CN113950134B - Dormancy prediction method, device, equipment and computer readable storage medium of base station - Google Patents
Dormancy prediction method, device, equipment and computer readable storage medium of base station Download PDFInfo
- Publication number
- CN113950134B CN113950134B CN202111224197.3A CN202111224197A CN113950134B CN 113950134 B CN113950134 B CN 113950134B CN 202111224197 A CN202111224197 A CN 202111224197A CN 113950134 B CN113950134 B CN 113950134B
- Authority
- CN
- China
- Prior art keywords
- base station
- chromosome
- power consumption
- determining
- total power
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W52/00—Power management, e.g. TPC [Transmission Power Control], power saving or power classes
- H04W52/02—Power saving arrangements
- H04W52/0203—Power saving arrangements in the radio access network or backbone network of wireless communication networks
- H04W52/0206—Power saving arrangements in the radio access network or backbone network of wireless communication networks in access points, e.g. base stations
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D30/00—Reducing energy consumption in communication networks
- Y02D30/70—Reducing energy consumption in communication networks in wireless communication networks
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Mobile Radio Communication Systems (AREA)
Abstract
The invention provides a dormancy prediction method, a dormancy prediction device, dormancy prediction equipment and a computer readable storage medium of a base station, comprising the following steps: obtaining each historical total power consumption of a base station to be tested, and carrying out chromosome coding to obtain a first chromosome corresponding to each historical total power consumption; performing genetic iterative operation on each first chromosome to obtain the minimum total power consumption of the base station to be tested in the predicted time period; determining the load power consumption of the base station to be tested in a predicted time period according to the minimum total power consumption; and when the load power consumption meets the first preset condition and the base station energy consumption meets the second preset condition, determining that the base station to be tested can sleep in the predicted time period. In the invention, the load power consumption represents the energy consumption of the 5G base station consumed by the user terminal, the base station energy consumption represents the energy consumption of the 5G base station, and the 5G base station energy consumption and the energy consumption consumed by the 5G base station terminal are combined to determine whether the 5G base station can sleep or not, so that the energy-saving technology provided by the invention has universality.
Description
Technical Field
The present invention relates to the field of base station dormancy technologies, and in particular, to a method, an apparatus, a device, and a computer readable storage medium for predicting dormancy of a base station.
Background
The deployment and popularity of 5G networks has given people various benefits to life and work, such as higher rates, wider bandwidths, lower latency, and higher connection densities, neglecting the cost of such high performance.
For example, a great amount of innovative services derived from the communication advantages of large bandwidth, high speed and low time delay of 5G need more data processing, so that the electricity consumption of the 5G base station is increased, and the full-load power of a single station of the 5G base station is approximately 3700W; in addition, the 5G base station adopts more antennas than the 4G base station, which indicates that the 5G base station needs more power consumption, and the power consumption of the 5G single station is 2.5-3.5 times of that of the 4G base station. As another example, an idle 5G base station carries very low, even no traffic, but consumes a lot of power: the 5G AAU (Active Antenna Unit, AAU active antenna unit)/RRU (Remote Radio Unit ) has an idle average power consumption of 633W, and BBU (Building Base band Unit, indoor baseband processing unit) has an idle average power consumption of nearly 300W. Therefore, energy conservation and emission reduction of the 5G base station are problems which need to be solved.
At present, the base station energy saving method sets a threshold value by simply and roughly judging the average value of the flow, the flow of the 5G base station is smaller than the threshold value, the cell of the 5G base station is automatically closed, and the cell is opened when the flow of the 5G base station is larger than the threshold value, so that the method is too simple and roughly, and the user cannot enjoy the 5G network service in time. Thus, these energy saving techniques can only be used in 5G base stations in areas where people are sparsely populated or in areas where customer perception requirements are low. Namely, the existing base station energy saving method has no universality.
Disclosure of Invention
The invention provides a dormancy prediction method, a dormancy prediction device, dormancy prediction equipment and a computer readable storage medium for solving the problem that the existing base station energy saving method does not have universality.
In one aspect, the present invention provides a sleep prediction method of a base station, including:
acquiring each historical total power consumption of a base station to be tested;
chromosome coding is carried out on each historical total power consumption to obtain a first chromosome corresponding to each historical total power consumption;
performing genetic iteration operation on each first chromosome to obtain the minimum total power consumption of the base station to be tested in a prediction time period;
determining the load power consumption and the base station energy consumption of the base station to be tested in the predicted time period according to the minimum total power consumption;
and when the load power consumption meets a first preset condition and the base station energy consumption meets a second preset condition, determining that the base station to be tested can sleep in the prediction time period.
Optionally, the step of performing genetic iterative operation on each first chromosome to obtain the minimum total power consumption of the base station to be tested in the predicted time period includes:
determining a first fitness function value corresponding to each first chromosome;
determining the selected probability of each first chromosome according to each first fitness function value, and determining a first dyeing gas corresponding to the selected probability larger than a preset threshold value as a target chromosome;
determining adaptive crossover probability and first adaptive mutation probability of the target chromosome according to each first fitness function value;
crossing each target chromosome according to each self-adaptive crossing probability, and mutating each target chromosome according to each first self-adaptive mutation probability to obtain each second chromosome;
determining a second fitness function value corresponding to each second chromosome;
when each second fitness function value does not meet the iteration condition, determining second adaptive variation probability of each second chromosome according to each second fitness function value;
and determining the minimum self-adaptive variation probability in each second self-adaptive variation probability, and determining the minimum total power consumption of the base station to be tested in a preset time period according to the minimum self-adaptive variation probability.
