CN114268965A - Mobile communication network coverage rate calculation method and device based on deep learning - Google Patents
Mobile communication network coverage rate calculation method and device based on deep learning Download PDFInfo
- Publication number
- CN114268965A CN114268965A CN202111575058.5A CN202111575058A CN114268965A CN 114268965 A CN114268965 A CN 114268965A CN 202111575058 A CN202111575058 A CN 202111575058A CN 114268965 A CN114268965 A CN 114268965A
- Authority
- CN
- China
- Prior art keywords
- sub
- area
- region
- pilot signal
- ith
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D30/00—Reducing energy consumption in communication networks
- Y02D30/70—Reducing energy consumption in communication networks in wireless communication networks
Landscapes
- Mobile Radio Communication Systems (AREA)
Abstract
The invention discloses a mobile communication network coverage rate calculation method and a mobile communication network coverage rate calculation device based on deep learning, wherein the method comprises the following steps: obtaining the pilot signal transmitting power of the ith sub-regionAnd pilot signal transmit power of each neighbor sub-region of the ith sub-regionWherein the parameter niTo representThe number of neighbor sub-regions of the ith sub-region; normalizing the pilot signal transmitting power of the ith sub-area and the neighbor sub-areas of the ith sub-area according to the following formula;whereinDenotes the pilot signal maximum transmission power of the r-th neighbor sub-region of the i-th sub-region,denotes the minimum transmit power of the pilot signal of the r-th neighbor sub-region,representing the normalized pilot signal transmit power of the r-th neighbor sub-region; and inputting the pilot signal transmitting power of the ith sub-area after normalization processing and each neighbor sub-area of the ith sub-area into the trained deep neural network to obtain the communication signal coverage rate of the sub-area network formed by the ith sub-area and each neighbor sub-area of the ith sub-area.
Description
Technical Field
The present application relates to the field of mobile communications technologies, and in particular, to a method and an apparatus for calculating a coverage rate of a mobile communication network based on deep learning.
Background
In the real-time optimization algorithm of the pilot signal transmission power of the mobile communication network, the coverage rate of the communication network is a problem which must be considered. At present, the existing mobile communication network coverage rate calculation method needs to process massive MR data, the processing time of the MR data is in direct proportion to the total amount of the MR data, when the MR data needing to be processed is more, the required calculation amount is correspondingly increased, and the calculation efficiency is lower.
Therefore, it is necessary to provide a new method for calculating the coverage of the mobile communication network, so as to ensure a high accuracy of the calculation result while being capable of calculating the coverage of the mobile communication network quickly.
Disclosure of Invention
The embodiment of the specification provides a method and a device for calculating the coverage rate of a mobile communication network based on deep learning, so that the coverage rate of the mobile communication network can be calculated quickly, and meanwhile, the calculation result is guaranteed to have higher accuracy.
In order to solve the above technical problem, the embodiments of the present specification are implemented as follows:
an embodiment of the present specification provides a method for calculating a coverage rate of a mobile communication network based on deep learning, where the method includes:
step S1, obtaining the pilot signal emission power of the ith sub-area in the areaAnd pilot signal transmission power of each neighbor sub-region of the ith sub-regionq=1,…,ni(ii) a Wherein the parameter niRepresenting the number of neighbor sub-regions of the ith sub-region;
step S2, normalizing the pilot signal transmitting power of the ith sub-area and the neighbor sub-areas of the ith sub-area according to the following formula;
whereinRepresents a pilot signal maximum transmit power of an r-th neighbor sub-region of the i-th sub-region,denotes the pilot signal minimum transmit power of the r-th neighbor sub-region of the i-th sub-region,indicating the normalized pilot signal transmitting power of the r-th neighbor sub-region of the ith sub-region;
step S3, inputting the pilot signal transmitting power of the ith sub-area and each neighbor sub-area of the ith sub-area after normalization processing into the trained deep neural network, and obtaining the communication signal coverage rate of the sub-area network formed by the ith sub-area and each neighbor sub-area of the ith sub-area
The deep neural network is used for calculating the communication signal coverage rate of the subarea network formed by the subareas according to the pilot signal transmitting power of the subareas.
