Load judgment method and device for wireless access network, equipment, medium and product

Abstract

The embodiment of the invention discloses a load judgment method and device for a wireless access network, equipment, a medium and a product. The load judgment method comprises the following steps: acquiring software resource data and hardware resource data of the wireless access network; according to the software resource data and the hardware resource data, determining resource configuration information corresponding to the wireless access network by using an association relationship between software and hardware in the wireless access network; obtaining load data of each hardware board card in the wireless access network; according to the network structure of the wireless access network, carrying out association summarization on the load data of each hardware board card to obtain a flow control load type index set; the flow control load index set is used for representing the network flow control state of the wireless access network; and based on the resource configuration information and the flow control load index set, determining an equipment load condition of the wireless access network by using a preset neural network algorithm. According to the technical scheme, quantitative evaluation of the load of the wireless access network is realized, and quantitative management of the hardware load is realized.

Description

TECHNICAL FIELD

[0001] Embodiments of the present disclosure relate to the field of mobile communication technology, and in particular to a load determination method and device for a radio access network, equipment, medium and product. BACKGROUND

[0002] The existing main networking form of a radio access network (RAN) is a distributed radio access network (D-RAN) and a centralized radio access network (C-RAN). The D-RAN is one of the traditional mainstream access network forms, in which a baseband processing unit (BBU) is separately placed in a cabinet, and a remote radio unit (RRU) / active antenna unit (AAU) and an antenna are hung on a tower. Although this networking method is mature, it has some shortcomings. For example, each site needs a separate BBU and a machine room, resulting in a large amount of machine room construction and equipment investment, and a high operation and maintenance cost. Moreover, the scalability and flexibility of the D-RAN are limited, and it is difficult to quickly adapt to the rapid growth of network capacity and diversified business requirements.

[0003] The C-RAN is also one of the mainstream forms of the existing 4G / 5G RAN. It realizes centralized management and flexible deployment of resources by centralizing BBU processing resources, or separating central units (CUs) or distributed units (DUs) and connecting remote radio frequency units through high-bandwidth and low-latency optical fibers or optical transmission networks. The C-RAN overcomes some shortcomings of the D-RAN to some extent, such as reducing the number of machine rooms and significantly reducing the operation and maintenance cost. However, the C-RAN still has some deficiencies. For example, the hanging capacity of the C-RAN at the present stage is usually planned based on regional classification (such as dense urban areas, general urban areas, and suburban areas) and the number of sites, and lacks actual quantitative standards and optimization methods. Each C-RAN machine room does not exceed 20 physical sites, and a BBU does not exceed 2-10 physical sites. This planning method is relatively extensive, and does not fully consider the actual traffic differences of each site coverage area, resulting in large differences in the cell standard, number and bandwidth of the hung cells, and making it difficult to achieve fine and efficient resource utilization and load management.

[0004] In actual operation, the load optimization of the existing network mainly aims at the physical resource block (PRB) utilization rate and traffic volume of a cell for high and low load optimization adjustment and idle-to-busy operation, but lacks effective quantification and optimization means for high load or overload of hardware devices and boards. High load of a baseband board or a master control board can cause flow control of the signaling plane and the user plane, resulting in a large number of signaling discards and user plane packet discards, which seriously affects user perception, and the grasp and optimization adjustment of the hardware device load are of great significance for fully understanding the device resource utilization and reducing investment.

[0005] In addition, as the network load changes continuously, the existing network lacks quantifiable RAN-side network optimization schemes. Different manufacturers, different types of devices, boards and different service models can cause different load and flow control state performances, making the optimization work more complex and difficult to quantify. At the same time, in order to meet the demand for promoting cost reduction and efficiency improvement, energy saving and emission reduction, optimization and adjustment of the wireless side network structure are imperative, and further integration of the wireless side site needs to be done in combination with the RAN-side resource load.

[0006] In terms of C-RAN networking strategy, multiple sites are hung under one BBU device, and more cells are hung under each baseband board, but when there is a sudden high load or long-term high load, due to the limitation of the processing capacity of the baseband board and the master control board, a large amount of flow control often occurs, including paging, radio resource control (RRC) access, handover and data packets, which affects network performance and user experience. On the other hand, some BBU devices, boards and even AAU / RRU devices may have long-term low load in the case of non-coverage layer due to changes in network environment, resulting in waste of hardware resources. SUMMARY

[0007] Embodiments of the present disclosure provide a load judgment method, device, equipment, medium and product of a radio access network, which realize quantitative evaluation of the load of the radio access network and quantitative management of the hardware load.

[0008] In a first aspect, a load judgment method of a radio access network is provided, comprising:

[0009] obtaining software resource data and hardware resource data of the radio access network;

[0010] determining resource configuration information corresponding to the radio access network according to the software resource data and the hardware resource data by using the association relationship between software and hardware in the radio access network; the resource configuration information at least includes type of a baseband processing unit, number of hardware boards and type information of the hardware boards;

[0011] obtain load data of each hardware board card in the radio access network;

[0012] correlate and aggregate the load data of each hardware board card according to a network structure of the radio access network to obtain a flow control load index set; the flow control load index set is used to represent a network flow control state of the radio access network;

[0013] determine a device load condition of the radio access network based on the resource configuration information and the flow control load index set by using a preset neural network algorithm.

[0014] In a second aspect, a load determination apparatus of a radio access network is provided, and the apparatus comprises:

[0015] a data obtaining module configured to obtain software resource data and hardware resource data of the radio access network;

[0016] a resource configuration information determining module configured to determine resource configuration information corresponding to the radio access network by using an association relationship between software and hardware in the radio access network according to the software resource data and the hardware resource data; the resource configuration information at least comprises type information of a baseband processing unit, quantity information of a hardware board card, and type information of the hardware board card;

[0017] a load data obtaining module configured to obtain load data of each hardware board card in the radio access network;

[0018] a flow control load index set determining module configured to correlate and aggregate the load data of each hardware board card according to a network structure of the radio access network to obtain a flow control load index set; the flow control load index set is used to represent a network flow control state of the radio access network;

[0019] a device load condition determining module configured to determine a device load condition of the radio access network based on the resource configuration information and the flow control load index set by using a preset neural network algorithm.

[0020] In a third aspect, an electronic device is provided, and the device comprises:

[0021] at least one processor; and

[0022] a memory communicatively connected to the at least one processor; wherein

[0023] the memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor to enable the at least one processor to execute the load determination method of the radio access network as described in the first aspect.

[0024] In a fourth aspect, a computer readable storage medium is provided, and the computer readable storage medium stores a computer program. The computer program, when executed by a processor, implements the method for judging load of a radio access network according to the first aspect.

[0025] In a fifth aspect, a computer program product is provided, and the computer program product comprises a computer program. The computer program, when executed by a processor, implements the method for judging load of a radio access network according to the first aspect.

[0026] The embodiments of the present disclosure disclose a method, device, equipment, medium and product for judging load of a radio access network. The method comprises: obtaining software resource data and hardware resource data of the radio access network; determining resource configuration information corresponding to the radio access network by using an association relationship between software and hardware in the radio access network according to the software resource data and the hardware resource data; the resource configuration information at least comprises type of a baseband processing unit, quantity of hardware boards and type information of the hardware boards; obtaining load data of each hardware board in the radio access network; associating and summarizing the load data of each hardware board according to a network structure of the radio access network to obtain a flow control load index set; the flow control load index set is used to represent a network flow control state of the radio access network; and determining a device load condition of the radio access network by using a preset neural network algorithm based on the resource configuration information and the flow control load index set. In the technical solution, the resource configuration information corresponding to the radio access network is determined by using the software resource data and the hardware resource data of the radio access network, the flow control load index set is determined by using the load data of each hardware board in the radio access network, the device load condition of the radio access network is determined by using the preset neural network algorithm based on the resource configuration information and the flow control load index set, the quantitative evaluation of the networking architecture on the radio access network side is realized, the quantitative management of the hardware load is realized, the network architecture optimization is effectively supported, the network potential risk is reduced, the hardware resource utilization efficiency is improved, and strong support is provided for the fine management and optimization of the existing network.

[0027] It should be understood that the content described in this part is not intended to identify key or important features of the embodiments of the present disclosure, nor is it used to limit the scope of the embodiments of the present disclosure. Other features of the embodiments of the present disclosure will become apparent from the following description. BRIEF DESCRIPTION OF DRAWINGS

[0028] In order to more clearly illustrate the technical solutions in the embodiments of the present disclosure, the drawings needed in the embodiment description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present disclosure, and other drawings can be obtained by those skilled in the art without creating any creative labor.

[0029] Figure 1 This is a flowchart of a method for determining the load of a wireless access network provided in Embodiment 1 of this disclosure;

[0030] Figure 2 This is a schematic diagram of a preset neural network structure provided in Embodiment 1 of this disclosure;

[0031] Figure 3 This is a schematic diagram of the structure of a load assessment device for a wireless access network provided in Embodiment 2 of this disclosure;

[0032] Figure 4 This is a schematic diagram of the structure of an electronic device provided in Embodiment 3 of this disclosure. Detailed Implementation

[0033] To enable those skilled in the art to better understand the solutions of the embodiments of this disclosure, the technical solutions of the embodiments of this disclosure will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the embodiments of this disclosure, and not all embodiments. Based on the embodiments of this disclosure, all other embodiments obtained by those of ordinary skill in the art without creative effort should fall within the protection scope of the embodiments of this disclosure.

[0034] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this disclosure are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this disclosure described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0035] Example 1

[0036] Figure 1 This is a flowchart of a method for determining the load of a wireless access network according to Embodiment 1 of this disclosure. This embodiment is applicable to determining the load of a wireless access network. The method can be executed by a load determination device for the wireless access network, which can be implemented in hardware and / or software. This load determination device can be configured in an electronic device, including but not limited to computers, PCs, electronic devices, and servers, which are devices with data processing capabilities. Figure 1As shown, the method includes:

[0037] S110: Acquire software resource data and hardware resource data of the wireless access network.

[0038] In this embodiment, the Radio Access Network (RAN) is a crucial component connecting User Equipment (UE) and the Core Network (CN) in a mobile communication system. The RAN is responsible for handling the transmission and management of radio signals, ensuring that UEs can access the network and communicate. The RAN can replace the Radio Access Network using radio base stations; a radio base station typically consists of three parts: a Base Unit (BBU), a Remote Utility Unit (RRU), and an antenna. The Access Activated Unit (AAU) usually includes the RRU and antenna components; the BBU is further divided into a Centralized Unit (CU) and a Distributed Unit (DU), with the CU component potentially centralized or cloud-based; the Distributed Radio Access Network (D-RAN) typically has one BBU corresponding to one site; the Centralized Radio Access Network (C-RAN) is typically a centralized radio access network, with one BBU or BBU pool corresponding to multiple sites, centralized BBU resources, and the AAU, RRU, and antenna components located close to the demand side.

[0039] Specifically, it can obtain software resource data and hardware resource data of the wireless access network. The software resource data can be the resource data corresponding to the software in the wireless access network, and the hardware resource data can be the resource data corresponding to the hardware in the wireless access network.

[0040] For example, the hardware resource data on the wireless RAN side includes BBU equipment information and related configurations, AAU / RRU and antenna equipment and related configurations; the software resource data includes 4G and / or 5G cell resources and corresponding related configurations. BBU equipment information mainly includes the BBU model, the number and model of the main control board, and the number and model of the baseband board. The maximum number of RRC users supported and the maximum signaling processing capacity vary significantly between different baseband board and main control board models. The cell capabilities of different baseband board models, or even the same baseband board model, are greatly influenced by the cell's network standard, duplex mode, cell bandwidth, cell frequency band, and number of channels. The AAU / RRU model, number of channels, and supported frequency bands are also strongly correlated with the corresponding cell coverage scenario. Software resources mainly refer to logical cells, and cell attributes include 4G and / or 5G networks, Time Division Duplex (TDD) / Frequency Division Duplex (FDD) standards, cell bandwidth, number of channels, frequency band, Tracking Area Code (TAC), latitude and longitude, cell public land mobile network list, macro base stations, and indoor distribution systems, among other information.

[0041] S120. Based on software resource data and hardware resource data, determine the resource configuration information corresponding to the wireless access network by utilizing the association between software and hardware in the wireless access network; the resource configuration information includes at least the type of baseband processing unit, the number of hardware boards, and the type information of the hardware boards.

[0042] It is known that after obtaining software and hardware resource data, it is also possible to obtain the relationships between software and hardware in the wireless access network. These relationships can include software-hardware compatibility, the operational status of software modules on specific hardware boards, and so on. By analyzing these relationships, a more accurate understanding of the actual configuration of the wireless access network can be obtained.

[0043] As described above, based on software and hardware resource data, the resource configuration information corresponding to the wireless access network can be determined by utilizing the correlation between software and hardware. This resource configuration information refers to a series of parameters and data used to manage and optimize the resource allocation of the wireless access network. This information ensures that the wireless network can efficiently handle different types of services, meet user needs, and optimize the utilization of network resources. For example, the resource configuration information may at least include the type of baseband processing unit, the number of hardware cards, and the type information of the hardware cards.

[0044] As can be seen, by summarizing the relationships between hardware and software resources, we can obtain information such as the BBU equipment model, baseband board model and quantity, main control board model and quantity, the number of 4G / 5G cells corresponding to the main control board and baseband board, and the corresponding cell bandwidth for RAN (such as C-RAN or D-RAN) sites.

[0045] It should be noted that the RAN deployment architecture can be C-RAN or D-RAN. One C-RAN or D-RAN typically corresponds to one BBU device. Each BBU device includes a BBU model and a BBU serial number. A BBU device can have different card slots, and different slots can correspond to different cards. Different BBU models have different baseband board and main control board slots. BBU model compatibility with cards is usually backward compatible. Newer BBU models support newer versions of main control boards and baseband boards, and the newer models have stronger processing capabilities. The cards responsible for handling signaling and services in a BBU device are mainly baseband boards and main control boards. One BBU typically has two slots for the main control board and five to six slots for the baseband board. The main control board and baseband board each have corresponding model and serial number. The baseband board has a corresponding Common Public Radio Interface (CPRI) or enhanced Common Public Radio Interface (eCPRI) port, connected to the remote radio unit (RRU / AAU) via high-bandwidth, low-latency fiber optic or optical transmission network. The coverage areas of the RRU and AAU correspond to logical cells. The hardware and software resources of C-RAN and D-RAN are associated through hardware board serial numbers, baseband board and main control board slot numbers, RRU / AAU port numbers, and cell ECGI. Due to differences in deployment methods and the significant variations in the number of connected sites, some C-RAN BBUs may experience high loads due to too many connected sites or cells and high traffic volume, while some D-RAN or C-RAN systems may experience low loads due to insufficient traffic. Whether 4G / 5G cells share a BBU can be determined by the BBU serial number; the cell can be associated with the baseband board by the RRU / AAU port number, the corresponding baseband board serial number and port number; whether 4G / 5G cells share a main control board, baseband board / AAU or RRU, etc., can be determined by the corresponding main control board, baseband board, AAU / RRU serial number.

[0046] As described above, the software resource data and hardware resource data that can be obtained through the association between software and hardware in the wireless access network may include BBU model, BBU serial number, base station name, 5G base station identifier (gNodeB ID) / 4G base station identifier (eNodeB ID), main control board cabinet slot number, main control board model, main control board serial number, baseband board cabinet slot number, baseband board model, baseband board serial number, baseband board common public radio interface (CPRI) port number / e-CPRI port number, RRU / AAU model, RRU / AAU serial number, RRU / AAU port number, RRU / AAU channel information, cell CGI / ECGI, frequency band, bandwidth, TDD / FDD duplex mode, 4G / 5G network standard, public terrestrial mobile network list, TAC information, whether 4G / 5G share a main control unit, and whether 4G / 5G share a baseband board, etc.

[0047] S130: Obtain the load data of each hardware board in the wireless access network.

[0048] In this embodiment, the wireless access network can be composed of multiple hardware boards, and the load data of each hardware board in the wireless access network can be acquired. The load data of each board can be used to reflect the operating status of each board. The load data of each hardware board may include the highest CPU utilization and average CPU utilization of the main control board (such as VBP_1_3 or 0-0-7 main control board) and baseband board corresponding to each gNodeBID / eNodeBID or specific cabinet slot number under the base station name. Cell-level flow control and load-related data such as RRC, paging, flow control, user plane, and utilization data can be acquired.

[0049] S140. Based on the network structure of the wireless access network, the load data of each hardware board is correlated and summarized to obtain a set of flow control load indicators; the set of flow control load indicators is used to represent the network flow control status of the wireless access network.

[0050] In this embodiment, each hardware board corresponds to its own load data. By considering the network structure of the wireless access network, such as the connection relationships between boards, data transmission paths, and their interactions when processing user data, the load data of each board can be analyzed hierarchically to obtain a set of flow control load indicators. For example, the hierarchical relationship and data flow between hardware boards can be determined based on the network structure of the wireless access network. For instance, in a typical wireless access network, the baseband board may handle user data, the main control board may handle signaling and control functions, and the RRU / AAU is responsible for transmitting and receiving wireless signals. There are clear data interactions and dependencies between these boards. Based on the hierarchical relationship between the hardware boards, the load data of each board is correlated hierarchically.

[0051] For example, the load data of the baseband board may affect the load of the main control board, and the load of the main control board in turn affects the performance of the entire base station. Through this hierarchical correlation, scattered load data can be integrated to form a set of flow control load indicators. The set of flow control load indicators can be used to evaluate the network flow control status of the radio access network. For example, the set of flow control load indicators may include the load status of different levels of boards such as each BBU, C-RAN site, D-RAN site, baseband board per slot, main control board, and cell, as well as RRC flow control, paging flow control, handover flow control, user plane flow control, and load status.

[0052] It's important to explain that the main control board is the core control component of the radio RAN-side BBU equipment, primarily responsible for RRC, NG, Xn / X2 interface control plane, and user plane information processing. The baseband board is the core component of the BBU equipment responsible for signal baseband processing, including signal modulation, demodulation, encoding, and decoding. It is mainly responsible for RRC control plane and cell air interface resource allocation-related control plane processing. The Call Attempts Per Second (CAPS) processing capacity of the main control board and baseband board determines the number of users and service processing capabilities that can be handled simultaneously, meaning it can handle a large number of concurrent signaling and data transmissions. When the main control board and baseband board exceed their load limits, device flow control occurs; that is, the device controls input and output flow to prevent overload and maintain stability. Device flow control targets signaling data, service data, and operation and maintenance data. When the device is subjected to flow surges, the device flow control function can reduce the risk of device reset, thereby improving device reliability; it can also reduce the risk of deterioration in access success rate and handover success rate. Different equipment manufacturers have different processing mechanisms for the main control board and baseband board, resulting in different indicators when overloaded. Some manufacturers' equipment shows more paging congestion and loss, while others' equipment shows a large number of RRC drops when overloaded. However, the common point is that overload of the main control board and baseband board usually causes signaling plane such as RRC access, paging, and handover to be dropped or fail, as well as user plane PDCP downlink packet drop.

[0053] Based on the signaling and user plane issues arising from overload of the main control board and baseband board, flow control load indicators can be preliminarily divided into 5 categories: 1) CPU utilization of each board of the main control board and baseband board, including maximum CPU utilization and average CPU utilization; 2) Flow control data such as RRC: mainly including the number of RRC establishment request messages dropped due to flow control, the number of RRC establishment rejections due to resource allocation failures, the number of RRC establishment request messages dropped due to exceeding the RRC request limit, RRC connection failure status, and the maximum and average number of RRC connections; 3) Paging data: number of paging received, number of paging dropped, and number of paging congestion; 4) Handover and measurement data: number of handover inbound messages dropped due to base station flow control (inter-system + intra-system), number of measurement reports dropped due to CPU overload, number of handover requests, number of handover failures, etc.; 5) User plane and utilization data: number of packets sent at the PDCP protocol layer, number of PDCP dropped packets, uplink and downlink PRB utilization, uplink and downlink traffic volume, etc.

[0054] S150: Based on resource allocation information and a set of flow control load indicators, a preset neural network algorithm is used to determine the device load status of the wireless access network.

[0055] It should be noted that the wireless access network can include network architectures such as 4G standalone BBU, 5G standalone BBU, and 4 / 5G shared BBU devices, and the characteristics of different flow control load scenarios also differ. Determining the RAN-side load level and different high-load scenarios requires information such as the baseband board model, main control board model, flow control load indicators corresponding to the baseband board and main control board, the downstream cell configuration, and load status of the RAN-side BBU devices. Different weights must be assigned to different pieces of information to effectively determine the RAN-side load situation. In this embodiment, after obtaining the resource configuration information and the set of flow control load indicators, the resource configuration information and the set of flow control load indicators can be input into a preset neural network algorithm to determine the device load status of the wireless access network.

[0056] Among them, the preset neural network algorithm can be a pre-trained neural network. For example, the preset neural network algorithm can be a multilayer perceptron (MLP) neural network algorithm. The MLP neural network algorithm can be used to train and learn on the existing network data, find out the hardware load of the RAN side, and make relevant predictions. It can give effective strategies by combining different hardware loads of the RAN side with hardware resource equipment models and actual network architecture. For low-load RAN side resources, resource integration can be performed by combining latitude and longitude.

[0057] This embodiment provides a method for determining the load of a wireless access network, including: acquiring software resource data and hardware resource data of the wireless access network; determining resource configuration information corresponding to the wireless access network based on the software resource data and hardware resource data, utilizing the correlation between software and hardware in the wireless access network; the resource configuration information includes at least the type of baseband processing unit, the number of hardware boards, and the type information of the hardware boards; acquiring load data of each hardware board in the wireless access network; correlating and summarizing the load data of each hardware board according to the network structure of the wireless access network to obtain a set of flow control load indicators; the set of flow control load indicators is used to represent the network flow control status of the wireless access network; and determining the device load status of the wireless access network using a preset neural network algorithm based on the resource configuration information and the set of flow control load indicators. This technical solution realizes the quantitative evaluation of the wireless access network architecture, achieves quantitative management of hardware load, effectively supports network architecture optimization, reduces potential network risks, improves hardware resource utilization efficiency, and provides strong support for the refined management and optimization of the existing network.

[0058] As an optional implementation of this embodiment, the wireless access network load determination method provided in this embodiment further includes:

[0059] 1) Determine the adjustment strategy based on the device load and the network structure of the wireless access network.

[0060] In this embodiment, under the same network standard, the processing and carrying capacities of different models of main control boards and baseband boards vary significantly. Even for the same board model, the processing capacity differs under different network standards. The baseband board is more complex, involving multiple factors such as network standard, duplex mode, frequency band, bandwidth, and number of channels, resulting in different numbers of cells connected to the same board model. Different network standards, duplex modes, frequency bands, and number of channels lead to significant differences in the amount of information exchange and processing mechanisms. Therefore, reasonable optimization and adjustments are needed, taking into account load conditions, connected cell configuration information, the signaling processing capabilities of the board model, and the specifications of the number of RRC connected users.

[0061] Specifically, after determining the device load of the wireless access network through neural networks, adjustment strategies can be determined based on the device load and network structure of the wireless access network.

[0062] Following the above description, common equipment load conditions can include typical high flow control loads, such as high load on the main control board and normal load on the baseband board. The adjustment strategy could be to replace the main control board with a high-performance one. Table 1 provides load data for one type of baseband board and main control board.

[0063] Table 1 Load data for baseband board and main control board

[0064]

[0065] Table 1 shows that at different times, the load on the two baseband boards of the RAN-side BBU equipment was not very high, with both maximum and average CPU utilization rates between 20% and 50%. However, the main control board had a high load, with average CPU utilization exceeding 84% and maximum utilization reaching 100%, resulting in relatively more signaling data loss. The main control board load is high, while the load on the two baseband boards is not very high, and there are only two baseband boards. The main control board issue is more prominent; a possible solution is to replace it with a newer model of main control board.

[0066] Common equipment load situations may include: too many C-RAN connected sites leading to high overall load; adjustment strategies could include adding BBU equipment, or ensuring normal service load after RAN-side splitting. Uneven load distribution between the two baseband boards within the BBU equipment at RAN-side sites can be addressed by changing the baseband board model and adjusting the correspondence between the baseband boards and cells. For example, at a venue hosting an event, a significant number of RRC connection request messages were dropped due to flow control between 7 PM and 9 PM, with the most severe flow control at 8 PM. The maximum number of users during these three periods were 8043, 8622, and 8334 respectively. By 9 PM, as the number of users decreased, the number of dropped RRC connection request messages due to flow control also decreased. The C-RAN site's BBU equipment has 5 baseband boards supporting 24 cells. Except for the baseband board in slot 0-0-1, the other baseband boards experience significant signaling loss and RRC failures due to user specification exceeding limits. Except for the baseband board in slot 0-0-1, the CPU utilization of the other baseband boards is above 70% at maximum and is also relatively high, generally between 40% and 62%. The main control board's average CPU utilization exceeds 70%, with maximum CPU utilization between 87% and 88%, indicating a very high load. The excessively high load on the main control board, along with the high peak and average loads of almost all baseband boards, indicates that the overall BBU load is too high and cannot handle the corresponding workload. Splitting the BBU is the most suitable solution.

[0067] Common equipment load conditions may also include: unbalanced load between the two baseband boards within the BBU equipment on the RAN side. The adjustment strategy can be to replace the baseband board model and adjust the correspondence between the baseband board and the cells. For example, the two cells connected to the 0-0-2 baseband board (LBBPD4) have very high loads, while the four cells connected to the 0-0-4 UBBPD9 board have relatively low loads. The 0-0-2 baseband board has a large number of connected users, some paging drops, PDCP packet drops, and a small number of RRC request drops. In this example, the 0-0-2 baseband board (LBBPD4) is an older version with relatively insufficient processing capacity, and the two connected cells have very high loads (which can be seen from the maximum number of users in the cell). The processing capacity of this baseband board cannot handle the corresponding load of the two cells; university areas are typically high-load scenarios. The adjustment strategy could be to replace the 0-0-2 baseband board with the UBBPD9 model, and simultaneously adjust the high and low load cell connections between the two baseband boards to balance the load on the two baseband boards; the problem would be resolved after the adjustment.

[0068] It should be noted that different flow control load scenarios involve different indicator periods, and the weights of different flow control load types also vary. The weights of relevant input information can be adjusted according to different needs, and the relevant weights can also be optimized in reverse based on the processing results to meet different needs and improve the accuracy of optimization strategies. Based on historical data and expert knowledge training, the corresponding weights are continuously adjusted and optimized to optimize and improve the MLP neural network model. Based on the output of the neural network model, combined with the network architecture and board model capabilities, optimization strategies are provided. Based on the load data of the RAN-side BBU equipment, corresponding optimization adjustment suggestions are given.

[0069] As described above, the RAN-side network architecture provides information on the network structure of various radio access networks for each BBU, including the main control board, baseband board model and quantity, number of connected cells, cell bandwidth, cell network standard, cell frequency band and bandwidth, cell duplex mode, cell channel configuration, and Public Land Mobile Network List (PLMNLIST). Based on RAN-side load indicators and MLP neural networks, the BBU load and the load of each level of board can be obtained. Combining the BBU model, board model, and network architecture, reasonable optimization solutions can be provided for high-load RAN-side hardware. Furthermore, low-load RAN-side resources can be effectively integrated.

[0070] 2) Update the network structure of the wireless access network according to the adjustment strategy.

[0071] Specifically, once the adjustment strategy is determined, the network structure of the wireless access network can be updated using the determined adjustment strategy.

[0072] For example, based on the BBU load classification and optimization suggestions output by the neural network, the RAN-side hardware resources and network structure can be optimized and adjusted. The BBU load classification is basically consistent with common high-load scenarios of flow control. Based on the optimization suggestions, operations such as BBU splitting, changing the model of the baseband board and main control board, and load balancing between boards can be carried out to adjust the network structure of the wireless access network. Combining the RAN-side flow control load changes and user behavior data, the neural network is used to perform intelligent optimization and adjustment to achieve dynamic allocation and optimization of resources.

[0073] As described above, abnormal loads require comprehensive analysis in conjunction with RAN equipment and peripheral equipment malfunctions or service anomalies. For abnormal load situations, if the flow control load problem occurs frequently over a long period and involves a large number of cells with flow control loads, flow control load analysis can be performed based on TAC. A comprehensive analysis of the load situation of the corresponding board in the flow control cell can be used to determine whether the TAC division is too large. If the abnormal load situation has a short time period and involves few sites and cells, it is usually more correlated with alarms.

[0074] As described above, for RAN-side equipment with low load, considering the need for site equipment simplification, latitude and longitude distances, and access ring conditions, D-RAN and C-RAN can be merged to save equipment room space and reduce energy consumption. Based on the number of main control boards and baseband boards configured on the RAN-side equipment, the board load of the baseband boards and main control boards, the downstream cell configuration, and the remaining board ports, RAN site consolidation and baseband board consolidation can be performed. Alternatively, different swaps can be made within the RAN-side BBU based on board load and board model to achieve a more rational allocation of RAN-side hardware resources and improve network processing performance.

[0075] As described above, for cell expansion and RAN planning for new access sites, the RAN-side sites to be accessed can be reasonably planned based on the resource configuration of the existing RAN-side hardware equipment, the flow control load situation and forecast, as well as the traffic of the expanded cell or new access site. The corresponding baseband board and baseband board model can also be planned. Cell expansion connection and C-RAN and D-RAN planning for new access sites can be carried out based on quantifiable data of the actual hardware resource load.

[0076] As an optional implementation of this embodiment, the wireless access network load determination method provided in this embodiment further includes:

[0077] 1) Obtain the training sample set, which includes historical resource configuration information, historical flow control load index set, and corresponding labels for actual equipment load conditions.

[0078] In this embodiment, a training sample set can be obtained, which includes historical resource configuration information, a set of historical flow control load indicators, and corresponding labels for actual device load conditions. The training sample set can be collected from past wireless access network operation records. The historical resource configuration information reflects the past hardware and software configuration of the network, the set of historical flow control load indicators records the network's flow control status under these configurations, and the labels for actual device load conditions are the actual observed device load conditions, used as a reference standard for model training.

[0079] 2) Input the historical resource configuration information and the set of historical flow control load indicators into the initial neural network algorithm to obtain the predicted equipment load.

[0080] Specifically, historical resource allocation information and historical flow control load index sets can be input into the initial neural network algorithm to obtain the equipment load situation predicted by the neural network.

[0081] 3) Based on the target error, the weight vectors of each neuron in the hidden layer of the initial neural network algorithm are updated and iterated through the backpropagation algorithm until the target error meets the preset condition, thus obtaining the preset neural network algorithm; the target error is the error between the predicted equipment load status and the actual equipment load status label.

[0082] In this embodiment, after obtaining the predicted equipment load, the error between the predicted equipment load and the actual equipment load label can be calculated to obtain the target error. Based on the target error, the weight vectors of each neuron in the hidden layer of the initial neural network algorithm are updated iteratively using the backpropagation algorithm until the target error meets a preset condition, thus obtaining the preset neural network algorithm.

[0083] Backpropagation is a commonly used neural network training method. It calculates the error between the predicted result and the true label, and then propagates this error back from the output layer to the hidden layers of the network, thereby adjusting the weight vectors of each neuron. This process is repeated until the target error meets a preset condition, that is, the error between the predicted device load status and the actual device load status label reaches an acceptable range.

[0084] As an optional implementation of this embodiment, the preset neural network algorithm includes: an input layer, a hidden layer, and an output layer; the hidden layer contains at least one neuron.

[0085] Taking the multilayer perceptron (MLP) neural network as an example, Figure 2 A schematic diagram of a pre-defined neural network provided in this application is shown below. Figure 2 As shown, an MLP neural network includes an input layer, an output layer, and multiple hidden layers. All layers are fully connected, with the output of one layer serving as the input to the next. The neural network includes... One input neuron, One output neuron, A multi-layer feedforward network structure with n hidden layer neurons. The output layer has n hidden layer neurons. The threshold of each neuron is used This indicates that the hidden layer is... Preset for each neuron Indicates. Input layer number 1 The first neuron and the hidden layer The connection weights between neurons are Hidden layer The nth neuron and the output layer The connection weights between neurons are The hidden layer is recorded. The input received by each neuron is Output layer The input received by each neuron is ,in For the hidden layer The output of each neuron. The input layer is the input end, where input data is read in; the hidden layer is the information processing end, and the number of hidden layers can be set; the output layer is the information output end. and The weights represent the input layer to the hidden layer and the hidden layer to the output layer, respectively.

[0086] As an optional implementation of this embodiment, the method provided in this embodiment further includes: determining the device load status of the wireless access network using a preset neural network algorithm based on the resource configuration information and the flow control load index set, including:

[0087] 1) The resource configuration information and the set of flow control load indicators are preprocessed to obtain the target data set.

[0088] Specifically, after obtaining the resource configuration information and the set of flow control load indicators, preprocessing can be performed on these information and indicators to determine the target data. This preprocessing includes data cleaning and normalization. The resource configuration information and the set of flow control load indicators can include RAN-side BBU device hardware / software association table information and corresponding key flow control indicators.

[0089] 2) Using the input layer, the target data set is transmitted to the hidden layer.

[0090] Specifically, after obtaining the target data set through preprocessing, the target data set can be transmitted to the hidden layer through the input layer. For example, the preprocessed hardware configuration information of each RAN-side BBU device and key flow control load indicators can be used as feature vectors. There are five categories of flow control load indicators, and key flow control indicators can be selected according to different equipment manufacturers as needed. When the hardware load is high, signaling flow control load indicators are more prominent, with RRC flow control load, paging flow control, and board CPU load indicators being relatively more prominent, and their weights can be set higher. Among them, the board CPU load indicator is set with the highest weight, followed by RRC flow control load and paging flow control, while other flow control load indicators are set with lower weights. When the hardware load is high for different equipment manufacturers, the performance of other flow control load indicators, except for board CPU load, varies; some RRC flow control load indicators are prominent, some paging load indicators are prominent, and some handover load indicators are prominent. Different weights can be adjusted according to the actual situation.

[0091] 3) By performing weighted summation and probability transformation on each target data in the target data set through each neuron of the hidden layer, the first load assessment value and the second load assessment value corresponding to each hardware board are obtained.

[0092] Specifically, each neuron can process a single feature. The neurons in the hidden layer can perform weighted summation and probability transformation on the target data in the target dataset to determine the first and second load assessment values ​​for each hardware board. The first load assessment value can represent the load status of each device in the wireless access network, while the second load assessment value can include detailed load metrics.

[0093] It should be explained that the hidden layer can also derive the flow control load of each baseband board and main control board of the BBU device based on the target data set (RAN-side configuration data and flow control load indicators) after preprocessing by the input layer, according to relevant weights and corresponding neural network algorithms.

[0094] Based on the above description, combined with the input layer data, the following can be obtained: single-board load indicators (baseband board flow control load indicators and / or main control board flow control load indicators). These include the maximum CPU utilization of each board over a period of time, the average CPU utilization during busy hours, the number of times the threshold is exceeded, the number of times the board's RRC / paging / switching flow control is exceeded / failure ratio, the maximum number of RRC users, PRB utilization, user plane drop ratio, and the number of times various indicators exceed the threshold ratio.

[0095] 4) Based on the first load assessment value and the second load assessment value, the output layer is used to determine the device load status of the wireless access network; the device load status includes at least: the load status of the baseband processing unit of the wireless access network, the flow control load data of the baseband board of the wireless access network, and the flow control load status of the main control board of the wireless access network.

[0096] Specifically, after obtaining the first and second load assessment values, RRC can obtain the device load status of the wireless access network by setting different weights for the load status of the main control board and the baseband board, based on information such as the flow control load indexes of each baseband board and the main control board in the hidden layer, and the number of times each main control board and baseband board exceed the threshold. The device load status includes at least: the load status of the baseband processing unit of the wireless access network, the flow control load data of the baseband board of the wireless access network, and the flow control load status of the main control board of the wireless access network.

[0097] As an optional implementation of this embodiment, it further includes: performing weighted summation and probability transformation on each target data in the target data set through each neuron of the hidden layer to obtain a first load assessment value and a second load assessment value corresponding to each hardware board, including:

[0098] 1) Based on the weight vectors corresponding to each neuron in the hidden layer, perform a weighted sum on each target data in the target data set to obtain the weighted sum corresponding to each hidden layer neuron.

[0099] Specifically, for each target data point in the target dataset, it is multiplied element-wise with the weight vectors corresponding to each neuron in the hidden layer and then summed to calculate the weighted sum of the target data point at each neuron. This completes the linear mapping from input features to the hidden layer state, ensuring that subsequent nonlinear transformations have an interpretable and reproducible data foundation.

[0100] 2) Use the activation function to perform probability transformation on each weighted sum to determine the first load assessment value corresponding to each board.

[0101] It is known that after obtaining the weighted sum corresponding to each hidden layer neuron, the weighted sum of the output of each hidden layer neuron can be sent one by one to a preset activation function (such as the softmax function). The function maps the sum to the probability interval, so that the mapping result directly reflects the first load evaluation value of the corresponding board.

[0102] 3) Determine the corresponding second load assessment value for each board based on the first load assessment value for each board.

[0103] Specifically, after determining the first load assessment value for each board, the second load assessment value for the corresponding board can be determined based on the first load assessment value for each board. The second assessment value may include detailed load indicators.

[0104] It should be explained that after obtaining the first load assessment value, i.e., the "board load score", for each board, the score can be processed according to the preset score-load mapping rules. On the one hand, the score range can be mapped to levels such as "light load, normal, heavy load, and overload" according to business needs. On the other hand, detailed load indicators such as the historical peak value, temperature margin, and remaining power supply capacity of each board can be introduced simultaneously to perform weighted correction or table lookup compensation on the first load assessment value, thereby calculating a second load assessment value that is more in line with the actual operating risk. This makes the final output load level retain the intuitiveness of the score while integrating fine-grained information from multiple dimensions, providing a decision-making basis that balances accuracy and interpretability for subsequent channel switching, task migration, or heat dissipation speed adjustment.

[0105] For example, data is passed from the input layer to the hidden layer. For each hidden layer neuron, a weighted sum of its inputs is calculated, and then the weighted sum is passed to the activation function. This process occurs at each hidden layer neuron. The activation function uses softmax to obtain the most reasonable load data. The load status of the hidden layer and the output layer can be the load score for each board, and the load status corresponding to the score can be set as needed. At the same time, detailed load indicators are output.

[0106] The technical solution provided in this embodiment obtains the equipment load status of the RAN side based on the wireless RAN side hardware and software resources and the corresponding flow control load index set, and provides RAN side network optimization and adjustment strategies. Specifically, it includes: acquiring RAN side hardware and software resource information and their interrelationships, and load data of each level of boards to determine the resource configuration information and flow control load index set corresponding to the wireless access network; employing an MLP multilayer neural network algorithm, using the resource configuration information and flow control load index set corresponding to the wireless access network as input layer parameters, using multiple hidden layers, and performing weight optimization according to different needs, thereby obtaining the hierarchical load status of hardware resources; and providing optimizable adjustment strategies based on the processing capabilities and network configuration of different hardware board models, thus providing RAN side network optimization strategies. The technical effects achieved by the above solutions include: 1) Constructing a classification and evaluation index system for RAN-side hardware and software resource flow control load, including CPU utilization of boards, flow control load indicators for RRC, paging, handover, and measurement signaling, and user plane flow control load indicators for packet abandonment and traffic volume; 2) Based on flow control load indicators, according to the hardware and software relationship, the flow control load indicators can be summarized at the level of individual main control boards, baseband boards, board types, and BBUs. Combined with the hierarchical hardware load indicators and their frequency of occurrence, the RAN hardware load situation (flow control load indicator set) can be determined; 3) Based on the MLP neural network algorithm, and based on the RAN-side hardware and software network and corresponding load indicators, adjusting and optimizing the weights of non-key flow control load indicators to obtain hierarchical load information and achieve quantitative management of hardware resources. Optimizable adjustment strategies are provided based on board models and network architecture. The above solutions can effectively identify RAN-side network hardware load problems, effectively improve user experience, and provide support and basis for RAN hardware resource optimization and integration, and idle resource reallocation. It should be noted that the above technical solutions can be used in different forms of traditional BBU equipment networking, as well as in different equipment forms such as cloud-radio access network (Cloud-RAN) and open radio access network (O-RAN).

[0107] The technical solution provided in this implementation establishes the correlation between RAN-side hardware and software resources and the scope of RAN-side resources, including 4G, 5G, and 5G reverse-opening 4G equipment. Hardware and software resources include BBU equipment models, board models and quantities, board slot usage and baseband board port numbers, RRU / AAU information, and corresponding cell-level related resources, forming a complete hardware and software resource association on the RAN side. This facilitates resource management and allocation, and RAN-side architecture adjustments. It allows for quick understanding and control of the existing hardware resources, remaining resources, and number of connected cells for each RAN-side BBU equipment. A complete indicator system for RAN hardware and software resource flow control load is constructed, including five categories of indicators such as board CPU load, signaling and user plane flow control load indicators, etc. The flow control load evaluation system includes board CPU utilization... This method includes user plane metrics such as RRC, paging, handover, and measurement signaling indicators, as well as PDCP packet drop and traffic. It also includes the number of times various metrics exceed thresholds within a certain time period to aid in judgment. It addresses the current situation where network load is measured solely by cell-level PRB utilization and the number of RRC users, improving the RAN-side hardware and software load evaluation system. It not only evaluates from the perspective of cell-level resource management but also supplements signaling and user plane load management as well as board load management. It can comprehensively evaluate the RAN-side load situation; it achieves quantitative data management of hardware load, supporting RAN-side network architecture optimization; based on this method, it pioneered quantitative management of BBU hardware load, improving the network's hardware and software load indicator system and quantitative management mechanism; it effectively avoids the current extensive RAN network management model and can be used for high-load optimization and adjustment of existing network hardware, reducing potential network risks. It can effectively evaluate the load situation of the current D-RAN and C-RAN structures, can be used for existing site merging and rectification, and as a quantitative reference for new network access planning. It can also be used for hardware resource evaluation during cell idle-to-busy replacement, greatly improving hardware resource utilization efficiency.

[0108] Example 2

[0109] Figure 3 This is a schematic diagram of the structure of a load assessment device for a wireless access network provided in Embodiment 2 of this disclosure; as shown... Figure 3 As shown, the device includes: a data acquisition module 210, a resource configuration information determination module 220, a load data acquisition module 230, a flow control load index set determination module 240, and an equipment load status determination module 250.

[0110] Among them, the data acquisition module 210 acquires software resource data and hardware resource data of the wireless access network;

[0111] The resource configuration information determination module 220 is used to determine the resource configuration information corresponding to the wireless access network based on the software resource data and hardware resource data, utilizing the association relationship between software and hardware in the wireless access network; the resource configuration information includes at least the type of baseband processing unit, the number of hardware boards, and the type information of the hardware boards;

[0112] The load data acquisition module 230 is used to acquire load data of each hardware board in the wireless access network;

[0113] The flow control load index set determination module 240 is used to correlate and summarize the load data of each hardware board according to the network structure of the wireless access network to obtain the flow control load index set; the flow control load index set is used to represent the network flow control status of the wireless access network.

[0114] The device load determination module 250 is used to determine the device load of the wireless access network based on the resource configuration information and the set of flow control load indicators, using a preset neural network algorithm.

[0115] Embodiment 2 of this disclosure provides a load assessment device for a wireless access network, which realizes quantitative evaluation of the network architecture on the wireless access network side, achieves quantitative management of hardware load, effectively supports network architecture optimization, reduces potential network risks, improves hardware resource utilization efficiency, and provides strong support for the refined management and optimization of the existing network.

[0116] Furthermore, the device also includes:

[0117] The adjustment strategy determination module is used to determine the adjustment strategy based on the device load and the network structure of the wireless access network.

[0118] The update module is used to update the network structure of the wireless access network according to the adjustment strategy.

[0119] Furthermore, the preset neural network algorithm includes: an input layer, a hidden layer, and an output layer; the hidden layer contains at least one neuron.

[0120] Furthermore, the equipment load determination module 250 also includes:

[0121] The target data set determination unit is used to preprocess the resource configuration information and the flow control load index set to obtain the target data set;

[0122] A transmission unit is used to transmit the target data set to the hidden layer using the input layer;

[0123] The evaluation value determination unit is used to perform weighted summation and probability transformation on each target data in the target data set through each neuron of the hidden layer to obtain the first load evaluation value and the second load evaluation value corresponding to each hardware board.

[0124] The device load determination unit is used to determine the device load status of the wireless access network based on the first load assessment value and the second load assessment value using the output layer; the device load status includes at least: the load status of the baseband processing unit of the wireless access network, the flow control load data of the baseband board of the wireless access network, and the flow control load status of the main control board of the wireless access network.

[0125] Furthermore, the evaluation value determination unit is also used for:

[0126] Based on the weight vectors corresponding to each neuron in the hidden layer, the target data in the target data set are weighted and summed to obtain the weighted sum corresponding to each hidden layer neuron.

[0127] The activation function is used to perform a probability transformation on each weighted sum to determine the first load assessment value for each board.

[0128] The second load assessment value for each board is determined based on the first load assessment value for each board.

[0129] Furthermore, the device also includes:

[0130] The training set acquisition module is used to acquire a training sample set, which includes historical resource configuration information, a set of historical flow control load indicators, and corresponding labels for actual equipment load conditions.

[0131] The prediction module is used to input the historical resource configuration information and the set of historical flow control load indicators into the initial neural network algorithm to obtain the predicted equipment load situation;

[0132] The pre-approval neural network algorithm determination module is used to update and iterate the weight vectors of each neuron in the hidden layer of the initial neural network algorithm based on the target error through the backpropagation algorithm until the target error meets the preset condition, thereby obtaining the preset neural network algorithm; the target error is the error between the predicted equipment load condition and the actual equipment load condition label.

[0133] The load assessment device for wireless access networks provided in this disclosure can execute the load assessment method for wireless access networks provided in any embodiment of this disclosure, and has the corresponding functional modules and beneficial effects for executing the method.

[0134] Example 3

[0135] Figure 4 A schematic diagram of the structure of an electronic device 10 that can be used to implement embodiments of the present disclosure is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the embodiments of the present disclosure described and / or claimed herein.

[0136] like Figure 4 As shown, the electronic device 10 includes at least one processor 11 and a memory, such as a read-only memory (ROM) 12 or a random access memory (RAM) 13, communicatively connected to the at least one processor 11. The memory stores computer programs executable by the at least one processor. The processor 11 can perform various appropriate actions and processes based on the computer program stored in the ROM 12 or loaded from storage unit 18 into the RAM 13. The RAM 13 can also store various programs and data required for the operation of the electronic device 10. The processor 11, ROM 12, and RAM 13 are interconnected via a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0137] Multiple components in electronic device 10 are connected to I / O interface 15, including: input unit 16, such as keyboard, mouse, etc.; output unit 17, such as various types of displays, speakers, etc.; storage unit 18, such as disk, optical disk, etc.; and communication unit 19, such as network card, modem, wireless transceiver, etc. Communication unit 19 allows electronic device 10 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0138] Processor 11 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microprocessor, etc. Processor 11 performs the various methods and processes described above, such as load assessment methods for wireless access networks.

[0139] In some embodiments, the load assessment method for a wireless access network can be implemented as a computer program tangibly contained in a computer-readable storage medium, such as storage unit 18. In some embodiments, part or all of the computer program can be loaded and / or installed on electronic device 10 via ROM 12 and / or communication unit 19. When the computer program is loaded into RAM 13 and executed by processor 11, one or more steps of the load assessment method for a wireless access network described above can be performed. Alternatively, in other embodiments, processor 11 can be configured to perform the load assessment method for a wireless access network by any other suitable means (e.g., by means of firmware).

[0140] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), payload-programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0141] Computer programs for implementing the methods of embodiments of this disclosure may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing apparatus, such that when executed by the processor, the computer programs cause the functions / operations specified in the flowcharts and / or block diagrams to be performed. The computer programs may be executed entirely on a machine, partially on a machine, or as a standalone software package, partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0142] In the context of embodiments of this disclosure, a computer-readable storage medium may be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, apparatus, or device. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0143] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the electronic device. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0144] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or middleware components (e.g., application servers), or frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., communication networks). Examples of communication networks include local area networks (LANs), wide area networks (WANs), blockchain networks, and the Internet.

[0145] A computing system can include clients and servers. Clients and servers are generally located far apart and typically interact through communication networks. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or cloud host, which is a hosting product within the cloud computing service system to address the shortcomings of traditional physical hosts and VPS services, such as high management difficulty and weak business scalability.

[0146] It should be understood that the various forms of processes shown above can be used to reorder, add, or delete steps. For example, the steps described in the embodiments of this disclosure can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of the embodiments of this disclosure can be achieved, and this document does not impose any limitations.

[0147] The specific embodiments described above do not constitute a limitation on the scope of protection of the embodiments disclosed herein. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the embodiments disclosed herein should be included within the scope of protection of the embodiments disclosed herein.

[0148] This disclosure also provides a computer program product, including a computer program and / or instructions, which, when executed by a processor, implements the load determination method for a wireless access network as provided in any embodiment of this application.

[0149] In implementing a computer program product, computer program code for performing the operations of the embodiments of this disclosure can be written in one or more programming languages ​​or a combination thereof. Programming languages ​​include object-oriented programming languages ​​such as Java, Smalltalk, and C++, as well as conventional procedural programming languages ​​such as C or similar languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0150] Note that the above are merely preferred embodiments and the technical principles applied in this disclosure. Those skilled in the art will understand that this disclosure is not limited to the specific embodiments described herein, and various obvious changes, readjustments, and substitutions can be made without departing from the protection scope of this disclosure. Therefore, although the embodiments of this disclosure have been described in detail above, this disclosure is not limited to the above embodiments. More other equivalent embodiments may be included without departing from the concept of this disclosure, and the scope of this disclosure is determined by the scope of the appended claims.

Claims

1. A method for determining the load of a wireless access network, characterized in that, include: Acquire software and hardware resource data of the wireless access network; Based on the software resource data and hardware resource data, the resource configuration information corresponding to the wireless access network is determined by utilizing the association between software and hardware in the wireless access network; the resource configuration information includes at least the type of baseband processing unit, the number of hardware boards, and the type information of the hardware boards; Obtain load data for each hardware board in the wireless access network; Based on the network structure of the wireless access network, the load data of each hardware board is correlated and summarized to obtain a set of flow control load indicators; the set of flow control load indicators is used to represent the network flow control status of the wireless access network. Based on the resource configuration information and the set of flow control load indicators, the device load of the wireless access network is determined using a preset neural network algorithm.

2. The method according to claim 1, characterized in that, The method further includes: The adjustment strategy is determined based on the device load and the network structure of the wireless access network. The network structure of the wireless access network is updated according to the adjustment strategy described above.

3. The method according to claim 1, characterized in that, The preset neural network algorithm includes: an input layer, a hidden layer, and an output layer; the hidden layer contains at least one neuron.

4. The method according to claim 3, characterized in that, The step of determining the device load status of the wireless access network based on the resource configuration information and the set of flow control load indicators using a preset neural network algorithm includes: The resource configuration information and the set of flow control load indicators are preprocessed to obtain the target data set; The target data set is transmitted to the hidden layer using the input layer; By performing weighted summation and probability transformation on each target data in the target data set through each neuron of the hidden layer, the first load assessment value and the second load assessment value corresponding to each hardware board are obtained respectively. Based on the first load assessment value and the second load assessment value, the output layer is used to determine the device load status of the wireless access network; the device load status includes at least: the load status of the baseband processing unit of the wireless access network, the flow control load data of the baseband board of the wireless access network, and the flow control load status of the main control board of the wireless access network.

5. The method according to claim 4, characterized in that, The step of performing weighted summation and probability transformation on each target data in the target data set through each neuron of the hidden layer to obtain the first load assessment value and the second load assessment value corresponding to each hardware board includes: Based on the weight vectors corresponding to each neuron in the hidden layer, the target data in the target data set are weighted and summed to obtain the weighted sum corresponding to each hidden layer neuron. The activation function is used to perform a probability transformation on each weighted sum to determine the first load assessment value for each board. The second load assessment value for each board is determined based on the first load assessment value for each board.

6. The method according to claim 5, characterized in that, The method further includes: Obtain a training sample set, which includes historical resource configuration information, a set of historical flow control load indicators, and corresponding labels for actual equipment load conditions. The historical resource configuration information and the set of historical flow control load indicators are input into the initial neural network algorithm to obtain the predicted equipment load. Based on the target error, the weight vectors of each neuron in the hidden layer of the initial neural network algorithm are updated and iterated through backpropagation until the target error meets the preset condition, thus obtaining the preset neural network algorithm; the target error is the error between the predicted equipment load condition and the actual equipment load condition label.

7. A load assessment device for a wireless access network, characterized in that, include: The data acquisition module acquires software and hardware resource data of the wireless access network. The resource configuration information determination module is used to determine the resource configuration information corresponding to the wireless access network based on the software resource data and hardware resource data, utilizing the association relationship between software and hardware in the wireless access network; the resource configuration information includes at least the type of baseband processing unit, the number of hardware boards, and the type information of the hardware boards; The load data acquisition module is used to acquire load data from various hardware boards in the wireless access network. The flow control load index set determination module is used to correlate and summarize the load data of each hardware board according to the network structure of the wireless access network to obtain the flow control load index set; the flow control load index set is used to represent the network flow control status of the wireless access network. The device load determination module is used to determine the device load of the wireless access network based on the resource configuration information and the set of flow control load indicators, using a preset neural network algorithm.

8. An electronic device, characterized in that, include: At least one processor; as well as, A memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform the load determination method for a wireless access network as described in any one of claims 1-6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by the processor, the program implements the load determination method for a wireless access network as described in any one of claims 1-6.

10. A computer program product, characterized in that, The computer program product includes a computer program that, when executed by a processor, implements the load determination method for a wireless access network as described in any one of claims 1-6.

Images