CN103369572A - A Load-Based Automatic Neighbor Relation Optimization Method in Long Term Evolution System - Google Patents

Abstract

The invention relates to a load-based automatic neighbor cell relationship optimization method in a long-term evolution system, which relates to the technical field of wireless communication. Including two stages, the first stage is the optimization of the initial establishment of ANR, that is, when there is no neighbor list in the base station, the second stage is the optimization of the ANR maintenance stage, that is, when there is a neighbor list in the base station; the long-term evolution system cell The base station equipment exchanges load status information through the X2 interface, etc., and according to the load status information of the cell itself and the load status information of adjacent cells, as well as the reference signal received power RSRP information of the cell and the surrounding adjacent cells reported by the mobile user terminal, in the adjacent cell In the area configuration and management module, the priority of adjacent cells is optimized to maintain load balance between adjacent cells, reduce unnecessary handovers, and ensure the communication quality of each cell in the system. The requirements for real-time or data synchronization are not very high, allowing a large effective time window.

Description

A Load-Based Automatic Neighbor Relation Optimization Method in Long Term Evolution System

technical field

The invention relates to the technical field of wireless communication, in particular to a load-based automatic neighbor cell relationship optimization method in a long-term evolution system.

Background technique

Automatic neighbor relation configuration (Automatic Neighbor Relation, ANR) means that the user equipment (User Equipment, UE) in the connected state assists in measuring the relevant information of surrounding cells, and reports it to the evolved base station (eNodeB, eNB) and the network side, and the eNB And the network side automatically completes the function of adding and optimizing the neighbor list.

The core of the ANR function is the management of the Neighbor Relation List (NRL) by the eNB. In the Long Term Evolution (LTE) system, the UE side can decode the ID of any measurement cell with a given frequency, and the eNB automatically maintains and updates the neighbor cell relationship list, and updates and maintains the neighbor cell list. Its functions include: Area detection function, adjacent cell addition/deletion function, adjacent cell modification function and maintenance management center (Operation and Maintenance, O&M) update adjacent cell information function. Among them, the adjacent cell detection function is mainly completed by the measurement function and the X2 interface configuration update, the adjacent cell addition/deletion function and the adjacent cell modification function are mainly completed by the internal information and the X2 interface configuration update.

The LTE system adopts advanced technologies such as Multiple Input Multiple Output (MIMO), wireless relay, and cell interference coordination. In order to reduce transmission delay, improve user data transmission speed, and ensure seamless user switching, a Mesh network structure is adopted . The eNBs are connected to each other through the X2 interface, and to the Evolved Packet Core (EPC) through the S1 interface. The EPC includes MME and S-GW. The eNB is connected to the Mobility Management Entity (MME) through the S1-MME interface. The -U interface is connected to the Signaling Gateway (the Serving Gateway, S-GW). Compared with the traditional cellular network, the base station controller in the traditional cellular network is eliminated in the LTE structure, and only EPC and eNB are included, so that the transmission rate of UE data is faster, the control of radio resources is more stable, and low latency and low complexity are achieved. degree and low cost.

The existing ANR technology mainly focuses on establishing and maintaining a neighbor relationship list under stable network conditions.

Chinese patent CN102378247A discloses a method and system for realizing automatic neighbor relationship measurement. When it is determined that the UE is in the public connection state, the network side sends automatic neighbor relationship (ANR) measurement control information to the UE, and the UE controls the measurement based on the obtained ANR. The ANR measurement of dedicated control is performed on the information, and the measurement result is reported to the network side. The invention realizes the automatic neighbor cell relationship reporting by using the UE in the public connection state to complete the dedicated control ANR measurement. Moreover, since the ANR measurement is completed in the public connection state of the UE, it saves air resources and improves the use of the UE. It is felt that there will be no service impact on terminals in the dedicated state.

Chinese patent CN102761897A discloses a method and system for realizing automatic neighbor cell relationship measurement, including that before the UE performs ANR measurement, the network side sends the ANR measurement control configuration parameters containing control parameters related to different types of ANR measurement scheduling to the UE, and the UE When performing ANR measurement, different types of neighboring unknown cells are measured according to the received related control parameters for different types of ANR measurement scheduling. In the invention, the UE measures different types of adjacent unknown cells according to the ratio of received ANR measurement control configuration parameters for different types of ANR measurement scheduling related control parameters. In this way, under certain network expectations, UE performs ANR measurement log in a more targeted manner, rather than purely following Best Effort, thereby improving the performance of ANR measurement log and performing ANR measurement with quality assurance.

Chinese patent CN102448128A discloses a method and system for detecting and adding a base station, a terminal, and an automatic neighboring cell relationship. The base station sends a detection instruction including a preset detection frequency point to the terminal, and then the terminal performs the detection at the preset detection frequency point according to the detection instruction. Cells are detected on the Internet, and the cell information of the cells whose detected signal strength meets the preset conditions is fed back to the base station, and the base station judges whether there is a new neighbor cell based on the cell information. That is, when the invention detects adjacent cells, it does not send the cell information of all cells detected on the detection frequency point to the base station, but only sends the cell information of the cells whose signal strength meets the preset conditions to the base station, so it can be used for Effective screening of detected cells can greatly reduce redundant information in the process of configuring neighbor relationship, simplify the process of maintaining neighbor relationship, and improve the efficiency of neighbor relationship maintenance. Find the neighboring cell to be handed over to improve the efficiency and success rate of terminal handover.

Contents of the invention

The purpose of the present invention is to provide a load-based automatic neighbor cell relationship optimization method in a long term evolution system.

The present invention comprises two stages, stage one is the optimization of ANR initial establishment, namely when there is no neighboring cell list in the base station, and stage two is the optimization of ANR maintenance stage, namely when there is a neighboring cell list in the base station;

The first stage includes the following steps:

1) The base station exchanges load information with neighboring base stations to obtain the load value of neighboring cells;

2) The base station side sends the automatic neighbor relation (ANR) measurement indication to the user equipment (UE) side, and obtains the reference signal receiving power (RSRP for short) measurement information reported by the user equipment (UE);

3) Optimize the neighboring cell list by reporting the reference signal received power (RSRP) measurement information obtained in step 1) and the user equipment (UE) obtained in step 2);

4) Base station judgment will detect that the cell is added to the neighbor cell relationship, and the base station O&M system manages ANR Blacklist and ANRWhitelist.

In step 1), the base station exchanges load information with neighboring base stations, and the specific method for obtaining the load value of neighboring cells can be: the base station and neighboring base stations use the X2 interface (X2 is the third-generation mobile communication technology specification organization 3GPP A defined logical interface between base stations and base stations) to exchange load information; the adjacent cells include all adjacent cells around this cell.

In step 2), the specific method for the base station side to send the automatic neighbor relation (ANR) measurement indication to the user equipment (UE) side may be:

a. The base station side sends the DRX command MAC control unit information message of the MAC (Medium Access Control) layer, and includes at least one field in the message to indicate the ANR measurement;

b. The base station side sends a dedicated control message of the MAC layer, and indicates the ANR measurement through the dedicated message;

c. The base station side sends a control message of the RRC (Radio Resource Control, radio resource control) layer, and instructs the MAC layer to send an ANR measurement instruction through the RRC layer message.

In step 2), the user equipment (UE) reports the reference signal received power (RSRP) measurement information, and the UE measures RSRP by continuously adding the reference signal (pilot signal) power in the subframe of the full measurement bandwidth. The measurement process is filtering sampling , r _n is the updated filter measurement result, r _n-1 is the old filter measurement result, q _n is the latest measurement result received by the end point, K is the filter coefficient provided by the network, and the filter sample is expressed as:

${\mathrm{<span\; class="notranslate">\; r</span>}}_{\mathrm{<span\; class="notranslate">\; no</span>}}<span\; class="notranslate">\; =</span><span\; class="notranslate">\; \{</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; -</span>{<span\; class="notranslate">\; (</span>\frac{\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; )</span>}^{\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; /</span>\mathrm{<span\; class="notranslate">\; 4</span>}}<span\; class="notranslate">\; \}</span><span\; class="notranslate">\; \&times;</span>{\mathrm{<span\; class="notranslate">\; r</span>}}_{\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; +</span>{<span\; class="notranslate">\; (</span>\frac{\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; )</span>}^{\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; /</span>\mathrm{<span\; class="notranslate">\; 4</span>}}<span\; class="notranslate">\; \&times;</span>{\mathrm{<span\; class="notranslate">\; q</span>}}_{\mathrm{<span\; class="notranslate">\; no</span>}}<span\; class="notranslate">\; .</span>$

In step 3), the cell uses the load value of the neighbor cell and the RSRP information reported by the UE as weights to update the priority of the neighbor list. The neighbor list is acquired by the detected cell reported by the UE and the eNB obtained Neighboring cell load values L _load are jointly used as weights to further optimize the priority of the neighboring cell list, then the neighboring cell list L _lin can be expressed as

L _lin =αRSRP+βL _load .

The second stage includes the following steps:

5) When the ANR is in the maintenance phase, each cell in the cell cluster periodically measures its own load status;

6) Send its own load status information to neighboring cells, and the receiving cell will dynamically update the load status information of neighboring cells;

7) The receiving cell dynamically updates the neighbor list according to the load information L _load of the neighbor cell, and further optimizes the priority of the neighbor list;

8) The base station decides to update the neighbor list of the base station O&M system;

9) The NRL on the base station side dynamically updates and maintains the neighbor cell list. When the UE sends a handover request to the eNB, the eNB will query the neighbor cell relationship list to determine the optimal target cell information;

10) Repeat steps 5) to 9);

11) When the load of the cell is too heavy and load transfer is required, select a light-loaded neighboring cell for load transfer. Specifically, it searches its own neighbor cell list, and detects that the load status of the neighbor cells in the cell cluster is a light-load cell or an overload cell, and then performs load transfer for the light-load neighbor cells.

Since the user's moving rate is low, and the load is a statistical value for a period of time, the present invention does not have very high requirements on real-time performance or data synchronization, allowing a larger effective time window.

There are at least the following problems in the prior art:

At present, the update of the neighbor relationship mainly depends on the measurement of the relevant information of the surrounding cells by the UE in the connected state. The UE does not need the guidance of the known neighbor list to collect physical layer information and related high-level information. It is a time-consuming job for the system to obtain GCI (Global Cell Identity, global cell ID). The UE has a certain moving speed. In a long measurement time, it is possible to miss or wrongly detect the cell only through the UE. In addition, in the existing UE measurement, the UE needs to be in the DRX (Discontinuous Reception, discontinuous reception) sleep period. Once there is data access, the UE may interrupt the ANR measurement at any time, so that the current ANR measurement data of the base station and the network side can be reduced. Not guaranteed. On the other hand, when the network load is too concentrated, load imbalance among local cells occurs, resulting in too much measurement information reported by UEs in some cells and too little measurement information reported by UEs in some cells. The collection and processing of too much information leads to The data screening with the base station is complicated, and the redundant transmission of a large amount of related data consumes energy and occupies the processing process; while the obtained measurement information is too small, there are large errors in the statistical results, so that the measurement accuracy of ANR cannot meet the requirements . Moreover, these technologies lack the guarantee of the handover success rate in the case of unbalanced load in local cells.

The present invention has the following advantages: the neighbor list is optimized according to the load state of the neighbor cell, the automatic update of the neighbor list can be realized smoothly, the network interaction load is reduced, the UE is avoided from being selected for unnecessary measurement, and the utilization rate of wireless resources is improved. In this way, the load transfer between cells is optimized, effective sector switching is ensured, ping-pong switching is avoided, the call drop rate is reduced, and a continuous optimal balance is achieved for the neighbor cell list and load balancing. Since the user's moving rate is low, and the load is a statistical value for a period of time, the present invention does not have very high requirements on real-time performance or data synchronization, allowing a relatively large effective time window.

Description of drawings

FIG. 1 is a schematic diagram of the first stage of a load-based ANR optimization method.

Fig. 2 is a schematic diagram of the second stage of a load-based ANR optimization method.

Figure 3 shows the load transfer based on the load-based ANR optimization method.

Detailed ways

The following embodiments will further illustrate the present invention in conjunction with the accompanying drawings.

An embodiment of the present invention provides an ANR-based optimization method. The present invention includes two stages. The first stage is applicable to the optimization when ANR is initially established.

1) As shown in Figure 1, the base station exchanges load information to obtain the load value of the neighboring cell. The base station and neighboring base stations exchange load information through the X2 interface (X2 is a logical interface between base stations and base stations defined in the third-generation mobile communication technology specification organization 3GPP).

Firstly, when planning neighboring cells, in order to ensure that the mobile station can freely switch between neighboring cells, the neighboring cell relationship list should include all surrounding neighboring cells. Specifically, each cell in the cell cluster completes the measurement of its own load status information, and at the same time sends its own load status information to neighboring cells, and the receiving cell stores the load status information of the neighboring cells in its own neighbor cell list.

The calculation of the load value can be calculated through the resource blocks currently used in the cell and the total resource blocks. Assume that the total physical resource blocks in the cell are M _PRE , the user application service rate is D _u , and the rate that user u can provide per resource block is R (SINR _u ), the load value L _load is calculated as follows:

${\mathrm{<span\; class="notranslate">\; L</span>}}_{\mathrm{<span\; class="notranslate">\; load</span>}}<span\; class="notranslate">\; =</span>\frac{\mathrm{<span\; class="notranslate">\; 1</span>}}{{\mathrm{<span\; class="notranslate">\; m</span>}}_{\mathrm{<span\; class="notranslate">\; PRE</span>}}}<span\; class="notranslate">\; \&Center\; Dot;</span>\underset{\mathrm{<span\; class="notranslate">\; x</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; u</span>}<span\; class="notranslate">\; )</span>}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}\frac{{\mathrm{<span\; class="notranslate">\; D.</span>}}_{\mathrm{<span\; class="notranslate">\; u</span>}}}{\mathrm{<span\; class="notranslate">\; R</span>}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; SINR</span>}}_{\mathrm{<span\; class="notranslate">\; u</span>}}<span\; class="notranslate">\; )</span>}$

Another calculation method of the load can be calculated by the total system capacity in the cell and the currently used capacity. Assume that the total system capacity of the cell is C _total , the current use capacity of the cell is C _current , and another calculation method of the load value L _load can be :

${\mathrm{<span\; class="notranslate">\; L</span>}}_{\mathrm{<span\; class="notranslate">\; load</span>}}<span\; class="notranslate">\; =</span>\frac{{\mathrm{<span\; class="notranslate">\; C</span>}}_{\mathrm{<span\; class="notranslate">\; current</span>}}}{{\mathrm{<span\; class="notranslate">\; C</span>}}_{\mathrm{<span\; class="notranslate">\; total</span>}}}$

The receiving cell can divide the cell into several grades according to the load information of neighboring cells. For example, assuming that the system overload threshold is 60%, the range of load corresponding to each load state is: 0% to 30% in the light load state, and 0% to 30% in the medium load state. 30% to 60% in the load state, 60% to 80% in the heavy load state, and 80% to 100% in the severe overload state. According to the load value of the cell, the cell is divided into several levels. For example, the conditions of the cells listed in Table 1.

Table 1 Neighborhood Load Conditions

Cell 1 light load Cell 2 medium load Cell 3 light load Cell 4 severe overload District 5 light load District 6 overload District 7 medium load

2) The base station sends an instruction to perform automatic neighbor relation (ANR) measurement to the user equipment (UE).

The UE starts the ANR measurement according to the ANR measurement instruction sent by the base station side. The base station side sends the indication of ANR measurement to the UE, including 3 ways:

a. The base station side sends the DRX command MAC control unit information message of the MAC (Medium Access Control) layer, and includes at least one field in the message to indicate the ANR measurement;

b. The base station side sends a dedicated control message of the MAC layer, and indicates the ANR measurement through the dedicated message;

c. The base station side sends a control message of the RRC (Radio Resource Control, radio resource control) layer, and instructs the MAC layer to send an ANR measurement instruction through the RRC layer message.