Preferably, the network structure of the deep neural network is specifically as follows:
the number of hidden layers of the deep neural network is m, wherein the number of neurons of the l-th hidden layer is Nl-1,l=1,2,...,m;
The connection weight of the kth neuron of the l-1 layer hidden layer of the deep neural network and the kth neuron of the l layer hidden layer is
preferably, the activation function of the hidden layer of the deep neural network is a Sigmoid activation function.
Preferably, the activation function of the output layer of the deep neural network is a Soft-max activation function.
Preferably, a random gradient descent method SGD is adopted, and a plurality of groups of training data are obtained to train the deep neural network based on the existing mobile communication network coverage rate calculation method based on MR data; any one of the sets of training data includes a number of pilot signal transmit powers and a signal coverage of the number of pilot signal transmit powers.
Preferably, the obtaining the ith sub-region and each neighbor sub-region of the ith sub-regionCommunication signal coverage rate of subarea network formed by areasThen, the method further comprises the following steps:
step S41, presetting a threshold value of the communication signal coverage as α, if soThen the subregion C is considerediNot satisfying the covering condition, and dividing the sub-area CiThe label is Ci *I belongs to I, and I is an identification set of the marked sub-region; taking all the marked sub-areas as nodes,if Ci *And Cs *Are neighbors of each other, then Ci *And Cs *Is connected, Ci *And Cs *One edge exists between the two edges, the set of all the edges is recorded as E, and a graph G is constructed, wherein G is (V, E); finding all connected branches { X in the graph G1,...,XtT is the number of connected branches, and a connected branch X is arrangedlThe node with the lowest coverage rate in t is 1Dividing the sub-regionThe transmitting power of the pilot signal is adjusted up by delta P;
and S42, calculating the communication signal coverage rate of all the sub-areas again by using a deep neural network algorithm according to the pilot signal transmitting power after being adjusted, marking new sub-areas which do not meet the coverage condition, and repeating S41 until all the sub-areas meet the coverage condition.
Meanwhile, the invention also provides a mobile communication network coverage rate calculation device based on deep learning, which is characterized by comprising the following steps:
pilot signal transmissionA power obtaining module for obtaining the pilot signal transmitting power of the ith sub-area in the areaAnd pilot signal transmission power of each neighbor sub-region of the ith sub-regionq=1,…,ni(ii) a Wherein the parameter niRepresenting the number of neighbor sub-regions of the ith sub-region;
a normalization module, configured to normalize the pilot signal transmission power of the ith sub-area and the pilot signal transmission power of the neighboring sub-areas of the ith sub-area according to the following formula;
whereinRepresents a pilot signal maximum transmit power of an r-th neighbor sub-region of the i-th sub-region,denotes the pilot signal minimum transmit power of the r-th neighbor sub-region of the i-th sub-region,indicating the normalized pilot signal transmitting power of the r-th neighbor sub-region of the ith sub-region;
a communication signal coverage rate calculation module, configured to input the pilot signal transmission power of the ith sub-area and each neighbor sub-area of the ith sub-area after normalization processing into the trained deep neural network, so as to obtain a communication signal coverage rate of a sub-area network formed by the ith sub-area and each neighbor sub-area of the ith sub-area
The deep neural network is used for calculating the communication signal coverage rate of the subarea network formed by the subareas according to the pilot signal transmitting power of the subareas.
At least one embodiment provided in this specification can achieve the following advantageous effects:
the technical scheme of the embodiment provides the coverage rate calculation and optimization algorithm based on the deep neural network on the basis of the existing communication network coverage rate calculation method, and the running time of the new algorithm does not depend on the MR data volume, so that the method is more suitable for real-time optimization of a large-scale mobile communication network, and can ensure that the calculation result has higher accuracy while the coverage rate of the mobile communication network is calculated quickly.
Drawings
In order to more clearly illustrate the embodiments of the present disclosure or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments described in the present application, and for those skilled in the art, other drawings can be obtained according to the drawings without any creative effort.
Fig. 1 is a schematic view of an application scenario of a deep learning based mobile communication network coverage calculation method provided in an embodiment of the present specification;
fig. 2 is a flowchart illustrating a method of a deep learning-based coverage calculation method for a mobile communication network according to an embodiment of the present disclosure;
fig. 3 is a network structure diagram of a neural network used in a method for calculating coverage of a mobile communication network based on deep learning according to an embodiment of the present disclosure;
fig. 4 is a schematic structural diagram of a mobile communication network coverage calculation device based on deep learning according to an embodiment of the present specification.