Optionally, after the step of determining the second fitness function value corresponding to each of the second chromosomes, the method further includes:
updating each second chromosome to be a first chromosome when each second fitness function value meets an iteration condition;
and returning to the step of determining the selected probability of each first chromosome according to each first fitness function value.
Optionally, the step of performing genetic iterative operation on each first chromosome to obtain the minimum total power consumption of the base station to be tested in the predicted time period includes:
inputting each first chromosome into a prediction model, wherein the prediction model is used for carrying out genetic iterative operation on each first chromosome;
and taking the numerical value output by the prediction model as the minimum total power consumption of the base station to be detected in the prediction time period.
Optionally, after the step of determining that the base station to be tested may sleep in the predicted period of time, the method further includes:
inquiring dynamic parameters of each 5G cell corresponding to the base station to be tested when the current time point is in the predicted time period;
and controlling the base station to be tested to sleep when each dynamic parameter is a normal parameter.
Optionally, after the step of controlling the base station to be tested to sleep, the method further includes:
when detecting that a 5G terminal is registered to a network, determining a target 5G cell in which the 5G terminal is located, wherein when the base station to be tested is in a dormant state, each 5G cell stops providing network services;
and activating the network service of the target 5G cell to enable the 5G terminal to be successfully registered into the network.
Optionally, after the step of activating the network service of the target 5G cell, the method further includes:
acquiring the activation times of the network service when the network service is not activated successfully, wherein when the network service is not activated successfully, an activation instruction is repeatedly sent to the target 5G cell;
and when the activation times reach preset times, sending a maintenance request of the target 5G cell to a preset terminal.
On the other hand, the invention also provides a dormancy prediction device of the base station, which comprises:
the acquisition module is used for acquiring each history total power consumption of the history of the base station to be tested;
the coding module is used for carrying out chromosome coding on each historical total power consumption to obtain a first chromosome corresponding to each historical total power consumption;
the operation module is used for carrying out genetic iterative operation on each first chromosome to obtain the minimum total power consumption of the base station to be tested in the prediction time period;
the determining module is used for determining the load power consumption and the base station energy consumption of the base station to be detected in the predicted time period according to the minimum total power consumption;
and the determining module is further configured to determine that the base station to be tested can sleep in the predicted time period when the load power consumption meets a first preset condition and the base station energy consumption meets a second preset condition.
On the other hand, the invention also provides a dormancy prediction device of the base station, which comprises: a memory and a processor;
the memory stores computer-executable instructions;
the processor executes the computer-executable instructions stored by the memory, causing the processor to perform the sleep prediction method of the base station as described above.
In another aspect, the present invention also provides a computer readable storage medium having stored therein computer executable instructions which when executed by a processor are configured to implement the sleep prediction method of a base station as described above.
According to the dormancy prediction method, the dormancy prediction device, the equipment and the computer readable storage medium of the base station, each historical total power consumption of the history of the base station to be detected is obtained, chromosome coding is carried out on each historical total power consumption to obtain first chromosomes corresponding to the historical total power consumption, genetic iterative operation is carried out on each first chromosome to obtain the minimum total power consumption of the base station to be detected in a prediction time period, then the load power consumption and the base station energy consumption of the base station to be detected in the prediction time period are determined according to the minimum total power consumption, and if the load power consumption meets a first preset condition and the base station energy consumption meets a second preset condition, the base station to be detected can be determined to sleep in the prediction time period. In the invention, the load power consumption represents the energy consumption consumed by the 5G base station for the user terminal, and the base station energy consumption represents the energy consumption consumed by the 5G base station itself.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the disclosure and together with the description, serve to explain the principles of the disclosure.
FIG. 1 is a system architecture diagram of a method for implementing sleep prediction for a base station in accordance with the present invention;
fig. 2 is a flowchart of a first embodiment of a sleep prediction method of a base station according to the present invention;
fig. 3 is a schematic diagram of a refinement flow of step S30 in a second embodiment of the sleep prediction method of the base station according to the present invention;
fig. 4 is a flowchart of a third embodiment of a sleep prediction method of a base station according to the present invention;
FIG. 5 is a block diagram illustrating a sleep prediction apparatus of a base station according to the present invention;
fig. 6 is a schematic hardware structure of a sleep prediction apparatus of a base station according to the present invention.
Specific embodiments of the present disclosure have been shown by way of the above drawings and will be described in more detail below. These drawings and the written description are not intended to limit the scope of the disclosed concepts in any way, but rather to illustrate the disclosed concepts to those skilled in the art by reference to specific embodiments.
Detailed Description
Reference will now be made in detail to exemplary embodiments, examples of which are illustrated in the accompanying drawings. When the following description refers to the accompanying drawings, the same numbers in different drawings refer to the same or similar elements, unless otherwise indicated. The implementations described in the following exemplary examples are not representative of all implementations consistent with the present disclosure. Rather, they are merely examples of apparatus and methods consistent with some aspects of the present disclosure as detailed in the accompanying claims.
The invention provides a dormancy prediction method of a base station, which can be realized through a system framework diagram shown in figure 1. As shown in fig. 1, the base station 100 is communicatively connected to a sleep prediction apparatus 200 of the base station. The solution of the coordinate pairs with the user and the base station as a pair of coordinates (x, y) is carried into the function F (x, y) to be solved. According to the survival of the fittest, the (x, y) pairs are defined such that the larger the function value F (x, y) is, the more suitable the environment is, so that the (x, y) pairs suitable for the environment are more likely to be reserved, while the (x, y) pairs not suitable for the environment are eliminated with a great probability, the reserved points are propagated to generate new points, and most of the points which are finally reserved after the evolution are points suitable for the environment, namely, the points are near the highest points. Such as (x=2.2, y=0.8), (x-1.6, y=2.4), (x=2.1, y=0.8), (x= -1.5, y=2.3), (x=2.2, y=0.8), (x= -1.5, y=2.3) etc. are one possible solution to the problem of maximizing, i.e. individuals in the genetic algorithm by which the individual solution of the problem is obtained and which is encoded in the computer as a vector representation (x, y), the set of such a set of possible solutions is the solved population, which for the base station is the set of base stations that determines whether to sleep. Based on this principle, the dormancy prediction apparatus 200 may encode the historical total power consumption of the base station 100 into chromosomes, so as to perform genetic iterative operations on each chromosome, i.e., predict whether the base station 100 may dormancy in a predicted time period.
The following describes the technical scheme of the present invention and how the technical scheme of the present application solves the above technical problems in detail with specific embodiments. The following embodiments may be combined with each other, and the same or similar concepts or processes may not be described in detail in some embodiments. Embodiments of the present invention will be described below with reference to the accompanying drawings.
Referring to fig. 2, fig. 2 is a first embodiment of a sleep prediction method of a base station according to the present invention, the sleep prediction method of the base station includes the following steps:
step S10, each historical total power consumption of the base station to be tested is obtained.
In the present embodiment, the execution subject is a sleep prediction apparatus of a base station, and for convenience of description, the following means are referred to as a sleep prediction apparatus of a base station. The base station will record its own total power consumption over various time periods, which is defined as the historical total power consumption, and the base station will send various historical total power consumption to the device. The device screens the total power consumption to obtain the historical total power consumption of the base station in the predicted time period, for example, the base station obtains the historical total power consumption of the base station in the range of 0:00am-6:00am in the last 21 days.
And step S20, carrying out chromosome coding on each historical total power consumption to obtain a first chromosome corresponding to each historical total power consumption.
After obtaining each historical total power consumption, the device performs dyeing variable coding on each historical total power consumption to obtain a chromosome corresponding to each historical total power consumption. In this embodiment, the chromosome corresponding to the history of total power consumption is defined as the first chromosome. Specifically, the device encodes the historical total power consumption in unsigned binary, e.g., 01101.
And step S30, performing genetic iteration operation on each first chromosome to obtain the minimum total power consumption of the base station to be tested in the prediction time period.
After obtaining each first chromosome, carrying out genetic iterative operation on each chromosome, thereby obtaining the minimum total power consumption of the base station to be tested in a vehicle time period.
Specifically, determining fitness functions of each first chromosome by using a roulette strategy, selecting first chromosomes with high fitness (the fitness functions are larger than a preset threshold value and the fitness of the chromosomes is high) from each first chromosome, eliminating the first chromosomes with low fitness, intersecting the selected first chromosomes according to intersecting probability and intersecting method to obtain intersecting chromosomes, and simultaneously mutating the selected first chromosomes according to a certain mutation probability and mutation method to obtain mutation chromosomes, wherein the mutation chromosomes and the intersecting chromosomes are new generation chromosomes, so that intersecting and mutation are carried out again until the fitness functions of all the chromosomes do not meet iteration conditions, and stopping genetic iteration. The iteration condition is, for example, that the fitness function of each chromosome is larger than a set value.
After the device completes iterative genetic operation of each first chromosome, the latest target chromosomes are obtained, and the device decodes the target chromosomes to obtain the predicted total power consumption of the base station to be detected in a future predicted time period. The device selects the minimum predicted total power consumption from the predicted total power consumption, so as to obtain the minimum total power consumption of the predicted base station in the predicted time period. It should be noted that, the period to which each historical total power consumption belongs is the prediction period, for example, the device obtains 0:00am-1:00am, and the prediction period is 0:00am-1:00am.
And S40, determining the load power consumption and the base station energy consumption of the base station to be tested in the predicted time period according to the minimum total power consumption.
For the base station, the minimum energy consumption and load value of the base station group are calculated to determine whether the base station is dormant or not, and the benefit function U of the user u Is a throughput function of the user in order to achieve faster transmission rates; while the benefit function U of the base station BS The objective function is to make the whole system more energy-efficient:
U u ＝R b (x)＝wlb(1+SINR b (x))；
wherein U is BS I.e. minimum total power consumption, beta b φ b In order for the load to consume power,is the energy consumption of the base station, alpha b Is a constant.
There is a certain relation between throughput and load power consumption, and U is determined u U and U BS In the case of (2), beta can be obtained b φ b and I.e. the device can get negativeLoad power consumption and base station power consumption.
Step S50, when the load power consumption meets the first preset condition and the base station energy consumption meets the second preset condition, determining that the base station to be tested can sleep in the predicted time period.
For the user, the optimal policy satisfiesFor the base station, the optimal strategy satisfies +.>Thus, the device is based on b * Obtaining a first preset condition and based on b * K * Obtaining a second preset condition. After the device obtains the load power consumption and the base station energy consumption of the base station to be detected, judging whether the load power consumption meets a first preset condition or not, and judging whether the base station energy consumption meets a second preset condition or not. If the load power consumption meets the first preset condition and the base station energy consumption meets the second preset condition, the base station to be tested can be determined to sleep in the predicted time period.
In addition, the first preset condition may be that the load power consumption is smaller than the preset power consumption and the base station energy consumption is smaller than the preset energy consumption, that is, after the load power consumption and the base station energy consumption of the base station to be detected are obtained, whether the load power consumption is smaller than the preset power consumption or not is judged, and whether the base station energy consumption is smaller than the preset energy consumption or not is judged. If the load power consumption is smaller than the preset power consumption and the base station energy consumption is smaller than the preset energy consumption, the base station to be tested can be determined to sleep in the predicted time period.
In the technical scheme provided by the embodiment, each historical total power consumption of the history of the base station to be tested is obtained, chromosome coding is performed on each historical total power consumption to obtain a first chromosome corresponding to the historical total power consumption, genetic iterative operation is performed on each first chromosome to obtain the minimum total power consumption of the base station to be tested in a prediction time period, then the load power consumption and the base station energy consumption of the base station to be tested in the prediction time period are determined according to the minimum total power consumption, and if the load power consumption meets a first preset condition and the base station energy consumption meets a second preset condition, the base station to be tested can be determined to sleep in the prediction time period. In the invention, the load power consumption represents the energy consumption consumed by the 5G base station for the user terminal, and the base station energy consumption represents the energy consumption consumed by the 5G base station itself.
Referring to fig. 3, fig. 3 is a second embodiment of a sleep prediction method of a base station according to the present invention, based on the first embodiment, step S30 includes:
step S31, determining a first fitness function value corresponding to each first chromosome.
And S32, determining the selected probability of each first chromosome according to each first fitness function value, and determining the first dyeing gas corresponding to the selected probability larger than a preset threshold value as a target chromosome.
And step S33, determining the adaptive crossover probability and the first adaptive mutation probability of the target chromosome according to the first fitness function values.
And step S34, crossing each target chromosome according to each self-adaptive crossing probability, and mutating each target chromosome according to each first self-adaptive mutation probability to obtain each second chromosome.
Step S35, determining a second fitness function value corresponding to each second chromosome.
And S36, determining second adaptive mutation probability of each second chromosome according to each second fitness function value when each second fitness function value does not meet the iteration condition.
And step S37, determining the minimum self-adaptive variation probability in each second self-adaptive variation probability, and determining the minimum total power consumption of the base station to be tested in a preset time period according to the minimum self-adaptive variation probability.
In the present embodiment, the minimum value of the objective function value U (x) is the value, and U (x) is the minimum total workConsumption, i.e. energy consumption of base stationsAnd load beta b φ b Minimum value. The fitness function Fit (f (x) whose objective function is transformed into a chromosome is therefore:
wherein Fit (f (x)) is a fitness function, C max Represents the minimum value of U (x) generated by cutting off the current evolution degree, C max As the degree of evolution changes. The apparatus determines a first fitness function Fit (f (x)) for each first chromosome.
For example, a first chromosome f (x 1 )＝13(011010)、f(x 2 ) =3 (111110), then Fit (f (x 1 ))＝13 2 ＝169，Fit(f(x 2 ))＝3 2 =9, wherein f (x 1 ) 13 in=13 (011010) indicates the number of parameters corresponding to the historical total power consumption, that is, N parameters make the total power consumption generated by the base station to be tested be the historical total power consumption.
The population obtained by the calculation of the fitness function of the device needs to be transferred to the next generation by selecting or copying the genes of the individuals with high fitness value in the population as the parents, and the individuals with low fitness value are eliminated. Thus, the device needs to determine the target chromosome among the respective first chromosomes. Specifically, the mapping relationship between the fitness function (fitness function value) of the chromosome and the selected probability is:
p i is the probability that individual (first chromosome) i is selected, U (x) i ) Is a fitness function of chromosome, x i Representing a chromosome.
The device divides the first fitness function value of each first chromosome by the sum of the first fitness function values to obtain the selected probability of each first chromosome. The device determines the first dyeing gas corresponding to the selected probability larger than the preset threshold value as a target chromosome.
Crossover probability is a key to the performance of genetic algorithms, and has a direct impact on the convergence of genetic algorithms. The greater the crossover probability, the faster the chromosome will cross, and the faster it will produce new individuals. Therefore, the cross probability can be automatically adjusted according to the difference of fitness function values in the calculation process, and the expression is as follows:
p c is an adaptive crossover probability; f (f) max The fitness function value is the largest for individuals in the population; f (f) avg An average fitness function value for each generation of population; f is a larger fitness function value for one of the two individuals producing the crossover; k (K) 1 、K 2 The interval is a constant of (0, 1). The device can determine the self-adaptive crossover probability of each target chromosome through the formula. The device crosses each target chromosome based on the crossover probability.
In addition, the device also needs to mutate each target chromosome. Specifically, the mutation operator operation simulates the phenomenon that a natural gene is subjected to gene mutation in the genetic process, and the adaptive mutation probability is as follows:
p m is the adaptive mutation probability; k (K) 3 、K 4 Is a constant of interval (0, 1). The device is based on p m The first adaptive mutation probability for each target chromosome is determined. The device mutates the chromosomes according to the first adaptive mutation probability, and the chromosomes obtained by mutation and crossover are defined as second chromosomes.
The device then calculates a second fitness function value for each second chromosome,if the second fitness function value does not meet the iteration condition, the device calculates the second adaptive mutation probability of each second chromosome through the second fitness function value, and determines the smallest p in the second adaptive mutation probability m I.e. U (x) can be determined, thereby determining the energy consumptionAnd load beta b φ b . The iteration condition is, for example, that at least one adaptive variation probability is smaller than a preset probability.
And when each second fitness function value meets the iteration condition, the device performs genetic iteration on each second chromosome updated to the first chromosome, and returns to perform the step of determining the probability of selection of each first chromosome according to each first fitness function value, namely, iterating each updated first chromosome.
In the technical scheme provided by the embodiment, the device performs genetic iteration operation on each first chromosome based on the genetic iteration operation, so that the minimum total power consumption of the base station to be detected in a prediction time period is accurately predicted.
In one embodiment, step S30 includes:
inputting each first chromosome into a prediction model, wherein the prediction model is used for carrying out genetic iterative operation on each first chromosome;
and taking the value output by the prediction model as the minimum total power consumption of the base station to be detected in the prediction time period.
In this embodiment, the prediction model is stored in the apparatus. The apparatus inputs each chromosome into a predictive model for performing genetic iterative operations on each first chromosome. The value output by the prediction model can be used as the minimum total power consumption of the base station to be detected in the prediction time period.
When the prediction model is adopted for prediction, the prediction model needs to be trained. Specifically, the device acquires a training data set, divides each training data in the training data set into a model training set and a test set according to the ratio of 3:1, trains the model by adopting the training data in the model training set and the genetic iteration described in the embodiment, tests the model by adopting the training data in the test set after training, and if the test result reaches the preset accuracy rate of more than 95%, the trained model is available, namely the model is a prediction model. If the accuracy rate is not up to the preset accuracy rate, acquiring new training data for training. It should be noted that, with the increase of the data set, the model is continuously trained through genetic iteration optimization, so that the prediction accuracy of the prediction model is more approximate to 99%.
Referring to fig. 4, fig. 4 is a third embodiment of a sleep prediction method for a base station according to the present invention, and after step S50, further includes:
and step S60, inquiring the dynamic parameters of each 5G cell corresponding to the base station to be tested when the current time reaches the predicted time period.
And step S70, controlling the base station to be tested to sleep when each dynamic parameter is a normal parameter.
In this embodiment, when it is predicted that the base station may sleep in the predicted time period and the current time point is in the predicted time period, the device needs to query the dynamic parameters of each 5G cell of the base station to be tested. Specifically, the device inquires the state of an NR DU cell (5G cell) of the base station to be tested, if the state of the NR DU cell is a non-parameter, then inquires the dynamic parameters of all NR DU cells under the base station to be tested, and if the dynamic parameters of all NR DU cells are normal, namely the dynamic parameters are all normal parameters, the base station to be tested can be controlled to sleep, namely all 5G cells of the base station to be tested are closed. In addition, when the current time point is in the prediction time period, the device detects whether the base station to be detected has a fault alarm or not, if the base station to be detected does not have the fault alarm, the base station to be detected is controlled to sleep, if the base station to be detected does not have the fault alarm, the base station to be detected cannot sleep, and the base station to be detected can go into sleep after the fault corresponding to the fault alarm is solved.
Because the user uses the wireless network to be random and mobile, the user needs to check and open the deep sleep cell in time to provide 5G network service when the 5G user moves into the area newly when the user is closed in idle time. When the base station to be tested is dormant, the base station to be tested continuously scans the name, the station number, the user type (4G/5G) and the data updating time, and if the last data updating time and the current time are monitored to be less than 2 minutes, the 5G terminal can be determined to be registered to the network. Because the base station to be tested is in a dormant state, each 5G cell stops providing network service, so the device needs to determine the target 5G cell in which the 5G terminal is located, and then activate the network service of the target 5G cell, so that the 5G terminal is successfully registered into the network, and other 5G cells are prevented from starting the network service. The activation of the target 5G cell may be an incoming ACT NRDUCEL.
After performing an activation operation (inputting ACT NRDUCEL or transmitting an activation instruction) on the target 5G base station, the network service may not be activated, and at this time, the apparatus needs to repeatedly transmit the activation instruction to the target 5G cell. When the network service is not activated, the device acquires the activation times of the network service, wherein the activation times represent the times that the network service is not activated. The device stores preset times, if the activation times reach the preset times, a maintenance request of the target 5G cell is sent to a preset terminal, namely a maintenance work order is sent to maintenance personnel.
In the technical scheme provided by the embodiment, when the current time point is in the predicted time period, the dynamic parameters of each 5G cell corresponding to the base station to be tested are queried, if the dynamic parameters are normal parameters, the base station to be tested can be controlled to sleep, and the situation that the faults of the 5G cells are not found and maintained due to the fact that the base station enters sleep is avoided.
The present invention also provides a sleep prediction apparatus of a base station, referring to fig. 5, a sleep prediction apparatus 500 of a base station includes:
an obtaining module 510, configured to obtain each historical total power consumption of the base station to be tested;
the encoding module 520 is configured to perform chromosome encoding on each historical total power consumption to obtain a first chromosome corresponding to each historical total power consumption;
an operation module 530, configured to perform genetic iterative operations on each first chromosome, to obtain minimum total power consumption of the base station to be tested in a prediction time period;
a determining module 540, configured to determine, according to the minimum total power consumption, load power consumption and base station power consumption of the base station to be tested in a predicted time period;
the determining module 540 is configured to determine that the base station to be tested can sleep in the predicted time period when the load power consumption satisfies the first preset condition and the base station power consumption satisfies the second preset condition.
In one embodiment, the sleep prediction apparatus 500 of a base station includes:
a determining module 540, configured to determine a first fitness function value corresponding to each first chromosome;
a determining module 540, configured to determine a selected probability of each first chromosome according to each first fitness function value, and determine a first dyeing gas corresponding to the selected probability greater than a preset threshold as a target chromosome;
a determining module 540, configured to determine an adaptive crossover probability and a first adaptive mutation probability of the target chromosome according to each first fitness function value;
an operation module 530, configured to cross each target chromosome according to each adaptive cross probability, and mutate each target chromosome according to each first adaptive mutation probability, so as to obtain each second chromosome;
a determining module 540, configured to determine a second fitness function value corresponding to each second chromosome;
a determining module 540, configured to determine, when each second fitness function value does not meet the iteration condition, a second adaptive variation probability of each second chromosome according to each second fitness function value;
the determining module 540 is configured to determine a minimum adaptive variation probability among the second adaptive variation probabilities, and determine a minimum total power consumption of the base station to be tested in a preset time period according to the minimum adaptive variation probability.
In one embodiment, the sleep prediction apparatus 500 of a base station includes:
the updating module is used for updating each second chromosome into the first chromosome when each second fitness function value meets the iteration condition;
and the execution module is used for returning to the step of determining the selection probability of each first chromosome according to each first fitness function value.
In one embodiment, the sleep prediction apparatus 500 of a base station includes:
the input module is used for inputting each first chromosome into the prediction model, wherein the prediction model is used for carrying out genetic iterative operation on each first chromosome;
and the determining module 540 is configured to take the value output by the prediction model as the minimum total power consumption of the base station to be tested in the prediction time period.
In one embodiment, the sleep prediction apparatus 500 of a base station includes:
the query module is used for querying dynamic parameters of each 5G cell corresponding to the base station to be tested when the current time point is in the predicted time period;
and the control module is used for controlling the base station to be tested to sleep when each dynamic parameter is a normal parameter.
In one embodiment, the sleep prediction apparatus 500 of a base station includes:
a determining module 540, configured to determine, when detecting that the 5G terminal is registered to the network, a target 5G cell in which the 5G terminal is located, where each 5G cell stops providing network services when the base station to be tested is in a dormant state;
and the activation module is used for activating the network service of the target 5G cell so as to enable the 5G terminal to be successfully registered into the network.
In one embodiment, the sleep prediction apparatus 500 of a base station includes:
an obtaining module 510, configured to obtain, when the network service is not activated successfully, the activation times of the network service, where when the network service is not activated successfully, the activation instructions are repeatedly sent to the target 5G cell;
and the sending module is used for sending a maintenance request of the target 5G cell to a preset terminal when the activation times reach the preset times.
Fig. 6 is a hardware configuration diagram of a sleep prediction apparatus of a base station according to an exemplary embodiment.
The sleep prediction apparatus 600 of the base station may include: a process 601, such as a CPU, a memory 602, and a transceiver 603. It will be appreciated by those skilled in the art that the structure shown in fig. 6 does not constitute a limitation of the sleep prediction apparatus of the base station, and may include more or fewer components than shown, or may combine certain components, or may be a different arrangement of components. The memory 602 may be implemented by any type or combination of volatile or nonvolatile memory devices such as Static Random Access Memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic or optical disk.
Processor 601 may invoke a computer program stored in memory 602 to perform all or part of the steps of the sleep prediction method for a base station described above.
The transceiver 603 is used for receiving information transmitted from an external device and transmitting information to the external device.
A non-transitory computer readable storage medium, which when executed by a processor of a sleep prediction device of a base station, causes the sleep prediction device of the base station to perform the sleep prediction method of the base station described above.
A computer program product comprising a computer program which, when executed by a processor of a sleep prediction device of a base station, enables the sleep prediction device of the base station to perform the above-described sleep prediction method of the base station.
Other embodiments of the disclosure will be apparent to those skilled in the art from consideration of the specification and practice of the disclosure disclosed herein. This application is intended to cover any adaptations, uses, or adaptations of the disclosure following, in general, the principles of the disclosure and including such departures from the present disclosure as come within known or customary practice within the art to which the disclosure pertains. It is intended that the specification and examples be considered as exemplary only, with a true scope and spirit of the disclosure being indicated by the following claims.
It is to be understood that the present disclosure is not limited to the precise arrangements and instrumentalities shown in the drawings, and that various modifications and changes may be effected without departing from the scope thereof. The scope of the present disclosure is limited only by the appended claims.
