CN102905317B - Mobile load balancing method used for multiple cells - Google Patents

Abstract

The invention discloses a mobility load balancing method applied to multiple cells, which mainly solves the problems of ping-pong load transfer and slow convergence speed of load balancing existing in the existing load balancing method. The implementation steps are: determining cell clusters and triggering load balancing; generating source cell sets and target cell sets in the cell clusters; prioritizing the source cell sets and calculating load transfer steps; prioritizing the target cell sets; The source cell with the highest priority selects the target cell in order of priority for load transfer; after the load transfer of the source cell with the highest priority ends, select the source cell with the second priority from the source cell set to continue the load transfer until the cell cluster All the source cells in the network have performed load transfer, and the load balancing process ends. The invention effectively solves the problem of ping-pong load transfer, and at the same time speeds up the convergence speed of load balancing, thereby improving system performance.

Description

A Mobility Load Balancing Method Applied to Multiple Cells

technical field

The invention relates to the technical field of wireless communication, in particular to a mobility load balancing method, which can be used for mobility load balancing between multiple cells in the advanced international mobile communication system IMT-Advanced.

Background technique

The advanced international mobile communication system IMT-Advanced is different from the traditional cellular network. It adopts a flat network structure instead of the traditional centralized control structure. As shown in Figure 1, the base stations directly exchange information through the X2 interface, and communicate accordingly through negotiation. radio resource allocation and handover.

The new features and technical requirements of the IMT-Advanced system make the traditional network management and optimization methods no longer efficient, and the self-organizing network SON technology can improve the self-organizing ability of the network, realize the self-configuration, self-optimization and self-healing of the network, and simplify the wireless network The design and operation and maintenance of the network greatly reduce the maintenance cost of the network, and can well meet the new features and technical requirements of the IMT-Advanced system.

The wireless cellular network in the IMT-Advanced system has limited resources, but due to the uneven geographical distribution of users and the randomness of service initiation, the load in the network will show an unbalanced distribution, resulting in limited resources that cannot be fully utilized. use. In order to solve the above problems, the mobility load balancing MLB technology, which is an important use case of SON, was introduced. This technology not only reduces the workload of network management and optimization, but also enables the heavier load of the two networks or two systems to partly The load is transferred to the other party, so as to balance the load distribution between the network or the system, improve the utilization rate of wireless resources, and improve the system performance.

The existing MLB technology is mainly to balance the load between a hotspot cell and its adjacent target cells in a homogeneous network. Load status, select the neighbor cell with the lightest load as the target cell for load transfer, until the load status of the hotspot cell is lower than the load balancing threshold, the load balancing ends. Since this method only considers selecting the cell with the lightest load as the target cell for load transfer, several hotspot cells may choose the same cell as the target cell. The load is transferred, resulting in a ping-pong transfer process of the load, which slows down the convergence speed of the load balancing; in addition, after the load balancing of a single hotspot cell is completed, because other hotspot cells do not alleviate the high load state, it is easy to cause a sudden traffic failure. Areas are blocked, thereby affecting the overall performance of the system.

Contents of the invention

The purpose of the present invention is to address the shortcomings of the above-mentioned existing MLB technology, and provide a mobility load balancing method applied to multiple cells, so as to realize load balancing smoothly, avoid load ping-pong transfer, accelerate the convergence speed of load balancing, and then improve system performance.

A kind of mobility load balancing method applied to multiple cells provided by the present invention comprises the following steps:

(1) During the load balancing execution cycle, each cell in the system detects its own load status, and exchanges load information with its neighbor cells, and compares its own load l with the load balancing threshold l _th , and the formula l≥l _th The cell and its neighbor cells form a cell cluster of size N, and trigger the load balancing process in the cell cluster;

(2) Divide the N cells in the cell cluster into θ source cells and N-θ target cells, that is, the cell with load l≥l _th -ε is the source cell, otherwise it is the target cell, where ε is the load of the cell Protection margin, whose value is a positive number less than the load balancing threshold l _th , and the source cell set θ _s is composed of θ source cells, and the target cell set θ _t is composed of N-θ target cells;

(3) Prioritize the source cell set θ _s according to the size of the load, and the source cell i with the highest priority performs load transfer first, and calculates the load transfer step length: Among them, Δl _i = l _i -(l _th -ε) is the load that needs to be unloaded by the source cell i with the highest priority, l _i is the load of the source cell i with the highest priority, and m is the maximum number of load transfers, the value is a positive integer;

(4) For the target cell set θ _t according to the target cell label Prioritized by the size of the label, the label Obtained according to the range where the load of the target cell j can receive Δl _j =l _j -(l _th -ε), where l _j is the load of the target cell j:

$\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; j</span>}}<span\; class="notranslate">\; =</span>\left\{\begin{array}{cc}\mathrm{<span\; class="notranslate">\; 0</span>}<span\; class="notranslate">\; ,</span>& \mathrm{<span\; class="notranslate">\; \&Delta;</span>}{\mathrm{<span\; class="notranslate">\; l</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}<span\; class="notranslate">\; <</span>{\mathrm{<span\; class="notranslate">\; \&pi;</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}\\ \mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; ,</span>& {\mathrm{<span\; class="notranslate">\; \&pi;</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; \&le;</span>\mathrm{<span\; class="notranslate">\; \&Delta;</span>}{\mathrm{<span\; class="notranslate">\; l</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}<span\; class="notranslate">\; <</span>\mathrm{<span\; class="notranslate">\; 2</span>}{\mathrm{<span\; class="notranslate">\; \&pi;</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}\\ <span\; class="notranslate">\; .</span>& \\ <span\; class="notranslate">\; .</span>& \\ <span\; class="notranslate">\; .</span>& \\ \mathrm{<span\; class="notranslate">\; m</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; ,</span>& <span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; m</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; )</span>{\mathrm{<span\; class="notranslate">\; \&pi;</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; \&le;</span>\mathrm{<span\; class="notranslate">\; \&Delta;</span>}{\mathrm{<span\; class="notranslate">\; l</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}<span\; class="notranslate">\; <</span>\mathrm{<span\; class="notranslate">\; m</span>}{\mathrm{<span\; class="notranslate">\; \&pi;</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}\\ \mathrm{<span\; class="notranslate">\; m</span>}<span\; class="notranslate">\; ,</span>& \mathrm{<span\; class="notranslate">\; m</span>}{\mathrm{<span\; class="notranslate">\; \&pi;</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; \&le;</span>\mathrm{<span\; class="notranslate">\; \&Delta;</span>}{\mathrm{<span\; class="notranslate">\; l</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}\end{array}\right.<span\; class="notranslate">\; ;</span>$

(5) The source cell i with the highest priority selects the target cell in turn according to the priority for load transfer;

(6) After the source cell i with the highest priority finishes the load transfer, return to step (3), and select the next source cell according to the priority from the source cell set θ _s to continue the load transfer until all the source cells in the cell cluster are After load transfer is performed, a load balancing process ends.

Compared with the prior art, the present invention has the following advantages:

(1) Through the setting of the cell load protection margin ε, the hotspot cell can be transferred in advance, so as to avoid the regional congestion caused by the sudden business; at the same time, the cell load protection margin ε also allows the target cell to leave a certain Load margin, even after receiving the load transferred by multiple hotspot cells, it can avoid the possibility of itself becoming a hotspot cell, thereby effectively avoiding ping-pong load transfer;

(2) Through the setting of the load transfer step size π, the mobility load balancing MLB is smoothly realized, and the number of load transfers in a single load transfer process and the number of load transfers of the entire system can be controlled, which speeds up the mobility load balancing MLB convergence speed.

Description of drawings

FIG. 1 is a schematic diagram of an architecture of an IMT-Advanced system in the prior art;

Fig. 2 is the overall flowchart of the mobility load balancing of the present invention;

Fig. 3 is a sub-flow chart of load transfer in the present invention.

Detailed ways

With reference to Fig. 2, a kind of mobility load balancing method that the present invention proposes being applied to multiple cells, concrete steps are as follows:

Step 1, determine cell clusters, and trigger load balancing.

1a) During the load balancing execution period, the base stations of each cell in the system obtain the current load status of the cell, and exchange load information with the base stations of neighboring cells through the X2 interface between the base stations;

1b) Each cell compares its own load l with the load balancing threshold l _th , where the load balancing threshold l _th is a normalized value set by the system, and 0<l _th <1, if the cell load l ≥ l _th , then the cell and its neighbor cells form a cell cluster of size N, and trigger the load balancing process in the cell cluster, otherwise, wait for the next load balancing execution cycle.

Step 2, generate source cell set θ _s and target cell set θ _t .

2a) The N cells in the cell cluster are respectively judged as follows: if the load of the cell l≥l _th -ε, then the cell is the source cell, otherwise it is the target cell, where ε is the load protection margin of the cell, and its value is a positive number less than l _th ;

2b) After the decision process, the N cells in the cell cluster are divided into θ source cells and N-θ target cells;

2c) The source cell set θ _s is composed of θ source cells, and the target cell set θ _t is composed of N-θ target cells.

Step 3: Prioritize the source cell set θ _s , and calculate the load transfer step.

3a) Prioritize the source cell set θ _s according to the load from large to small, so as to ensure that the source cell with the largest load gets the highest priority, and the load transfer operation is performed first;

3b) Calculate the load transfer step size π _i from the source cell i with the highest priority:

First, calculate the load to be offloaded by the source cell i with the highest priority Δl _i =l _i -(l _th -ε), where l _i is the load of the source cell i with the highest priority;

Second, calculate the load transfer step size: Among them, m is the maximum number of load transfers, and the value is a positive integer.

Step 4: Prioritize the target cell set θ _t .

4a) Calculate the acceptable load of the target cell j Δl _j = l _j -(l _th -ε), where l _j is the load of the target cell j;

4b) Obtain the label of the target cell j according to the range where Δl _j is located

$\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; j</span>}}<span\; class="notranslate">\; =</span>\left\{\begin{array}{cc}\mathrm{<span\; class="notranslate">\; 0</span>}<span\; class="notranslate">\; ,</span>& \mathrm{<span\; class="notranslate">\; \&Delta;</span>}{\mathrm{<span\; class="notranslate">\; l</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}<span\; class="notranslate">\; <</span>{\mathrm{<span\; class="notranslate">\; \&pi;</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}\\ \mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; ,</span>& {\mathrm{<span\; class="notranslate">\; \&pi;</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; \&le;</span>\mathrm{<span\; class="notranslate">\; \&Delta;</span>}{\mathrm{<span\; class="notranslate">\; l</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}<span\; class="notranslate">\; <</span>\mathrm{<span\; class="notranslate">\; 2</span>}{\mathrm{<span\; class="notranslate">\; \&pi;</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}\\ <span\; class="notranslate">\; .</span>& \\ <span\; class="notranslate">\; .</span>& \\ <span\; class="notranslate">\; .</span>& \\ \mathrm{<span\; class="notranslate">\; m</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; ,</span>& <span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; m</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; )</span>{\mathrm{<span\; class="notranslate">\; \&pi;</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; \&le;</span>\mathrm{<span\; class="notranslate">\; \&Delta;</span>}{\mathrm{<span\; class="notranslate">\; l</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}<span\; class="notranslate">\; <</span>\mathrm{<span\; class="notranslate">\; m</span>}{\mathrm{<span\; class="notranslate">\; \&pi;</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}\\ \mathrm{<span\; class="notranslate">\; m</span>}<span\; class="notranslate">\; ,</span>& \mathrm{<span\; class="notranslate">\; m</span>}{\mathrm{<span\; class="notranslate">\; \&pi;</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; \&le;</span>\mathrm{<span\; class="notranslate">\; \&Delta;</span>}{\mathrm{<span\; class="notranslate">\; l</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}\end{array}\right.<span\; class="notranslate">\; ;</span>$

4c) The target cell set θ _t according to the target cell label Prioritize the size of the load to ensure that the target cell with the strongest load receiving capability receives the load first, and realizes the rational utilization of resources.

Step 5, load transfer is performed by the source cell i with the highest priority.

Referring to Figure 3, the specific implementation of this step is as follows:

5a) The source cell i with the highest priority is selected from the target cell set θ _t according to the label Select the target cell j from large to small for load transfer, and the load transferred at one time is Where mπ _i is the load that needs to be transferred by the source cell i with the highest priority, is the maximum load that the target cell j can receive at one time;

5b) According to the utility value formula, the target cell j calculates the current utility value of the cell and the corresponding utility value after the cell receives the load of Δ _j , and compares the two values. If the utility value of the cell after receiving the load is greater than current utility value, then go to step 5d), otherwise go to step 5c);

Among them, the utility value formula is expressed as follows:

$\stackrel{<span\; class="notranslate">\; ~</span>}{\mathrm{<span\; class="notranslate">\; u</span>}}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; u</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; C</span>}<span\; class="notranslate">\; =</span>\left\{\begin{array}{c}\mathrm{<span\; class="notranslate">\; a</span>}<span\; class="notranslate">\; \&Center\; Dot;</span>{\mathrm{<span\; class="notranslate">\; log</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; l</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; l</span>}<span\; class="notranslate">\; <</span>{\mathrm{<span\; class="notranslate">\; l</span>}}_{\mathrm{<span\; class="notranslate">\; the\; th</span>}}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; \&epsiv;</span>}\\ \mathrm{<span\; class="notranslate">\; a</span>}<span\; class="notranslate">\; \&Center\; Dot;</span>{\mathrm{<span\; class="notranslate">\; log</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; l</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; -</span>\frac{\mathrm{<span\; class="notranslate">\; b</span>}}{\mathrm{<span\; class="notranslate">\; \&epsiv;</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; l</span>}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; l</span>}}_{\mathrm{<span\; class="notranslate">\; the\; th</span>}}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; \&epsiv;</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; )</span><span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; l</span>}}_{\mathrm{<span\; class="notranslate">\; the\; th</span>}}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; \&epsiv;</span>}<span\; class="notranslate">\; \&le;</span>\mathrm{<span\; class="notranslate">\; l</span>}<span\; class="notranslate">\; <</span>{\mathrm{<span\; class="notranslate">\; l</span>}}_{\mathrm{<span\; class="notranslate">\; the\; th</span>}}\\ \mathrm{<span\; class="notranslate">\; a</span>}<span\; class="notranslate">\; \&Center\; Dot;</span>{\mathrm{<span\; class="notranslate">\; log</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; l</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; +</span>{<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; l</span>}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; l</span>}}_{\mathrm{<span\; class="notranslate">\; the\; th</span>}}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; \&epsiv;</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; )</span>}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; b</span>}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; \&epsiv;</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; l</span>}}_{\mathrm{<span\; class="notranslate">\; the\; th</span>}}<span\; class="notranslate">\; \&le;</span>\mathrm{<span\; class="notranslate">\; l</span>}\end{array}\right.$

in, Represents the utility value of the community, U represents the gross income of the community, C represents the cost of the community, a is a parameter related to the gross profit U of the community, and the value is 0 to 1, b is a parameter related to the cost C of the community, which takes The value is 0 to 1;

5c) The source cell i with the highest priority selects the next target cell from the target cell set θ _t to continue load transfer. If there is no next target cell in the target cell set θ _t , the load transfer of the source cell i with the highest priority The process ends; otherwise, return to step 5a);

5d) According to the utility value formula, the source cell i with the highest priority calculates the current utility value of the cell and the corresponding utility value after the cell transfers the load of Δ _j , and compares the two values, if the cell transfers the load is greater than the current utility value, then the source cell i with the highest priority transfers a load of Δ _j to the target cell j, where the utility value formula is the same as the utility value formula in step 5b);

5e) The source cell i with the highest priority makes the following judgment: if Then update the value of mπ _i Return to step 5c); otherwise, the load transfer process of the source cell i with the highest priority ends.

In step 6, a load balancing end judgment is made.

If all the source cells in the cell cluster have performed load transfer, the load balancing process ends, otherwise, return to step 3, and select the next source cell from the source cell set θ _s according to the priority to continue the load transfer.

The above are specific examples of the present invention, and do not constitute any limitation to the present invention. Obviously, different changes can be made under the idea of the present invention, but these changes all belong to the protection scope of the present invention.

Claims (4)

1. A mobility load balancing method applied to multiple cells comprises the following steps:

(1) in the load balancing execution period, each cell in the system detects the load state of the cell, interacts load information with the adjacent cells, and simultaneously carries the load l of the cell and a load balancing threshold l_thComparing the values of l and l_thThe cell and the adjacent cell form a cell cluster with the size of N, and a load balancing process is triggered in the cell cluster;

(2) dividing N cells in a cell cluster into theta source cells and N-theta source cellsTarget cell, i.e. load l ≧ l_thThe cell of epsilon is the source cell and vice versa is the target cell, where epsilon is the cell load protection margin and has a value less than the load balancing threshold/_thAnd theta source cells constitute a source cell set theta_sA target cell set theta consisting of N-theta target cells_t；

(3) To source cell set theta_sCarrying out priority sequencing according to the size of load, carrying out load transfer from a source cell i with the highest priority firstly, and calculating the load transfer step length:wherein Δ l_i＝l_i-(l_thε) the amount of load that needs to be unloaded for the source cell i with the highest priority, l_iThe load of a source cell i with the highest priority is obtained, m is the maximum load transfer frequency, and the value is a positive integer;

(4) for target cell set theta_tAccording to target cell designationIs prioritized, the indexReceivable load amount delta l according to target cell j_j＝l_j-(l_thε) in which l_jLoad amount for target cell j:

(5) the source cell i with the highest priority selects the target cells in sequence according to the priority to carry out load transfer;

(6) after the source cell i with the highest priority finishes load transfer, the step returns to the step (3)) From the source cell set theta_sAnd selecting the next source cell according to the priority to continue load transfer until all the source cells in the cell cluster perform load transfer, and ending the load balancing process.

2. The method as claimed in claim 1, wherein each cell in the system in step 1) detects its own load status and interacts load information with its neighboring cells, and the base station of each cell obtains the current load status of the cell and interacts load information with the base stations of its neighboring cells through an X2 interface between the base stations.

3. The method as claimed in claim 1, wherein the source cell i with the highest priority in step 5) selects the target cells in turn according to the priorities for load transfer, and the method comprises the following steps:

5a) the source cell i with the highest priority is selected from the target cell set theta_tIn accordance with the reference numberSelecting a target cell j from large to small for load transfer, wherein the load capacity of one transfer isWherein m pi_iThe amount of load to be transferred for the source cell i with the highest priority,the maximum acceptable load capacity at one time of the target cell j is obtained;

5b) respectively calculating the current utility value of the target cell j and the corresponding utility value after the cell receives the load with the size of delta j according to a utility value formula, comparing the two values, and executing the step 5 d) if the utility value after the cell receives the load is larger than the current utility value, or executing the step 5 c);

5c) the source cell i with the highest priority is selected from the target cell set theta_tSelecting next target cell to continue load transfer, if the target cell set theta_tIf the next target cell does not exist, the load transfer process of the source cell i with the highest priority is finished; otherwise, returning to the step 5 a);

5d) respectively calculating the current utility value of the cell and the cell transfer size delta according to the utility value formula by the source cell i with the highest priority_jThe corresponding utility value after the load amount is obtained, the two values are compared, and if the utility value after the load transfer of the cell is larger than the current utility value, the transfer of the source cell i with the highest priority to the target cell j is delta_jThe amount of load of (a);

5e) the source cell i with the highest priority makes the following decisions: if it isThen update m pi_iHas a value ofReturning to the step 5 c); otherwise, the load transfer process of the source cell i with the highest priority is ended.

4. A mobility load balancing method applied to multi-cells according to claim 3, wherein the utility value formula in step 5 b) and step 5 d) is expressed as follows:

wherein,the utility value of the cell is represented, U represents the gross benefit of the cell, C represents the cost of the cell, a is a parameter related to the gross benefit U of the cell and takes a value of 0 to 1, b is a parameter related to the cost C of the cell and takes a valueFrom 0 to 1.