Detailed Description
To make the objects, technical solutions and advantages of one or more embodiments of the present disclosure more apparent, the technical solutions of one or more embodiments of the present disclosure will be described in detail and completely with reference to the specific embodiments of the present disclosure and the accompanying drawings. It is to be understood that the embodiments described are only a few embodiments of the present specification, and not all embodiments. All other embodiments that can be derived by a person skilled in the art from the embodiments given herein without making any creative effort fall within the scope of protection of one or more embodiments of the present specification.
To clearly introduce the technical solution of this embodiment, an application scenario of the technical solution of this embodiment is described below, and fig. 1 is a schematic diagram of an application scenario of a mobile communication network coverage rate calculation method based on deep learning provided in this embodiment of this specification. As shown in fig. 1, a certain area S (e.g. a certain city) includes several sub-areas covered with mobile communication networks, a set B is a set formed by the sub-areas covered with mobile communication networks (communication signals may be 2g, 3g, 4g, or 5g), the set B includes L sub-areas (in this scenario, it is assumed that L is equal to 11) in total, and specifically includes B1、b2、b3、b4、b5、b6、b7、b8、b9、b10And b11As shown in fig. 1, each sub-area biAll have a set of neighbor sub-regions NiI.e. with this sub-region biSets of positionally adjacent subregions, in subregions b1For the sake of example, sub-region b1Of the neighbor sub-region N1Comprising a sub-region b2、b3、b4、b7、b10And sub-region b11. In the technical scheme of the embodiment, each sub-area biAre equipped with communication base stations which transmit mobile signals for user terminal equipment in a sub-area. Meanwhile, the embodiment of the invention assumes that the neighbor relation of the two sub-regions is symmetrical, namely if the sub-region biIs a sub-region bjIs then sub-region bjAnd necessarily also sub-region biOf the network.
In the real-time optimization algorithm of the pilot signal transmission power of the mobile communication network, the signal coverage of the communication network is a problem which must be considered. At present, the existing mobile communication network coverage rate calculation method needs to process massive MR data, the processing time of the MR data is in direct proportion to the total amount of the MR data, when the MR data needing to be processed is more, the required calculation amount is correspondingly increased, and the calculation efficiency is lower.
The scheme provides a coverage rate calculation and optimization algorithm based on a deep neural network on the basis of the existing coverage rate calculation method. The running time of the new algorithm is independent of the MR data volume, so that the method is more suitable for real-time optimization of a large-scale mobile communication network.
The technical concept of the technical solution of the present invention is explained below, and the present invention obtains a large amount of input/output data of pilot channel transmission power and pilot channel signal coverage of each sub-region by using the existing method for calculating mobile communication network coverage based on MR data (see the patent applied by the inventor before, for example, "a method for calculating mobile communication network coverage based on MR data" with patent No. 202111305615.1), trains the deep neural network by using these data, and further models the complex relationship between input (i.e. transmission power of pilot channel of each sub-region) and output (pilot channel signal coverage). Based on the trained deep neural network, a rapid calculation method of the coverage rate of the mobile communication network is provided, and finally, an optimization method of the coverage rate of the mobile communication network is provided by combining a graph theory method.
To describe the technical solution of the present invention in detail, as shown in fig. 2, fig. 2 is a schematic flow chart of a method for calculating a coverage of a mobile communication network based on deep learning according to an embodiment of the present disclosure, and as shown in fig. 2, from a program perspective, an execution subject of the flow may be a program installed on an application server or an application client.
As shown in fig. 2, the process may include the following steps:
step S202: obtaining the pilot signal transmitting power of the ith sub-area in the areaAnd pilot signal transmission power of each neighbor sub-region of the ith sub-regionq=1,…,ni(ii) a Wherein the parameter niRepresenting the number of neighbor sub-regions of the ith sub-region.
Step S204: normalizing the pilot signal transmitting power of the ith sub-area and the neighbor sub-areas of the ith sub-area according to the following formula;
whereinRepresents a pilot signal maximum transmit power of an r-th neighbor sub-region of the i-th sub-region,denotes the pilot signal minimum transmit power of the r-th neighbor sub-region of the i-th sub-region,represents the normalized pilot signal transmit power of the r-th neighbor sub-region of the i-th sub-region.
