CN104168634A - Distributed user location awareness cell closing method for LTE-A cellular network - Google Patents

Abstract

The invention discloses a distributed user location awareness cell closing method for an LTE-A cellular network, and belongs to the field of LTE-A downlink cooperative communication energy resource management in wireless communication. The distributed user location awareness cell closing method for the LTE-A cellular network is characterized in that cells which are small in number of users and low in utilization rate can be closed preferentially according to the number of users in peripheral communication cells, and thus the closing probability of all the cells with the low utilization rate can be improved. According to the distribution of the users in the communication cells, the optional adjacent base station range for receiving all transfer users in one cell is decreased, and therefore the algorithm complexity caused by the adoption of a minimax principle is reduced. The distributed user location awareness cell closing method for the LTE-A cellular network has the advantages that under the condition that the communication quality of the users is guaranteed, the user number threshold value participating in cell closing judgment can be adjusted dynamically according to the number of the users in the peripheral cells, the cell closing number is increased, and more energy can be saved; combined with the positions of the users in the cells, the search scope of an optimal receiving set is decreased, and hence the algorithm complexity is lowered.

Description

For the distributed user location aware community method for closing of LTE-A cellular network

Technical field

The present invention relates to a kind of distributing cell method for closing based on customer location, belong to the association area of LTE-A down collaboration communication energy resources management in radio communication.

Background technology

Pass through the fast development of decades, under the driving of ever-increasing data traffic demand, Information & Communication Technology (Information and Communication Technology, ICT) is just towards the future development of more focusing on performance, and relevant energy consumption also increases thereupon.A period of time recently, recognizing after the following fact, energy resource consumption is promoted to an important issue and extensive discussions: the energy resource consumption of ICT industry accounts for 3% of global energy consumption, the discharge capacity of greenhouse gas accounts for 2～4% of global emissions, and the energy consumption of ICT industry is with annual 15～20% speed increment, and within every 5 years, energy consumption is doubled.In conventional wireless cellular network, base station (Base Stations, BS) energy consumption accounts for the more than 70% of total energy consumption, and it should be noted that the load of bearing when base station is lighter or when non-loaded, its energy consumption still can reach 90% of its energy consumption peak value.This organizes, and digital watch is clear to be operated and will become an applicable approach that reduces energy consumption base station.

In order to meet the design principle of can guarantee service quality at any time (Quality of Service, QoS), operator can affix one's name to a lot of base stations in order to eliminate communication blind district at a slice intra-zone conventionally.Although this with increase number of base stations exchange for systematic function promote means very effective, owing to there is the base station of a large amount of low utilization rates, expended a large amount of energy.In conjunction with the tidal effect of communication service, can adopt the low utilization rate of closed portion base station in low business hours section and region, and not close base station by all the other the mode of covering is provided, significantly reduce base station energy consumption.In 3GPP Rel-9, specify that communication cell can, according to the statistical parameter such as flow load, state duration, determine by related algorithm whether community is closed.According to judgement mode, the method that community is closed can be divided into two large classes, and a class is only to close judgement by the signaling exchange between near base station, is distributed method for closing; Another kind of is between base station, to select a group head to carry out unified operation, is centralized method for closing.Distributed method for closing is comparatively flexible, can adjust shut down decision according to local region information.Centralized method for closing need to be obtained whole cell informations, and time decision is longer.

In the pertinent literatures such as " Distributed Inter-BS Cooperation Aided Energy Efficient Load Balancing for Cellular Networks ", " Traffic-and-Interference Aware Base Station Switching for Green Cellular Networks ", all the energy resources management of cellular communications networks is had to comparatively deep introduction and research.Although previous literature has also added different criterions in order to more reasonably to close judgement in having considered communication service tidal regime, but lack deep understanding and consideration for the user distribution characteristic in communication cell, this causes part utilance Di community to fail closing, thereby the energy-saving effect that community is closed is affected.Therefore, the present invention is directed to based on OFDM (Orthogonal Frequency Division Multiplexing, the energy resources management of LTE-A cellular network communication community OFDM) conducts in-depth research, analyze the community of previous literature put forward the methods and closed judgment condition, and the algorithm of selective reception shifting flows base station, and combine the user distribution characteristic in community, a kind of Xin community method for closing has been proposed.

List of references

1、Distributed?Inter-BS?Cooperation?Aided?Energy?Efficient?Load?Balancing?for?Cellular?Networks,IEEE?Transactions?on?Wireless?Communications,(Volume:12,Issue:11).

2、Traffic-and-Interference?Aware?Base?Station?Switching?for?Green?Cellular?Networks,IEEE18th?International?Workshop?on?Computer?Aided?Modeling?and?Design?of?Communication?Links?and?Networks(CAMAD),2013.

Summary of the invention

Main purpose of the present invention is, at the communication network of disposing in order to tackle peak value period traffic carrying capacity, in the time of its non-peak value period, cause in the situation of ample resources waste, for the sight of user's random distribution in cellular communications networks, to promote communication system efficiency as target, the LTE-A downlink communication environment building based on OFDM completes the power-save operation for poor efficiency community.Solve at non-peak value period communication network and expended the problem of unnecessary resource, and belonged to the communication quality of the transferring user of closing community by reducing algorithm complex and guarantee in conjunction with customer position information originally.

For the distributed user location aware community method for closing of LTE-A cellular network, its feature is: by the number of users information in conjunction with center cell and neighbor cell thereof and cell flow information, combine the poor efficiency community that decision is applicable to closing, according to center cell customer location, dynamically dwindle the hunting zone of transferring user optimum reception set, thereby reach the object that improves system capacity utilance and reduce algorithm complex.The present invention realizes successively according to the following steps:

For the distributed user location aware community method for closing of LTE-A cellular network, it is characterized in that comprising the steps:

Step 1: system initialization;

The communication network model of setting in the present invention is many Cellular Ad hoc networks of LTE-A model, and communication environment is set as the LTE-A downlink communication link based on OFDM technology; Each communication cell is divided into three sectors, and adopts the directional antenna with 120 ° of arguments, and 25 different Resource Block are distributed in each sector; All users are random distribution in 19 communication cells, and hypothesis holding position is constant; All users all always with its under base station carry out exchanges data with variable Rate not; Base station is assigned in its frequency resource block used for the transmitting power of different user, and adjusts according to its selected message transmission rate;

Step 2: for unique user UE _i, it receives Signal to Interference plus Noise Ratio Signal to Interference plus Noise Ratio, and SINR is:

${\mathrm{\γ}}_i=\frac{P_{i,\mathrm{recieved}}}{I_{i,\mathrm{intra}}+I_{i,\mathrm{inter}}+P_N}$

Wherein P _{i, recieved}the useful signal received power receiving for user side; I _{i, intra}for co-channel interference in community; I _{i, inter}for minizone co-channel interference; P _nfor additive white Gaussian noise, because this method has adopted OFDM technology, therefore in communication cell, there is not co-channel interference;

Step 3: according to known SINR, UE _ispectrum efficiency Spectral Efficiency, SE can be expressed as:

${\mathrm{\&psi;}}_i=\left\{\begin{array}{ccc}0& \mathrm{if}& {\mathrm{\&gamma;}}_i<{\mathrm{\&gamma;}}_{\min }\\ \mathrm{\&xi;}{\log }_2(1+{\mathrm{\&gamma;}}_i)& \mathrm{if}& {\mathrm{\&gamma;}}_{\min }\&le;{\mathrm{\&gamma;}}_i<{\mathrm{\&gamma;}}_{\max }\\ {\mathrm{\&psi;}}_{\max }& \mathrm{if}& {\mathrm{\&gamma;}}_i\&GreaterEqual;{\mathrm{\&gamma;}}_{\max }\end{array}\right.$

Wherein 0≤ξ≤1st, the decay factor that comprises active loss; γ _minthe minimum SINR that can be used for calculating spectrum efficiency; γ _maxthe maximum SINR that can be used for calculating spectrum efficiency; ψ _maxit is available maximum spectral efficiency;

Step 4: according to known SE, UE _ishared Resource Block Resource Blocks, RB number can be expressed as:

${\mathrm{\&beta;}}_i=\left[\frac{R}{W_{\mathrm{RB}}{\mathrm{\&psi;}}_i}\right]$

Wherein R _iuE _ithe message transmission rate of asking; W _rBit is a bandwidth that Resource Block is shared; Operator [x] represents x to carry out rounding operation;

Step 5: obtaining base station BS _jafter the shared RB number of all users, BS _jload factor Load Factor, LF can be expressed as:

${\mathrm{\&rho;}}_j=\frac{\mathrm{\&Sigma;}{\mathrm{\&beta;}}_{i,j}}{R{B}_{\mathrm{total}}}$

Wherein β _i,jto be subordinated to BS _juE _ithe RB number of asking; RB _totalbS _jthe all RB numbers that have;

Step 6: each base station is adjudicated to consideration successively according to the closing sequence pre-establishing:

Step 6.1: judge BS _jwhen in leading decision, whether enter for the first time judgement and consider;

Step 6.2: if, according to the number of users and the BS that do not close community around _jflow load consider whether it is applicable to closing;

Step 6.3: if not, only according to BS _jflow load consider whether it is applicable to closing;

Step 7: considering to close community BS according to user _jinterior distribution situation, gets rid of and receives the lower neighbor cell of transferring user possibility, thereby determines reception BS _juser's reception base station scope;

Step 8: from receiving the scope of base station, in the situation that keeping the original transmission rate of user, select one group of optimal set receive transferring user and close BS according to the very big algorithm of minimization _jif there is no available reception set, BS _jbe held open;

Step 9: repeating step 2 is to step 8 until all base station Close Alls that is applicable to closing.

Brief description of the drawings

Fig. 1 LTE-A self-organizing cellular network model;

Fig. 2 co-channel interference situation schematic diagram;

Fig. 3 number of users is combined consideration flow chart with flow information;

The algorithm complex of Fig. 4 based on customer location reduces concept map;

Fig. 5 user's random distribution schematic diagram;

Fig. 6 dynamic disturbance effect comparison diagram;

Fig. 7 combines consideration effect comparison diagram;

Fig. 8 the inventive method complexity reduces effect comparison diagram;

The different consideration scheme of Fig. 9 effect comparison diagram;

Figure 10 closes base station number and system total energy consumption contacts figure;

Figure 11 FB(flow block) of the present invention.

Embodiment

Below in conjunction with Figure of description, the specific embodiment of the present invention is illustrated:

Referring to shown in Fig. 1, is the LTE-A self-organizing cellular network concept map of the present invention's employing.Arbitrarily base station is only adjacent base station and carries out the communication interaction of user profile and flow information.Although there is no above information exchange between non-conterminous base station, but can influence each other by common adjacent base station.

Referring to shown in Fig. 2, is co-channel interference situation schematic diagram.Only the figure illustrates in same time, used with the community of source base station same resource block and can produce co-channel interference to the user of source base station service.

Refer to shown in Fig. 3, for number of users is combined consideration flow chart with flow information.This figure has explained the cooperation flow process of number of users information and flow information.

Refer to shown in Fig. 4, for the algorithm complex based on customer location reduces concept map.This figure illustration receive the idea of set contraction scope according to the location positioning of user in community.

Referring to shown in Fig. 5, is user's random distribution schematic diagram.In figure represent base station location, represent customer location

Step (1), for 19 Cellular Ad hoc networks models, sets its station spacing and is rice.And based on the consideration for frequency resource management, each communication cell is divided into three sectors, all adopt the directional antenna with 120 ° of arguments.25 different Resource Block are distributed in each sector, and the bandwidth of each Resource Block is 180kHz.All users are random distribution in 19 communication cells, and hypothesis holding position is constant, and always with its under base station carry out exchanges data with variable Rate not.For saving frequency resource, set each user and be only allowed to occupy a Resource Block, therefore base station will be adjusted according to this user-selected message transmission rate for user's transmitting power.Base station maximum transmission power is set as 43dBm, and sets and distribute to each downlink data transmission channel, and each user's minimum emissive power and maximum transmission power are respectively 15dBm and 30dBm.Setting the distribution power of each user's acquiescence is minimum emissive power.

Step (2), for unique user UE _i, its reception SINR is:

${\mathrm{\&gamma;}}_i=\frac{P_{i,\mathrm{recieved}}}{I_{i,\mathrm{intra}}+I_{i,\mathrm{inter}}+P_N}$

Wherein P _{i, recieved}for the useful signal received power that user side receives, I _{i, intra}for co-channel interference in community, I _{i, inter}for minizone co-channel interference, P _nfor noise density is-additive white Gaussian noise of 174dBm/Hz.Because this method has adopted OFDM technology, therefore in communication cell, there is not co-channel interference, i.e. I _{i, intra}=0, minizone co-channel interference is in same time, with UE _iuse same resource block to its influential all user's ∑ UE _{i' ≠ i}affiliated base station for UE _ipower stack, i.e. ∑ P _{i' ≠ i, recieved}.In order to obtain P _{i, recieved}, the present invention has adopted non-visual link (Non-Line-Of-Sight, NLOS) city macrocell (Urban Macrocell, the UMa) channel model of WINNER+:

PL＝(44.9-6.55log ₁₀(h _BS))log ₁₀(d)+5.83log ₁₀(h _BS)+14.78+34.97log ₁₀(f _c)

Wherein base station height h _bSbe made as 25 meters, carrier frequency f _cbe made as 2GHz, d is that base station is to the geometric distance between user.In conjunction with above formula and other relevant parameters, P _{i, recieved}can be represented as:

dBm(P _i,recieved)＝P _i+AG-PL-PeL-NF

Wherein P _ifor distributing to UE _ipower, AG is antenna gain, PeL is penetration loss, the noise factor that NF is user side, dBm (x) represents the dBm form of x.

Step (3), according to known SINR, UE _ispectrum efficiency SE can be expressed as:

${\mathrm{\&psi;}}_i=\left\{\begin{array}{ccc}0& \mathrm{if}& {\mathrm{\&gamma;}}_i<{\mathrm{\&gamma;}}_{\min }\\ \mathrm{\&xi;}{\log }_2(1+{\mathrm{\&gamma;}}_i)& \mathrm{if}& {\mathrm{\&gamma;}}_{\min }\&le;{\mathrm{\&gamma;}}_i<{\mathrm{\&gamma;}}_{\max }\\ {\mathrm{\&psi;}}_{\max }& \mathrm{if}& {\mathrm{\&gamma;}}_i\&GreaterEqual;{\mathrm{\&gamma;}}_{\max }\end{array}\right.$

Wherein 0≤ξ≤1st, the decay factor that comprises active loss, γ _minthe minimum SINR that can be used for calculating spectrum efficiency, γ _maxthe maximum SINR that can be used for calculating spectrum efficiency, ψ _maxit is available maximum spectral efficiency;

Step (4), according to known SE, UE _ishared Resource Block RB can be expressed as:

${\mathrm{\&beta;}}_i=\left[\frac{R}{W_{\mathrm{RB}}{\mathrm{\&psi;}}_i}\right]$

Wherein R _iuE _ithe message transmission rate of asking; W _rBit is a bandwidth that Resource Block is shared; Operator [x] represents x to carry out rounding operation; According to the setting in step (1), β _imust equal 1, therefore at β _i≠ 1 o'clock, need to be to its P _iin allowed band, adjust and regain SINR, SE and again carry out the calculating of number of resource blocks object.

Step (5), at β _iand P _iafter all no longer needing to adjust, BS _jload factor LF can be expressed as:

${\mathrm{\&rho;}}_j=\frac{\mathrm{\&Sigma;}{\mathrm{\&beta;}}_{i,j}}{R{B}_{\mathrm{total}}}$

Wherein β _i,jto be subordinated to BS _juE _ithe RB number of asking; RB _totalbS _jthe all RB numbers that have.

Step (6), is generating after the flow information of all base stations, according to the closing sequence pre-establishing, each base station is adjudicated to consideration successively:

Step (6.1), judges BS _jwhen in leading decision, whether enter for the first time judgement and consider.

Step (6.2), considers if enter for the first time judgement, the number of users of first adding up peripheral cell, and it is averaged, if BS _jnumber of users be less than this mean value, and BS _jflow load be less than the threshold value of closing of setting, prepare to start search and can accept its whole shifting flows and the reception collection of base stations without any QoS loss.If BS _jnumber of users be not less than peripheral cell user's mean value, and BS _jflow load be less than the threshold value of closing of setting, first consider to close other base stations.If BS _jflow load be not less than the threshold value of closing of setting, no longer consider to close this base station.

Step (6.3), considers if not entering for the first time judgement, only according to BS _jflow load consider whether it is applicable to closing, even BS _jflow load be less than the threshold value of closing of setting, prepare to start search and can accept its whole shifting flows and the reception collection of base stations without any QoS loss.

Step (7), calculates BS _jnumber of users in corresponding region, interior neighbor cell if number of users is zero in a certain region, is got rid of the reception set that contains this corresponding community, region from hunting zone, and the attempting of minimizing receives number of sets and can be represented as:

$\begin{array}{c}{T}_{\mathrm{dec}}=T_{\mathrm{tran}}-T_{\mathrm{loca}}\\ ={\mathrm{\&Sigma;}}_{n=1}^{n_{\max }}{\mathrm{\&Sigma;}}_{l=1}^{l_{\mathrm{total}}}{\mathrm{RS}}_{n,l}-{\mathrm{\&Sigma;}}_{n=1}^{n_{\max }-n_{\mathrm{exc}}}{\mathrm{\&Sigma;}}_{l=1}^{l_{\mathrm{total}}}\\ ={\mathrm{\&Sigma;}}_{n=1}^{n_{\max }}{C}_{n_{\max }}^n-{\mathrm{\&Sigma;}}_{n=1}^{n_{\max }-n_{\mathrm{exc}}}{C}_{n_{\max }-n_{\mathrm{exc}}}^n\end{array}{\mathrm{RS}}_{n,l}$

Wherein T _tranfor the reception number of sets of utilizing full way of search to obtain, T _locafor getting rid of the reception number of sets obtaining behind number of users Wei Ling community, corresponding region, RS _n,lfor l the reception set that contains n community, n _maxfor the maximum number of cells containing in all reception set, l _totalfor the reception set sum that contains n community, n _excfor BS _jthe number of cells being around excluded, $C_n^m=\frac{n!}{(n-m)!m!}.$

Step (8), from receiving the scope of base station, in the situation that not losing user communication quality, selects one group of optimal set receive transferring user and close BS according to the very big algorithm of minimization _jif there is no available reception set, BS _jbe held open:

Step (8.1), calculates RS _n,lin each base station estimate load factor Estimated Load Factor, ELF:

ρ _j',elf＝ρ _j'+ρ _j→j'≤ρ _max(j'≠j)

Wherein ρ _j'for BS _jadjacent base station BS _j'the load of itself, ρ _{j → j'}for BS _jbe transferred to BS _j'load, ρ _maxfor the maximum of base station can be accepted load threshold.If ρ _j ^'+ ρ _{j → j'}> ρ _max, judge that this reception set is unavailable.

If this reception set is available, in pair set all base stations estimate load summation, to obtain receiving whole flow loads of this set after whole transferring users:

$S_{n,l}={\mathrm{\&Sigma;}}_{j\&Element;{\mathrm{RS}}_{n,l}}{\mathrm{\&rho;}}_{j,\mathrm{elf}}$

Step (8.2), measures maximum to the load flow of all reception set with same base number, and records this reception set:

$S_{n,\max }=\max \left({\mathrm{\&Sigma;}}_{l=1}^{l_{\max }}{S}_{n,l}\right)$

Wherein l _maxfor the sum of the reception set that contains n community.Get the peaked object of load flow and be and make full use of as much as possible the ability to accept that receives base station, and improve other and have the possibility of closing compared with the base station of low discharge load, because receive after transferring user, transferring user may cause this reception base station not close.

Step (8.3), in order transferring user to be linked into try one's best in the middle of few base station, therefore all load flow maximums of obtaining in step (8.2) are got to minimum value, and the reception set that has this minimum load flow is the optimum set that receives:

$S_{\min i\max }=\min \left({\mathrm{\&Sigma;}}_{n=1}^{n_{\max }}{S}_{n,\max }\right)$

Step (9), repeating step (2) is to step (8) until all base station Close Alls that is applicable to closing.

Can express based on Fig. 6 dynamic disturbance effect comparison diagram: the dynamic disturbance comprising in the present invention is considered,, along with the way of upgrading transmitting power and interfere information is closed in each base station, can improve significantly the energy-saving effect that community method for closing brings.In figure, each emulated data draws by 100 different user's random distribution scenes.In figure represent not consider the method for closing (non dynamic interference, non_di) of dynamic disturbance, represent to consider the method for closing (dynamic interference, di) of dynamic disturbance.

Combining consideration effect comparison diagram based on Fig. 7 can express: consideration is combined in the number of users restriction comprising in the present invention with flow information can increase the number of closing community to a certain extent.And it should be noted that the energy-saving effect obtaining is different for fixing number of users threshold value, and it is lower to present number of users threshold value, the general trend that number is more is closed in base station.But too low Threshold also can cause the unstable of energy-saving effect, and fixing threshold value lacks the adaptability for different distributions situation.The consecutive mean number of users method for limiting proposing in the present invention can dynamically be adjusted threshold value and obtain certain energy saving profit according to the number of users of peripheral cell, increases the number of closing base station.In figure represent di method for closing, represent the di method for closing (dynamic interference number 15, di_num_15) of considering that number of users restriction and threshold value are 15, represent the di method for closing (dynamic interference number 17, di_num_17) of considering that number of users restriction and threshold value are 17, represent the di method for closing (dynamic interference number 19, di_num_19) of considering that number of users restriction and threshold value are 19, represent the di method for closing (dynamic interference number 21, di_num_21) of considering that number of users restriction and threshold value are 21, represent to consider the di method for closing (dynamic interference number average, di_num_average) of consecutive mean number of users restriction.

Reducing effect comparison diagram based on Fig. 8 the inventive method complexity can express: the algorithm complex minimizing method based on closing customer location in community comprising in the present invention does not significantly have influence on the energy-saving effect of system.In figure represent di method for closing, represent the di method for closing (dynamic interference predict, di_predict) of the minimizing algorithm complex of having considered customer location.

Can express based on the different consideration scheme of Fig. 9 effect comparison diagram: the consideration of combining of dynamic disturbance, number of users, customer location and flow information can be in reducing algorithm complex, the number that in elevator system, community is closed.As can be seen from the figure, in di_num_average method and di_pre_num_aver method, close number of base stations and be greater than the situation of 5 and all reached 85% left and right, di method and di_predict method remain on 70% left and right.Although comprise the consideration that reduces algorithm complex in di_num_average method, obviously do not have influence on the number of closing community.In figure represent di method for closing, represent di_predict method for closing, represent di_num_average method for closing, represent to consider the di method for closing (dynamic interference predict and number average, di_pre_num_aver) of number of users restriction and customer location.

Closing number and system total energy consumption based on Figure 10 base station contacts figure and can express: along with cell base station is closed increasing of number, system total energy consumption also declines thereupon, has further verified the validity of community method for closing.In figure represent the system energy consumption under different consideration methods, represent that the base station under different consideration methods closes number.In figure, di_predict is abbreviated as di_p, and di_num_average is abbreviated as di_n, and di_pre_num_aver is abbreviated as di_p_n.

Claims (1)

1. for the distributed user location aware community method for closing of LTE-A cellular network, it is characterized in that comprising the steps:

Step 1: system initialization;

The communication network model of setting in the present invention is many Cellular Ad hoc networks of LTE-A model, and communication environment is set as the LTE-A downlink communication link based on OFDM technology; Each communication cell is divided into three sectors, and adopts the directional antenna with 120 ° of arguments, and 25 different Resource Block are distributed in each sector; All users are random distribution in 19 communication cells, and hypothesis holding position is constant; All users all always with its under base station carry out exchanges data with variable Rate not; Base station is assigned in its frequency resource block used for the transmitting power of different user, and adjusts according to its selected message transmission rate;

Step 2: for unique user UE _i, it receives Signal to Interference plus Noise Ratio Signal to Interference plus Noise Ratio, and SINR is:

${\mathrm{\&gamma;}}_i=\frac{P_{i,\mathrm{recieved}}}{I_{i,\mathrm{intra}}+I_{i,\mathrm{inter}}+P_N}$

Wherein P _{i, recieved}the useful signal received power receiving for user side; I _{i, intra}for co-channel interference in community; I _{i, inter}for minizone co-channel interference; P _nfor additive white Gaussian noise, because this method has adopted OFDM technology, therefore in communication cell, there is not co-channel interference;

Step 3: according to known SINR, UE _ispectrum efficiency Spectral Efficiency, SE can be expressed as:

${\mathrm{\&psi;}}_i=\left\{\begin{array}{ccc}0& \mathrm{if}& {\mathrm{\&gamma;}}_i<{\mathrm{\&gamma;}}_{\min }\\ \mathrm{\&xi;}{\log }_2(1+{\mathrm{\&gamma;}}_i)& \mathrm{if}& {\mathrm{\&gamma;}}_{\min }\&le;{\mathrm{\&gamma;}}_i<{\mathrm{\&gamma;}}_{\max }\\ {\mathrm{\&psi;}}_{\max }& \mathrm{if}& {\mathrm{\&gamma;}}_i\&GreaterEqual;{\mathrm{\&gamma;}}_{\max }\end{array}\right.$

Wherein 0≤ξ≤1st, the decay factor that comprises active loss; γ _minthe minimum SINR that can be used for calculating spectrum efficiency; γ _maxthe maximum SINR that can be used for calculating spectrum efficiency; ψ _maxit is available maximum spectral efficiency;

Step 4: according to known SE, UE _ishared Resource Block Resource Blocks, RB number can be expressed as:

${\mathrm{\&beta;}}_i=\left[\frac{R}{W_{\mathrm{RB}}{\mathrm{\&psi;}}_i}\right]$

Wherein R _iuE _ithe message transmission rate of asking; W _rBit is a bandwidth that Resource Block is shared; Operator [x] represents x to carry out rounding operation;

Step 5: obtaining base station BS _jafter the shared RB number of all users, BS _jload factor Load Factor, LF can be expressed as:

${\mathrm{\&rho;}}_j=\frac{\mathrm{\&Sigma;}{\mathrm{\&beta;}}_{i,j}}{R{B}_{\mathrm{total}}}$

Wherein β _i,jto be subordinated to BS _juE _ithe RB number of asking; RB _totalbS _jthe all RB numbers that have;

Step 6: each base station is adjudicated to consideration successively according to the closing sequence pre-establishing:

Step 6.1: judge BS _jwhen in leading decision, whether enter for the first time judgement and consider;

Step 6.2: if, according to the number of users and the BS that do not close community around _jflow load consider whether it is applicable to closing;

Step 6.3: if not, only according to BS _jflow load consider whether it is applicable to closing;

Step 7: considering to close community BS according to user _jinterior distribution situation, gets rid of and receives the lower neighbor cell of transferring user possibility, thereby determines reception BS _juser's reception base station scope;

Step 8: from receiving the scope of base station, in the situation that keeping the original transmission rate of user, select one group of optimal set receive transferring user and close BS according to the very big algorithm of minimization _jif there is no available reception set, BS _jbe held open;

Step 9: repeating step 2 is to step 8 until all base station Close Alls that is applicable to closing.