A 5G load balancing method based on adjustment coefficient

Abstract

The present invention discloses a 5G load balancing method based on an adjustment coefficient. The problem of cell congestion and low resource utilization in the prior art is solved. The method includes obtaining information on cell load and signal level values; calculating the physical resource block space margin of each cell relative to a reference value according to the cell resource quantity; determining the best and worst cells; obtaining users in the overlapping area of ​​the best and worst cells, selecting cells whose total physical resource blocks are less than the physical resource block space margin of the best cell and transferring them to the best cell, and updating the cell data; setting the balancing condition and performing convergence judgment. The present invention can obtain the number and margin of physical resource blocks occupied by each cell, calculate the corresponding comprehensive adjustment value; thereby searching for the best and worst cells, traversing the worst cell users that meet the resource space margin requirements in the best cell, and implementing load transfer, thereby solving the problem of low 5G system utilization and overload of some cells caused by inconsistent loads between cells.

Description

Technical Field

[0001] The present invention relates to the field of 5G technology, and in particular to a 5G load balancing method based on an adjustment coefficient. Background Art

[0002] The 3GPP working group defines load balancing as a part of wireless resource management, and its purpose is to provide service quality assurance for more users, because load balancing can solve the congestion problem of hotspot cells and improve the resource utilization of the system. 5G is no exception, especially when the 5G hotspot cells and the surrounding neighboring cells are in a serious traffic imbalance state. The patent number is ZL201410324613.0, and the name is a Chinese invention patent for a load balancing method based on a neighboring cell set. It discloses that by analyzing the load space that the minimum load cell can accommodate, the number of users to be transferred in the neighboring cell set is reversed, thereby solving the problem of traffic imbalance. However, NSLB still has many problems. First, the definition of the optimal cell is limited to the consideration of the load; second, the small share of each load transfer between cells leads to low convergence efficiency. Summary of the invention

[0003] The present invention mainly solves the problems of cell congestion and low resource utilization in the prior art, and provides a 5G load balancing method based on an adjustment coefficient.

[0004] The present invention also solves the problem that the optimal cell definition of the existing balancing method is limited to the load and the convergence efficiency is low, and provides a 5G load balancing method based on the adjustment coefficient.

[0005] The above technical problem of the present invention is mainly solved by the following technical solution: a 5G load balancing method based on an adjustment coefficient, comprising the following steps:

[0006] Step 1: Obtain cell load and signal level value information;

[0007] Step 2: Calculate the physical resource block space margin of each cell relative to the benchmark value based on the cell resource quantity;

[0008] Step 3: Calculate the reverse load and signal level scores of the cell, calculate the comprehensive adjustment value of each cell based on the allocation of the adjustment coefficient, and determine the best and worst cells;

[0009] Step 4: Obtain users in the overlapping area of ​​the best and worst cells, select a cell whose total physical resource blocks are less than the physical resource block space margin of the best cell, transfer the cell to the best cell, and update the cell data;

[0010] Step 5: Set the equilibrium condition and make a convergence judgment. If the convergence condition is not met, repeat steps 2 to 5 until the equilibrium condition is met.

[0011] The present invention can directly obtain the number and margin of physical resource blocks occupied by each cell, and then calculate the corresponding comprehensive adjustment value; thereby searching for the best and worst cells, traversing the worst cell users that can meet the resource space margin requirements in the best cell, and implementing load transfer, thereby solving the problem of low 5G system utilization and overload of some cells caused by inconsistent loads between cells, and freeing up more resources to provide users with 5QI (5G QoS Indicator) protection.

[0012] As a preferred solution, the step 1 specifically includes:

[0013] Get n neighboring cells NBS n ={gNB 1 ,gNB 2 ,…,gNB n}Load {Cld 1 ,Cld 2 ,…,Cld n}, total number of physical resource blocks {PrT 1 ,PrT 2 ,…,PrT n}, and the signal level value {RSRP 1 ,RSRP 2 ,…,RSRP n}information.

[0014] As a preferred solution, the step 2 specifically includes:

[0015] (2-1): For NBS n ={gNB 1 ,gNB 2 ,…,gNB n}, calculate the corresponding physical resource block occupancy Pro i =ceil(Cld i *PrT i ), where ceil() represents the ceiling function;

[0016] (2-2): Calculate the physical resource block balance reference value of all cells

[0017] (2-3): Calculate the physical resource block space margin Prv of each cell i =Pro Av -Pro i , where i=1,2,…,n.

[0018] As a preferred solution, the step three specifically includes:

[0019] (3-1): For NBS n ={gNB 1 ,gNB 2 ,…,gNB n Each cell gNB in i , calculate the reverse load Cld′ i =1-Cld i ;

[0020] Calculate the mathematical expectation of the signal level Calculate the signal level score of each cell separately

[0021] (3-2): Set the adjustment coefficient A C , calculate the comprehensive adjustment value Cv i =A C *Cld′ i +(1-A C )*RW i , where 0 C ≤1;

[0022] (3-3): In NBS n Find the one with the largest Cv i Value of cell gNB mx and minimum Cv i Value of cell gNB mn , respectively determined as the best cell and the worst cell. i If there are multiple cells with the same value, you can choose any one of them.

[0023] As a preferred solution, the step 4 specifically includes:

[0024] (4-1): Obtain the best cell gNB mx and the worst cell gNB mn All users in the coverage overlap area UsrN = {Usr 1 ,Usr 2 ,…,Usr mxn ,}, and the number of physical resource blocks requested by the user ProN = {Proxn 1 ,Proxn 2 ,…,Proxn mxn};

[0025] (4-2): Arrange the users UsrN in ascending order of the number of physical resource blocks ProN;

[0026] (4-3): Get the conditions that meet the requirements ​The maximum PreN value, where PreN≤mxn; migrate all PreN users to the optimal cell gNB mx middle;

[0027] (4-4): Update the optimal cell gNB mx Physical resource block occupancyPro mx =Pro mx +TProN, worst cell gNB mn Physical resource block occupancyPro mn =Pro mn -TProN;

[0028] According to the updated Pro mx and Pro mn Value update cell gNB mx Load

[0029]

[0030] Update cell gNB mn Load

[0031]

[0032] As a preferred solution, the specific process of step 5 includes:

[0033] (5-1): Set the acceptable equilibrium state β 0 , convergence target value ε, for all cells NBS n ={gNB 1 ,gNB 2 ,…,gNB n}, calculate the load balancing factor

[0034]

[0035] (5-2): Determine whether the convergence condition is met

[0036]

[0037] If satisfied, end the step; if not satisfied, repeat steps 2 to 5 until the equilibrium condition is met, end the step.

[0038] Therefore, the advantages of the present invention are: it can directly obtain the number and margin of physical resource blocks occupied by each cell, and then calculate the corresponding comprehensive adjustment value; thereby searching for the best and worst cells, traversing the worst cell users that can meet the resource space margin requirements in the best cell, and implementing load transfer, thereby solving the problem of low 5G system utilization and overload of some cells caused by inconsistent loads between cells, and making use of more resources to provide users with 5QI protection. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 It is a schematic diagram of the process structure of the present invention;

[0040] Figure 2 It is a comparison diagram of the convergence efficiency of the method of the present invention and other methods in simulation experiments;

[0041] Figure 3 This is a comparison chart of the maximum overloaded cell in the simulation experiment between the method of the present invention and other methods;

[0042] Figure 4 It is a comparison chart of cell load changes in simulation experiments between the method of the present invention and other methods. DETAILED DESCRIPTION

[0043] The technical solution of the present invention is further specifically described below through embodiments and in conjunction with the accompanying drawings.

[0044] Example:

[0045] This embodiment provides a 5G load balancing method based on an adjustment coefficient, such as Figure 1 As shown, the following steps are included:

[0046] Step 1: Obtain cell load and signal level information; specifically including:

[0047] Get n neighboring cells NBS n ={gNB 1 ,gNB 2 ,…,gNB n}Load {Cld 1 ,Cld 2 ,…,Cld n}, total number of physical resource blocks {PrT 1 ,PrT 2 ,…,PrT n}, and the signal level value {RSRP 1 ,RSRP 2 ,…,RSRP n}information.

[0048] Step 2: Calculate the physical resource block space margin of each cell relative to the benchmark value based on the cell resource quantity; specifically including:

[0049] (2-1): For NBS n ={gNB 1 ,gNB 2 ,…,gNB n}, calculate the corresponding physical resource block occupancy Pro i = Ceil(Cld i *PrT i ), where ceil() represents the ceiling function;

[0050] (2-2): Calculate the physical resource block balance reference value of all cells

[0051] (2-3): Calculate the physical resource block space margin Prv of each cell i =Pro Av -Pro i , where i=1,2,…,n.

[0052] Step 3: Calculate the reverse load and signal level scores of the cell, calculate the comprehensive adjustment value of each cell based on the allocation of the adjustment coefficient, and determine the best and worst cells; specifically include:

[0053] (3-1): For NBS n ={gNB 1 ,gNB 2 ,…,gNB n Each cell gNB in i , calculate the reverse load Cld′ i =1-Cld i ;

[0054] Calculate the mathematical expectation of the signal level Calculate the signal level score of each cell separately

[0055] (3-2): Set the adjustment coefficient A C , calculate the comprehensive adjustment value Cv i =A C *Cld′ i +(1-A C )*RW i , where 0 C ≤1;

[0056] (3-3): In NBS n Find the one with the largest Cv i Value of cell gNB​mx and minimum Cv i Value of cell gNB mn , respectively determined as the best cell and the worst cell. i If there are multiple cells with the same value, you can choose any one of them.

[0057] Step 4: Obtain users in the overlapping area of ​​the best and worst cells, select a cell whose total physical resource blocks are less than the physical resource block space margin of the best cell and transfer them to the best cell, and update the cell data; specifically including:

[0058] (4-1): Obtain the best cell gNB mx and the worst cell gNB mn All users in the coverage overlap area UsrN = {Usr 1 ,Usr 2 ,…,Usr mxn ,}, and the number of physical resource blocks requested by the user ProN = {Proxn 1 ,Proxn 2 ,…,Proxn mxn};

[0059] (4-2): Arrange the users UsrN in ascending order of the number of physical resource blocks ProN;

[0060] (4-3): Get the conditions that meet the requirements The maximum PreN value, where PreN≤mxn; migrate all PreN users to the optimal cell gNB mx middle;

[0061] (4-4): Update the optimal cell gNB mx Physical resource block occupancyPro mx =Pro mx +TProN, worst cell gNB mn Physical resource block occupancyPro mn =Pro mn -TProN;

[0062] According to the updated Pro mx and Pro mn Value update cell gNB mx Load

[0063]

[0064] Update cell gNB mn Load

[0065]

[0066] Step 5: Set the equilibrium condition and make a convergence judgment. If the convergence condition is not met, repeat steps 2 to 5 until the equilibrium condition is met. The specific process includes:

[0067] (5-1): Set the acceptable equilibrium state β 0 , convergence target value ε, for all cells NBS n ={gNB 1 ,gNB 2 ,…,gNB n}, calculate the load balancing factor

[0068]

[0069] (5-2): Determine whether the convergence condition is met

[0070]

[0071] If satisfied, end the step; if not satisfied, repeat steps 2 to 5 until the equilibrium condition is met, end the step.

[0072] The present embodiment is specifically described below with an example. Taking n=5 as an example, the cell load of the 5G system is shown in Table 1.

[0073] Table 1 Business carried out

[0074]

[0075] Table 2 Basic data

[0076]

[0077] A 5G load balancing method based on an adjustment coefficient comprises the following steps:

[0078] Step 1: Obtain cell load and signal level value information; the load information and signal level value information are shown in Table 1.

[0079] Step 2: Calculate the physical resource block space margin of each cell relative to the benchmark value based on the cell resource quantity;

[0080] About NBS n ={gNB 1 ,gNB 2 ,…,gNB n}, calculate the corresponding physical resource block occupancy

[0081] Pro i =ceil(Cld i *PrT i ) = {132,69,172,88,238};

[0082] Calculate the equalization reference value of all cell physical resource blocks

[0083]

[0084] Calculate the physical resource block space margin of each cell

[0085] Prv i =Prb Av -Pro i ={8,71,-32,52,-98};

[0086] Step 3: Calculate the reverse load and signal level scores of the cell, calculate the comprehensive adjustment value of each cell based on the allocation of the adjustment coefficient, and determine the best and worst cells;

[0087] About NBS n ={gNB 1 ,gNB 2 ,…,gNB n Each cell gNB in i , calculate the reverse load

[0088] Cld′ i =1-Cld 9 ={0.52,0.75,0.37,0.68,0.13};

[0089] Calculate the mathematical expectation of the signal level Calculate the signal level score of each cell separately

[0090]

[0091] Set adjustment factor A C , calculate the comprehensive adjustment value

[0092] Cv i =A C *Cld′ i +(1-A C )*RW i ={0.59,0.82,0.48,0.74,0.34},

[0093] In NBS n Find the one with the largest Cv i Value of cell gNB mx and minimum Cv i Value of cell gNB mn , determine the best cell and the worst cell. In this example, the best cell and the worst cell are gNB 2 and gNB 5 .

[0094] Step 4: Obtain users in the overlapping area of ​​the best and worst cells, select a cell whose total physical resource blocks are less than the physical resource block space margin of the best cell, transfer the cell to the best cell, and update the cell data;

[0095] Get the best cell gNB mx and the worst cell gNB mn All users in the coverage overlap area UsrN = {Usr 1 ,Usr 2 ,…,Usr mxn ,}; as shown in Table 3:

[0096] Community Apply for the number of physical resource blocks Usr1 17 Usr2 53 Usr3 34 Usr4 26 Usr5 55

[0097] The number of physical resource blocks it applies for is

[0098] ProN = {Proxn 1 ,Proxn 2 ,…,Proxn mxn} = {17,53,34,26,55};

[0099] Arrange the users UsrN in ascending order according to the number of physical resource blocks ProN to obtain

[0100] {Usr 1 ,Usr 4 ,Usr 3 ,Usr 2 ,Usr 5 ,},

[0101] The corresponding numbers of physical resource blocks are {17, 26, 34, 53, 55};

[0102] Get the conditions that meet the requirements The maximum PreN value is 2, that is, the first two users {Usr 1 ,Usr 4}, these two users are transferred from gNB 5 Migration to gNB 2 middle,

[0103] Update the best cell gNB 2 Physical resource block occupancyPro 2 =Pro 2 +43=112, the worst cell gNB 5 Physical resource block occupancyPro 5 =Pro 5 -43 = 195; according to the updated Pro 2 and Pro5 Value update cell gNB 2 Load Update cell gNB 5 Load 0.71.

[0104] Step 5: Set the equilibrium condition and make a convergence judgment. If the convergence condition is not met, repeat steps 2 to 5 until the equilibrium condition is met.

[0105] For NBS n ={gNB 1 ,gNB 2 ,…,gNB 5}, calculate the load balancing factor

[0106]

[0107] Does not meet the convergence condition abs(β-β W )=0.0712≤0.01%, repeat steps 2 to 5 until the equilibrium condition is met, and then end the step.

[0108] The following simulation experiment is used to illustrate this.

[0109] The ACLB method of this embodiment and the previous NSLB method are simulated on the MATLAB platform. The basic data information is shown in Tables 1 to 3 above. Certain users are randomly added and random services are configured. The results are shown in Tables 1 to 3. Figures 2 to 4 shown.

[0110] like Figure 2 The load balancing convergence process of the two methods shown in Figure 2 is shown in Figure 2. By setting the optimal adjustment coefficient A C , the ACLB of the present invention converges faster than the previous NSLB, can search for the best and worst cells in a shorter time, and perform load transfer;

[0111] like Figure 3 The figure shows the user deloading situation of the maximum overloaded cell under different adjustment coefficients of ACLB. The larger the adjustment coefficient, that is, the overemphasis on the influence of cell load, the closer ACLB is to the NSLB state. In fact, the NSLB algorithm is ACLB under ACLB. C = a special case of 1;

[0112] like Figure 4 The ACLB in different adjustment coefficient states shows that the minimum load base station is subject to load transfer. Figure 3 Similar to A C The smaller it is, the more the target optimal cell will consider the signal level in addition to the load, which can make customers perceive better when the load transfer amount is equal.

[0113] The specific embodiments described herein are merely examples of the spirit of the present invention. Those skilled in the art may make various modifications or additions to the specific embodiments described or replace them in similar ways, but they will not deviate from the spirit of the present invention or exceed the scope defined by the appended claims.

Claims

1. A 5G load balancing method based on an adjustment coefficient, characterized in that: The following steps are involved: Step 1: Obtain cell load and signal level value information; Step 2: Calculate the physical resource block space margin of each cell relative to the benchmark value based on the cell resource quantity; Step 3: Calculate the cell reverse load Cld′ i Sum signal level score RW i , set the adjustment coefficient A c , calculate the comprehensive adjustment value Cv i =A c *Cld′ i +(1-A c )*RW i , where 0 c ≤1, find the one with the maximum Cv i Value and minimum Cv i The cell with the best value is determined as the optimal cell gNB mx and the worst cell gNB mn ;​ Step 4: Obtain the number of users and physical resource blocks applied for by the users in the overlapping area of ​​the best and worst cells, select a cell whose total number of physical resource blocks is less than the physical resource block space margin of the best cell and transfer it to the best cell, and update the cell data; Step 5: Set the equilibrium condition and make a convergence judgment. If the convergence condition is not met, repeat steps 2 to 5 until the equilibrium condition is met.

2. According to the 5G load balancing method based on the adjustment coefficient according to claim 1, it is characterized in that Step 1 specifically includes: Get n neighboring cells NBS n ={gNB1,gNB2,…,gNB n } of load {Cld1,Cld2,…,Cld n }, the total number of physical resource blocks {PrT1,PrT2,…,PrT n }, and signal level values ​​{RSRP1, RSRP2,…, RSRP n }information.

3. A 5G load balancing method based on adjustment coefficient according to claim 2, characterized in that The step 2 specifically includes: (2-1): For NBS n ={gNB1,gNB2,…,gNB n }, calculate the corresponding physical resource block occupancy Pro i =ceil(Cld i *PrT i ), where ceil() represents the ceiling function; (2-2): Calculate the physical resource block balance reference value of all cells (2-3): Calculate the physical resource block space margin Prv of each cell i =Pro Av -Pro i , where i=1,2,…,n.

4. A 5G load balancing method based on adjustment coefficient according to claim 3, characterized in that Step three specifically includes: (3-1): For NBS n ={gNB1,gNB2,…,gNB n Each cell gNB in i , calculate the reverse load Cld′ i =1-Cld i ; Calculate the mathematical expectation of the signal level Calculate the signal level score of each cell separately (3-2): Set the adjustment coefficient A c , calculate the comprehensive adjustment value Cv i =A c *Cld′ i +(1-A c )*RW i , where 0 c ≤1;​ (3-3): In NBS n Find the one with the largest Cv i Value of cell gNB mx and minimum Cv i Value of cell gNB mn , respectively determined as the best cell and the worst cell.

5. A 5G load balancing method based on adjustment coefficient according to claim 4, characterized in that The step 4 specifically includes: (4-1): Obtain the best cell gNB mx and the worst cell gNB mn Covering all users in the overlapping area UsrN = {Usr1, Usr2, ..., Usr mxn }, and the number of physical resource blocks requested by the user ProN = {Proxn1, Proxn2, …, Proxn mxn }; (4-2): Arrange the users UsrN in ascending order of the number of physical resource blocks ProN; (4-3): Get the conditions that meet the requirements The maximum PreN value, where PreN≤mxn, Prv mx is the physical resource block space margin of the optimal cell; all PreN users are migrated to the optimal cell gNB mx middle; (4-4): Update the optimal cell gNB mx Physical resource block occupancyPro mx =Pro mx +TProN, worst cell gNB mn Physical resource block occupancyPro mn =Pro mn -TProN; According to the updated Pro mx and Pro mn Value update cell gNB mx Load Cld mx =Pro mx / PrT mx , Update cell gNB mn Load Cld mn =Pro mn / PrT mn 。 6. A 5G load balancing method based on adjustment coefficient according to claim 5, characterized in that The specific process of step five includes: (5-1): Set the acceptable equilibrium state β0, converge to the target value ε, for all cells NBS n ={gNB1,gNB2,…,gNB n }, calculate the load balancing factor (5-2): Determine whether the convergence condition is met If satisfied, end the step; if not satisfied, repeat steps 2 to 5 until the equilibrium condition is met, end the step.

Images