Step S206: inputting the pilot signal transmitting power of the ith sub-area after normalization processing and each neighbor sub-area of the ith sub-area into the trained deep neural network to obtain the communication signal coverage rate of the sub-area network formed by the ith sub-area and each neighbor sub-area of the ith sub-area
The deep neural network is used for calculating the communication signal coverage rate of the subarea network formed by the subareas according to the pilot signal transmitting power of the subareas.
The technical scheme of the embodiment provides the coverage rate calculation and optimization algorithm based on the deep neural network on the basis of the existing communication network coverage rate calculation method, and the running time of the new algorithm does not depend on the MR data volume, so that the method is more suitable for real-time optimization of a large-scale mobile communication network, and can ensure that the calculation result has higher accuracy while the coverage rate of the mobile communication network is calculated quickly.
Based on the method of fig. 2, the present specification also provides some specific embodiments of the method, which are described below.
In the technical solution of the optional embodiment, the network structure of the deep neural network is specifically as follows:
as shown in fig. 3, the number of hidden layers of the deep neural network is m, where the number of neurons in the l-th hidden layer is Nl-1,l=1,2,...,m;
The connection weight of the kth neuron of the l-1 layer hidden layer of the deep neural network and the kth neuron of the l layer hidden layer is
in the technical solution of an optional embodiment, the activation function of the hidden layer of the deep neural network is a Sigmoid activation function.
In the technical scheme of the optional embodiment, the activation function of the output layer of the deep neural network is a Soft-max activation function.
In particular, considering the practical application of increasing the total power of the pilot channel transmission of any cell, the signal coverage of the pilot channel of the mobile communication network is increased, i.e. the signal coverage of the pilot channel of the mobile communication network is increasedIs aboutj is 0, …, n. a monotone increasing function, and in order to build a more reasonably practical model, the built deep neural network should be a monotone neural network model. Therefore, let the weight of the k-th neuron in layer l-1 and the j-th neuron in layer l beLet the bias of the jth neuron in the l layer beThe hidden layer activation function is a Sigmoid function, so that monotonicity can be kept, that is:
typically, the l-th layer has Nl-1Output of a neuron, i.e., the jth neuron at layer l +1Comprises the following steps:
Input and output are more compactly represented using a matrix: layer I has Nl-1One neuron, and the l +1 th layer has NlEach neuron, the weight from the l-th layer to the l + 1-th layer forms an Nl-1×NlMatrix W ofl+1The bias of the l +1 th layer is NlVector b of x 1l+1. Output of the l-th layer is Nl-1Vector a of x 1lThe output of the l +1 th layer is NlVector a of x 1l+1Then a isl+1Comprises the following steps:
al+1=σ(Wl+1al+bl+1)
thus, the final output am+2Comprises the following steps:
am+2=σ(Wm+2(σ(Wm+1(…)+bm+1))+bm+2)
then inputting a training data set, training a neural network, wherein the forward propagation calculates the output of the training samples, and measuring the loss between the output of the training samples and the real value by using a cross entropy loss function J (W, b, x, y), wherein x is input data, y is the actual output of the samples, W is a weight matrix of all the hidden layers and the output layers, and b is the bias of all the hidden layers and the output layers.
J(W,b,x,y)=-ylnam+2-(1-y)ln(1-am+2)
Minimizing the loss function J (W, b, x, y) is required to make the predicted output value of the training sample as close as possible to the actual output value of the sample. Optimizing W, b for each layer iteration by gradient descent method to minimize loss function, and recording linear output of layer I before activation as Nl-1Vector z of x 1l,zl=Wlal-1+blThe output layer gradient case is as follows:
wherein, order:
then:
further up layer by layer recursion, for the l-th layer:
while
Therefore, the gradient of each layer w and b can be solved by recursion through a mathematical induction method, iterative optimization is carried out until the variation value of w and b is smaller than a threshold value, and the output w and b are proper weight and bias.
In the technical scheme of the optional embodiment, a random gradient descent method SGD is adopted, and a plurality of groups of training data are obtained to train the deep neural network based on the existing mobile communication network coverage rate calculation method based on MR data; any one of the sets of training data includes a number of pilot signal transmit powers and a signal coverage of the number of pilot signal transmit powers.
Then, using the trained and verified deep neural network, for each cell CiTransmitting total power according to input pilot channelj-0.. n, calculating the signal coverage of the pilot channel. The specific process is as follows:
the weight value of the k-th neuron of the trained l layer and the j-th neuron of the l +1 layer is set asLet the bias of the jth neuron in the l +1 th layer beThe matrix form is weight matrix from the l layer to the l +1 layerBias of the l +1 th layer is NlVector of x 1The output of the l +1 th layerIf l is 1, then
And (3) final output:
in the technical solution of the optional embodiment, the communication signal coverage of the sub-area network formed by the ith sub-area and each neighboring sub-area of the ith sub-area is obtainedThen, the method further comprises the following steps:
step S41, presetting TongThe threshold value of the signal coverage is alpha ifThen the subregion C is considerediNot satisfying the covering condition, and dividing the sub-area CiThe label is Ci *I belongs to I, and I is an identification set of the marked sub-region; taking all the marked sub-areas as nodes,if Ci *And Cs *Are neighbors of each other, then Ci *And Cs *Is connected, Ci *And Cs *One edge exists between the two edges, the set of all the edges is recorded as E, and a graph G is constructed, wherein G is (V, E); finding all connected branches { X in the graph G1,...,XtT is the number of connected branches, and a connected branch X is arrangedlThe node with the lowest coverage rate in t is 1Dividing the sub-regionThe transmitting power of the pilot signal is adjusted up by delta P;
and S42, calculating the communication signal coverage rate of all the sub-areas again by using a deep neural network algorithm according to the pilot signal transmitting power after being adjusted, marking new sub-areas which do not meet the coverage condition, and repeating S41 until all the sub-areas meet the coverage condition.
Meanwhile, the present invention also provides a mobile communication network coverage rate calculation device based on deep learning, as shown in fig. 4, the device includes:
a pilot signal transmission power obtaining module 402, configured to obtain the pilot signal transmission power of the ith sub-zone in the zoneAnd each neighbor sub-of the ith sub-regionPilot signal transmit power of a regionq=1,...,ni(ii) a Wherein the parameter niRepresenting the number of neighbor sub-regions of the ith sub-region.
A normalization module 404, configured to normalize the pilot signal transmission power of the ith sub-area and the pilot signal transmission power of the neighboring sub-areas of the ith sub-area according to the following formula;
whereinRepresents a pilot signal maximum transmit power of an r-th neighbor sub-region of the i-th sub-region,denotes the pilot signal minimum transmit power of the r-th neighbor sub-region of the i-th sub-region,indicating the normalized pilot signal transmitting power of the r-th neighbor sub-region of the ith sub-region;
a communication signal coverage rate calculation module 406, configured to input the pilot signal transmission power of the ith sub-region and each neighbor sub-region of the ith sub-region after normalization processing into the trained deep neural network, so as to obtain a communication signal coverage rate of a sub-region network formed by the ith sub-region and each neighbor sub-region of the ith sub-region
The deep neural network is used for calculating the communication signal coverage rate of the subarea network formed by the subareas according to the pilot signal transmitting power of the subareas.
The foregoing description has been directed to specific embodiments of this disclosure. Other embodiments are within the scope of the following claims. In some cases, the actions or steps recited in the claims may be performed in a different order than in the embodiments and still achieve desirable results. In addition, the processes depicted in the accompanying figures do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing may also be possible or may be advantageous.
In the 90 s of the 20 th century, improvements in a technology could clearly distinguish between improvements in hardware (e.g., improvements in circuit structures such as diodes, transistors, switches, etc.) and improvements in software (improvements in process flow). However, as technology advances, many of today's process flow improvements have been seen as direct improvements in hardware circuit architecture. Designers almost always obtain the corresponding hardware circuit structure by programming an improved method flow into the hardware circuit. Thus, it cannot be said that an improvement in the process flow cannot be realized by hardware physical modules. For example, a Programmable Logic Device (PLD), such as a Field Programmable Gate Array (FPGA), is an integrated circuit whose Logic functions are determined by programming the Device by a user. A digital character system is "integrated" on a PLD by the designer's own programming without requiring the chip manufacturer to design and fabricate a dedicated integrated circuit chip. Furthermore, nowadays, instead of manually manufacturing an Integrated Circuit chip, such Programming is often implemented by "logic compiler" software, which is similar to a software compiler used in program development and writing, but the original code before compiling is also written by a specific Programming Language, which is called Hardware Description Language (HDL), and HDL is not only one but many, such as abll (advanced desktop Expression Language), ahdl (alternate Hardware Description Language), traffic, pl (core universal Programming Language), HDCal (jhdware Description Language), lang, Lola, HDL, laspam, hardward Description Language (vhr Description Language), and vhjlang (Hardware Description Language), which are currently used in most general. It will also be apparent to those skilled in the art that hardware circuitry that implements the logical method flows can be readily obtained by merely slightly programming the method flows into an integrated circuit using the hardware description languages described above.
The controller may be implemented in any suitable manner, for example, the controller may take the form of, for example, a microprocessor or processor and a computer-readable medium storing computer-readable program code (e.g., software or firmware) executable by the (micro) processor, logic gates, switches, an Application Specific Integrated Circuit (ASIC), a programmable logic controller, and an embedded microcontroller, examples of which include, but are not limited to, the following microcontrollers: ARC 625D, AtmelAT91SAM, Microchip PIC18F26K20, and Silicone Labs C8051F320, the memory controller may also be implemented as part of the control logic for the memory. Those skilled in the art will also appreciate that, in addition to implementing the controller as pure computer readable program code, the same functionality can be implemented by logically programming method steps such that the controller is in the form of logic gates, switches, application specific integrated circuits, programmable logic controllers, embedded microcontrollers and the like. Such a controller may thus be considered a hardware component, and the means included therein for performing the various functions may also be considered as a structure within the hardware component. Or even means for performing the functions may be regarded as being both a software module for performing the method and a structure within a hardware component.
The systems, devices, modules or units illustrated in the above embodiments may be implemented by a computer chip or an entity, or by a product with certain functions. One typical implementation device is a computer. In particular, the computer may be, for example, a personal computer, a laptop computer, a cellular telephone, a camera phone, a smartphone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or a combination of any of these devices.
For convenience of description, the above devices are described as being divided into various units by function, and are described separately. Of course, the functionality of the units may be implemented in one or more software and/or hardware when implementing the present application.
As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
In a typical configuration, a computing device includes one or more processors (CPUs), input/output interfaces, network interfaces, and memory.
The memory may include forms of volatile memory in a computer readable medium, Random Access Memory (RAM) and/or non-volatile memory, such as Read Only Memory (ROM) or flash memory (flash RAM). Memory is an example of a computer-readable medium.
Computer-readable media, including both non-transitory and non-transitory, removable and non-removable media, may implement information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), Static Random Access Memory (SRAM), Dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), Read Only Memory (ROM), Electrically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), Digital Versatile Disks (DVD) or other optical storage, magnetic cassettes, magnetic tape disk storage or other magnetic storage devices, or any other non-transmission medium which can be used to store information which can be accessed by a computing device. As defined herein, a computer readable medium does not include a transitory computer readable medium such as a modulated data signal and a carrier wave.
It should also be noted that the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element.
As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The application may be described in the general context of computer-executable instructions, such as program modules, being executed by a computer. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform particular tasks or implement particular abstract data types. The application may also be practiced in distributed computing environments where tasks are performed by remote processing devices that are linked through a communications network. In a distributed computing environment, program modules may be located in both local and remote computer storage media including memory storage devices.
The above description is only an example of the present application and is not intended to limit the present application. Various modifications and changes may occur to those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the scope of the claims of the present application.
Claims (7)
1. A method for calculating the coverage rate of a mobile communication network based on deep learning is characterized by comprising the following steps:
step S1, obtaining the pilot signal emission power of the ith sub-area in the areaAnd pilot signal transmission power of each neighbor sub-region of the ith sub-regionWherein the parameter niRepresenting the number of neighbor sub-regions of the ith sub-region;
step S2, normalizing the pilot signal transmitting power of the ith sub-area and the neighbor sub-areas of the ith sub-area according to the following formula;
whereinRepresents a pilot signal maximum transmit power of an r-th neighbor sub-region of the i-th sub-region,denotes the pilot signal minimum transmit power of the r-th neighbor sub-region of the i-th sub-region,indicating the normalized pilot signal transmitting power of the r-th neighbor sub-region of the ith sub-region;
step S3, inputting the pilot signal transmitting power of the ith sub-area and each neighbor sub-area of the ith sub-area after normalization processing into the trained deep neural network, and obtaining the communication signal coverage rate of the sub-area network formed by the ith sub-area and each neighbor sub-area of the ith sub-area
The deep neural network is used for calculating the communication signal coverage rate of the subarea network formed by the subareas according to the pilot signal transmitting power of the subareas.
2. The method for calculating the coverage rate of the mobile communication network based on the deep learning of claim 1, wherein the network structure of the deep neural network is as follows:
the number of hidden layers of the deep neural network is m, wherein the number of neurons of the l-th hidden layer is Nl-1,l=1,2,...,m;
The connection weight of the kth neuron of the l-1 layer hidden layer of the deep neural network and the kth neuron of the l layer hidden layer is
3. the deep learning-based mobile communication network coverage calculation method according to claim 2, wherein the activation function of the hidden layer of the deep neural network is a Sigmoid activation function.
4. The method for calculating the coverage rate of the mobile communication network based on the deep learning of claim 2, wherein the activation function of the output layer of the deep neural network is a Soft-max activation function.
5. The method for calculating the coverage rate of the mobile communication network based on the deep learning of claim 2, wherein a random gradient descent (SGD) method is adopted, and a plurality of groups of training data are obtained to train the deep neural network based on the existing mobile communication network coverage rate calculation method based on MR data; any one of the sets of training data includes a number of pilot signal transmit powers and a signal coverage of the number of pilot signal transmit powers.
6. The method according to claim 1, wherein the communication signal coverage of the sub-area network composed of the ith sub-area and each neighboring sub-area of the ith sub-area is obtainedThen, the method further comprises the following steps:
step S41, presetting a threshold value of the communication signal coverage as α, if soThen the subregion C is considerediNot satisfying the covering condition, and dividing the sub-area CiMarking asI is an identification set of marked sub-regions; taking all the marked sub-areas as nodes,if it isAndfor mutual neighbors, thenAndis communicated with the air inlet pipe and the air outlet pipe,andone edge exists between the two edges, the set of all the edges is recorded as E, and a graph G is constructed, wherein G is (V, E); finding all connected branches { X in the graph G1,...,XtT is the number of connected branches, and a connected branch X is arrangedlThe node with the lowest coverage rate in t is 1Dividing the sub-regionThe transmitting power of the pilot signal is adjusted up by delta P;
and S42, calculating the communication signal coverage rate of all the sub-areas again by using a deep neural network algorithm according to the pilot signal transmitting power after being adjusted, marking new sub-areas which do not meet the coverage condition, and repeating S41 until all the sub-areas meet the coverage condition.
7. An apparatus for computing a coverage rate of a mobile communication network based on deep learning, the apparatus comprising:
a pilot signal emission power obtaining module for obtaining the pilot signal emission power of the ith sub-zone in the zoneAnd pilot signal transmission power of each neighbor sub-region of the ith sub-regionWhereinParameter niRepresenting the number of neighbor sub-regions of the ith sub-region;
a normalization module, configured to normalize the pilot signal transmission power of the ith sub-area and the pilot signal transmission power of the neighboring sub-areas of the ith sub-area according to the following formula;
whereinRepresents a pilot signal maximum transmit power of an r-th neighbor sub-region of the i-th sub-region,denotes the pilot signal minimum transmit power of the r-th neighbor sub-region of the i-th sub-region,indicating the normalized pilot signal transmitting power of the r-th neighbor sub-region of the ith sub-region;
a communication signal coverage rate calculation module, configured to input the pilot signal transmission power of the ith sub-area and each neighbor sub-area of the ith sub-area after normalization processing into the trained deep neural network, so as to obtain a communication signal coverage rate of a sub-area network formed by the ith sub-area and each neighbor sub-area of the ith sub-area
The deep neural network is used for calculating the communication signal coverage rate of the subarea network formed by the subareas according to the pilot signal transmitting power of the subareas.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202111575058.5A CN114268965B (en) | 2021-12-21 | 2021-12-21 | Mobile communication network coverage rate calculation method and device based on deep learning |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202111575058.5A CN114268965B (en) | 2021-12-21 | 2021-12-21 | Mobile communication network coverage rate calculation method and device based on deep learning |
Publications (2)
Publication Number | Publication Date |
---|---|
CN114268965A true CN114268965A (en) | 2022-04-01 |
CN114268965B CN114268965B (en) | 2023-09-08 |
Family
ID=80828395
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202111575058.5A Active CN114268965B (en) | 2021-12-21 | 2021-12-21 | Mobile communication network coverage rate calculation method and device based on deep learning |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN114268965B (en) |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2016115394A1 (en) * | 2015-01-16 | 2016-07-21 | Hughes Network Systems, Llc | Method and apparatus for beam selection for a multibeam multi-satellite communications system |
CN108834100A (en) * | 2018-06-15 | 2018-11-16 | 九江学院 | A kind of efficient broadcast routing method |
CN111226254A (en) * | 2017-07-10 | 2020-06-02 | 埃奈迪斯公司 | Reliability of network maps |
CN111882041A (en) * | 2020-07-31 | 2020-11-03 | 国网重庆市电力公司电力科学研究院 | Power grid attack detection method and device based on improved RNN (neural network) |
CN112770398A (en) * | 2020-12-18 | 2021-05-07 | 北京科技大学 | Far-end radio frequency end power control method based on convolutional neural network |
CN113706333A (en) * | 2021-08-10 | 2021-11-26 | 国网电子商务有限公司 | Method and system for automatically generating topology island of power distribution network |
-
2021
- 2021-12-21 CN CN202111575058.5A patent/CN114268965B/en active Active
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2016115394A1 (en) * | 2015-01-16 | 2016-07-21 | Hughes Network Systems, Llc | Method and apparatus for beam selection for a multibeam multi-satellite communications system |
CN111226254A (en) * | 2017-07-10 | 2020-06-02 | 埃奈迪斯公司 | Reliability of network maps |
CN108834100A (en) * | 2018-06-15 | 2018-11-16 | 九江学院 | A kind of efficient broadcast routing method |
CN111882041A (en) * | 2020-07-31 | 2020-11-03 | 国网重庆市电力公司电力科学研究院 | Power grid attack detection method and device based on improved RNN (neural network) |
CN112770398A (en) * | 2020-12-18 | 2021-05-07 | 北京科技大学 | Far-end radio frequency end power control method based on convolutional neural network |
CN113706333A (en) * | 2021-08-10 | 2021-11-26 | 国网电子商务有限公司 | Method and system for automatically generating topology island of power distribution network |
Non-Patent Citations (1)
Title |
---|
马晓雷,刘元安,高锦春,沈雅琴: "无线Ad Hoc网络功率控制透明传输调度算法", 北京邮电大学学报, no. 05 * |
Also Published As
Publication number | Publication date |
---|---|
CN114268965B (en) | 2023-09-08 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Wu et al. | $ L1 $-norm batch normalization for efficient training of deep neural networks | |
Chen et al. | ReGAN: A pipelined ReRAM-based accelerator for generative adversarial networks | |
Moons et al. | Embedded deep learning | |
US20210065058A1 (en) | Method, apparatus, device and readable medium for transfer learning in machine learning | |
CN112329680B (en) | Semi-supervised remote sensing image target detection and segmentation method based on class activation graph | |
CN113435585B (en) | Service processing method, device and equipment | |
Vong et al. | Imbalanced learning for air pollution by meta-cognitive online sequential extreme learning machine | |
WO2019040672A1 (en) | Systems and methods for determining circuit-level effects on classifier accuracy | |
CN112200132A (en) | Data processing method, device and equipment based on privacy protection | |
Yan et al. | Small watershed stream-flow forecasting based on LSTM | |
CN117223009A (en) | Performance scaling of data stream deep neural network hardware accelerators | |
CN111458030A (en) | Infrared human body temperature measurement calibration method and device | |
CN115712866A (en) | Data processing method, device and equipment | |
CN111652286A (en) | Object identification method, device and medium based on graph embedding | |
CN116304720A (en) | Cost model training method and device, storage medium and electronic equipment | |
US20200167402A1 (en) | Analog circuit for softmax function | |
CN113837635A (en) | Risk detection processing method, device and equipment | |
CN114268965A (en) | Mobile communication network coverage rate calculation method and device based on deep learning | |
CN117036829A (en) | Method and system for achieving label enhancement based on prototype learning for identifying fine granularity of blade | |
CN111191090A (en) | Method, device, equipment and storage medium for determining service data presentation graph type | |
CN111914083A (en) | Statement processing method, device and storage medium | |
CN115618748A (en) | Model optimization method, device, equipment and storage medium | |
CN115496162A (en) | Model training method, device and equipment | |
CN111062477B (en) | Data processing method, device and storage medium | |
CN114861665B (en) | Method and device for training reinforcement learning model and determining data relation |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |