ₁<j ₂, and j ₁, j ₂∈ 1,2,3 ..., N _e, if x _irepresent the weights identifying the limit that less point is got in two end points of back, a cut-out limit, if vectorial X=(x ₁, x ₂, x ₃..., x _ne),

$\mathrm{Fitness}\left(X\right)=\underset{j\&Element;V}{\mathrm{\&Sigma;}}{(M_j-\frac{\underset{j\&Element;V}{\mathrm{\&Sigma;}}{M}_j}{\left|V\right|})}^2+\mathrm{\&alpha;}\underset{i\&Element;E}{\mathrm{\&Sigma;}}|x_i-x_{0i}|,$

Wherein, $M_j=\underset{i\&Element;E}{\mathrm{\&Sigma;}}(m_j+m_{\mathrm{ij}}),\underset{i\&Element;E}{\mathrm{\&Sigma;}}|x_i-x_{0i}|$ The required terminal number switched in feasible solution during for getting minimum value to Fitness (X), α is 10 ^-3positive number, the initial segmentation state of adjacent map is X=(x ₀₁, x ₀₂, x ₀₃..., x _0Ne), calculate the feasible solution of Fitness (X) being got to minimum value;

Allocation units, for splitting described adjacent map according to described feasible solution, are divided to corresponding community according to each point after segmentation and the weights on residual limit that are connected to each point by the changeable terminal in described objective network.

6. the SiteServer LBS of mobile radio communication according to claim 5, it is characterized in that, described indexing unit is used for received signal strength in described selection area to be less than the terminal of predetermined power, or the terminal label that signal to noise ratio is less than default signal to noise ratio is changeable terminal.

7. the SiteServer LBS of mobile radio communication according to claim 5, is characterized in that, described computing unit calculates feasible solution when getting minimum value to Fitness (X) according to quantum cluster ion algorithm.

8. the SiteServer LBS of mobile radio communication according to any one of claim 5 to 7, is characterized in that, also comprise:

Updating block, divides the quantity of terminal in rear community to the changeable terminal in described objective network for basis, and the Reference Signal Received Power that each terminal is uploaded, and upgrades the bigoted parameter of individual character of community.