Claims (9)
1. A method for predicting dormancy of a base station, comprising:
acquiring each historical total power consumption of a base station to be tested;
chromosome coding is carried out on each historical total power consumption to obtain a first chromosome corresponding to each historical total power consumption;
performing genetic iteration operation on each first chromosome to obtain the minimum total power consumption of the base station to be tested in a prediction time period;
determining the load power consumption and the base station energy consumption of the base station to be tested in the predicted time period according to the minimum total power consumption;
when the load power consumption meets a first preset condition and the base station energy consumption meets a second preset condition, determining that the base station to be tested can sleep in the prediction time period;
the step of performing genetic iterative operation on each first chromosome to obtain the minimum total power consumption of the base station to be detected in the prediction time period comprises the following steps:
determining a first fitness function value corresponding to each first chromosome;
determining the selected probability of each first chromosome according to each first fitness function value, and determining a first dyeing gas corresponding to the selected probability larger than a preset threshold value as a target chromosome;
determining adaptive crossover probability and first adaptive mutation probability of the target chromosome according to each first fitness function value;
crossing each target chromosome according to each self-adaptive crossing probability, and mutating each target chromosome according to each first self-adaptive mutation probability to obtain each second chromosome;
determining a second fitness function value corresponding to each second chromosome;
when each second fitness function value does not meet the iteration condition, determining second adaptive variation probability of each second chromosome according to each second fitness function value;
and determining the minimum self-adaptive variation probability in each second self-adaptive variation probability, and determining the minimum total power consumption of the base station to be tested in a preset time period according to the minimum self-adaptive variation probability.
2. The method of claim 1, wherein after the step of determining the second fitness function value corresponding to each of the second chromosomes, further comprising:
updating each second chromosome to be a first chromosome when each second fitness function value meets an iteration condition;
and returning to the step of determining the selected probability of each first chromosome according to each first fitness function value.
3. The method for predicting dormancy of a base station according to claim 1, wherein said step of performing genetic iterative operations on each of said first chromosomes to obtain a minimum total power consumption of said base station to be measured in a predicted time period comprises:
inputting each first chromosome into a prediction model, wherein the prediction model is used for carrying out genetic iterative operation on each first chromosome;
and taking the numerical value output by the prediction model as the minimum total power consumption of the base station to be detected in the prediction time period.
4. The method for predicting dormancy of a base station according to claim 1, wherein after said step of determining that said base station under test can dormancy for said predicted time period, further comprises:
inquiring dynamic parameters of each 5G cell corresponding to the base station to be tested when the current time point is in the predicted time period;
and controlling the base station to be tested to sleep when each dynamic parameter is a normal parameter.
5. The method for predicting dormancy of a base station according to claim 4, wherein after said step of controlling said base station to be dormant, further comprising:
when detecting that a 5G terminal is registered to a network, determining a target 5G cell in which the 5G terminal is located, wherein when the base station to be tested is in a dormant state, each 5G cell stops providing network services;
and activating the network service of the target 5G cell to enable the 5G terminal to be successfully registered into the network.
6. The method for predicting dormancy of a base station according to claim 5, wherein after said step of activating the network service of the target 5G cell, further comprises:
acquiring the activation times of the network service when the network service is not activated successfully, wherein when the network service is not activated successfully, an activation instruction is repeatedly sent to the target 5G cell;
and when the activation times reach preset times, sending a maintenance request of the target 5G cell to a preset terminal.
7. A sleep prediction apparatus for a base station, comprising:
the acquisition module is used for acquiring each history total power consumption of the history of the base station to be tested;
the coding module is used for carrying out chromosome coding on each historical total power consumption to obtain a first chromosome corresponding to each historical total power consumption;
the operation module is used for carrying out genetic iterative operation on each first chromosome to obtain the minimum total power consumption of the base station to be tested in the prediction time period;
the determining module is used for determining the load power consumption and the base station energy consumption of the base station to be detected in the predicted time period according to the minimum total power consumption;
the determining module is further configured to determine that the base station to be tested can sleep in the predicted time period when the load power consumption meets a first preset condition and the base station energy consumption meets a second preset condition;
the step of performing genetic iterative operation on each first chromosome to obtain the minimum total power consumption of the base station to be detected in the prediction time period comprises the following steps:
the determining module is further used for determining a first fitness function value corresponding to each first chromosome;
the determining module is further configured to determine a selected probability of each first chromosome according to each first fitness function value, and determine a first dyeing gas corresponding to the selected probability greater than a preset threshold as a target chromosome;
the determining module is further configured to determine an adaptive crossover probability and a first adaptive mutation probability of the target chromosome according to each first fitness function value;
the operation module is further configured to cross each target chromosome according to each adaptive cross probability, and mutate each target chromosome according to each first adaptive mutation probability to obtain each second chromosome;
the determining module is further used for determining a second fitness function value corresponding to each second chromosome;
the determining module is further configured to determine a second adaptive variation probability of each second chromosome according to each second fitness function value when each second fitness function value does not meet an iteration condition;
the determining module is further configured to determine a minimum adaptive variation probability among the second adaptive variation probabilities, and determine a minimum total power consumption of the base station to be tested in a preset time period according to the minimum adaptive variation probability.
8. A sleep prediction apparatus for a base station, comprising: a memory and a processor;
the memory stores computer-executable instructions;
the processor executes the computer-executable instructions stored in the memory, causing the processor to perform the sleep prediction method of a base station as claimed in any one of claims 1 to 6.
9. A computer readable storage medium, wherein computer executable instructions are stored in the computer readable storage medium, which when executed by a processor is configured to implement the method of predicting dormancy of a base station according to any one of claims 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111224197.3A CN113950134B (en) | 2021-10-20 | 2021-10-20 | Dormancy prediction method, device, equipment and computer readable storage medium of base station |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111224197.3A CN113950134B (en) | 2021-10-20 | 2021-10-20 | Dormancy prediction method, device, equipment and computer readable storage medium of base station |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113950134A CN113950134A (en) | 2022-01-18 |
| CN113950134B true CN113950134B (en) | 2023-07-25 |
Family
ID=79331926
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111224197.3A Active CN113950134B (en) | 2021-10-20 | 2021-10-20 | Dormancy prediction method, device, equipment and computer readable storage medium of base station |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113950134B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114828042A (en) * | 2022-04-07 | 2022-07-29 | 中国联合网络通信集团有限公司 | Base station system control method, device, equipment, base station system and storage medium |
| CN115426030B (en) * | 2022-09-06 | 2023-05-05 | 广州爱浦路网络技术有限公司 | Satellite energy saving method and satellite energy saving device based on big data |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106413057A (en) * | 2016-10-09 | 2017-02-15 | 北京邮电大学 | Wireless access network energy consumption optimizing method based on SDWN (Software Defined Wireless Networking) framework |
| CN108848520A (en) * | 2018-05-28 | 2018-11-20 | 西安交通大学 | A kind of base station dormancy method based on volume forecasting and base station state |
| CN112822747A (en) * | 2021-03-02 | 2021-05-18 | 吉林大学 | Routing strategy based on genetic algorithm and ant colony algorithm in wireless sensor network |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102892188B (en) * | 2012-10-09 | 2016-07-06 | 中兴通讯股份有限公司 | Based on the ascending power control method of genetic algorithm and device in communication network |
| US9980218B2 (en) * | 2015-02-27 | 2018-05-22 | Huawei Technologies Canada Co., Ltd. | System and method for user terminal-aware cell switch-off |
| CN107493584A (en) * | 2017-07-31 | 2017-12-19 | 清华大学 | A kind of method and apparatus for determining base station dormancy scheme |
-
2021
- 2021-10-20 CN CN202111224197.3A patent/CN113950134B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106413057A (en) * | 2016-10-09 | 2017-02-15 | 北京邮电大学 | Wireless access network energy consumption optimizing method based on SDWN (Software Defined Wireless Networking) framework |
| CN108848520A (en) * | 2018-05-28 | 2018-11-20 | 西安交通大学 | A kind of base station dormancy method based on volume forecasting and base station state |
| CN112822747A (en) * | 2021-03-02 | 2021-05-18 | 吉林大学 | Routing strategy based on genetic algorithm and ant colony algorithm in wireless sensor network |
Non-Patent Citations (1)
| Title |
|---|
| 5G网络技术对提升4G网络性能的研究;刘奕;;数码世界(第04期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113950134A (en) | 2022-01-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Li et al. | Multi-agent deep reinforcement learning based spectrum allocation for D2D underlay communications | |
| Zhao et al. | Deep reinforcement learning for user association and resource allocation in heterogeneous cellular networks | |
| US10945145B2 (en) | Network reconfiguration using genetic algorithm-based predictive models | |
| Yang et al. | Learning-based energy-efficient resource management by heterogeneous RF/VLC for ultra-reliable low-latency industrial IoT networks | |
| CN113950134B (en) | Dormancy prediction method, device, equipment and computer readable storage medium of base station | |
| Razavi et al. | Self-optimization of capacity and coverage in LTE networks using a fuzzy reinforcement learning approach | |
| CN111641681A (en) | Internet of things service unloading decision method based on edge calculation and deep reinforcement learning | |
| CN103906076B (en) | A kind of method that distributed self-adaption adjusts small base station transmitting power bias | |
| Lee et al. | Affinity propagation for energy-efficient BS operations in green cellular networks | |
| WO2014173316A1 (en) | Coverage and capacity joint optimization method, device, and system | |
| Shodamola et al. | A machine learning based framework for KPI maximization in emerging networks using mobility parameters | |
| CN115066006A (en) | Base station dormancy method, equipment and medium based on reinforcement learning | |
| WO2021083230A1 (en) | Power adjusting method and access network device | |
| Sharma et al. | Multiobjective reinforcement learning based energy consumption in C-RAN enabled massive MIMO | |
| Zhi et al. | AP deployment optimization in non-uniform service areas: A genetic algorithm approach | |
| CN106954268A (en) | Access network resource distribution method under a kind of SDN frameworks | |
| CN107995034B (en) | Energy and service cooperation method for dense cellular network | |
| Wang et al. | Evolutionary game based heterogeneous wireless network selection with multiple traffics in 5G | |
| CN109660995A (en) | AP intelligent power saving control method and equipment in WiFi system based on machine learning algorithm | |
| Li et al. | Qos-aware mobile service optimization in multi-access mobile edge computing environments | |
| Chincoli et al. | Predictive power control in wireless sensor networks | |
| CN113938978A (en) | Heterogeneous wireless sensor path finding method based on reinforcement learning | |
| Han et al. | Power Control Research for Device-to-Device Wireless Network Underlying Reinforcement Learning | |
| CN112584386A (en) | 5G C-RAN resource prediction and allocation method and system | |
| Xu et al. | Energy efficiency and delay determinacy tradeoff in energy harvesting-powered zero-touch deterministic industrial M2M communications |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |