Method, server, and storage medium for parameter adjustment

Abstract

Embodiments of the present invention relate to the field of communications, and disclose a method for parameter adjustment, a server, and a storage medium. In the present invention, an index value of a first performance index of a device is obtained; the merits and demerits of each candidate value of a target operating parameter of the device under the index value of the first performance index are found; the merits and demerits of the candidate value refer to the degree of merits and demerits of setting the target operating parameter to the candidate value under the index value of the first performance index; according to the merits and demerits of each of the found candidate values, one of the candidate values is selected as the set value of the adjusted target operating parameter. In this way, the purpose of automatically adjusting to obtain a parameter value with higher accuracy is achieved.

Description

Technical Field

[0001] Embodiments of the present invention relate to the field of communications, and in particular, to a method for parameter adjustment, a server, and a storage medium. Background Art

[0002] Wireless communication systems include LTE (Long Term Evolution) and NR (New Radio) systems, etc. The base station therein is an important part of the wireless communication system. With the increasing complexity of base station design, the number of parameters of the base station is also increasing. For considerations such as flexibility, personalized requirements, special scenario requirements, and further performance optimization, the base station provides parameters that can be adjusted on demand. However, network optimization engineers need to select the required parameters from them through processes such as analysis, adjustment, and observation, and perform manual adjustment.

[0003] The inventors found that there are at least the following problems in the prior art: Adjusting parameters manually requires a large amount of human resources, and the parameters selected manually may be inaccurate. Summary of the Invention

[0004] Embodiments of the present invention aim to provide a method for parameter adjustment, a server, and a storage medium, achieving the purpose of automatically adjusting to obtain parameters with high accuracy.

[0005] To solve the above technical problems, an embodiment of the present invention provides a method for parameter adjustment, including the following steps: obtaining the value of the first performance indicator of the device; finding the superiority and inferiority degrees of each candidate value of the target operating parameter of the device under the value of the first performance indicator; the superiority and inferiority degree of the candidate value refers to the superiority and inferiority degree of setting the target operating parameter as the candidate value under the value of the first performance indicator; according to the found superiority and inferiority degrees of each candidate value, selecting one candidate value as the set value of the adjusted target operating parameter.

[0006] An embodiment of the present invention also provides a server, including: at least one processor; and a memory communicatively connected to the at least one processor; wherein, the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the method for parameter adjustment as described above.

[0007] An embodiment of the present invention also provides a computer-readable storage medium storing a computer program, and when the computer program is executed by a processor, the method for parameter adjustment as described above is implemented.

[0008] In the embodiment of the present invention, compared with the prior art, the method first obtains the index value of the first performance index of the device, and selects a candidate value as the set value of the adjusted target operating parameter according to the superiority and inferiority of each candidate value of the target operating parameter of the device under the index value of the first performance index. Since the candidate value selected according to the superiority and inferiority is more accurate, the set value of the adjusted target parameter is more accurate, thus achieving the purpose of automatically adjusting to obtain a parameter with higher accuracy.

[0009] In addition, the process of finding the superiority and inferiority of each candidate value of the operating parameter of the device under the index value of the first performance index includes: determining the current index value range to which the index value of the first performance index belongs; according to the corresponding relationship between the index value range and the superiority and inferiority of each candidate value of the target operating parameter, finding the superiority and inferiority of each candidate value of the target operating parameter corresponding to the current index value range as the superiority and inferiority of each candidate value of the target operating parameter of the device under the index value of the first performance index. When the corresponding relationship between the index value range and the superiority and inferiority of each candidate value of the operating parameter is preset in the device, the method of this embodiment can simply and accurately determine the superiority and inferiority of each candidate value of the target operating parameter of the device under the index value of the first performance index.

[0010] In addition, after selecting a candidate value as the set value of the adjusted target operating parameter according to the superiority and inferiority of the found candidate values, the method further includes: updating the superiority and inferiority of the selected candidate value under the index value of the first performance index. By continuously updating the superiority and inferiority of the candidate values of the target operating parameter of the device under the index values of different first performance indexes, the superiority and inferiority become more accurate, and each time a candidate value is selected as the set value of the adjusted target operating parameter, it is based on the most recently updated superiority and inferiority, improving the accuracy of the automatically adjusted parameter.

[0011] In addition, before the goodness or badness of the candidate values selected under the updated metric value of the first performance metric, the following steps are also included: If the set value of the target operating parameter after adjustment does not match the set value of the target operating parameter before adjustment, obtain the metric value of the second performance metric of the device; when the metric value of the second performance metric indicates that the performance of the device is qualified, the step of updating the goodness or badness of the candidate values selected under the updated metric value of the first performance metric includes: obtaining the metric value of the third performance metric of the device; and updating the goodness or badness of the candidate values selected under the updated metric value of the first performance metric according to the metric value of the third performance metric. When the metric value of the second performance metric indicates that the performance of the device is qualified, it means that it is feasible for the device to operate based on the set value of the target operating parameter after adjustment. Then, continue to operate the device with the set value of the target operating parameter after adjustment, and update the goodness or badness of the candidate values selected in this state according to the obtained metric value of the third performance metric, so that the updated goodness or badness is more accurate.

[0012] In addition, after obtaining the metric value of the second performance metric of the device, the following steps are also included: when the metric value of the second performance metric indicates that the performance of the device is unqualified, roll back the set value of the target operating parameter to the set value of the target operating parameter before adjustment; the step of updating the goodness or badness of the candidate values selected under the updated metric value of the first performance metric includes: reducing the goodness or badness of the candidate values selected under the updated metric value of the first performance metric. When the metric value of the second performance metric indicates that the performance of the device is unqualified, it means that it is not feasible for the device to operate based on the set value of the target operating parameter after adjustment. Rolling back the set value of the target operating parameter to the set value of the target operating parameter before adjustment can make the device operate based on the set value of the target parameter before adjustment, keep the original operating state of the device, and at this time, it is necessary to reduce the goodness or badness of the candidate values selected under the updated metric value of the first performance metric to make the updated goodness or badness more accurate.

