5G base station intelligent dormancy and wake-up energy saving method based on edge computing

Abstract

The invention relates to the technical field of information communication, in particular to a 5G base station intelligent dormancy and wake-up energy-saving method based on edge computing, which is applied to a competition base station and an arbitration base station in a distributed base station cluster. And the competition base station judges whether to participate in competition based on the service load, the channel quality and the energy consumption information, and broadcasts competition information to the distributed base station cluster. And after receiving the competition information, the arbitration base station searches the alternative base stations with similar historical service modes in the arbitration area of the arbitration base station, ranks the base stations according to idle loads, and decides whether to vote to support the corresponding competition base stations or not based on the sequence positions of the alternative base stations. And after each arbitration base station votes, summarizing the competition base station with the most votes as a service base station, and broadcasting the service base station to the base station cluster so as to adjust the operation state of the base station. Unnecessary wakeup and energy consumption of the base station are effectively reduced, the energy efficiency of the 5G network is improved, the service quality is ensured, and intelligent energy-saving optimization is realized.

Description

Technical Field

[0001] The present invention relates to the field of information and communication technologies, and particularly to an intelligent sleep and wake-up energy-saving method for 5G base stations based on edge computing. Background Art

[0002] With the wide deployment of 5G networks, mobile communication operators are facing unprecedented energy consumption challenges. Since 5G base stations adopt large-scale antenna arrays, millimeter-wave communications, and higher spectral bandwidths, the power consumption per unit base station is much higher than that of traditional 4G base stations. Practical tests show that the power consumption of a single 5G base station is about 3-4 times that of a 4G base station, and even reaches 3.7-3.9 kW under full load, resulting in a significant increase in the proportion of the power expenditure of the communication network in the operating cost (OPEX) of the operator.

[0003] For example, the Chinese patent application with the publication number CN115604793A provides a method and system for waking up a sleeping base station. The method includes: the base station receives a signal to be detected in the sleeping state and determines the signal-to-interference-plus-noise ratio of the signal to be detected; wherein, the base station receiver is turned on in the sleeping state; based on the signal-to-interference-plus-noise ratio of the signal to be detected, it is determined that the base station receives a signal sent by a user, and the base station is switched to the wake-up state. It can accurately identify whether the signal received by the sleeping base station is a signal sent by a user, reduce the influence of interference and noise on the identification result, and wake up the sleeping base station in time when it is determined that a signal sent by a user is received.

[0004] The above existing technologies all have the problems raised in this background art: inefficient resource allocation leads to frequent state switching of some base stations, increasing additional energy consumption. To solve the above problems, the present application designs an intelligent sleep and wake-up energy-saving method for 5G base stations based on edge computing. Summary of the Invention

[0005] The technical problem to be solved by the present invention is to provide, in view of the deficiencies of the prior art, an intelligent sleep and wake-up energy-saving method for 5G base stations based on edge computing, which is applied to competing base stations and arbitration base stations in a distributed base station cluster. The competing base stations judge whether to participate in the competition based on service load, channel quality, and energy consumption information, and broadcast competition information to the distributed base station cluster. After receiving the competition information, the arbitration base stations search for alternative base stations with similar historical service patterns within their own arbitration areas, sort the base stations according to the idle load, and decide whether to vote in support of the corresponding competing base stations based on the sequence positions of the alternative base stations. After each arbitration base station votes, the competing base station with the most votes is summarized as the serving base station and broadcast to the base station cluster to adjust the operating state of the base stations. It effectively reduces unnecessary wake-up and energy consumption of the base stations, improves the energy efficiency of the 5G network, ensures service quality, and realizes intelligent energy-saving optimization.

[0006] To achieve the above object, the present invention provides the following technical solutions: A 5G base station intelligent sleep and wake-up energy-saving method based on edge computing is applied to competing base stations in a distributed base station cluster. The distributed base station cluster further includes a plurality of arbitration base stations, and each arbitration base station is one of the base stations in the distributed base station cluster that are not currently participating in the competition. Each arbitration base station corresponds to an arbitration area composed of a plurality of base stations. Each base station is configured with an edge computing node. The 5G base station intelligent sleep and wake-up energy-saving method includes: Receiving a network response request of a user terminal broadcast by the distributed base station cluster; Generating competition information according to competition metrics, and broadcasting the competition information and historical processing information to the distributed base station cluster, so that each arbitration base station votes for a serving base station based on the received competition information and historical processing information of the competing base stations; If the competing base station is selected as the serving base station, switch its own state to or maintain the wake-up state.

[0007] The competition metrics include service load, channel quality, and base station energy consumption. The generating competition information according to the competition metrics includes: Obtaining the current service load, where the current service load includes PRB occupancy rate, number of user connections, and throughput, and calculating a service load index according to a preset weight; Obtaining channel quality metrics, where the channel quality metrics include RSRP, SINR, and path loss, and calculating a channel quality level according to a channel quality threshold; Obtaining the current power consumption, energy efficiency per bit, and switching energy consumption overhead, and calculating a base station energy consumption index; Judging whether it is necessary to compete for the network response request according to the service load index, channel quality level, and base station energy consumption index. If necessary, constructing competition information. If not, broadcasting base station identification information to the distributed base station cluster to serve as an arbitration base station.

[0008] Judging whether it is necessary to compete for the network response request according to the service load index, channel quality level, and base station energy consumption index includes: Comparing the service load index with a first threshold to judge whether the current base station service load is higher than the first threshold, where the first threshold is adjusted by the edge computing node according to historical load data; If the service load index is higher than the first threshold, it is determined that it is necessary to compete for the response request; If the service load index is less than or equal to the first threshold, further compare the channel quality level with a second threshold. When the channel quality level is greater than or equal to the second threshold, further compare the base station energy consumption index with a third threshold; otherwise, abandon the competition. The second threshold is determined by all arbitration base stations in the previous round based on the average received signal quality reported by the user terminal. If the base station energy consumption index is less than the third threshold, it is determined that the response request needs to be competed for; otherwise, abandon the competition.

[0009] A 5G base station intelligent sleep and wake-up energy-saving method based on edge computing is applied to an arbitration base station in a distributed base station cluster. Each arbitration base station corresponds to an arbitration area composed of multiple base stations. The distributed base station cluster further includes multiple competing base stations. Each arbitration base station is one of the base stations in the distributed base station cluster that are not currently participating in the competition. Each base station is configured with an edge computing node. The 5G base station intelligent sleep and wake-up energy-saving method includes: Receive the competition information and historical processing information sent by each competing base station that is not within its own arbitration area. Vote for a serving base station according to the competition information and historical processing information corresponding to each competing base station. Broadcast the base station identifier of the serving base station to the distributed base station cluster, so that the corresponding competing base stations can switch or maintain their own states to the wake-up state.

[0010] The voting for a serving base station according to the competition information and historical processing information corresponding to each competing base station includes: For each competing base station, determine whether there is a corresponding alternative base station within its own arbitration area according to the competition information and historical processing information corresponding to each competing base station. Sort all the base stations within its own arbitration area in descending order of idle load to obtain an idle load sequence. For each competing base station with an alternative base station, determine the sequence position of the corresponding alternative base station. If the sequence position of the corresponding alternative base station is after the set position, confirm to vote for this competing base station. Receive the voting votes of each competing base station sent by other arbitration base stations, and connect with the voting votes of each competing node determined by itself to summarize and determine the serving base station.

[0011] The determination of whether there is a corresponding alternative base station within its own arbitration area includes: Calculate the similarity between the historical processing information of this competing base station and the historical processing information of each base station within its own arbitration area according to the historical processing information of the competing base station. Select the base station with the highest similarity to the historical processing information of each competing base station within its own arbitration area as the corresponding alternative base station according to the similarity.

[0012] The summarization determines a serving base station, including: Each arbitration base station generates voting information according to the voting result determined by itself, and the voting information includes the identifiers of each competing base station that has received the vote of this arbitration base station; Broadcast its own voting information to the distributed base station cluster; Receive the voting information broadcast by other arbitration base stations, and combine its own voting information to count the number of votes obtained by each competing base station; Take the competing base station with the most votes as the serving base station.

[0013] When multiple competing base stations obtain the same highest number of votes, the summarization to determine the serving base station further includes: Obtain the service load index, channel quality level, and base station energy consumption index in the competition information; Successively execute the preference rules until a unique serving base station is determined.

[0014] The successive execution of the preference rules until a unique serving base station is determined includes: Compare the service load indexes of competing base stations with the same number of votes, and select the base station with the highest service load index as the serving base station; If the service load indexes are the same, then compare the channel quality levels, and select the base station with the highest channel quality level as the serving base station; If the channel quality levels are the same, then compare the base station energy consumption indexes, and select the base station with the lowest base station energy consumption index as the serving base station; If the base station energy consumption indexes are the same, then select the base station with the highest historical user connection success rate as the serving base station; If the historical user connection success rates are the same, then select the base station with the shortest response delay from the last wake-up from the dormant state to the active state as the serving base station.

[0015] When the number of arbitration base stations does not meet the preset base threshold, the 5G base station intelligent sleep and wake-up energy-saving method further includes: According to the competition information of each competing base station, perform a reverse order sorting on the competition information to generate a reverse order list; Sequentially convert the competing base stations in the reverse order list into arbitration base stations until the number of arbitration base stations meets the preset base threshold.

[0016] Compared with the prior art, the beneficial effects of the present invention are: Based on the similarity of historical business models, the present invention matches a corresponding alternative base station for each competing base station within the arbitration area, and determines whether to vote in support of the competing base station in combination with the sorted position of the idle load of the alternative base station. Compared with the traditional arbitration method that only relies on static competition information, the present invention can more accurately evaluate the suitability of the competing base station for the services in the arbitration area, ensuring that the selected serving base station can not only meet the service requirements but also optimize resource scheduling. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Other features, objects, and advantages of the present invention will become more apparent by reading the following detailed description of non-limiting embodiments with reference to the accompanying drawings: Figure 1 FIG. is a schematic flowchart of a method for intelligent sleep and wake-up energy saving of 5G base stations based on edge computing applied to competing base stations in Embodiment 1 of the present invention; Figure 2 FIG. is a schematic flowchart of a method for intelligent sleep and wake-up energy saving of 5G base stations based on edge computing applied to arbitration base stations in Embodiment 2 of the present invention; Figure 3 FIG. is a schematic diagram of the arbitration area in Embodiment 2 of the present invention; Figure 4 FIG. is a schematic diagram of arbitration voting in Embodiment 2 of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0018] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments.

[0019] Embodiment 1 Please refer to Figure 1 , an embodiment provided by the present invention: A method for intelligent sleep and wake-up energy saving of 5G base stations based on edge computing applied to competing base stations, which is applied to competing base stations in a distributed base station cluster. The distributed base station cluster further includes a plurality of arbitration base stations, each arbitration base station is one of the base stations in the distributed base station cluster that are not currently participating in the competition, each arbitration base station corresponds to an arbitration area composed of a plurality of base stations, and each base station is configured with an edge computing node. The method for intelligent sleep and wake-up energy saving of 5G base stations includes: S1: Receive the network response request of the user terminal broadcast by the distributed base station cluster; In this embodiment, each competing base station periodically monitors the network status broadcast of the distributed base station cluster, which includes the network response request of the user terminal. This network response request can be a service connection request initiated by the user, such as a data uplink or downlink request, or a service optimization signaling sent by the network itself, such as an adjustment request triggered by the network side detecting a sudden increase in local traffic.

[0020] Specifically, each competing base station can timely obtain the current network status and provide a basis for subsequent competition decisions, ensuring that the selected serving base station can meet user needs and avoiding unnecessary base station wake-up or sleep failures. Compared with the traditional fixed time interval strategy, it can more accurately reflect the actual situation of network load and improve the scheduling efficiency of network resources.

[0021] S2: Generate competition information according to competition metrics; In this embodiment, after each competing base station receives a network response request, it calculates competition information based on competition metrics such as current traffic load, channel quality, and base station energy consumption, and establishes a fair competition mechanism among multiple base stations to ensure that among multiple candidate base stations, a base station with strong service carrying capacity, good channel quality, and low energy consumption can be selected to provide services. Compared with the traditional method of making decisions only based on traffic load or channel quality alone, it comprehensively considers three factors: load balancing, signal coverage, and energy saving, which can effectively improve the scientific nature of resource scheduling and reduce energy consumption at the same time.

[0022] S3: Broadcast the competition information and historical processing information to the distributed base station cluster so that each arbitration base station votes for the serving base station based on the received competition information and historical processing information of the competing base stations; In this embodiment, after the competing base station generates competition information, it broadcasts it together with the historical processing information to the distributed base station cluster for the arbitration base station to vote. The historical processing information mainly includes: Historical user connection success rate: Measuring the stability of the base station in providing services in the past; Historical average throughput: Used to evaluate the carrying capacity of the base station under high traffic load; Historical channel quality fluctuation: Reflecting the long-term stability of the base station channel environment.

[0023] Furthermore, after receiving the competition information and historical processing information of all competing base stations, the arbitration base station votes to select the serving base station based on the intelligent voting mechanism. The voting process includes: Calculate the similarity of the historical processing information between the competing base station and the base stations within its arbitration area, and select the base station with the highest similarity as the alternative base station.

[0024] Sort in descending order according to the idle load, and decide whether to vote for the competing base station according to the ranking position of the alternative base station.

[0025] After completing the independent voting, it broadcasts the voting results to the entire base station cluster and finally determines the competing base station with the most votes as the final serving base station.

[0026] S4: If selected as the serving base station, switch or maintain its own state to the wake-up state; In this embodiment, the selected serving base station will perform state switching, including waking up from the sleep state and maintaining the current wake-up state, which is specifically determined according to whether the base station is active. For other unselected competing base stations, only their current states need to be maintained, avoiding the long-term high energy consumption problem caused by the fixed operation mode of traditional base stations. At the same time, unnecessary wake-up operations are also prevented, improving the overall energy efficiency of the base stations.

[0027] The specific steps of S2 are as follows: S2.1: Obtain the current traffic load, where the current traffic load includes the PRB occupancy rate, the number of user connections, and the throughput, and calculate the traffic load index according to the preset weights; Specifically, in different scenarios, the measurement criteria for traffic load should be different. Relying solely on a single indicator (such as the PRB occupancy rate) is likely to lead to misjudgment. For example, in some scenarios, although the PRB occupancy rate is high, the number of user connections is small, indicating that the resource utilization efficiency of the base station is low and it should not be directly regarded as a high load. Therefore, by calculating the traffic load index through weighting, the actual traffic pressure of the base station can be more comprehensively reflected, providing a more reasonable basis for competition decisions.

[0028] In this embodiment, competing base stations need to first obtain the current traffic load information to determine their own network resource usage and calculate the traffic load index. The traffic load information includes the PRB (Physical Resource Block) occupancy rate, the number of user connections, and the throughput. Among them, the PRB occupancy rate reflects the usage of the available radio resources of the base station, the number of user connections represents the number of users served by the current base station, and the throughput measures the overall data processing ability of the base station.

[0029] Furthermore, the base station will perform weighted summation on these three key parameters according to the preset weights to generate the traffic load index. The setting of the weights is dynamically adjusted according to factors such as the base station's historical traffic pattern, geographical location, and service type. For example, for base stations in hot spots, the throughput weight is higher, while for base stations with a wide coverage but few connected users, the weight of the PRB occupancy rate is higher.

[0030] S2.2: Obtain the channel quality indicators, where the channel quality indicators include RSRP, SINR, and path loss, and calculate the channel quality level according to the channel quality threshold; Specifically, relying solely on a single channel quality parameter (such as RSRP) may not accurately reflect the communication ability of the base station. For example, in a dense cell deployment environment, the RSRP may be high, but due to severe interference from neighboring cells, the SINR may be low, resulting in poor actual communication quality. Therefore, by integrating multiple channel quality metrics and combining them with channel quality thresholds for grading, the service ability of the base station can be more accurately reflected, thereby avoiding selecting a base station with poor channel quality as the serving base station and improving the overall network performance.

[0031] In this embodiment, the base station first obtains its own RSRP and SINR through a periodic measurement and reporting mechanism, where the RSRP is used to measure the received power of the reference signal sent by the base station at the user end, and the SINR is used to measure the interference situation of the wireless channel. The base station also calculates the overall channel attenuation by combining a path loss model (based on the wireless propagation environment, such as urban, suburban, or indoor scenarios). The measured values are smoothed over a period of time to ensure the stability of the channel quality assessment.

[0032] Furthermore, when calculating the channel quality level, the base station compares the measured RSRP, SINR, and path loss with preset channel quality thresholds to determine the channel quality level.

[0033] The base station matches the RSRP value with a preset RSRP threshold table to determine the RSRP level: RSRP > -80 dBm, denoted as RSRP_level = 5 (excellent); -90 dBm ≤ RSRP ≤ -80 dBm, denoted as RSRP_level = 4 (good); -100 dBm ≤ RSRP ≤ -90 dBm, denoted as RSRP_level = 3 (average); -110 dBm ≤ RSRP ≤ -100 dBm, denoted as RSRP_level = 2 (poor); RSRP < -110 dBm, denoted as RSRP_level = 1 (extremely poor).

[0034] The base station matches the SINR value with a preset SINR threshold table to determine the SINR level: SINR > 20 dB, denoted as SINR_level = 5 (excellent); 10 dB ≤ SINR ≤ 20 dB, denoted as SINR_level = 4 (good); 5 dB ≤ SINR ≤ 10 dB, denoted as SINR_level = 3 (average); 0 dB ≤ SINR ≤ 5 dB, denoted as SINR_Level = 2 (poor); SINR < 0 dB, denoted as SINR_Level = 1 (extremely poor).

[0035] After the base station calculates the path loss, it is matched with the preset path loss threshold table: Path loss < 90 dB, denoted as path loss_Level = 5 (excellent); 90 dB ≤ path loss ≤ 100 dB, denoted as path loss_Level = 4 (good); 100 dB ≤ path loss ≤ 110 dB, denoted as path loss_Level = 3 (average); 110 dB ≤ path loss ≤ 120 dB, denoted as path loss_Level = 2 (poor); path loss > 120 dB, denoted as path loss_Level = 1 (extremely poor).

[0036] Furthermore, based on the weighted calculation method, the RSRP_Level, SINR_Level, and path loss_Level are weighted and summed to calculate the final channel quality level.

[0037] S2.3: Obtain the current power consumption, energy efficiency per bit, and switching energy consumption overhead, and calculate the base station energy consumption index; In this embodiment, competing base stations need to evaluate their own energy consumption situations in order to select a base station with higher energy efficiency as the serving base station while ensuring service quality. The energy consumption calculation of the base station mainly involves three aspects: current power consumption, energy efficiency per bit, and switching energy consumption overhead.

[0038] Furthermore, the current power consumption of the base station is comprehensively determined by the power consumption of components such as the power amplifier (PA), baseband processing unit (BBU), and cooling system of the base station, and is usually measured in real time through a built-in power management module.

[0039] Furthermore, the energy efficiency per bit is an indicator to measure the energy efficiency of the base station. The calculation method is the ratio of the total throughput of the base station to the total power consumption within a unit time. The higher this value, the more data the base station can transmit under unit power consumption, and the higher the energy efficiency. Through the adaptive energy efficiency analysis algorithm and combined with historical service data, the current energy efficiency per bit is calculated to determine whether it has the ability of efficient energy consumption management.

[0040] Furthermore, the switching energy consumption overhead mainly refers to the energy loss during the process of the base station waking up from the sleep state to the normal working state. The base station predicts whether its energy consumption meets the energy-saving optimization goal if it is selected as the serving base station currently by statistically analyzing the average power consumption and wake-up time of the recent several wake-up operations.

[0041] S2.4: Determine whether to compete for the network response request based on the service load index, channel quality level, and base station energy consumption index. If so, construct competition information; if not, broadcast the base station identification information to the distributed base station cluster to serve as the arbitration base station. In this embodiment, after obtaining the service load index, channel quality level, and base station energy consumption index, the competing base station needs to comprehensively determine whether it is suitable to compete for the network response request, and construct competition information or convert to an arbitration base station according to the decision result. The competition information includes the service load index, channel quality level, and base station energy consumption index. The construction method of the competition information adopts a standardized coding format to reduce the amount of broadcast data and improve the information transmission efficiency. In addition, the competition information will be protected by a security mechanism to prevent information tampering and improve the security and reliability of the system.

[0042] Furthermore, the arbitration base station is randomly selected and dynamically adjusted according to a preset base threshold.

[0043] Specifically, after the competition decision is completed, the base stations that did not participate in the competition will be randomly selected as arbitration base stations. The number of selected base stations is restricted by the base threshold to ensure that the number of arbitration base stations can meet the voting requirements. The minimum number of the base threshold is set by the arbitration mechanism in the distributed base station cluster and is usually adjusted based on historical network data. If the number of arbitration base stations is insufficient, they will be supplemented from the competing base stations in the order of the lowest service load priority.

[0044] The specific steps of S2.4 are as follows: S2.4.1: Compare the service load index with the first threshold to determine whether the current base station service load is higher than the first threshold, where the first threshold is adjusted by the edge computing node according to historical load data. In this embodiment, after each competing base station receives the network response request from the user terminal, it first needs to evaluate its own service load level to determine whether there are sufficient resources to participate in the competition. The service load index is an important parameter for measuring the current resource usage of the base station, and its calculation is based on key indicators such as PRB (Physical Resource Block) occupancy rate, number of user connections, and throughput. To make the judgment of the service load index more accurate, the base station does not use a fixed threshold, but dynamically adjusts it based on historical load data, and the edge computing node is responsible for calculating the first threshold.

[0045] Specifically, the edge computing node will regularly count the historical load situation of the base station, including the average service load index within a certain past time window and the load situations of other base stations in the same geographical area, to ensure that the first threshold can adapt to different service models.

[0046] Exemplarily, during peak hours, the edge computing node will increase the first threshold to ensure that only truly high-load base stations participate in the competition; while during low-traffic hours, the first threshold will be decreased to encourage more base stations to participate in the competition and improve the network resource utilization rate. Adapt to different traffic scenarios, avoid the competition imbalance problem caused by the fixed threshold strategy, reduce unnecessary competition, and optimize the overall network energy efficiency.

[0047] S2.4.2: If the service load index is higher than the first threshold, it is determined that the response request needs to be competed for. In this embodiment, if the service load index exceeds the first threshold, it indicates that the base station has a strong service carrying capacity and a large traffic demand. Therefore, this base station should actively compete to become the serving base station. In this case, the base station will immediately construct competition information and broadcast it to the distributed base station cluster to participate in the selection process of the serving base station. The competition information includes key parameters such as the current service load index, channel quality level, and base station energy consumption index of the base station for the arbitration base station to evaluate. Ensure that high-load base stations participate first, so as to ensure that service demands are met in a timely manner, reduce resource waste, and improve the overall service ability of the 5G network. In addition, by preferentially selecting high-service-load base stations as competitors, the spectrum resource utilization rate can also be improved, the energy consumption overhead of low-load base stations can be reduced, and the network can still maintain a high operating efficiency under high-load conditions.

[0048] S2.4.3: If the service load index is less than or equal to the first threshold, the channel quality level is further compared with the second threshold. When the channel quality level is greater than or equal to the second threshold, the base station energy consumption index is further compared with the third threshold, otherwise the competition is abandoned, where the second threshold is determined by all arbitration base stations in the previous round based on the average received signal quality reported by the user terminals. In this embodiment, if the service load index of the base station does not exceed the first threshold, it does not necessarily mean that the base station cannot compete, but rather the channel quality needs to be further evaluated for excellence. The arbitration base stations in the previous round calculate based on the received signal quality reported by all user terminals to determine the second threshold in the current environment.

[0049] Exemplarily, in an environment with high wireless interference, the arbitration base station will increase the second threshold to ensure that only base stations with excellent channel quality can participate in the competition; while in an environment with relatively abundant wireless resources, the second threshold is appropriately decreased to encourage more base stations to compete and improve the network coverage ability.

[0050] Specifically, after receiving the second threshold, the base station will compare its own channel quality level with this threshold. If its channel quality level is greater than or equal to the second threshold, the base station still has the qualification to compete and enters the next energy consumption screening; otherwise, the base station will give up the competition and turn to the candidate sequence of the arbitration base station. Ensure that the selected serving base station has good wireless channel quality, thereby improving the user experience, reducing network congestion and disconnection risks. Make the network resource allocation more reasonable and improve the overall communication performance.

[0051] S2.4.4: If the energy consumption index of the base station is less than the third threshold, it is determined that the response request needs to be competed for; otherwise, give up the competition; In this embodiment, if the service load index of the base station is less than or equal to the first threshold, but the channel quality level is higher than the second threshold, the energy consumption situation of the base station also needs to be considered to ensure the energy saving of network operation.

[0052] Specifically, the current power consumption reflects the current operating energy consumption of the base station, the energy efficiency per bit measures the service transmission ability of the base station under unit energy consumption, and the switching energy consumption overhead evaluates the additional energy consumption of the base station from sleep to operation. The base station will compare the calculated energy consumption index with the third threshold, which is set by analyzing historical energy consumption data and the characteristics of the base station equipment.

[0053] Exemplarily, for remote base stations with weak power grid supply capabilities, the third threshold is set lower to limit the frequent wake-up of high-energy-consuming base stations; while for base stations in urban areas, due to sufficient power resources, the third threshold is set higher to give priority to service requirements.

[0054] Furthermore, when the energy consumption index of the base station is lower than the third threshold, it indicates that the base station has high energy efficiency, so it can participate in the competition and broadcast competition information; on the contrary, if the energy consumption index of the base station is greater than or equal to the third threshold, even if the service load and channel quality meet the requirements, it will not participate in the competition, but choose to give up the competition and turn to the arbitration base station. This screening mechanism ensures finding the best balance between service requirements and energy efficiency, avoiding increased power consumption due to the frequent competition of high-energy-consuming base stations, while ensuring the network service quality. Adapt to different power supply situations, improve the energy-saving effect of 5G base stations, reduce operating costs, and achieve more efficient base station resource scheduling.

[0055] Embodiment 2 Please refer to Figure 2, the present invention provides an embodiment: a 5G base station intelligent sleep and wake-up energy-saving method based on edge computing applied to an arbitration base station, which is applied to an arbitration base station in a distributed base station cluster. Each arbitration base station corresponds to an arbitration area composed of multiple base stations. The distributed base station cluster further includes multiple competing base stations. Each arbitration base station is one of the base stations that are currently not participating in the competition in the distributed base station cluster. Each base station is configured with an edge computing node. The specific steps of the 5G base station intelligent sleep and wake-up energy-saving method are as follows: A1: Receive the competition information and historical processing information sent by each competing base station not within its own arbitration area; In this embodiment, after each arbitration base station enters the arbitration state, it will listen to and receive the competition information and historical processing information sent by all competing base stations not in its arbitration area. The competition information includes the service load index, channel quality level, and base station energy consumption index, which are used to measure the current operating state and service capabilities of the competing base stations; the historical processing information includes past service records, historical user connection success rates, and historical average throughputs, which are used to evaluate the long-term stability and reliability of the competing base stations.

[0056] A2: Vote for a serving base station according to the competition information and historical processing information corresponding to each competing base station; In this embodiment, after the arbitration base station receives the competition information and historical processing information of the competing base stations, it will select the optimal serving base station based on an intelligent voting mechanism. The voting process includes multiple stages to ensure that the finally selected base station can meet the service requirements and achieve the energy-saving goal. Compared with the traditional fixed strategy, the present application can adapt to different network load conditions. By analyzing historical information + intelligent sorting and voting, the selected serving base station can meet the service requirements and effectively save energy. The introduction of the alternative base station strategy can reduce unnecessary base station wake-up, thereby further reducing the overall network energy consumption and improving the energy-saving benefit of the 5G network.

[0057] A3: Broadcast the base station identifier of the serving base station to the distributed base station cluster, so that the corresponding competing base stations switch or maintain their own states to the wake-up state; In this embodiment, after the arbitration base station completes the voting, it will broadcast the finally selected service base station identifier to the entire distributed base station cluster to notify all competing base stations of the voting results. For the selected service base station, it will switch or remain in the wake-up state to continue providing services to users, while the unselected competing base stations will maintain their current states. To ensure the smoothness of the base station state transition, the broadcast voting results also include a load transfer strategy to ensure that users can seamlessly migrate to the target base station during the base station handover process. For example, if a base station switches from the sleep state to the active state, the surrounding cells will adjust the load balancing parameters to migrate some user traffic to this base station to make full use of its resources and improve the overall network efficiency. In addition, during the broadcast process, the arbitration base station will adopt a differential broadcast mechanism to send status change information only to the affected base stations instead of broadcasting it to the entire network, thereby reducing communication overhead and improving network efficiency.

[0058] Specifically, although the prior art can perform simple voting based on the static information reported by competing base stations to achieve the sleep and wake-up decisions of 5G base stations, there are still obvious deficiencies in practical applications. For example, after receiving the information of competing base stations outside the region, the existing arbitration base stations often lack effective means to evaluate the necessity of such competing base stations in the actual service scenario, especially unable to accurately judge whether the base stations outside the region can solve the actual service bearing requirements within the arbitration region, resulting in possible incorrect voting decisions when the service load fluctuates greatly or resources are tense. This method not only easily causes unreasonable allocation of base station resources but also leads to frequent wake-up and sleep of base stations, seriously affecting the network energy efficiency and service quality.

[0059] Furthermore, in this embodiment, an intelligent voting method based on the "idle load sequence position of the alternative base station" is proposed to solve the above problems. Specifically, when the arbitration base station conducts voting, it first uses historical processing information and competition information to find alternative base stations with similar historical service performance inside the arbitration region for each competing base station outside the region; then, it decides whether to vote for the corresponding competing base station according to the position of the alternative base station in the idle load sequence of the base stations in this region. For example, when the historical processing information of a competing base station A is highly similar to that of base station B in the arbitration region, base station B will be selected as the alternative base station for base station A. After the arbitration base station sorts the idle loads of all base stations in this region, if the idle load ranking of the alternative base station B is relatively low (that is, the resources of the alternative base station itself are relatively tense or busy), it means that if the competing base station A is selected as the service base station, it can effectively relieve the service load pressure in this region, so it is confirmed to vote for the competing base station A; on the contrary, if the idle load ranking of the alternative base station B is relatively high (that is, the idle resources of the alternative base station itself are relatively sufficient), it means that there are already base stations inside the arbitration region that can undertake the service requirements of this region, and there is no need to introduce additional external competing base stations to share the services, so it does not vote for the competing base station A.

[0060] Please refer to Figure 3 and Figure 4 For example, assume that there is a business peak within a certain arbitration area currently, and there are competing base stations X and Y outside the area participating in the competition. Since the traditional arbitration method only votes based on the static indicators of the base stations, it may wrongly ignore the actual service value of base stations X and Y, or wrongly select an inappropriate base station to wake up. In this embodiment, the arbitration base station first determines the base station M most similar to base station X and the base station N most similar to base station Y in the area as substitute base stations according to historical data, and then analyzes the idle load ranking of base stations N and M. If the ranking of base station N in the idle load sequence within the area is after the set position, it indicates that the existing base station resources in the area are strained. If base station X is selected, the business pressure can be effectively shared. Therefore, the arbitration base station confirms to vote for competing base station X. On the contrary, if the idle load ranking of substitute base station M is before the set position, it means that there are sufficient base station resources in this area to meet the business requirements, and the participation of base station Y at this time will not bring obvious benefits but will cause additional energy consumption. Therefore, the arbitration base station will reasonably give up voting for base station Y.

[0061] Obviously, the method based on the position of the idle load sequence of the substitute base station proposed in this embodiment effectively solves the technical defect that the existing arbitration decision only relies on static competition information and cannot accurately evaluate the business adaptability of base stations outside the area. Through this method, the arbitration base station can accurately judge the actual value of the participation of base stations outside the area in the business bearing within the area, so as to accurately control the wake-up and sleep behaviors of the base stations, avoid resource waste and energy consumption increase, and effectively improve the network service quality and energy efficiency level.

[0062] The specific steps of A2 are as follows: A2.1: For each competing base station, determine whether there is a corresponding substitute base station in its own arbitration area according to the competition information and historical processing information corresponding to each competing base station; In this embodiment, since the competing base station itself is located outside the area, it is difficult for the arbitration base station to directly judge the actual performance of its service ability in this area. By finding substitute base stations similar in historical data, the true business performance of the competing base station can be effectively inferred, so as to more accurately judge whether the business in this area should be assigned to the competing base station, and improve the accuracy and rationality of the selection of service base stations.

[0063] Specifically, the arbitration base station first obtains the historical processing information provided by the competing base station, including but not limited to key indicators such as the business processing mode, user connection success rate, network load trend, and response delay in the past period of time; Furthermore, the arbitration base station utilizes the historical service data of all base stations within its own area stored in its edge computing nodes, and through intelligent analysis methods, such as cosine similarity, Euclidean distance, or other multi-dimensional data similarity analysis methods, calculates the similarity between the competing base station and the historical processing data of each base station within its own area one by one; finally, the arbitration base station selects the base station within its own area with the highest similarity and the similarity exceeding a predetermined threshold as the corresponding alternative base station.

[0064] Furthermore, if no base station with a similarity exceeding the threshold is found, it is considered that there is no suitable alternative base station for the competing base station within its own area. A vote is directly conducted on the competing base station.

[0065] A2.2: Sort all the base stations within the self-arbitration area in descending order of idle load to obtain an idle load sequence; In this embodiment, the arbitration base station needs to reasonably arrange service allocation according to the resource usage status of the base stations within the area. Therefore, the arbitration base station first regularly collects and statistically analyzes the idle load information of each base station within the area through its configured edge computing nodes in real time.

[0066] Specifically, the utilization rate of physical resource blocks (PRBs), the number of user connections, throughput data, and the utilization rate of the CPU or baseband unit are monitored in real time, and based on this information, the real-time idle load index of each base station is calculated, indicating the capacity of the remaining resources of the base station. Subsequently, the arbitration base station sorts the idle load indexes of all base stations to obtain a complete idle load sequence. The base station with a larger load index has a higher degree of resource idleness and a larger remaining capacity. When it is necessary to evaluate the necessity of competing base stations outside the area to participate in the competition, the remaining resources of the base stations within the area can be clearly and intuitively understood, which helps to accurately judge the actual service carrying value of the competing base stations in the subsequent steps and avoid problems such as waste or resource shortage caused by unreasonable resource allocation.

[0067] A2.3: For each competing base station with an alternative base station, determine the sequence position of the corresponding alternative base station. If the sequence position of the corresponding alternative base station is after the set position, confirm to vote for the competing base station; Specifically, the arbitration base station first defines a preset load sequence threshold position, which is used as the threshold benchmark for indicating the resource tension degree in the area. For example, if the threshold position is set to the median position of the sequence, the base stations after this position represent the base stations with relatively tight resources or relatively high load; if the alternative base station is located after the set threshold position, it means that the base station resources in the arbitration area are relatively busy. At this time, if the competing base station can be selected as the serving base station, it can effectively relieve the service pressure in this area. Therefore, the arbitration base station confirms to vote for the corresponding competing base station outside the area. On the contrary, if the alternative base station is located before the threshold position, that is, the resources are relatively abundant, it indicates that there are already enough resources in the current area to handle the service requirements, and there is no need to introduce competing base stations outside the area to participate in service bearing additionally. Therefore, it gives up voting for this competing base station. This judgment strategy based on the load sorting position of the alternative base station can intuitively reflect the resource tension degree in the current area and the actual necessity of the competing base station to participate in service, avoid the defects of making decisions simply based on static competition indicators, and effectively improve the accuracy of the voting decision and the optimization utilization efficiency of network resources.

[0068] A2.4: Receive the voting votes of each competing base station sent by other arbitration base stations, connect with the voting votes of each competing node determined by itself, and summarize to determine the serving base station; Specifically, the arbitration base station first encodes its voting results for each competing base station, including the identifier of the competing base station supported by the vote, and broadcasts its voting information to the entire distributed base station cluster through a predefined distributed communication protocol, such as the X2 interface or other dedicated signaling channels. Then, the arbitration base station receives and parses the voting information from other arbitration base stations, and performs real-time summarization of the voting data through the edge computing node to form the vote count result for each competing base station. The arbitration base station ensures the consistency and accuracy of the voting results through real-time data synchronization. If multiple competing base stations obtain the same number of votes during the summarization process, the arbitration base station further executes multi-level preference rules in sequence according to information such as the service load index, channel quality level, base station energy consumption index, historical user connection success rate, and response delay provided by the competing base stations, and finally determines the unique serving base station. Make full use of the information sharing and collaborative decision-making capabilities of each base station in the entire distributed arbitration mechanism, avoid inaccurate results caused by the judgment deviation of a single base station, improve the accuracy, reliability, and objectivity of the serving base station selection, so as to allocate network resources more reasonably, ensure efficient service bearing, and achieve significant energy-saving effects at the same time.

[0069] The specific steps of A2.4 are as follows: A2.4.1: Each arbitration base station generates voting information according to its determined voting results, and the voting information includes the identifiers of the competing base stations that obtain the vote of this arbitration base station; A2.4.2: Broadcast its own voting information to the distributed base station cluster; A2.4.3: Receive the voting information broadcast by other arbitration base stations, and combine with its own voting information to count the number of votes obtained by each competing base station; A2.4.4: Take the competing base station with the most votes as the serving base station; When multiple competing base stations obtain the same maximum number of votes, A2.4 further includes: A2.4.5: Obtain the traffic load index, channel quality level, and base station energy consumption index in the competition information; A2.4.6: Compare the traffic load indexes of competing base stations with the same number of votes, and select the base station with the highest traffic load index as the serving base station; A2.4.7: If the traffic load indexes are the same, then compare the channel quality levels, and select the base station with the highest channel quality level as the serving base station; A2.4.8: If the channel quality levels are the same, then compare the base station energy consumption indexes, and select the base station with the lowest base station energy consumption index as the serving base station; A2.4.9: If the base station energy consumption indexes are the same, then select the base station with the highest historical user connection success rate as the serving base station; A2.4.10: If the historical user connection success rates are the same, then select the base station with the shortest response delay from the last wake-up from the dormant state to the active state as the serving base station.

[0070] Although the embodiments of the present invention have been shown and described above, it can be understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those of ordinary skill in the art can make changes, modifications, substitutions, and variations to the above embodiments within the scope of the present invention.

Claims

1. A 5G base station intelligent sleep and wake-up energy-saving method based on edge computing is applied to a competing base station in a distributed base station cluster, wherein the distributed base station cluster further includes a plurality of arbitration base stations, each of which is one of the base stations in the distributed base station cluster that is not currently participating in the competition, and each arbitration base station corresponds to an arbitration area composed of a plurality of base stations, each of which is configured with an edge computing node, characterized in that: The 5G base station intelligent sleep and wake-up energy saving method includes: Receiving a network response request of a user terminal broadcasted by the distributed base station cluster; Generating competition information according to the competition indicator, broadcasting the competition information and historical processing information to the distributed base station cluster, so that each arbitration base station votes for the serving base station based on the received competition information and historical processing information of the competing base stations; If the competing base station is selected as the serving base station, its own state is switched to or maintained in the awake state.

2. According to the 5G base station intelligent sleep and wake-up energy saving method based on edge computing in claim 1, it is characterized in that: The competition indicators include service load, channel quality and base station energy consumption, and the generating of competition information according to the competition indicators includes: Obtaining a current service load, wherein the current service load includes a PRB occupancy rate, a number of user connections, and a throughput, and calculating a service load index according to a preset weight; Acquire a channel quality indicator, wherein the channel quality indicator includes RSRP, SINR and path loss, and calculate a channel quality level according to a channel quality threshold; Obtain current power consumption, energy efficiency per bit, and switch energy consumption overhead, and calculate the base station energy consumption index; Determine whether it is necessary to compete for the network response request according to the service load index, channel quality level and base station energy consumption index. If necessary, construct competition information. If not, broadcast base station identification information to the distributed base station cluster as an arbitration base station.

3. According to the 5G base station intelligent sleep and wake-up energy saving method based on edge computing in claim 2, it is characterized in that: Judging whether it is necessary to compete for the network response request according to the service load index, the channel quality level and the base station energy consumption index includes: Comparing the service load index with a first threshold to determine whether the current base station service load is higher than the first threshold, wherein the first threshold is adjusted by the edge computing node according to historical load data; If the traffic load index is higher than a first threshold, determining that the response request needs to be competed; If the service load index is less than or equal to the first threshold, the channel quality level is further compared with a second threshold; when the channel quality level is greater than or equal to the second threshold, the base station energy consumption index is further compared with a third threshold; otherwise, the competition is abandoned, wherein the second threshold is determined by an average value of received signal qualities reported by all arbitration base stations in the previous round based on user terminals; If the base station energy consumption index is less than the third threshold, it is determined that the response request needs to be competed; otherwise, the competition is abandoned.

4. A 5G base station intelligent sleep and wake-up energy-saving method based on edge computing is applied to an arbitration base station in a distributed base station cluster, each arbitration base station corresponds to an arbitration area composed of multiple base stations, and the distributed base station cluster also includes multiple competing base stations. Each arbitration base station is one of the base stations in the distributed base station cluster that is not currently participating in the competition. Each base station is configured with an edge computing node, characterized in that: The 5G base station intelligent sleep and wake-up energy saving method includes: receiving contention information and historical processing information sent by each contention base station that is not in its own arbitration area; According to the competition information and historical processing information corresponding to each competing base station, the serving base station is voted out; The base station identifier of the serving base station is broadcast to the distributed base station cluster, so that the corresponding competing base station switches its own state to or maintains the awake state.

5. According to claim 4, the 5G base station intelligent sleep and wake-up energy saving method based on edge computing is characterized in that: The step of voting out a serving base station according to the competition information and historical processing information corresponding to each competing base station includes: For each competing base station, determine whether there is a corresponding replacement base station in its own arbitration area according to the competition information and historical processing information corresponding to each competing base station; Sorting all base stations in the arbitration area according to idle load from large to small to obtain an idle load sequence; For each competing base station that has a replacement base station, determine the sequence position of the corresponding replacement base station, and if the sequence position of the corresponding replacement base station is after the set position, confirm to vote for the competing base station; Receive the votes of each competing base station sent by other arbitration base stations, connect the votes of each competing node determined by itself, and summarize to determine the serving base station.

6. According to claim 5, the 5G base station intelligent sleep and wake-up energy saving method based on edge computing is characterized in that: The determining whether there is a corresponding replacement base station in the arbitration area includes: Calculate the similarity between the historical processing information of the competing base station and each base station in its own arbitration area according to the historical processing information of the competing base station; According to the similarity, a base station having the highest similarity with the historical processing information of each competing base station is selected from its own arbitration area as the corresponding replacement base station.

7. According to claim 5, the 5G base station intelligent sleep and wake-up energy saving method based on edge computing is characterized in that: The summarizing and determining the serving base station includes: Each arbitration base station generates voting information according to the voting result determined by itself, wherein the voting information includes the identification of each competing base station that obtains the vote of the arbitration base station; Broadcast its own voting information to the distributed base station cluster; Receive voting information broadcast by other arbitration base stations, and combine it with its own voting information to count the number of votes obtained by each competing base station; The competing base station with the largest number of votes is used as the serving base station.

8. According to claim 7, the 5G base station intelligent sleep and wake-up energy saving method based on edge computing is characterized in that: When multiple competing base stations obtain the same highest number of votes, the aggregation determines the serving base station, further comprising: Acquire the service load index, channel quality level and base station energy consumption index in the competition information; The optimization rules are executed sequentially until a unique serving base station is determined.

9. According to claim 8, the 5G base station intelligent sleep and wake-up energy saving method based on edge computing is characterized in that: The sequentially executing the optimization rules until a unique serving base station is determined includes: Compare the service load indexes of competing base stations with the same number of votes, and select the base station with the highest service load index as the serving base station; If the service load indexes are the same, the channel quality levels are compared and the base station with the highest channel quality level is selected as the serving base station; If the channel quality levels are the same, the base station energy consumption indexes are compared, and the base station with the lowest base station energy consumption index is selected as the serving base station; If the base station energy consumption indexes are the same, the base station with the highest historical user connection success rate is selected as the service base station; If the historical user connection success rates are the same, the base station with the shortest response delay from the sleep state to the active state is selected as the serving base station.

10. According to claim 4, the 5G base station intelligent sleep and wake-up energy saving method based on edge computing is characterized in that: When the number of arbitration base stations does not meet the preset cardinality threshold, the 5G base station intelligent sleep and wake-up energy saving method further includes: According to the contention information of each competing base station, the contention information is sorted in reverse order to generate a reverse order list; The competing base stations in the reverse order list are converted into arbitration base stations in sequence until the number of arbitration base stations meets a preset cardinality threshold.

Images