KR101141426B1 - Mobility management method using adaptive handovers based on load balancing for 3g lte systems - Google Patents

Abstract

PURPOSE: A mobility management method for third generation long-term evolution system through adaptive handover is provided to solve a load distribution problem through adaptive handover. CONSTITUTION: A user device executes handover in a service cell(S18). The user device satisfies a condition which is greater than a handover hysteresis threshold value of a neighboring cell. The user device determines a load condition of a cell after executing handover(S19). The user device terminates the execution of the handover in case the ratio of available resource of the entire cell is smaller than a load condition.

Description

MOBILITY MANAGEMENT METHOD USING ADAPTIVE HANDOVERS BASED ON LOAD BALANCING FOR 3G LTE SYSTEMS} Using Load Balancing-Based Adaptive Handover

The present invention relates to a mobility management method suitable for 3G LTE, and more particularly, to a mobility management method for a third generation long-term evolution (3G LTE) system using an adaptive handover based on a load balancing scheme. It is about.

The Third Generation Partnership Project (3GPP) launched a Long-Term Evolution (LTE) project in 2004 to gain a competitive edge in both performance and cost. It was. The project must provide a variety of multiple services and ensure high data rates of quality of service (QoS). This long term evolution (LTE) enables a many-to-many interface between an enhanced Node B (eNodeB) and a core network node (access gateway: aGW), called S1-flex. . In addition, the concept of a pool area, so that a user equipment (UE) can be provided by at least one mobility management entity (MME) / user plane entity (User Plane Entity) This is introduced. These new features must be taken into account in the management of third generation long-term evolution systems, such as mobility and resource management.

Load balancing problems arise when some serving cells have insufficient resources to meet user needs while neighboring cells continue to have sufficient resources to accommodate new calls. Since this problem increases the blocking probability of a new cell or a handover cell, it degrades the quality of service (QoS) in a cell under heavy load. In addition, this problem reduces the efficiency of resource usage since no available resources are used in the neighboring cells. By considering a variety of new situations based on the S1-flex interface and the full domain concept, the mobility and resource management of third-generation long-term evolution systems must support load balancing.

Two general methods of channel borrowing and load distributing have been widely used to solve the load balancing problem. In a load balancing method based on channel borrowing, the overload cell attempts to borrow the necessary resources from the cell with the light load of the neighbor. Since the channel reuse factor of the third generation long term evolution system is 1 due to the Orthogonal Frequency Division Multiplex Access (OFDMA) technique, the channel borrowing technique is not applicable or sufficient for the third generation long term evolution system. The load balancing algorithm attempts to distribute the traffic load of heavily loaded cells by controlling the transmit power of the base station or forcing some mobile terminals to hand over to neighboring lightly loaded cells. Since controlling the transmit power of the base station can cause a lot of interference in the mobile terminal, a load balancing technique is more suitable for the third generation long term evolution system. Recently, as a load balancing algorithm, an adaptive handover time scheme that dynamically controls a handover time according to a load state of a cell has been proposed. In this technique, handover to a heavily loaded cell is delayed by a slow handover time algorithm, and it is possible for a heavily loaded cell to execute handover quickly by a fast handover time algorithm. However, this technique increases the number of unnecessary handovers when executing a quick handover, and requires additional signal processing for exchanging load information between cells at every handover. In particular, this technique should delay or block handover when the target cell is already a heavily loaded cell. Therefore, there is a need for an appropriate mobility management technique in the third generation long-term evolution system to solve the load balancing problem.

An object of the present invention is to provide a mobility management method suitable for a third generation long-term evolution system (3G LTE) that can solve a load balancing problem by using an adaptive handover.

As a means for solving the technical problem, the present invention, the step of determining the load condition of the cell currently being serviced; When the load condition of the currently serving cell becomes an overload condition, for each of the neighbor cells of the currently busy cell, the neighbor cells according to the handover hysteresis threshold of the currently serving cell and the load condition of the neighbor cells; Calculating each handover hysteresis threshold; Transmitting a handover hysteresis threshold of each of the neighbor cells to user devices serviced in the currently serving cell; And a user satisfying a condition in which a difference between a received signal strength by at least one of the neighboring cells and the received signal strength by the currently serving cell is greater than a handover hysteresis threshold of the at least one neighboring cell among the user devices. The apparatus provides a mobility management method for a third generation long-term evolution system using adaptive handover, comprising: performing a handover to at least one of the neighbor cells in the currently serving cell.

In one embodiment of the present invention, the step of transmitting to the user devices includes a handover hysteresis threshold of each of the neighbor cells in a measurement control message to include the measurement control message in the currently serving cell. May be a step of transmitting data.

In an embodiment of the present invention, the determining of the load condition may include: wherein a ratio of the amount of available resources to the total amount of resources of the currently serving cell is smaller than a first predetermined threshold for determining an overload condition. In this case, the current serving cell may be determined as an overload condition.

In one embodiment of the present invention, calculating the handover hysteresis threshold of each of the neighbor cells may be calculating the handover hysteresis threshold of each of the neighbor cells by the following equation.

[Equation]

In the above equation, Th _Hys (0) represents a handover hysteresis threshold of the currently serving cell, i is a variable indicating a neighboring cell, and New_thres (i) represents a handover hysteresis threshold of a neighboring cell i. , v _AR (i) represents the available resources of the neighboring cell i, v _TR (i) represents the total resources of the neighboring cell i, Th _{Post _ LB} represents a preset threshold to determine the light load condition, Th _{Avail _ LB} represents a preset threshold for determining the hysteresis threshold of the neighboring cell and is a value between a preset threshold for determining the overload condition and a preset threshold for determining the light load condition. Can have

In one embodiment of the present invention, executing the handover may further include performing a prediction policy for determining whether at least one of the neighbor cells is an optimal cell for handover.

In one embodiment of the present invention, the information management on the load may be performed by a radio resource management (RRM) function.

One embodiment of the present invention, after the step of performing the handover step of determining the load condition of the currently serving cell; And terminating the step of executing the handover if the ratio of the amount of available resources to the total amount of resources of the currently serving cell is greater than a second preset threshold to determine a light load condition. can do.

According to the present invention, there is an effect of increasing the efficiency of resource usage and improving the quality of service in a 3G LTE system.

1 is a flowchart illustrating an adaptive handover method based on load balancing applied to the present invention.
2 illustrates the relationship between parameters used in a load balancing technique in accordance with the present invention.
3 illustrates an example of unnecessary handover that may occur by a load balancing technique in accordance with the present invention.
4 illustrates an adaptive handover scheme based on load information exchange that can be applied to third generation long term evolution systems.
5 shows a simulation model of a cellular network designed for simulating the mobility management method according to the present invention.
6 is a diagram showing a traffic model applied to the simulation of the mobility management method according to the present invention.
7 is a graph showing the result of the handover call blocking rate calculated as a result of the simulation;
8 is a graph showing the results of the handover call drop rate calculated as a result of the simulation;
9 is a graph showing the results of the new call blocking rate calculated as a result of the simulation.
10 is a graph showing the result of the handover count calculated as a result of the simulation;
11 is a graph showing the results of the ping-pong handover rate calculated as a result of the simulation.

Various embodiments of the present invention will now be described in detail with reference to the accompanying drawings. However, embodiments of the present invention may be modified in various other forms, and the scope of the present invention is not limited to the embodiments described below. The embodiments of the present invention are provided to more fully describe the present invention to those skilled in the art. Accordingly, it should be noted that the shapes, sizes, etc. of the components shown in the drawings may be exaggerated for clarity.

Load balancing algorithm

The present invention relates to a mobility management scheme using adaptive handover based on load balancing of a third generation long-term evolution system (3G LTE) system. The term 'adaptive handover' means that the handover hysteresis threshold is dynamically updated according to the load condition of the cell. In a typical handover technique, the handover is determined according to the decrease in the received signal strength (RSS) received at the base station, and the hysteresis threshold is fixed. This does not take into account the load status of the currently serving cell and the target cell. According to the present invention, by considering both signal strength and load information, it is possible to handover to the optimal cell for the user with high quality of service. According to the present invention, since only the handover hysteresis threshold is considered without changing the structure of the system itself, it does not increase the complexity of the system.

The adaptive handover method of the present invention executes a load balancing process according to the load condition variation of each cell. The load condition change of the cell may be caused by a new call request or resource reconfiguration as well as a handover request. If the cell does not have enough remaining resources, it will request a load balancing procedure to force handover some user equipment to the neighbor cell. Thus, a heavily loaded cell transfers the load to its neighboring less loaded cell to adaptively distribute the load. In this technique, the conditions of multiple mobility management entities (MMEs) may be considered. As a last resort, handover to another MME cell is allowed, and processing time can be long due to MME variation.

1 is a flowchart illustrating a load balancing technique applied to a mobility management method for a third generation long-term evolution system using an adaptive handover according to the present invention. As shown in FIG. 1, in the mobility management method for the third generation long-term evolution system using the adaptive handover according to the present invention, when the load condition of the cell currently being served is changed (S11), the load of the cell currently being served is loaded. The condition is determined (S12), and if the currently serving cell is an overload condition, the handover hysteresis threshold for each of the neighboring cells of the currently serving cell is determined (S13), and the handover hysteresis threshold for each determined neighboring cell is determined. In the form of a measurement control message, the transmission is transmitted to the user equipments served in the currently serving cell (S14). Subsequently, the user device transmits the received signal strength for each of the neighbor cells to a cell that is currently serving in the form of a measurement control message (S15), and received by at least one of the neighbor cells among the user devices. At least one of the neighboring cells in the cell in which the user equipment satisfies a condition S16 in which the difference between the signal strength and the received signal strength by the currently serving cell is greater than the handover hysteresis threshold of the at least one neighboring cell. The handover is executed as one (S18). In addition, the load condition of the currently serving cell is determined again (S19), and the ratio of the amount of available resources to the total amount of resources of the currently serving cell is greater than the second preset threshold to determine the light load condition. In case of ending the handover and restoring to the initially set handover hysteresis threshold (S20).

The following symbols are used to describe the load balancing scheme in accordance with the present invention: the amount of available resources (v _AR (i)) for cell i (i = 0 means the cell currently being served) and Amount (v _TR (i)). The ratio of these two values can be used to compare with a preset threshold to determine if the current cell needs load balancing. The load balancing algorithm is triggered and terminated by preset thresholds Th _{Pre _ LB} and Th _{Post _ LB} respectively. The threshold Th _{Avail _ LB} represents a condition for allowing load balancing. 2 is a diagram illustrating the relationship between the parameters. If the ratio of the amount of available resources to the total amount of resources in the current cell is less than the threshold (Th _{Pre _ LB} ) (that is, v _AR (0) / v _TR (0) <Th _{Pre _ LB} ), the current cell Is considered to be in an overload condition and the load balancing process is triggered. According to the load information of neighboring cells, the overload cell dynamically sets different handover hysteresis thresholds for each neighboring cell. The new handover hysteresis threshold may be calculated as follows.

[Equation]

In the above equation, Th _Hys (0) is the handover hysteresis threshold of the current cell before load balancing, and New_thres (i) is the new handover hysteresis threshold for the neighboring cell i.

While load balancing is applied, the current cell first updates the threshold Th _Hys (0, i). The threshold Th _Hys (0, i) is the new handover hysteresis threshold of the current cell for neighbor cell i. The current cell sends this threshold (Th _Hys (0, i)) along with a measurement control message to the user equipment in service. The user device then updates this new handover hysteresis threshold. Suppose that user device l sends a measurement report message to the current cell as the event is triggered or periodically, and receives signals from the current cell and neighbor cell j of user device l (RS _l (0) and If the strength of the box represented by the RSS _l (j)) satisfy the _{RSS l (j) -RSS l (} 0)> Th Hys (0, i), l is a user equipment currently executing a handover from cell to cell j . In FIG. 1, the symbols '*' and '**' indicate a message that includes a new handover hysteresis threshold that is different from the original message. Here, new messages and multiple handover hysteresis thresholds for load balancing may require more resources. However, the bandwidth required for additional threshold management is extremely small compared to that required for user data services. Thus, each cell can be considered to be able to perform the above process with only negligible overhead.

As shown in FIG. 1, when the ratio of available resources to total resources of the cell reaches a threshold for termination (that is, v _AR (0) / v _TR (0)> Th _{Post_LB} ), the load balancing process is terminated. . If the cell meets this condition, the load of the cell is considered light enough and the cell restores the handover hysteresis threshold to its original value by sending another measurement control message to the serving user equipment.

Adaptive handover can increase the number of ping-pong handovers because some handovers may be executed too early when the handover hysteresis threshold is reduced. 3 shows an example of unnecessary handover. In Figure 3, by reducing the handover hysteresis threshold for cell 2 when performing load balancing, cell 1 extends the handover area from the original boundary to the new boundary. Suppose that cell 1 is a cell serving user equipment 1 and 2 and the two user equipments are located in the newly extended handover area. Subsequently, user devices 1 and 2 may perform a handover to cell 2 and may present their resources to solve the load balancing problem. However, if user device 2 is moving from cell 2 to cell 1, user device 2 should perform another handover from cell 2 to cell 1 when user device 2 enters the handover area of cell 2. If the time interval between the two handovers of the user device 2 is very short, the user device 2 performs an unnecessary handover and the performance of load balancing is also degraded. To solve this problem, a function of an optional function named "DirPred (l, j)" may be used as shown in FIG. 1 (S17). This DirPred (l, j) function A function of predicting whether the user device 1 moves in the cell j direction, using a linear prediction method based on the signal strength of the base station and a database-based prediction method based on the temporal and spatial movement characteristics of the user device. User device 1 If it is predicted to move in the cell j direction, it outputs 'yes', otherwise it outputs 'no'.

Many prediction techniques used handover estimation methods for fast and even handover. In the present invention, one of the following two prediction policies is used to simply determine whether the target cell is the optimal cell of the user equipment. Prediction policy I means that the user equipment executes handover only when the target cell is an optimal handover cell. Prediction policy II is a target cell when the number of user devices to which the adaptive handover is applied is not sufficient for load balancing. Even if this is not the optimal handover cell, some user equipment performs handover.

Handover  process

In the third generation long term evolution system, load information management is performed by Radio Resource Management (RRM). This RRM function may be located in the primary access gateway or master enhanced node B (eNodeB) in the centralized scheme or in each enhanced node B in the distributed scheme. In practice, the location of these functions is not very important because the load information and signaling costs are not very high for both centralized and distributed management. 4 is a diagram illustrating an adaptive handover scheme based on load information exchange applicable to a third generation long-term evolution system. In the present invention, a common RRM (CRRM) located in an access gateway (aGW) 41 may be used to manage load information. When the load state of the enhanced Node B 42 is changed, the enhanced Node B 42 updates the load information in the CRRM (S41). If the overload condition is satisfied at the enhanced Node B 42 (S42), the enhanced Node B 42 sends a load information request message to the access gateway 41 (S43), and a load information response message from the access gateway 41. By receiving (S44) to obtain the load information of the neighbor cell. According to the received load information, the enhanced Node B 42 updates the handover hysteresis threshold to the user device 43 (S45) and adapts the hand for load balancing until the load condition returns to a light load condition. The over process is executed (S46).

simulation

The inventors of the present invention have designed a system for implementing the mobility management method according to the present invention with a formal description technique using Specification and Description Language (SDL). To demonstrate the improved effect of the load balancing algorithm according to the present invention, three simulation systems were designed: a system without load balancing, a system with load balancing only, and a load balancing system using motion prediction of user equipment. . Both prediction policies (prediction policy I and prediction policy II) are used to implement the prediction technique. The following five performance parameters are used for performance testing.

Handover call blocking rate: The ratio of the number of blocked handover calls to the total number of handover calls.

Handover call drop rate: The ratio of the number of handover calls dropped to the total number of handover calls.

New call blocking rate: The ratio of the number of blocked calls to the total number of new calls.

Ping-pong handover rate: The ratio of the number of unnecessary handover calls to total handover calls.

Simulation model

5 illustrates a simulation model of a cellular network designed for simulating a mobility management method according to the present invention. As shown in FIG. 5, the simulation model of the cellular network area includes 21 macro cells distributed in two pool areas having the same radius of 1 km. In FIG. 5, five cells indicated by reference numerals 51 to 55 correspond to the movement zone 50, and are areas in which the user device can move. It is assumed that the sixteen peripheral cells surrounding the moving zone 50 are always in light load conditions to support the prediction technique.

Table 1 below shows various simulation parameters.

parameter value Pool Zone Count 2 Cell count 21 Cell radius 1000 m Cell capacity 30-130 PRB User device count 0-250 User device mobility Random waypoint User device speed 0-100 km User device active time Erlang distribution (average 1200 s) User device idle time Erlenmeyer distribution (200 s average) Traffic model Fish Egg Distribution (Average 3-8 PRB) Normal (fixed) handover threshold 5 dB Dynamic handover threshold 0-5 dB Th _{Pre_LB} 0.2 Th _{Avail_LB} 0.3 Th _{Post_LB} 0.4 Simulation time 10 hours

As shown in Table 1 above, this simulation allows a maximum of 250 user devices in the simulation area. Traffic of each user device follows the Erlang distribution as shown in FIG. 6. The ratio of active time to idle time for generating an overload cell during the simulation is 6: 1. The movement pattern of each user device is a random waypoint model. In a random waypoint model, the direction of movement is generated by a uniform distribution in the range of 0 ° and less than 360 °. In each linear path, the speed of the user device varies as follows: 0.3x (0-10%), 0.5x (10-20%), x (20-80%), 0.5x (80-90%) And 0.3 × (90-100%). Where x is the maximum velocity and the range in angle brackets indicates the position in the path.

Table 2 below shows three simulation scenarios designed by the inventors of the present invention.

scenario scenario scenario Cell capacity (PRB) 60 30-130 60 User device count 100-250 200 200 Traffic Average (PRB) 6 6 3-8

Simulation result

As the number of user devices increases and the capacity of the cell decreases or the traffic average increases, the number of overload cells increases. Since the results of each scenario are similar, only the results of the scenarios will be described.

7 is a graph showing the results of the handover call blocking rate calculated as a result of the simulation. As shown in FIG. 7, when the number of user devices increases to 250, handover call blocking is about 15% better than the conventional handover technique without load balancing, compared to a typical handover technique without load balancing. Indicates the rate. In FIG. 7, the handover call blocking rate of the present invention to which the prediction policy I is applied represents about 4% higher handover call blocking rate compared to the present invention without the prediction technique. This is because some user devices did not perform adaptive handover for load balancing by motion prediction. However, the results of applying Prediction Policy II are very similar to those of the present invention without the prediction technique. This is because the prediction policy II does not stop the adaptive handover until the end condition of the load balancing is satisfied.

8 is a graph showing the result of the handover call drop rate calculated as a simulation result, and FIG. 9 is a graph showing the result of the new call blocking rate calculated as a simulation result. As shown in FIG. 8 and FIG. 9, the handover call drop rate and the new call blocking rate show similar results to the handover call blocking rate shown in FIG.

10 is a graph showing the result of the handover count calculated as a result of the simulation. As shown in FIG. 10, the number of handovers of the present invention is slightly higher than that of a typical handover technique without load balancing due to the additional adaptive handover for load balancing. This can be regarded as a cost to the more efficient resource management of the present invention.

11 is a graph showing the results of the ping-pong handover rate calculated as a result of the simulation. As described above, the present invention without applying the prediction technique shows a ping pong handover rate of about 6% higher than without applying load balancing at the heaviest load state. As expected, it can be seen that the prediction technique reduces the ping-pong handover rate. This simulation result shows that the prediction technique is useful for reducing unnecessary handovers. 11 also shows that predictive policy II is not as efficient as predictive policy II. This is because user devices continue to perform handover when the overload condition is not resolved.

According to the simulation results as described above, it can be seen that the present invention can increase the efficiency of resource use and improve the service quality of the third generation long-term evolution system.

In the detailed description of the present invention, specific embodiments have been described, but various modifications may be made without departing from the scope of the present invention. Therefore, the scope of the present invention should not be limited to the described embodiments, but should be defined by the following claims and their equivalents.

S11: Change load condition
S13: Determination of Handover Hysteresis Threshold for Each Neighbor Cell
S14: Send measurement control message to user device
S15: Send measurement report message to cell
S18: Handover Execution
S20: Restore initially set handover hysteresis threshold

Claims (7)

Determining a load condition of a currently serving cell;
When the load condition of the currently serving cell becomes an overload condition, for each of the neighbor cells of the currently busy cell, the neighbor cells according to the handover hysteresis threshold of the currently serving cell and the load condition of the neighboring cells; Calculating each handover hysteresis threshold;
Transmitting a handover hysteresis threshold of each of the neighbor cells to user devices serviced in the currently serving cell;
Among the user devices, a user device satisfying a condition in which a difference between a received signal strength by at least one of the neighbor cells and a received signal strength by the currently serving cell is greater than a handover hysteresis threshold of the at least one neighbor cell. Performing a handover to at least one of the neighbor cells in the currently serving cell;
Re-determining a load condition of a cell that is currently serving after the step of executing the handover; And
Terminating the step of executing the handover if the ratio of the amount of available resources to the total amount of resources of the currently serving cell is greater than a second preset threshold to determine a light load condition;
A mobility management method for a third generation long-term evolution system using load balancing based adaptive handover comprising a.
The method of claim 1,
The step of transmitting to the user devices,
And including the handover hysteresis threshold of each of the neighbor cells in a measurement control message to transmit the measurement control message to user devices serviced in the currently serving cell. Mobility management method for third generation long-term evolution system using.
The method of claim 1,
Determining the load condition,
If the ratio of the amount of available resources to the total amount of resources of the currently serving cell is less than a first predetermined threshold for determining an overload condition, determining the currently serving cell as an overload condition. A mobility management method for a third generation long-term evolution system using load balancing based adaptive handover.
The method of claim 1,
Computing the handover hysteresis threshold of each of the neighbor cells,
Computing handover hysteresis threshold value of each of the neighbor cells by the following equation, the mobility management method for the third generation long-term evolutionary system using the load balancing based adaptive handover.
[Equation]

(Th _Hys (0) is the handover hysteresis threshold of the currently serving cell, i is a variable indicating a neighboring cell, New_thres (i) is a handover hysteresis threshold of neighboring cell i, and v _AR (i) is a neighbor) Available resources of cell i, v _TR (i) is the total resources of neighboring cell i, Th _{Post _ LB} is a preset threshold for determining light load conditions, Th _{Avail _ LB} is the hysteresis threshold of the neighbor cell Having a value between a preset threshold for determining the overload condition and a preset threshold for determining the light load condition with a preset threshold)
The method of claim 1,
The executing of the handover may further include performing a prediction policy for determining whether at least one of the neighbor cells is an optimal cell for handover. 3. Mobility management method for generation long-term evolution system.
The method of claim 1,
The information management on the load is a mobility management method for the third generation long-term evolutionary system using load balancing based adaptive handover, characterized in that performed by a Radio Resource Management (RRM) function.
delete

Images