In step 2), the user equipment (UE) performs ANR measurement, and reports RSRP (Reference Signal Receiving Power, reference signal receiving power) information to the base station side. The UE measures RSRP by continuously adding the reference signal (pilot signal) power in the subframe of the full measurement bandwidth. The measurement process is filter sampling, r _n is the updated filter measurement result, and r _n-1 is the old filter measurement result , q _n is the latest measurement result received by the end point, K is the filter coefficient provided by the network, and the filter sampling is expressed as:

${\mathrm{<span\; class="notranslate">\; r</span>}}_{\mathrm{<span\; class="notranslate">\; no</span>}}<span\; class="notranslate">\; =</span><span\; class="notranslate">\; \{</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; -</span>{<span\; class="notranslate">\; (</span>\frac{\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; )</span>}^{\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; /</span>\mathrm{<span\; class="notranslate">\; 4</span>}}<span\; class="notranslate">\; \}</span><span\; class="notranslate">\; \&times;</span>{\mathrm{<span\; class="notranslate">\; r</span>}}_{\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; +</span>{<span\; class="notranslate">\; (</span>\frac{\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; )</span>}^{\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; /</span>\mathrm{<span\; class="notranslate">\; 4</span>}}<span\; class="notranslate">\; \&times;</span>{\mathrm{<span\; class="notranslate">\; q</span>}}_{\mathrm{<span\; class="notranslate">\; no</span>}}$

3) The adjacent cell load value obtained in step 1) and the user equipment (UE) obtained in step 2) are reported to the reference signal received power (RSRP) measurement information to optimize the adjacent cell list. The cell uses the load value of the neighbor cell and the RSRP information reported by the UE as weight values to update the priority of the neighbor cell list. The neighbor cell list is acquired by the detection cell reported by the UE, and the neighbor cell load value Lload obtained by the eNB is used as a weight to further optimize the priority of the neighbor cell list. Then the neighbor list L _lin can be expressed as

L _lin =αRSRP+βL _load

4) The base station judgment will detect that the cell is added to the neighbor cell relationship, and the base station O&M system manages ANR Blacklist and ANRWhitelist.

5) When the load of the cell is too heavy, it needs to carry out load transfer, search its own neighbor cell list, and detect that the load status of the neighbor cell in the cell cluster is a light load cell or an overload cell, then it can select a light load neighbor cell to carry out load shifting.

The embodiment of the present invention has the following advantages: after the neighbor cell list is optimized based on the load state, the base station can effectively screen the monitored cells, greatly reducing the redundant information in the neighbor cell relationship configuration process, simplifying the neighbor cell relationship maintenance process, and improving Maintenance efficiency. When an overloaded cell needs to perform load transfer, the base station preferentially selects a UE with a high priority in the neighbor cell list and a lightly loaded neighbor cell for handover, which can effectively and flexibly perform cell load transfer, basically achieving dynamic load balance between cells, ensuring Handover success rate, optimize the entire network performance. For example, the cell 4 in the graph is originally seriously overloaded, but due to the adjustment in the neighbor cell relationship list, the load transfer is completed after the cell selection is performed when the user switches, and the cell 4 transfers the load to all the adjacent cells The above-mentioned cell 3 and the above-mentioned cell 5, so the serious overload situation of the cell 4 can be alleviated, and the load of the cell 6 is also transferred to the cell 5 at this time, and the load of the cell 5 has increased. The loads of the cells 4 and 6 are reduced, and the adjacent cell list is updated through the change of the load. The priority of the adjacent cell is reduced when the load is heavy, and the priority of the cell is increased when the load is reduced, so that the adjacent cells can be optimized in time. priority, when the user needs to switch at the next moment, it can better provide a switching target with high signal strength and good load to ensure the user's call quality.

An embodiment of the present invention provides an ANR-based optimization method. The present invention includes two stages, and the second stage is suitable for optimization in the ANR maintenance stage, as shown in FIG. 2 .

6) When ANR is in the maintenance phase, each cell in the cell cluster periodically measures its own load status;

7) Send its own load status information to neighboring cells, and the receiving cell dynamically updates the load status information of neighboring cells;

8) The receiving cell dynamically updates the neighbor list according to the load information L _load of the neighbor cell, and further optimizes the priority of the neighbor list;

9) The base station decides to update the neighbor list of the base station O&M system;

10) The NRL on the base station side dynamically updates and maintains the neighbor cell list. When the UE sends a handover request to the eNB, the eNB will query the neighbor cell relationship list to determine the optimal target cell information;

11) Repeat steps 6) to 10);

12) When the load of the cell is too heavy and load transfer is required, select a light-loaded neighboring cell for load transfer. Specifically, it searches its own neighbor cell list, and detects that the load status of the neighbor cells in the cell cluster is a light-load cell or an overload cell, and then performs load transfer for the light-load neighbor cells.

Since the user's moving rate is low, and the load is a statistical value for a period of time, the present invention does not have very high requirements on real-time performance or data synchronization, allowing a larger effective time window.

The embodiment of the present invention uses FIG. 3 to describe in more detail the detailed steps of the load-based optimization of neighboring cells when the load is unbalanced.

1. Step 401, the base station side periodically detects the cell's own load information;

2. Step 403, the source cell and the neighboring base station exchange their respective load information through the X2 interface;

3. Step 404. At the same time, the source cell judges whether to trigger mobile load balancing. If the load of the cell exceeds the system load threshold C (L _load > C), execute step 405. If the load of the cell is lower than the system load Threshold C (L _load <C), the base station still periodically detects the load information of the cell itself, and executes step 401;

4. Step 404, the source cell updates the neighbor list priority according to the load of the neighbor cell;

5. Step 405, the source cell selects a target mobile load balancing cell in the neighbor cell list. First, determine the amount of load that needs to be transferred according to its own load information, and then determine the target load balancing cell for load transfer according to the priority of adjacent cells in the neighbor cell list;

6. Step 406, the source cell negotiates handover parameters with the target cell. According to the amount of load that needs to be transferred, negotiate with the target load balancing cell to change the respective handover parameters, which mainly include the handover threshold and hysteresis margin;

7. Step 407, the source cell selects a suitable user for load transfer. According to the amount of load to be transferred, request the target load balancing cell to reserve corresponding resources, and select qualified users to perform handover according to the changed handover threshold and hysteresis margin;

8. After the user transfer is completed, go to step 401.

The embodiment of the present invention has the following advantages. The neighbor cell list based on the load state can be continuously optimized dynamically and automatically for a long time, which reduces unnecessary UE measurement, saves signaling overhead, and improves the utilization rate of radio resource management. The base station can Effective screening of cells to be monitored reduces redundant information in the process of configuring neighbor cell relationships, simplifies the process of maintaining neighbor cell relationships, and improves maintenance efficiency. When an overloaded cell needs to perform load transfer, the base station will preferentially select the neighbor cell with a high priority in the neighbor cell list and a light load for load transfer to ensure the success rate of the handover, thereby achieving continuous and stable load transfer and reducing the load control module or mobile load Balance the workload of the modules, and the entire network performance is automatically optimized for a long time.

Claims (5)