[0013] In addition, the first performance metric and the second performance metric are performance metrics of this cell, and the third performance metric includes performance metrics of the same type of this cell and co-frequency adjacent cells; wherein, the device is a base station. When updating the goodness or badness of the candidate values selected under the updated metric value of the first performance metric, not only the metric value of the performance metric of this cell is considered, but also the metric value of the performance metric of co-frequency adjacent cells is considered, which further improves the accuracy of the updated goodness or badness.

[0014] In addition, updating the goodness or badness of the candidate value selected based on the metric value of the third performance metric for the metric value of the first performance metric includes: obtaining an adjustment value for the goodness or badness based on the metric value of the third performance metric; and updating the goodness or badness of the candidate value selected based on the metric value of the first performance metric according to the adjustment value for the goodness or badness. The adjustment value for the goodness or badness obtained based on the metric value of the third performance metric is more accurate, and then updating the goodness or badness according to the adjustment value for the goodness or badness makes the updated goodness or badness more accurate.

[0015] In addition, reducing the goodness or badness of the candidate value selected based on the metric value of the first performance metric includes: reducing the goodness or badness of the candidate value selected based on the metric value of the first performance metric according to a preset adjustment value for the goodness or badness. In the case where the set value of the target operating parameter reverts to the set value of the target operating parameter before adjustment, there is also a specific method to reduce the goodness or badness of the candidate value selected based on the metric value of the first performance metric, and directly reducing it according to the preset adjustment value for the goodness or badness improves the update efficiency.

[0016] In addition, selecting one candidate value from the candidate values of the target operating parameter of the device based on the goodness or badness of each candidate value of the first performance metric includes: determining the selection method of the candidate value based on the ε-greedy algorithm; the selection method is random selection or selection according to the goodness or badness; and selecting one candidate value from the candidate values according to the selection method. Based on the ε-greedy algorithm, the selection method of the candidate value can be accurately determined, and then selecting one candidate value according to different selection methods can more accurately select a suitable candidate value.

[0017] In addition, the device is a base station, the first performance metric is the uplink block error rate of the cell, and the target operating parameter is the uplink transmission power; or the first performance metric is the resource block utilization rate of the cell, and the target operating parameter is a parameter characterizing the cell user migration management. In the actual application of parameter adjustment, the uplink block error rate of the cell affects the selection of the uplink transmission power, so it meets the needs of actual application that the first performance metric is the uplink block error rate of the cell and the target operating parameter is the uplink transmission power; the resource block utilization rate of the cell affects the selection of the parameter for cell user migration management, so it meets the needs of actual application that the first performance metric is the resource block utilization rate of the cell and the target operating parameter is the parameter for cell user migration management. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] One or more embodiments are exemplarily illustrated by the pictures in the corresponding drawings. These exemplary illustrations do not limit the embodiments. Elements with the same reference numerals in the drawings are represented as similar elements, unless otherwise stated, and the drawings in the figures do not constitute a proportional limitation.

[0019] Figure 1 is a flowchart of the method for parameter adjustment according to the first embodiment of the present invention;

[0020] Figure 2 is a flowchart of a specific implementation manner of step 102 according to the first embodiment of the present invention;

[0021] Figure 3 is a flowchart of a specific implementation manner of step 103 according to the first embodiment of the present invention;

[0022] Figure 4 is a flowchart of the method for parameter adjustment according to the second embodiment of the present invention;

[0023] Figure 5 is a flowchart of the method for parameter adjustment according to the third embodiment of the present invention;

[0024] Figure 6 is a schematic structural diagram of the server according to the fourth embodiment of the present invention. Detailed implementation manners

[0025] To make the objectives, technical solutions and advantages of the embodiments of the present invention clearer, the embodiments of the present invention will be described in detail below with reference to the accompanying drawings. However, those of ordinary skill in the art can understand that in the embodiments of the present invention, many technical details are provided for the reader to better understand the present application. However, even without these technical details and various changes and modifications based on the following embodiments, the technical solutions required to be protected by the present application can still be implemented. The following division of each embodiment is for convenience of description and should not constitute any limitation to the specific implementation manner of the present invention. Each embodiment can be combined and cross-referenced with each other on the premise of not conflicting with each other.

[0026] The first embodiment of the present invention relates to a method for parameter adjustment, which is applied to devices, such as base stations, etc. The specific process is as Figure 1 shown and includes:

[0027] Step 101, obtaining the index value of the first performance index of the device.

[0028] Specifically, the device automatically obtains the index value of the first performance index of the device within a most recent preset time. The preset time can be set according to actual needs. For example, the preset time is 15 minutes. The first performance index can be a type of performance index. For example, the first performance index is the uplink block error rate (BLER) of the cell 本 , obtaining the BLER 本The index value; the first performance indicator can also be multiple types of performance indicators. In this case, the index values of multiple types of performance indicators are obtained respectively. For example, if the first performance indicator is the resource block utilization rate, i.e., the PRB utilization rate, and the control channel element utilization rate, i.e., the CCE utilization rate, the index values of the PRB utilization rate and the CCE utilization rate are obtained respectively.

[0029] Step 102: Find the goodness or badness of each candidate value of the target operating parameter of the device under the index value of the first performance indicator. The goodness or badness of a candidate value refers to the degree of goodness or badness of setting the target operating parameter as the candidate value under the index value of the first performance indicator.

[0030] Specifically, the goodness or badness can be represented by a numerical value or various levels such as good or bad. In the following embodiments of this embodiment, the goodness or badness is represented by a numerical value, and the larger the numerical value, the better, but this is not limited thereto. When the device is initially operating, under the index values of different first performance indicators, the initial values of the goodness or badness of each candidate value of the target operating parameter of the device can all be 0, or different numerical values can be assigned according to actual experience; therefore, under the index values of different first performance indicators, the goodness or badness of each candidate value of the target operating parameter of the corresponding device may also be different, and it is necessary to find the goodness or badness of each candidate value of the target operating parameter of the device under the index value of the first performance indicator. In an example, the device is a base station, and the first performance indicator is the uplink block error rate (BLER) of the cell. 本 The target operating parameter is the uplink transmit power P0. In an example, the device is a base station, and the first performance indicator is the resource block utilization rate of the cell, i.e., the PRB utilization rate, and the target operating parameter is a parameter characterizing the cell user migration management. For example, the parameters characterizing the cell user migration management are three operations: kicking users, no operation, and attracting users. Kicking users means performing an operation to reduce the number of users connected to this cell, and attracting users means performing an operation to increase the number of users connected to this cell. Among them, the base station can preset to implement the kicking user operation or the attracting user operation through interoperation parameters. In this case, the implementation method is to send an instruction to encourage users to connect to other cells to the selected users to implement the kicking user operation, and to send an instruction to encourage users to connect to this cell to the selected users to implement the attracting user operation; it can also preset to implement the kicking user operation or the attracting user operation by changing the transmit power of the cell. In this case, the implementation method is to reduce the transmit power of the cell to implement the kicking user operation, and to increase the transmit power of the cell to implement the attracting user operation.

[0031] In one example, directly based on the metric value of the first performance metric, the superiority and inferiority degrees of each candidate value of the target operating parameter of the device can be found under the metric value of the first performance metric. For example, if the first performance metric is a type of performance metric, and under the metric value a of the first performance metric, the superiority and inferiority degrees of each candidate value of the target operating parameter of the device are 1, 3, 4; under the metric value b of the first performance metric, the superiority and inferiority degrees of each candidate value of the target operating parameter of the device are 2, 5, 3; under the metric value c of the first performance metric, the superiority and inferiority degrees of each candidate value of the target operating parameter of the device are 4, 5, 2...; when the obtained first performance metric value is a, then directly find that under the metric value a of the first performance metric, the superiority and inferiority degrees of each candidate value of the target operating parameter of the device are 1, 3, 4. If the first performance metric is two types of performance metrics M and N, and under the metric values a and b of the first performance metrics M and N respectively, the superiority and inferiority degrees of each candidate value of the target operating parameter of the device are 1, 3, 4; under the metric values b and c of the first performance metrics M and N, the superiority and inferiority degrees of each candidate value of the target operating parameter of the device are 2, 5, 3; under the metric values c and d of the first performance metric, the superiority and inferiority degrees of each candidate value of the target operating parameter of the device are 4, 5, 2...; when the obtained metric values of the first performance metrics M and N are a and b, then directly find that under the metric values a and b of the first performance metrics M and N, the superiority and inferiority degrees of each candidate value of the target operating parameter of the device are 1, 3, 4.

[0032] In one example, the specific flowchart for finding the superiority and inferiority degrees of each candidate value of the operating parameter of the device under the metric value of the first performance metric is as Figure 2 shown, including:

[0033] Step 1021, determine the current metric value range to which the metric value of the first performance metric belongs.

[0034] Step 1022, according to the correspondence between the metric value range and the superiority and inferiority degrees of each candidate value of the target operating parameter, find the superiority and inferiority degrees of each candidate value of the target operating parameter corresponding to the current metric value range, as the superiority and inferiority degrees of each candidate value of the target operating parameter of the device under the metric value of the first performance metric.

[0035] Specifically, the device pre-sets the correspondence between the metric value range and the superiority and inferiority degrees of each candidate value of the target operating parameter. The metric value range can be a continuous range, such as: 5% < BLER 本 ≤ 10%, or can be a set composed of discrete values, such as: {5%, 6%, 7%, 8%}. In one example, the first performance metric is a type of performance metric, and the first performance metric is BLER 本, the target operating parameter is P0, and the candidate values of P0 are -100dBm, -94dBm, -88dBm, -82dBm, -76dBm. The correspondence between the index value range and the merits of the candidate values of the target operating parameter is shown in the following table:

[0036]

[0037] If the BLER 本 has an index value of 11%, then the current index value range it belongs to is 10% < BLER 本 ≤15%. Then, according to the correspondence between the index value range in the above table and the merits of the candidate values of the target operating parameter, it is found that the merits of the candidate values of the target operating parameter corresponding to the current index value range are 1, 8, 4, 6, 2.

[0038] In an example, the first performance index is a type of performance index. The first performance index is the PRB utilization rate, and the target operating parameter is the parameter characterizing the cell user migration management, that is, the candidate values of the parameter characterizing the cell user migration management are three operations: kicking users, no operation, and attracting users. The correspondence between the index value range and the merits of the candidate values of the three operations of the target operating parameter is shown in the following table. Among them, the index value range is reflected in the form of a state in the table, but actually, when stored in the computer, it is the interval of the PRB utilization rate, such as the PRB utilization rate is [0%, 30%), the PRB utilization rate is [30%, 60%), the PRB utilization rate is [60%, 80%), the PRB utilization rate is [80%, 100%]. And according to actual experience, the PRB utilization rate of [0%, 30%) can be considered that the device is in an idle state, the PRB utilization rate of [30%, 60%) can be considered that the device is in a relatively idle state, the PRB utilization rate of [60%, 80%) can be considered that the device is in a relatively busy state, and the PRB utilization rate of [80%, 100%] can be considered that the device is in a busy state:

[0039] Kick user No operation Attract user Idle 3 0 2 Relatively idle 1 3 5 Relatively busy 4 3 1 Busy 6 2 2

[0040] If the index value of the PRB utilization rate is 40%, then the current index value range it belongs to is [30%, 60%), which is reflected as the device being in a relatively idle state in the above table. Then, according to the correspondence between the index value range in the above table and the merits of the candidate values of the target operating parameter, it is found that the merits of the candidate values of the target operating parameter corresponding to the current index value range are 1, 3, 5.

[0041] In one example, the first performance metric is of two types, namely the PRB utilization rate and the CCE utilization rate. The target operating parameter is a parameter characterizing cell user migration management. The candidate values of the parameter characterizing cell user migration management are three operations: kicking users, no operation, and attracting users. The correspondence between the index value intervals and the merits of the candidate values of the three operations of the target operating parameter is shown in the following table. Among them, the index value intervals are reflected in the table in the form of states, but actually when stored in the computer, the intervals of the PRB utilization rate and the CCE utilization rate are stored, such as the PRB utilization rate being in the range of [0%, 40%) and the CCE utilization rate being in the range of [0%, 50%), the PRB utilization rate being in the range of [0%, 40%) and the CCE utilization rate being in the range of [50%, 100%), the PRB utilization rate being in the range of [40%, 100%) and the CCE utilization rate being in the range of [0%, 50%), the PRB utilization rate being in the range of [40%, 100%) and the CCE utilization rate being in the range of [50%, 100%). According to actual experience, the PRB utilization rate being in the range of [0%, 40%) and the CCE utilization rate being in the range of [0%, 50%) can be considered as the device being in an idle state, the PRB utilization rate being in the range of [0%, 40%) and the CCE utilization rate being in the range of [50%, 100%) can be considered as the device being in a relatively idle state, the PRB utilization rate being in the range of [40%, 100%) and the CCE utilization rate being in the range of [0%, 50%) can be considered as the device being in a relatively busy state, and the PRB utilization rate being in the range of [40%, 100%) and the CCE utilization rate being in the range of [50%, 100%) can be considered as the device being in a busy state:

[0042] Kick user No operation Attract user Idle 3 0 2 Relatively idle 1 3 5 Relatively busy 4 3 1 Busy 6 2 2

[0043] If the index value of the PRB utilization rate is 30% and the CCE utilization rate is 20%, then the current index value intervals it belongs to are [0%, 40%) and [0%, 50%). Reflected in the above table, it is that the device is in an idle state. Then, according to the correspondence between the index value intervals in the above table and the merits of the candidate values of the target operating parameter, it is found that the merits of the candidate values of the target operating parameter corresponding to the current index value intervals are 3, 0, and 2.

[0044] Step 103: Select a candidate value according to the merits of the candidate values found, as the set value of the adjusted target operating parameter.

[0045] In one example, selecting a candidate value according to the merits of the candidate values found includes: directly selecting the candidate value with the highest merit from the merits of the candidate values found, or selecting any one of the two candidate values with the highest merit from the merits of the candidate values found, and so on.

[0046] In one example, according to the metric values of the first performance metric, the flowchart for selecting a candidate value from the candidate values of the target operating parameters of the device is as follows Figure 3 shown, including:

[0047] Step 1031: Based on the ε-greedy algorithm, determine the selection method for the candidate value; the selection method is random selection or selection according to the goodness or badness degree.

[0048] Specifically, first generate a random number x, where 0 ≤ x ≤ 1; if x is less than ε, determine that the selection method for the candidate value is random selection; if x is not less than ε, determine that the selection method for the candidate value is selection according to the goodness or badness degree; where The method of parameter adjustment is executed periodically, and T is the current execution count. For example: T = 255, ε = 0.067, if x is 0.02, the selection method is random selection, and if x is 0.08, the selection method is selection according to the goodness or badness degree.

[0049] Step 1032: According to the selection method, select a candidate value from the candidate values.

[0050] In one example, if the selection method is selection according to the goodness or badness degree, select the candidate value with the highest goodness or badness degree from the candidate values. For example: in the current state, the goodness or badness degrees of the candidate values of the target operating parameters of the device are 1, 8, 4, 6, 2 respectively, then select the candidate value with a goodness or badness degree of 8. If the selection method is random selection, then select any candidate value.

[0051] In this embodiment, first obtain the metric value of the first performance metric of the device, search for the goodness or badness degrees of the candidate values of the target operating parameters of the device under the metric value of the first performance metric, and select a candidate value as the set value of the adjusted target operating parameter according to the obtained goodness or badness degrees of the candidate values. Since the candidate value selected according to the goodness or badness degree is more accurate, the set value of the adjusted target parameter is more accurate, thus achieving the purpose of automatically adjusting to obtain parameters with higher accuracy.

[0052] The second embodiment of the present invention relates to a method for parameter adjustment. The second embodiment is substantially the same as the first embodiment, and the main difference is that: continuously update the goodness or badness degrees of the candidate values of the target operating parameters of the device in different states. The specific flowchart is as follows Figure 4 shown, including:

[0053] Step 201: Obtain the metric value of the first performance metric of the device.

[0054] Step 202: Search for the goodness or badness degrees of the candidate values of the target operating parameters of the device under the metric value of the first performance metric; the goodness or badness degree of the candidate value refers to the goodness or badness degree of setting the target operating parameter as the candidate value under the metric value of the first performance metric.

[0055] Step 203: Select a candidate value as the set value of the adjusted target operating parameter according to the quality degrees of the found candidate values.

[0056] Steps 201 - 203 are similar to steps 101 - 103 and will not be elaborated here.

[0057] Step 204: Update the quality degrees of the candidate values selected under the index value of the first performance index.

[0058] In one example, an adjustment value of the quality degree can be preset, and the adjustment value of the quality degree is added to the original quality degree to obtain a new quality degree, and the new quality degree is updated as the quality degree of the candidate value selected under the index value of the first performance index. For example: the preset adjustment value of the quality degree is 2, and the original quality degree is 3, then the new quality degree is 5, and the new quality degree 5 is updated as the quality degree of the candidate value selected under the index value of the first performance index.

[0059] In one example, a new quality degree can also be preset, and the new quality degree is directly updated as the quality degree of the candidate value selected under the index value of the first performance index. For example: the preset new quality degree is 5, then the new quality degree 5 is updated as the quality degree of the candidate value selected under the index value of the first performance index.

[0060] In this embodiment, by continuously updating the quality degrees of the candidate values of the target operating parameters of the device under the index values of different first performance indexes, the quality degrees are made more accurate, and each time a candidate value is selected as the set value of the adjusted target operating parameter, it is based on the most recently updated quality degree, which improves the accuracy of the automatically adjusted parameters.

[0061] The third embodiment of the present invention relates to a method for parameter adjustment. The third embodiment is substantially the same as the second embodiment, and the main difference is that: if the set value of the adjusted target operating parameter does not match the set value of the target operating parameter before adjustment, it is also necessary to evaluate whether it is feasible for the device to operate based on the set value of the adjusted target operating parameter according to the index value of the second performance index, and then perform corresponding operations. The specific flowchart is as Figure 5 shown, including:

[0062] Step 301: Obtain the index value of the first performance index of the device.

[0063] Step 302: Search for the quality degrees of the candidate values of the target operating parameter of the device under the index value of the first performance index; the quality degree of the candidate value refers to the quality degree of setting the target operating parameter as the candidate value under the index value of the first performance index.

[0064] Step 303: Select a candidate value as the set value of the adjusted target operating parameter according to the merits of each candidate value found.

[0065] Steps 301 - 303 are similar to Steps 101 - 103 and will not be elaborated here.

[0066] Step 304: If the set value of the adjusted target operating parameter does not match the set value of the target operating parameter before adjustment, obtain the index value of the second performance index of the device.

[0067] Specifically, when the set value of the adjusted target operating parameter is inconsistent with the set value of the target operating parameter before adjustment, it indicates that they do not match, and the index value of the second performance index of the device is obtained; when the set value of the adjusted target operating parameter is consistent with the set value of the target operating parameter before adjustment, it indicates that they match. If it is the first round of parameter adjustment, an initial value is preset for the set value of the target operating parameter, and the set value of the target operating parameter before the first round of adjustment is the initial value. The second performance index includes but is not limited to one or any combination of the following: Radio Resource Control (RRC) connection establishment success rate, the establishment success rate of the bearer of the user plane (ERAB), radio access success rate, handover success rate, radio disconnection rate, ERAB disconnection rate.

[0068] Step 305: Determine whether the index value of the second performance index indicates that the performance of the device is qualified. If it is qualified, first enter Step 306 and then enter Step 307; if it is unqualified, first enter Step 308 and then enter Step 309.

[0069] Specifically, the index values of the second performance index are compared with the corresponding preset thresholds according to the types respectively. If the index values of the preset number are lower than the corresponding preset thresholds, it indicates that the performance of the device is unqualified, otherwise, it indicates that the performance of the device is qualified. The preset number can be set according to actual needs. For example: If the index values of the second performance index include: RRC connection establishment success rate of 50%, ERAB establishment success rate of 70%, radio access success rate of 65%, handover success rate of 66%, and the corresponding preset thresholds are: 60%, 65%, 60%, 55%, 60% respectively, if the preset number is 1, and at this time there is one index value lower than the corresponding preset threshold, it indicates that the performance of the device is unqualified.

[0070] In an example, if the judgment results of more than the preset number are all unqualified, restore the merits of each candidate value of the target operating parameter of the device in different states to their initial values. The preset number can be set according to actual needs and is not specifically limited in this embodiment.

[0071] Step 306: Obtain the index value of the third performance index of the device.

[0072] Step 307: Update the quality of the candidate values selected under the metric value of the first performance metric according to the metric value of the third performance metric.

[0073] In one example, updating the quality of the candidate values selected under the metric value of the first performance metric according to the metric value of the third performance metric includes: obtaining an adjustment value for the quality according to the metric value of the third performance metric; and updating the quality of the candidate values selected under the metric value of the first performance metric according to the adjustment value for the quality. Specifically, the metric value of the third performance metric corresponds to preset different adjustment values for the quality, or the third performance metric is substituted into a preset calculation formula to calculate the adjustment value for the quality, and then the adjustment value for the quality and the original quality are added together to obtain a new quality, and the new quality is updated as the quality of the candidate values selected in the current state. For example: when the metric value of the third performance metric is 5%, the corresponding adjustment value for the quality is 2; when the metric value of the third performance metric is 10%, the corresponding adjustment value for the quality is 3.

[0074] In one example, the first performance metric and the second performance metric are performance metrics of the local cell, and the third performance metric includes performance metrics of the same type of the local cell and co-frequency adjacent cells; wherein, the device is a base station. For example: the first performance metric is the uplink block error rate (BLER) of the local cell 本 , the second performance metric is the ERAB drop rate of the local cell, and the third performance metric includes the total traffic Payload of the local cell re-obtained 本 and the total traffic Payload of co-frequency adjacent cells 邻 . In one example, the third performance metric is of the same type as the first performance metric. For example: the first performance metric is BLER 本 , and the third performance metric is the uplink block error rate (BLER) of the local cell re-obtained within a most recent preset time 本 and the uplink block error rate (BLER) of co-frequency adjacent cells 邻 .

[0075] Step 308: Roll back the set value of the target operating parameter to the set value of the target operating parameter before adjustment.

[0076] Specifically, if the set value of the target operating parameter P0 before adjustment is -88 dBm and the set value of the target operating parameter after adjustment is -94 dBm, then the set value of the target operating parameter P0 is rolled back to -88 dBm.

[0077] Step 309: Reduce the quality of the candidate values selected under the metric value of the first performance metric.

[0078] In one example, the goodness of candidate values selected at a reduced value of the first performance metric includes: reducing the goodness of candidate values selected at a reduced value of the first performance metric according to a preset adjustment value of the goodness. The preset adjustment value of the goodness is set according to actual needs and is not specifically limited in this embodiment. When the preset adjustment value of the goodness is a positive value, the original goodness is subtracted by the preset adjustment value of the goodness to obtain a new goodness, and the new goodness is updated as the goodness of the candidate value selected in the current state. For example: if the original goodness is 5 and the preset adjustment value of the goodness is 1, then the new goodness is 4, and the new goodness 4 is updated as the goodness of the candidate value selected in the current state. When the preset adjustment value of the goodness is a negative value, the preset adjustment value of the goodness is added to the original goodness to obtain a new goodness, and the new goodness is updated as the goodness of the candidate value selected in the current state. For example: if the original goodness is 5 and the preset adjustment value of the goodness is -1, then the new goodness is 4, and the new goodness 4 is updated as the goodness of the candidate value selected at the value of the first performance metric.

[0079] Two specific examples will be described below for illustration.

[0080] In the first specific example, the device is a base station and the first performance metric is the uplink block error rate (BLER) of the local cell 本 , which is obtained within the most recent preset time period. The preset time period is 15 minutes. Initialize the target uplink BLER 目标值 = 10%, the target operating parameter is P0, the candidate values of P0 are -100 dBm, -94 dBm, -88 dBm, -82 dBm, -76 dBm, and the initial set value of P0 is -88 dBm. The corresponding relationship between the index value range and the goodness of each candidate value of the target operating parameter is shown in the following table, and the initial value of the goodness of each candidate value of the target operating parameter of the device at the value of the first performance metric is 0. The values in the table are goodness:

[0081]

[0082] The following will be described in conjunction with the current execution count T being 256. The corresponding relationship between the index value range and the goodness of each candidate value of the target operating parameter is shown in the following table. The values in the table are the updated goodness:

[0083]

[0084] Step 1: Determine the current index value range to which it belongs according to the index value obtained within the most recent 15 minutes of the performance metric. If the obtained index value of the performance metric is 11%, determine that the current index value range to which it belongs is 10% < BLER 本≤15% state, in the third row of the above table.

[0085] Step 2, find the current index value range 10% < BLER 本 ≤15% and the merits and demerits of each candidate value of the target operating parameter are 1, 8, 4, 6, 2 respectively. Then, according to the ε-greedy algorithm, determine the selection method of the candidate value. If the generated random number x is 0.08 and T = 256, Since x is not less than ε, the selection method is to select according to the merits and demerits. Select the candidate value with the highest merit and demerit from the candidate values of the target operating parameter. Then select the candidate value with a merit and demerit of 8, that is, P0 is -94 dBm, and the set value of the adjusted target operating parameter is -94 dBm.

[0086] If the set value of the target operating parameter P0 before adjustment is -88 dBm, and the set value of the adjusted target operating parameter is inconsistent with the set value of the target operating parameter before adjustment, it indicates that the two do not match, and enter Step 3.

[0087] If the set value of the target operating parameter P0 before adjustment is -94 dBm, and the set value of the adjusted target operating parameter is consistent with the set value of the target operating parameter before adjustment, it indicates that the two match, and directly enter Step 4.

[0088] Step 3, after the base station operates based on P0 = -94 dBm for 10 s, obtain the index value of the second performance index of the device. If the RRC connection establishment success rate is 68%, the ERAB establishment success rate is 70%, the radio access success rate is 65%, and the handover success rate is 66%, and the corresponding preset thresholds are: 60%, 65%, 60%, 55%, 60% respectively. At this time, all the index values are higher than the corresponding preset thresholds, which indicates that the performance of the device is qualified, and enter Step 4.

[0089] If the RRC connection establishment success rate is 50%, the ERAB establishment success rate is 70%, the radio access success rate is 65%, and the handover success rate is 66%, and the corresponding preset thresholds are: 60%, 65%, 60%, 55%, 60% respectively. If one index value is lower than the corresponding preset threshold, it indicates that the performance of the device is unqualified. Roll back the set value of the target operating parameter to the set value of the target operating parameter before adjustment, that is, roll back the set value of the target operating parameter to -88 dBm, and the base station operates based on P0 = -88 dBm. And add the preset adjustment value -1 of the merit and demerit to the original merit and demerit 8 to get the new merit and demerit 7, and cover 8 in the table with the merit and demerit 7, and then re-enter Step 1.

[0090] Step 4, calculate the adjustment value of the merit and demerit according to the following formula:

[0091]

[0092] Wherein, R is the adjustment value of the goodness and badness degree, β is the preset initial weight value of adjacent cells, and N is the number of co-frequency adjacent cells. is the uplink block error rate of the serving cell obtained in the most recent 15 minutes. is the uplink block error rate of the serving cell to be obtained in the next 15 minutes, BLER 目标值 is the initialized target uplink block error rate. is the uplink block error rate of the co-frequency adjacent cell with index n obtained in the most recent 15 minutes. is the uplink block error rate of the co-frequency adjacent cell with index n to be obtained in the next 15 minutes. If the calculated value of R is 1, add the value 1 of R and the original goodness and badness degree 8 to obtain the new goodness and badness degree 9, overwrite 8 in the table with the goodness and badness degree 9, and then re-enter step one.

[0093] In the second specific example, the device is a base station, the first performance indicator is the resource block utilization rate PRB utilization rate, which is obtained in the most recent preset duration. The preset duration is 15 minutes, and the target operating parameter is a parameter representing the cell user migration management, namely, kicking users, attracting users, and no operation. The initial setting value of the target operating parameter is no operation. Among them, different operations are achieved by changing the transmit power of the cell. Kicking users means reducing the transmit power of the cell, attracting users means increasing the transmit power of the cell, and no operation means keeping it unchanged. The change amplitude is 5 dBm. The index value range is reflected in the form of a state in the table. According to practical experience, when the PRB utilization rate is in the range of [0%, 30%), it is considered that the device is in an idle state; when the PRB utilization rate is in the range of [30%, 60%), it is considered that the device is in a relatively idle state; when the PRB utilization rate is in the range of [60%, 80%), it is considered that the device is in a relatively busy state; when the PRB utilization rate is in the range of [80%, 100%], it is considered that the device is in a busy state. The corresponding relationship between the index value range and the goodness and badness degrees of the candidate values of the target operating parameter is shown in the following table, and the initial values of the goodness and badness degrees of the candidate values of the target operating parameter under the index value of the first performance indicator are all 0. The values in the table are the goodness and badness degrees:

[0094] a = Kick user a = No operation a = Attract user s = Idle 0 0 0 s = Relatively idle 0 0 0 s = Relatively busy 0 0 0 s = Busy 0 0 0

[0095] The following will be described in combination with the current execution count T being 256. The corresponding relationship between the index value range and the goodness and badness degrees of the candidate values of the target operating parameter is shown in the following table. The values in the table are the updated goodness and badness degrees:

[0096] Kick user No operation Attract user Idle 3 0 2 Relatively idle 1 3 5 Relatively busy 4 3 1 Busy 6 2 2

[0097] Step 1: Determine the current metric value range to which it belongs based on the metric value of the PRB utilization rate obtained within the most recent 15 minutes. If the metric value of the obtained PRB utilization rate is 70%, determine that the current metric value range to which it belongs is [60%, 80%), that is, the device is in a relatively busy state and is in the third row of the above table.

[0098] Step 2: Find that the merits and demerits of each candidate value of the target operating parameter corresponding to the device in a relatively busy state are 4, 3, and 1 respectively. Then, according to the ε-greedy algorithm, determine the selection method of the candidate value. If the generated random number x is 0.08 and T = 256, Since x is not less than ε, the selection method is to select according to the merits and demerits. Select the candidate value with the highest merit and demerit from the candidate values of the target operating parameter. Then select the candidate value with a merit and demerit of 4, that is, kick the user. The set value of the adjusted target operating parameter is to kick the user. If the transmit power of the current cell is 35 dBm, then adjust the transmit power of the cell to 30 dBm, and the base station operates based on the transmit power of the cell being 30 dBm.

[0099] If the set value of the target operating parameter before adjustment is no operation and the set value of the adjusted target operating parameter is inconsistent with the set value of the target operating parameter before adjustment, it indicates that the two do not match, and proceed to Step 3.

[0100] If the set value of the target operating parameter before adjustment is to kick the user and the set value of the adjusted target operating parameter is consistent with the set value of the target operating parameter before adjustment, it indicates that the two match, and directly proceed to Step 4.

[0101] Step 3: After the base station operates based on the transmit power of the cell being 30 dBm for 10 s, obtain the metric value of the second performance metric of the device. If the RRC connection establishment success rate is 68%, the ERAB establishment success rate is 70%, the radio access success rate is 65%, and the handover success rate is 66%, and the corresponding preset thresholds are: 60%, 65%, 60%, 55%, 60% respectively. At this time, all metric values are higher than the corresponding preset thresholds, indicating that the performance of the device is qualified, and proceed to Step 4.

[0102] If the RRC connection establishment success rate is 50%, the ERAB establishment success rate is 70%, the radio access success rate is 65%, and the handover success rate is 66%, while the corresponding preset thresholds are 60%, 65%, 60%, 55%, and 60% respectively, and if one of the metric values is lower than the corresponding preset threshold, it indicates that the performance of the device is unqualified. The set value of the target operating parameter is reverted to the set value of the target operating parameter before adjustment, that is, the set value of the target operating parameter is reverted to no operation, and the base station adjusts the transmit power of the cell to operate at 35 dBm. And the adjustment value of the preset goodness - of - fit - degree -1 is added to the original goodness - of - fit - degree 4 to obtain the new goodness - of - fit - degree 3, and the goodness - of - fit - degree 3 overwrites 4 in the table, and then re - enters Step 1.

[0103] Step 4, calculate the adjustment value of the goodness - of - fit - degree according to the following formula:

[0104] Formula 1,

[0105] Formula 2,

[0106] where, is the adjustment value of the goodness - of - fit - degree, is the total traffic of this cell obtained in the most recent 15 minutes, is the total traffic of this cell obtained in the next 15 minutes, is the total traffic of the co - frequency adjacent cell with index n obtained in the most recent 15 minutes, is the total traffic of the co - frequency adjacent cell with index n obtained in the next 15 minutes, is the maximum value of the goodness - of - fit - degree of each candidate value of the target operating parameter of the device in the next 15 - minute current state. Initialize the learning rate α = 0.1 and the discount rate γ = 0.9.

[0107] If the calculated adjustment value of the goodness - of - fit - degree is 2, add value 1 to the original goodness - of - fit - degree 4 to obtain the new goodness - of - fit - degree 6, overwrite 4 in the table with the goodness - of - fit - degree 6, and then re - enter Step 1.

[0108] In this embodiment, when the index value of the second performance index indicates that the performance of the device is qualified, it means that it is feasible for the device to operate based on the set value of the adjusted target operating parameter. Then, continue to operate with the set value of the adjusted target operating parameter, and update the quality of the candidate values selected in this state according to the obtained index value of the third performance index, so that the updated quality is more accurate. When the index value of the second performance index indicates that the performance of the device is unqualified, it means that it is not feasible for the device to operate based on the set value of the adjusted target operating parameter. Roll back the set value of the target operating parameter to the set value of the target operating parameter before adjustment, which can make the device operate based on the set value of the target parameter before adjustment, keep the original operating state of the device, and at this time, it is necessary to reduce the quality of the candidate values selected under the index value of the first performance index to make the updated quality more accurate. The fourth embodiment of the present invention relates to a server, as Figure 6 shown, including at least one processor 402; and a memory 401 communicatively connected to the at least one processor; wherein, the memory 401 stores instructions executable by the at least one processor 402, and the instructions are executed by the at least one processor 402 so that the at least one processor 402 can execute the embodiments of the above parameter adjustment method.

[0109] Wherein, the memory 401 and the processor 402 are connected by a bus. The bus can include any number of interconnected buses and bridges, and the bus connects various circuits of one or more processors 402 and the memory 401 together. The bus can also connect various other circuits such as peripheral devices, voltage regulators, and power management circuits, which are well known in the art, so they will not be further described herein. The bus interface provides an interface between the bus and the transceiver. The transceiver can be one element or multiple elements, such as multiple receivers and transmitters, and provides a unit for communicating with various other devices on the transmission medium. The data processed by the processor 402 is transmitted over the wireless medium through the antenna. Further, the antenna also receives data and transmits the data to the processor 402.

[0110] The processor 402 is responsible for managing the bus and general processing, and can also provide various functions, including timing, peripheral interface, voltage regulation, power management, and other control functions. The memory 401 can be used to store the data used by the processor 402 when executing operations.

[0111] The fifth embodiment of the present invention relates to a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, the above method embodiments are implemented.

[0112] That is, those skilled in the art can understand that all or part of the steps in the methods of the above embodiments can be completed by instructing relevant hardware through a program. The program is stored in a storage medium, including several instructions to enable a device (which can be a single-chip microcomputer, a chip, etc.) or a processor to execute all or part of the steps of the methods described in various embodiments of the present application. The aforementioned storage medium includes: various media that can store program codes such as USB flash drives, mobile hard disks, read-only memories (ROMs), random access memories (RAMs), magnetic disks, or optical discs.

[0113] Those of ordinary skill in the art can understand that the above embodiments are specific embodiments for implementing the present invention. In actual applications, various changes can be made in form and details without departing from the spirit and scope of the present invention.

Claims

1. A method for parameter adjustment, characterized in that, Including: Obtaining the index value of the first performance index of the device; Searching for the goodness or badness of each candidate value of the target operating parameter of the device under the index value of the first performance index; The goodness or badness of the candidate value refers to the goodness or badness degree of setting the target operating parameter to the candidate value under the index value of the first performance index; Selecting one of the candidate values as the set value of the adjusted target operating parameter according to the goodness or badness of each found candidate value; If the set value of the adjusted target operating parameter does not match the set value of the target operating parameter before adjustment, obtaining the index value of the second performance index of the device; When the index value of the second performance index indicates that the performance of the device is qualified, obtaining the index value of the third performance index of the device; Updating the goodness or badness of the candidate value selected under the index value of the first performance index according to the index value of the third performance index, where the first performance index and the second performance index are performance indexes of the local cell, and the third performance index includes performance indexes of the same type of the local cell and co-frequency adjacent cells.

2. The method for parameter adjustment according to claim 1, characterized in that The searching for the goodness or badness of each candidate value of the operating parameter of the device under the index value of the first performance index includes: Determining the current index value interval to which the index value of the first performance index belongs; Searching for the goodness or badness of each candidate value of the target operating parameter corresponding to the current index value interval according to the corresponding relationship between the index value interval and the goodness or badness of each candidate value of the target operating parameter, as the goodness or badness of each candidate value of the target operating parameter of the device under the index value of the first performance index.

3. The method for parameter adjustment according to claim 1, characterized in that, The updating the goodness or badness of the candidate value selected under the index value of the first performance index according to the index value of the third performance index includes: Obtaining an adjustment value of the goodness or badness according to the index value of the third performance index; Updating the goodness or badness of the candidate value selected under the index value of the first performance index according to the adjustment value of the goodness or badness.

4. The method for parameter adjustment according to claim 3, characterized in that, After obtaining the index value of the second performance index of the device, it further includes: When the index value of the second performance index indicates that the performance of the device is unqualified, rolling back the set value of the target operating parameter to the set value of the target operating parameter before adjustment; The updating the goodness or badness of the candidate value selected under the index value of the first performance index includes: reducing the goodness or badness of the candidate value selected under the index value of the first performance index.

5. The method for parameter adjustment according to claim 3, wherein The device is a base station.

6. The method for parameter adjustment according to claim 4, characterized in that, The reducing the goodness or badness of the candidate value selected under the index value of the first performance index includes: Reducing the goodness or badness of the candidate value selected under the index value of the first performance index according to a preset adjustment value of the goodness or badness.

7. The method for parameter adjustment according to claim 1, wherein The selecting one of the candidate values according to the goodness or badness of each candidate value of the target operating parameter of the device under the index value of the first performance index includes: Determining the selection method of the candidate value based on the ε-greedy algorithm; the selection method is random selection or selection according to the goodness or badness; Selecting one of the candidate values from the candidate values according to the selection method.

8. The method for parameter adjustment according to claim 1, characterized in that, The device is a base station, the first performance indicator is the uplink block error rate of the cell, and the target operating parameter is the uplink transmission power; or; the first performance indicator is the resource block utilization rate of the cell, and the target operating parameter is a parameter characterizing the user migration management of the cell.

9. A server, characterized in that, It includes: At least one processor; And, A memory communicatively connected to the at least one processor; wherein, The memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the parameter adjustment method as described in any one of claims 1 to 8.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the parameter adjustment method as described in any one of claims 1 to 8.

Images