1. A load-based automatic neighbor relation optimization method in a long-term evolution system, characterized in that it comprises two stages, stage one is the optimization of the initial establishment of ANR, that is, when there is no neighbor list in the base station, stage two is ANR Optimization in the maintenance phase, i.e. when there is a neighbor list in the base station; The first stage includes the following steps: 1) The base station exchanges load information with neighboring base stations to obtain the load value of neighboring cells; 2) The base station side sends an indication of automatically configuring neighbor cell relationship measurement to the user equipment side, and obtains the reference signal received power measurement information reported by the user equipment; 3) Optimize the neighbor cell list by reporting the reference signal received power measurement information obtained by the neighbor cell load value obtained in step 1) and the user equipment obtained in step 2); 4) Base station judgment will detect that the cell is added to the neighbor cell relationship, and the base station O&M system manages ANR Blacklist and ANRWhitelist; The second stage includes the following steps: 5) When the ANR is in the maintenance phase, each cell in the cell cluster periodically measures its own load status; 6) Send its own load status information to neighboring cells, and the receiving cell will dynamically update the load status information of neighboring cells; 7) The receiving cell dynamically updates the neighbor list according to the load information L _load of the neighbor cell, and further optimizes the priority of the neighbor list; 8) The base station decides to update the neighbor list of the base station O&M system; 9) The NRL on the base station side dynamically updates and maintains the neighbor cell list. When the UE sends a handover request to the eNB, the eNB will query the neighbor cell relationship list to determine the optimal target cell information; 10) Repeat steps 5) to 9); 11) When the load of the cell is too heavy and load transfer is required, select a light-loaded neighboring cell for load transfer. Specifically, it searches its own neighbor cell list, and detects that the load status of the neighbor cells in the cell cluster is a light-load cell or an overload cell, and then performs load transfer for the light-load neighbor cells. 2. A load-based automatic neighbor cell relationship optimization method in the long-term evolution system according to claim 1, characterized in that in stage 1 step 1), the base station exchanges load information with the neighbor base station to obtain the neighbor cell The specific method of the load value is: the base station and the adjacent base station exchange load information through the X2 interface; the adjacent cell includes all adjacent cells around the cell, and the X2 is a type defined in the third-generation mobile communication technology specification organization 3GPP The logical interface between the base station and the base station. 3. A load-based automatic neighbor relationship optimization method in a long-term evolution system according to claim 1, characterized in that in step 2) of stage 1, the base station side sends the automatically configured neighbor relationship measurement to the user equipment side The specific method indicated is: a. The base station side sends the DRX command MAC control element information message of the MAC layer, and includes at least one field in the message to indicate the ANR measurement; b. The base station side sends a dedicated control message of the MAC layer, and indicates the ANR measurement through the dedicated message; c. The base station side sends an RRC layer control message, and instructs the MAC layer to send an ANR measurement instruction through the RRC layer message. 4. A load-based automatic neighbor cell relationship optimization method in a long-term evolution system according to claim 1, characterized in that in step 2) of the first stage, the user equipment reports the reference signal reception power measurement information, and the user equipment passes through the complete The subframes of the measurement bandwidth are continuously added to the reference signal power to measure RSRP. The measurement process is filter sampling, r _n is the updated filter measurement result, r _n-1 is the old filter measurement result, and q _n is the latest received by the terminal. The measurement result, K is the filter coefficient provided by the network, and the filter sampling is expressed as: ${\mathrm{<span\; class="notranslate">\; r</span>}}_{\mathrm{<span\; class="notranslate">\; no</span>}}<span\; class="notranslate">\; =</span><span\; class="notranslate">\; \{</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; -</span>{<span\; class="notranslate">\; (</span>\frac{\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; )</span>}^{\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; /</span>\mathrm{<span\; class="notranslate">\; 4</span>}}<span\; class="notranslate">\; \}</span><span\; class="notranslate">\; \&times;</span>{\mathrm{<span\; class="notranslate">\; r</span>}}_{\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; +</span>{<span\; class="notranslate">\; (</span>\frac{\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; )</span>}^{\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; /</span>\mathrm{<span\; class="notranslate">\; 4</span>}}<span\; class="notranslate">\; \&times;</span>{\mathrm{<span\; class="notranslate">\; q</span>}}_{\mathrm{<span\; class="notranslate">\; no</span>}}<span\; class="notranslate">\; .</span>$ 5. A load-based automatic neighbor cell relationship optimization method in the long term evolution system according to claim 1, characterized in that in step 3) of stage one, the cell compares the load value of the neighbor cell with the RSRP information reported by the UE At the same time, it is used as a weight to update the priority of the neighbor list. The neighbor list is obtained by detecting the cell reported by the UE, and the neighbor load value L _load obtained by the eNB is used together as a weight to further optimize the priority of the neighbor list. , then the neighbor list L _lin is expressed as L _lin =αRSRP+βL _load .

Images