Energy efficiency optimization method for low power consumption power ad hoc network

Abstract

The application relates to a low-power-consumption energy efficiency optimization method for a power ad hoc network, a power system, computer equipment, a computer readable storage medium and a computer program product. The method comprises the following steps: based on a power intelligent sensor network, an energy collection model and a signal-to-noise ratio model are constructed; according to the energy collection model and the signal-to-noise ratio model, a target function is constructed with the maximum network energy efficiency as a target, and constraint conditions of the target function are set; a polyhedral model is constructed, sample data of an independent and identically distributed imperfect channel state are used to learn the polyhedral model, and channel gain information is output; according to the channel gain information, the target function is solved based on the constraint conditions of the target function, distribution power of each pair of sensors on different sub-channels and binary selection information of each pair of sensors on the sub-channels are determined, and a resource allocation scheme is obtained. The method can reduce the influence of imperfect CSI, maximize the network throughput, and guarantee the energy efficiency and data transmission quality.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of power grid optimization, in particular to a low-power-consumption power ad hoc network energy efficiency optimization method, a power system, a computer device, a computer readable storage medium and a computer program product. BACKGROUND

[0002] In recent years, China attaches great importance to digital development and has started the vigorous promotion of "Digital China" construction, which puts forward higher requirements for the application of digital technologies including intelligent sensing in the field of electric power industry. Based on this, the grand vision of the next generation of smart grid, "transparent grid", is proposed: the future of the power grid is a transparent grid, which effectively integrates sensor technology, information technology, data communication technology, electronic control technology, artificial intelligence, etc. in the power system, and can realize the transparency of device state, operation state and transaction state.

[0003] At present, the main online monitoring system adopts a large number of discrete, single sensors to monitor the operation state of the power system, which cannot realize effective integration and information sharing at the sensor level, making the development of multi-parameter information fusion sharing and comprehensive diagnosis slow, and cannot meet the development requirements of smart grid. The multi-integrated sensor adopts miniaturized integration, multi-physical quantity synchronous detection, full-scale measurement, and on-demand analog / digital mode to provide technical solutions, integrates sensitive elements, signal conditioning circuits, microprocessors and communication interface modules together, and has the intelligent trend of multi-environment physical quantity sensing and sensor information fusion. The intelligentization and informatization of smart grid need to configure a large number of sensors and data acquisition modules, how to organically integrate sensors and corresponding data acquisition and transmission systems to form a multi-physical quantity integrated sensor array and multi-dimensional information comprehensive analysis technology for efficient state analysis and fault diagnosis of equipment is a hot research direction of digital power grid construction at present.

[0004] By 2020, the total mileage of China's power transmission lines will increase from 1.15 million kilometers in 2014 to more than 1.59 million kilometers, and the mileage of power transmission lines will continue to increase. The contradiction between the rapid growth of power grid operation and maintenance line mileage and the relatively insufficient number of power grid operation and maintenance personnel gradually appears. With the development of smart grid construction and artificial intelligence technology, power transmission line inspection has transitioned from traditional "human patrol" to "human patrol + machine patrol", but the flight capability of unmanned aerial vehicles is limited by weather and endurance batteries, and there are also problems such as limited monitoring quantity, high technical requirements for operators, short patrol distance, influence of no-fly zones, etc., which cannot fully cover the monitoring needs of power transmission lines in a real-time manner; there is a problem of how to realize data fusion of multi-integrated sensors and long-distance transmission back to the server.

[0005] Therefore, in order to further improve the observability of power grid operation and enhance the information perception ability of transmission line, it is necessary to study multi-sensor integration technology to realize all-around and multi-angle monitoring of transmission line. Through the study of multi-physical quantity integrated sensor technology, the massive deployment of integrated sensor elements is realized. The study of multi-physical quantity integrated sensor distributed self-organizing network data communication technology completes the networking communication between multiple multi-physical quantity integrated sensor devices. The study of multi-physical quantity data analysis, feature extraction technology and auxiliary decision-making technology of integrated sensor, the development of multi-physical quantity data fusion application system, and the development of transmission line dynamic capacity monitoring software module, promote the application of artificial intelligence technology in massive monitoring data of transmission line, solve the actual needs of operation and maintenance of transmission line, realize the real-time return of measurement data with high efficiency and low delay to server processing and real-time monitoring of the working state of transmission line, achieve the goal of automatic, information-based and digitalized intelligent monitoring of transmission line, and finally realize the goal of transparent perception of power grid.

[0006] The multi-physical quantity integrated sensor of transmission line (multi-bundle, supporting AC line and spacer above four bundles) is a kind of intelligent housekeeper for panoramic real-time monitoring and intelligent early warning of overhead transmission line. The wide-range induction self-power supply technology is adopted to ensure the continuous and reliable operation of the device. The distributed wide-range induction power supply module provides continuous and reliable power supply for the sensor, realizes real-time online intelligent monitoring of low-load line, and has the advantages of easy expansion and high reliability.

[0007] The multi-bundle version can be directly installed on the 500kV transmission line by integrating with the spacer. The double-bundle and single-conductor version adopts an expandable and lightweight integrated design and can be directly installed on the 110 / 220kV transmission line. The power measurement gripper is suitable for various types of bundle conductors. The independent measurement power gripper structure is added, which is connected to the monitoring host through flexible connection lines, integrates the sub-conductor current and temperature collection module, and ensures that the device and the conductor are at the same potential. The modular integrated monitoring host includes visible light / ultraviolet / infrared camera, environmental temperature and humidity sensor, air pressure sensor, acceleration sensor, attitude sensor, high-precision Beidou positioning module, etc. The device supports the access of Internet of Things platform, file server and video server, and has the functions of online fusion perception and intelligent identification and alarm of conductor temperature / current / sag / dance, environmental temperature / humidity / pressure / altitude, channel visible light image / video, ultraviolet / infrared (optional) and other physical quantities.

[0008] In recent years, with the continuous improvement of the level of urban intelligence, various sensing devices and automation control systems in smart grid have been widely applied. Smart grid realizes the efficient operation and management of power systems by integrating advanced sensing, communication and control technologies. Wireless communication-enabled power smart sensor networks can provide diversified monitoring and management services for urban power systems. However, with the surge of diversified services, power smart sensor networks face the problem of spectrum resource scarcity. Cognitive radio (CR) based on power smart sensor networks is considered as a key technology to alleviate the problem of spectrum scarcity. CR allows unlicensed secondary user sensors to share spectrum with licensed primary user sensors, thereby meeting the spectrum needs of power smart sensor networks.

[0009] At the same time, the CR-based power smart sensor network system needs efficient data exchange between different sensor nodes, so it is necessary to improve the network flexibility of the power smart sensor network. Device to device (D2D) communication technology is a key technology for building a highly flexible network, which realizes direct service communication without data forwarding through the base station, greatly improving the efficiency of flexible networking.

[0010] With the high-speed communication of D2D technology in power smart sensor networks, business traffic will increase rapidly, and the energy consumption of sensor nodes will also increase, which makes energy efficiency an important research topic in smart grid scenarios. In order to prolong the working time of sensor nodes, energy harvesting (EH) technology is widely used in power smart sensor networks. EH technology allows sensor nodes to obtain energy from the environment, such as solar energy, wind energy, vibration energy, etc., thereby reducing dependence on external power sources. By combining energy harvesting technology, power smart sensor networks can maximize network energy efficiency (EE) while ensuring data transmission quality.

[0011] Currently, there are two aspects of research on power smart sensor networks combining D2D technology, energy harvesting technology and spectrum sharing: one focuses on the impact of energy on transmission efficiency; the other focuses on the quality of data transmission. In order to improve transmission efficiency, traditional research mostly considers the energy problem of D2D technology, and pays less attention to the quality of data transmission, which may lead to a decrease in data transmission reliability. There are also some researches that only focus on the quality of D2D data transmission, and pay less attention to energy efficiency, which may lead to increased energy consumption and reduced working time of sensor nodes.

[0012] In addition, the existence of imperfect channel state information (CSI) caused by the complexity of the environment affects the optimization effect of the smart grid. At present, few studies focus on the channel information uncertainty problem in the power smart sensor network system. Therefore, the main challenge faced by the power smart sensor network using D2D technology, energy harvesting technology and spectrum sharing is how to ensure energy efficiency and data transmission quality under the condition of uncertain channel information. SUMMARY

[0013] Therefore, it is necessary to provide a low-power energy-efficient optimization method for power ad hoc networks, a power system, a computer device, a computer readable storage medium and a computer program product to reduce the influence of imperfect CSI, maximize network throughput, and ensure energy efficiency and data transmission quality.

[0014] In a first aspect, the present application provides a low-power energy-efficient optimization method for power ad hoc networks, comprising:

[0015] Based on the power smart sensor network, an energy harvesting model and a signal-to-noise ratio model are constructed. The power smart sensor network includes a main sensor network and a secondary user network. The main sensor network includes a base station and M main sensors, and the secondary user network is composed of N pairs of secondary sensors for device-to-device communication. The energy harvesting model is used to represent the energy collected at the nth pair of secondary sensors. The signal-to-noise ratio model is used to represent the signal-to-noise ratio of the nth pair of secondary sensors.

[0016] According to the energy harvesting model and the signal-to-noise ratio model, a target function is constructed to maximize network energy efficiency, and the constraint conditions of the target function are set.

[0017] A polyhedral model is constructed, and the polyhedral model is learned using sample data of independent and identically distributed imperfect channel states to output channel gain information.

[0018] According to the channel gain information, the target function is solved based on the constraint conditions of the target function to determine the allocation power of each pair of secondary sensors on different sub-channels and the binary selection information of each pair of secondary sensors on the sub-channels, and a resource allocation scheme is obtained.

[0019] In one embodiment, the energy harvesting model is represented by the following formula (1):

[0020]

[0021] In formula (1), P M , α and β are constants, P M represents the maximum harvesting power when the actual energy harvesting circuit reaches saturation, and α and β are determined by the characteristics of the actual energy harvesting circuit. denotes the input power of the nth pair of secondary sensors, and is represented by the following equation (2):

[0022]

[0023] In equation (2), θ denotes a power allocation factor, denotes the transmission power of the nth pair of secondary sensors on the subchannel occupied by the mth primary sensor, denotes the channel gain of the mth primary sensor on the subchannel, denotes the channel gain of the nth pair of secondary sensors to the mth primary sensor, denotes the channel gain of the primary sensor.

[0024] In one embodiment, the signal-to-noise ratio model is represented by the following equation (3):

[0025]

[0026] In equation (3), denotes the transmission power of the nth pair of secondary sensors on the subchannel occupied by the mth primary sensor, denotes the channel gain of the mth primary sensor on the subchannel, denotes the channel gain of the nth pair of secondary sensors to the mth primary sensor, denotes the channel gain of the primary sensor; α i,m is a user association factor between the ith pair of secondary sensors and the mth primary sensor, and when α i,m = 1, it indicates that the ith pair of secondary sensors occupies the channel of the mth primary sensor, and when α i,m = 0, it indicates that the ith pair of secondary sensors does not occupy the channel of the mth primary sensor; σ 2 denotes the noise power.

[0027] In one embodiment, the objective function is represented by the following equation (4):

[0028]

[0029] In equation (4), denotes the transmission power of the nth pair of secondary sensors on the subchannel occupied by the mth primary sensor; denotes the data rate in the secondary user network, and is represented by the following equation (5):

[0030]

[0031] In equation (5), the signal-to-noise ratio of the nth pair of secondary sensors;

[0032] The constraint conditions of the objective function include a signal-to-noise ratio constraint condition, an energy collection limit condition, a secondary sensor power limit condition, and a subchannel constraint condition.

[0033] The signal-to-noise ratio constraint condition is represented by the following formula (6):

[0034]

[0035] In formula (6), γ th is a set threshold value;

[0036] The energy collection limit condition is represented by the following formula (7):

[0037] E min ≤ E eh (7).

[0038] In formula (7), E min is the minimum energy required by the secondary sensor;

[0039] The secondary sensor power limit condition is represented by the following formula (8):

[0040]

[0041] In formula (8), P max is the power limit value of the secondary sensor;

[0042] The subchannel constraint condition is represented by the following formula (9):

[0043]

[0044] In formula (9), α n，m is a user association factor between the nth pair of secondary sensors and the mth primary sensor.

[0045] In one embodiment, the polyhedral model is represented by the following formula (10):

[0046]

[0047] In formula (10), C represents the center of the polyhedron; λ>0 represents the size of the polyhedron, and S represents the uncertain channel state region learned from sample data of independent and identically distributed imperfect channel states; the channel gain information output by the polyhedral model is and represents the channel gain from the nth pair of secondary sensors to the mth primary sensor; represents the channel gain of the primary sensor.

[0048] In one of the embodiments, the polyhedral model is learned by using sample data of independent and identically distributed imperfect channel states, and output channel gain information includes:

[0049] The sample data of independent and identically distributed imperfect channel states is divided into two parts to obtain first sample data and second sample data;

[0050] The first sample data is used to approximate the shape of the high probability region, and the center of the polyhedron is set by the sample mean value of the first sample data, and the parameters of and are determined;

[0051] The size of the polyhedron is calibrated by using the second sample data to determine the parameter value of λ;

[0052] In the determined polyhedral model, the channel gain information is determined and output and

[0053] In a second aspect, the application further provides a power system, wherein a power intelligent sensor network is arranged in the power system, the power intelligent sensor network includes a main sensor network and a secondary user network, the main sensor network includes a base station and M main sensors, and the secondary user network is composed of N pairs of secondary sensors for device-to-device communication.

[0054] The secondary sensor performs resource configuration based on the resource allocation scheme determined by the method of the first aspect.

[0055] In a third aspect, the application further provides a computer device, including a memory and a processor, the memory stores a computer program, and the processor implements the steps of the method of the first aspect when executing the computer program.

[0056] In a fourth aspect, the application further provides a computer readable storage medium, which stores a computer program, and the computer program is executed by a processor to implement the steps of the method of the first aspect.

[0057] In a fifth aspect, the application further provides a computer program product, including a computer program, and the computer program is executed by a processor to implement the steps of the method of the first aspect.

[0058] The energy efficiency optimization method, the power system, the computer device, the computer readable storage medium and the computer program product of the low-power power ad hoc network optimize the energy efficiency of the low-power power ad hoc network based on the power intelligent sensor network, construct an energy collection model and a signal-to-noise ratio model, the power intelligent sensor network includes a main sensor network and a secondary user network, the main sensor network includes a base station and M main sensors, and the secondary user network is composed of N pairs of secondary sensors for device-to-device communication; the energy collection model is used to represent the energy collected at the nth pair of secondary sensors; the signal-to-noise ratio model is used to represent the signal-to-noise ratio of the nth pair of secondary sensors; according to the energy collection model and the signal-to-noise ratio model, a target function is constructed with the maximum network energy efficiency as the target, and the constraint condition of the target function is set; a polyhedral model is constructed, the polyhedral model is learned by using sample data of independent and identically distributed imperfect channel states, and channel gain information is output; according to the channel gain information, the constraint condition of the target function is used to solve the target function, the allocation power of each pair of secondary sensors on different sub-channels and the binary selection information of each pair of secondary sensors to the sub-channels are determined, and a resource allocation scheme is obtained. In this way, the energy collection process of the secondary sensor is modeled, the energy collection optimization model under the condition of imperfect CSI is established, and the energy efficiency of the network is maximized; in view of the problem that the target function is difficult to solve due to the imperfect CSI, the influence of the imperfect CSI is reduced by constructing the polyhedral model and learning the sample data, a more optimal resource allocation scheme is output, the network throughput is maximized, and the energy efficiency and data transmission quality are ensured. BRIEF DESCRIPTION OF DRAWINGS

[0059] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the related art, the drawings needed to be used in the description of the embodiments of the present application or the related art will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and other related drawings can be obtained by those skilled in the art without creating labor.

[0060] Figure 1 A flowchart of the energy efficiency optimization method of the low-power power ad hoc network in an embodiment;

[0061] Figure 2 A schematic diagram of the composition of the power intelligent sensor network in an embodiment;

[0062] Figure 3 A schematic diagram of the relationship between the noise power and the total throughput of the secondary network in an embodiment;

[0063] Figure 4 A schematic diagram of the relationship between the noise power and the energy efficiency of the secondary network in an embodiment;

[0064] Figure 5A diagram showing the relationship between the SINR threshold of a secondary sensor and the secondary network throughput in one embodiment;

[0065] Figure 6 A diagram showing the relationship between the SINR threshold of a secondary sensor and the secondary network energy efficiency in one embodiment;

[0066] Figure 7 An internal structure diagram of a computer device in one embodiment. DETAILED DESCRIPTION

[0067] In order to make the purposes, technical solutions and advantages of the present application clearer, further detailed description will be made to the present application in combination with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application, and are not used to limit the present application.

[0068] In one exemplary embodiment, as shown in Figure 1 , an energy efficiency optimization method for low-power power ad hoc network is provided, comprising:

[0069] In step 102, based on the power intelligent sensor network, an energy collection model and a signal-to-noise ratio model are constructed; the power intelligent sensor network includes a primary sensor network and a secondary user network, the primary sensor network includes a base station and M primary sensors, and the secondary user network is composed of N pairs of secondary sensors for device-to-device communication; the energy collection model is used to represent the energy collected at the nth pair of secondary sensors; and the signal-to-noise ratio model is used to represent the signal-to-noise ratio of the nth pair of secondary sensors.

[0070] It can be understood that the power intelligent sensor network is deployed in the power system to form a system with power transmission line information sensing capability. Through the power intelligent sensor, electrical quantities, physical quantities, environmental quantities, state quantities and behavior quantities and other information of each link of power production can be collected, and a bridge between the physical power system and the cyber-physical fusion system is built.

[0071] As shown in Figure 2 , in Figure 2 , the communication scenario, the power intelligent sensor network includes two main communication networks: a primary sensor network and a secondary user network. The primary sensor network is composed of a base station (BS) and a primary sensor (PS), and the primary sensor obtains fixed spectrum resources through spectrum license. The secondary user network is composed of secondary sensors (DS) for device-to-device (D2D) communication. Since the secondary user network lacks authorized frequency bands, it relies on sharing the resources of the primary sensor network. As shown in and represent the number of PS and D2D pairs, respectively. The PS, BS and D2D pairs will cause mutual interference in the communication process, as shown in Figure 2The main user (PU) interference link and the secondary user (SU) interference link in the system, and the interference between D2D pairs is mainly concerned in the embodiment. Each DS as a D2D communication transmitter is equipped with an EH (Energy Harvesting) rectifier circuit, and the task of the DS includes an EH phase and an information transmission phase. In the EH phase, the transmitting end of each D2D pair captures RF (Radio Frequency) energy during the PS information transmission. In the information transmission phase, the DS transmits information using the collected energy.

[0072] Optionally, a linear energy harvesting model is constructed, and the harvested energy is modeled as a linear function of the input power , that is, where 0≤η≤1 represents the EH conversion efficiency. Specifically, based on the expression of the input power , a linear relationship between the harvested energy and the input power is constructed to build the energy harvesting model.

[0073] Optionally, since the EH circuit usually exhibits nonlinear EH characteristics, a nonlinear energy harvesting model is constructed, and the harvested energy is modeled as a nonlinear function of the input power . Specifically, based on the expression of the input power , a nonlinear relationship between the harvested energy and the input power is constructed to build the energy harvesting model.

[0074] Optionally, considering the co-channel interference (i.e., the interference between D2D pairs) generated when the DS occupies the PS channel, a signal-to-noise ratio model is constructed.

[0075] Step 104, according to the energy harvesting model and the signal-to-noise ratio model, a target function is constructed with the goal of maximizing network energy efficiency, and the constraint conditions of the target function are set.

[0076] where the system performance is optimized by maximizing the network energy efficiency (EE) while satisfying various types of constraints. Optionally, considering the signal-to-noise ratio constraint, the energy harvesting requirement, the transmission power limit, and other constraint conditions, an optimization model is established with the goal of improving the network energy efficiency.

[0077] Step 106, a polytope model is constructed, and the polytope model is learned using sample data of independent and identically distributed imperfect channel states to output channel gain information.

[0078] where the high probability region (HPR) is used to represent uncertainty, and the HPR is used to optimize system performance. In order to determine the HPR, by collecting samples of independent and identically distributed imperfect CSI channels, it is helpful to characterize the uncertainty set and derive the HPR of CSI errors. By polytope integration to encapsulate and model the CSI-related uncertainty, the uncertainty of the channel state can be effectively simulated.

[0079] Step 108, according to the channel gain information, based on the constraint conditions of the objective function, solving the objective function to determine the allocation power of each pair of secondary sensors on different sub-channels and the binary selection information of each pair of secondary sensors to the sub-channels, and obtaining the resource allocation scheme.

[0080] Wherein, by considering the channel gain information learned by the imperfect CSI, based on various constraint conditions, solving the objective function to determine the resource allocation scheme that can maximize the network energy efficiency under the constraint conditions, including the allocation power of each pair of secondary sensors on different sub-channels (that is, the optimal transmission power of the transmitting end in the pair of secondary sensors) and the binary selection information a of each pair of secondary sensors to the sub-channels n，m , which is used to determine which sub-channel is selected by the primary sensor. When a n,m = 1, it means that the nth pair of secondary sensors occupies the channel of the mth primary sensor, and when a n,m = 0, it means that the nth pair of secondary sensors does not occupy the channel of the mth primary sensor.

[0081] In the above energy efficiency optimization method of the low-power power ad hoc network, the energy collection process of the secondary sensor is modeled, and an energy collection optimization model under the condition of imperfect CSI is established to maximize the energy efficiency of the network; in view of the problem that the imperfect CSI makes it difficult to solve the objective function, the influence of the imperfect CSI is reduced by constructing a polyhedral model and learning sample data, and a more optimal resource allocation scheme is output to maximize the network throughput and ensure the energy efficiency and data transmission quality.

[0082] In an exemplary embodiment, the energy collection model is represented by the following formula (1):

[0083]

[0084] In formula (1), P M , a and β are constants, P M represents the maximum harvested power when the actual energy collection circuit reaches saturation, and a and β are determined by the characteristics of the actual energy collection circuit; represents the input power of the nth pair of secondary sensors, which is represented by the following formula (2):

[0085]

[0086] In formula (2), θ represents a power allocation factor, represents the transmission power of the nth pair of secondary sensors on the sub-channel of the mth primary sensor, represents the channel gain of the mth primary sensor on the sub-channel, denotes the channel gain from the nth pair of secondary sensors to the mth primary sensor, denotes the channel gain of the primary sensor.

[0087] Compared with the linear energy harvesting model, the energy harvesting model constructed in the embodiment can express the nonlinear EH characteristics of the actual EH circuit, so that the optimization process is more in line with the actual situation, and the optimization effect of the smart grid is improved. In specific implementation, a suitable power allocation factor is set to find a balance between energy harvesting and information transmission, so as to maximize the energy efficiency (EE) and information transmission efficiency of the network.

[0088] In an exemplary embodiment, the signal-to-noise ratio model is represented by the following formula (3):

[0089]

[0090] In formula (3), denotes the transmission power of the nth pair of secondary sensors on the subchannel occupied by the mth primary sensor, denotes the channel gain of the mth primary sensor on the subchannel, denotes the channel gain from the nth pair of secondary sensors to the mth primary sensor, denotes the channel gain of the primary sensor; α i,m is a user association factor between the ith pair of secondary sensors and the mth primary sensor, when α i,m = 1, it indicates that the ith pair of secondary sensors occupies the channel of the mth primary sensor, when α i,m = 0, it indicates that the ith pair of secondary sensors does not occupy the channel of the mth primary sensor; σ 2 denotes the noise power.

[0091] In the information transmission stage, the DS uses the collected energy to perform information transmission; considering the co-frequency interference generated when the DS occupies the PS channel, a signal-to-noise ratio model as shown in formula (3) is constructed. denotes the transmission power of the ith pair of secondary sensors on the subchannel occupied by the mth primary sensor, α i,m denotes the user association factor between the ith pair of secondary sensors and the mth primary sensor, denotes the channel gain from the ith pair of secondary sensors to the mth primary sensor; the ith pair of secondary sensors is one of the N pairs of secondary sensors other than the nth pair of secondary sensors.

[0092] In an exemplary embodiment, the objective function is represented by the following formula (4):

[0093]

[0094] In formula (4), represents the nth pair of secondary sensors in occupying the subchannel of the mth primary sensor

[0095] transmit power; represents the data rate in the secondary user network, which is represented by the following formula (5):

[0096]

[0097] In formula (5), the signal-to-noise ratio of the nth pair of secondary sensors;

[0098] The constraint conditions of the objective function include a signal-to-noise ratio constraint condition, an energy harvesting limit condition, a secondary sensor power limit condition, and a subchannel constraint condition; wherein:

[0099] The signal-to-noise ratio constraint condition is represented by the following formula (6):

[0100]

[0101] In formula (6), γ th is a set threshold value;

[0102] The energy harvesting limit condition is represented by the following formula (7):

[0103] E min ≤ E eh (7) ;

[0104] In formula (7), E min is the minimum energy required by the secondary sensor;

[0105] The secondary sensor power limit condition is represented by the following formula (8):

[0106]

[0107] In formula (8), P max is the power limit value of the secondary sensor;

[0108] The subchannel constraint condition is represented by the following formula (9):

[0109]

[0110] In formula (9), α n，m is a user association factor between the nth pair of secondary sensors and the mth primary sensor.

[0111] The SINR of the DS is ensured to exceed a set threshold by setting the signal-to-noise ratio constraint, ensuring the data transmission quality of the DS. The energy collected is greater than the energy consumed by setting the energy harvesting limit condition, so as to ensure that the DS can continuously operate and complete the communication task. The transmission power of the DS is limited by setting the secondary sensor power limit condition, so as to not exceed the power limit value and meet the actual operation demand. The nth pair of secondary sensors only occupies a subchannel of the primary sensor by setting the subchannel constraint condition.

[0112] In an exemplary embodiment, the polyhedral model is represented by the following formula (10):

[0113]

[0114] In formula (10), represents the center of the polyhedron; λ>0 represents the size of the polyhedron, and S represents the uncertain channel state region learned from the sample data of the independent and identically distributed imperfect channel state; the channel gain information output by the polyhedral model is and represents the channel gain from the nth pair of secondary sensors to the mth primary sensor. represents the channel gain of the primary sensor.

[0115] Wherein, the problem of solving the objective function is a non-convex optimization problem. Since the objective function is a fractional form, i.e., the ratio of the total data rate of all DSs to the total transmission power, and the constraints involve nonlinear factors of SINR, EH and resource allocation, it is difficult to solve directly. Based on this, the high probability region (HPR) is used to represent the uncertainty in this embodiment, and the HPR is used to optimize the system performance. In order to determine the HPR, the samples of the independent and identically distributed imperfect CSI channel are collected (wherein, a m ∈R 2 ), which helps to characterize the uncertainty set and derive the HPR of CSI error. By polyhedral integration to encapsulate and model the CSI-related uncertainty, the uncertainty of the channel state can be effectively simulated. The polyhedral integration modeling provides a general framework that can be combined with various optimization tools and techniques, which can enhance the flexibility and robustness of system optimization.

[0116] In an exemplary embodiment, learning a polyhedral model with sample data of independent and identically distributed imperfect channel states, outputting channel gain information, includes: dividing the sample data of independent and identically distributed imperfect channel states into two parts to obtain first sample data and second sample data; using the first sample data to approximate the shape of the high probability region, and setting the center of the polyhedron through the sample mean of the first sample data, determining a parameter value and a parameter value; using the second sample data to calibrate the size of the polyhedron, determining a parameter value; and determining and outputting channel gain information in the polyhedral model with the determined parameter value and

[0117] In this embodiment, in order to solve the problem that the objective function is difficult to solve, a learning robust algorithm based on polyhedron (hereinafter referred to as PLRA algorithm) is provided to process the imperfect CSI problem. First, the user channel set is modeled to construct a polyhedral model as shown in formula (10); then, a specific learning algorithm is used to determine the parameters of the polyhedron, which includes two steps: shape learning and size calibration. By dividing the sample data into two parts, i.e. and is used to approximate the shape of the HPR, and is used to calibrate the size of the uncertainty set to meet the optimization model and have a coverage range.

[0118] In an optional implementation, shape learning includes the following steps: setting and as the origin of the polyhedral uncertainty set, and according to the following formula (11), and are calculated as the sample mean:

[0119]

[0120] It can be understood that the polyhedron is selected in this embodiment because it is easy to process and thus produces robust optimization. After performing the shape learning step of the polyhedral uncertainty set, in order to ensure that the obtained HPR parameters have good adaptability and generalizability, the size calibration step is performed next.

[0121] In an optional implementation, size calibration includes the following steps: using the data set to calibrate the polyhedral uncertainty set so that they contain points p with a confidence level of The core method involves estimating the quantile of the transformed data sample. Specifically, a transformation function is defined according to the following formula (12): ​

[0122]

[0123] Among them, t p (a) The random vector space Mapped to This transformation quantizes the components of 'a' relative to the reference value. and The deviation.

[0124] Using from the sample dataset The derived transformation t p (a) estimate Quantiles are used to determine the size of the HPR. The size can be set based on the sample dataset. t p The distribution of (a) is estimated. Quantiles. To formalize them, they are defined according to the following formula (13). Quantiles:

[0125]

[0126] Using this definition, The upper limit of the 1-K confidence level for quantiles can be derived using the confidence interval estimation method. Furthermore, to achieve the desired confidence level, the minimum sample size D-D1 must satisfy the constraint shown in formula (14):

[0127]

[0128] calculate t for each sample p The function values ​​of (a) are sorted in ascending order as t. p (1) ≤…≤t p (D) Take the first A value can be considered as t p The upper limit of (a), where Based on the given upper limit, the set can be set according to the following formula (15). Size:

[0129]

[0130] In determining the uncertainty set of a polyhedron After determining the value, the channel gain can be obtained using the aforementioned formula (10). and

[0131] It can be understood that after obtaining more accurate channel gains through the learning of the polyhedral model, in order to maximize the network efficiency under the model, a polyhedral-based learning resource algorithm is used to solve the model. The optimization problem is nonlinear because the objective function in the optimization model is affected by random parameters. In order to find the optimal solution, a convex optimization tool can be used to simplify the problem and solve the maximum network energy efficiency. After solving the objective function, a set of parameter values can be obtained as a resource allocation scheme, including the allocated power of each node n on different sub-channels, and the binary selection of each node n to the sub-channels, aiming to maximize the network energy efficiency while meeting the signal-to-noise ratio threshold, energy demand limit, and power upper and lower limit constraints, etc.

[0132] In an exemplary embodiment, referring to Table 1, Table 1 shows the pseudo code for robust optimization of resource allocation for polyhedral learning. The specific steps are as follows: independent and identically distributed samples are collected under incomplete channel state information, and the samples are divided into two parts, one part is used for shape learning, and the other part is used for size calibration. Second, set the size of HPR, and perform shape learning and size calibration steps to learn the channel gain parameters. Finally, the CVX tool in Python is used to solve the model to obtain the optimal transmit power of the user.

[0133] Table 1:

[0134]

[0135] In an exemplary embodiment, numerical and simulation results are provided to illustrate the performance of the PLRA algorithm provided by the embodiments of the present application, and are compared with the ellipsoid-based learning algorithm (ELRA), the robust optimization algorithm (RA) and the baseline algorithm (baseline) (hereinafter, the algorithm compared with the ELRA algorithm is referred to as the "comparison algorithm"). The coverage range of the BS is a circle with a radius of 600m, the BS bandwidth B is 5MHz, the number of DUs (Distributed Unit) is 40, and the number of data sets is 1000. The remaining simulation parameters are set with reference to Table 2.

[0136] Table 2:

[0137] Parameter Value Number of DSs N 40 BS bandwidth B 5 MHz Confidence level θ 0.98 Rate probability threshold probability 0.8 Maximum power P max ]]> 5W Maximum harvest power P M ]]> 24 mW α 150 β 0.014

[0138] Referring to Figure 3 , Figure 3 The relationship between the noise power σ 2 and the total rate of DS is illustrated. As can be seen from Figure 3 , as the noise power σ 2with the increase of noise power, the total data rate of DS decreases, it is clear that the PLRA algorithm shows superior performance over the comparative algorithms. In power smart sensor networks, DS clusters need to share the same frequency band under limited spectrum resources, resulting in spectrum resource competition and noise increase. Due to the existence of imperfect CSI in the network, this can cause errors during transmission and communication. The PLRA algorithm provided in the application can effectively solve the problem of imperfect CSI, and through the modeled channel, the reliability and efficiency of the system can be improved.

[0139] Referring to Figure 4 , Figure 4 represents the relationship between the noise power σ 2 and the energy efficiency EE. With the increase of noise power, the total data rate of DS tends to decrease. It can be clearly seen from Figure 4 that the PLRA algorithm is significantly superior to the comparative algorithms in maintaining higher data rates under different noise conditions. Due to the decline in signal quality, more retransmission and error correction processes will occur in the network, increasing the overall energy consumption. As more resources are consumed, the EE of the network gradually decreases in order to achieve reliable information transmission. Figure 4 It is shown that the PLRA effectively alleviates these adverse effects and can ensure better performance optimization in energy efficiency compared to other algorithms despite higher noise power, highlighting the ability of the PLRA algorithm to effectively optimize network performance under challenging conditions.

[0140] Referring to Figure 5 , Figure 5 shows the relationship between the SINR threshold γ th of DS in power smart sensor networks and the secondary network throughput. It can be seen from Figure 5 that the PLRA algorithm is significantly superior to other algorithms. With the increase of SINR threshold, the total data rate of DS tends to rise. This is because a higher SINR threshold requires a stronger signal relative to interference, thereby reducing the impact of interference on signal quality. Therefore, the system can more effectively utilize the available spectrum and bandwidth. This optimized spectrum utilization directly contributes to improving the overall throughput of the network. By setting a higher SINR threshold, the system can better manage network resources and adopt more effective information transmission techniques, ultimately improving the data rate achieved by DS and thus improving the robustness of the communication link, reducing the error rate, and improving the overall performance of information transmission.

[0141] Referring to Figure 6 , Figure 6 shows the relationship between the SINR threshold γ th of DS in power smart sensor networks and the energy efficiency EE of the secondary network. It can be seen from Figure 6 that when the SINR threshold γ thAs the EE of DS increases, and the EE of the PLRA algorithm is greater than the EE of the comparative algorithm, it proves that the performance of the proposed PLRA algorithm is superior to the comparative algorithm. In the power smart sensor network with imperfect CSI, the increase of the SINR threshold leads to the increase of the signal quality requirement, which reduces the impact of interference on the signal. A higher SINR threshold allows the system to transmit data under higher signal quality conditions, which helps to reduce the probability of transmission errors and data loss, and improves the overall performance of the network. Since the PLRA algorithm can more effectively optimize resource allocation and signal processing, it can still optimize higher network energy efficiency even in the presence of imperfect CSI information, which shows that it has significant advantages in improving network performance and optimizing energy use. This result further verifies the effectiveness of the PLRA algorithm in handling optimization tasks in complex network environments.

[0142] It can be understood that the energy efficiency optimization method of the low-power power ad hoc network provided by the embodiments of the present application aims to achieve efficient EE in the power smart sensor network in the EH scenario, which can effectively improve the network throughput of the power smart sensor network, and is superior to the comparative algorithm. First, the energy collection of the sensor node is modeled; on this basis, a robust optimization model with imperfect CSI is established by combining power control and rate requirement constraints, and the model aims to maximize network energy efficiency; and a robust learning algorithm based on a polyhedron is proposed, which reduces the impact of imperfect CSI by modeling and learning the channel, and finally obtains a more accurate solution. The simulation results verify that the method can effectively improve the performance of the network.

[0143] It should be understood that although each step in the flowchart involved in each embodiment as described above is displayed in sequence according to the direction of the arrow, these steps are not necessarily executed in sequence according to the direction of the arrow. Unless otherwise specified herein, the execution of these steps is not strictly limited in sequence, and these steps can be executed in other orders. Moreover, at least part of the steps in the flowchart involved in each embodiment as described above can include multiple steps or stages, which are not necessarily executed at the same time, but can be executed at different times, and the execution sequence of these steps or stages is not necessarily sequential, but can be executed in rotation or alternation with at least part of other steps or steps or stages in other steps.

[0144] In an exemplary embodiment, based on the same inventive concept, the embodiments of the present application also provide a power system, wherein a power smart sensor network is arranged in the power system, the power smart sensor network includes a primary sensor network and a secondary user network, the primary sensor network includes a base station and M primary sensors, and the secondary user network is composed of N pairs of secondary sensors for device-to-device communication.

[0145] The secondary sensor is configured according to the resource allocation scheme determined by the energy efficiency optimization method of the low-power consumption power ad hoc network provided in the embodiments of the present application.

[0146] In an alternative implementation, a computer device arranged in the power system or a computer device arranged at a remote end performs optimization analysis based on the energy efficiency optimization method of the low-power consumption power ad hoc network provided in the embodiments of the present application, outputs a resource allocation scheme, and provides the resource allocation scheme to each pair of secondary sensors in the power smart sensor network through a communication network, and the transmitting end of the pair of secondary sensors captures energy according to the resource allocation scheme and transmits information based on the collected energy.

[0147] In an exemplary embodiment, a computer device, which can be a server, has an internal structure as shown in Figure 7 The computer device includes a processor, a memory, an input / output interface (I / O) and a communication interface. The processor, the memory and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. The processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for running the operating system and the computer program in the non-volatile storage medium. The input / output interface of the computer device is configured to exchange information between the processor and external devices. The communication interface of the computer device is configured to communicate with external terminals through a network connection. The computer program is executed by the processor to implement an energy efficiency optimization method of a low-power consumption power ad hoc network.

[0148] Those skilled in the art can understand that Figure 7 The structure shown in the above

[0149] In one exemplary embodiment, a computer device is provided, comprising a memory and a processor, the memory storing a computer program, the processor implementing the following steps when executing the computer program: based on a power intelligent sensor network, constructing an energy collection model and a signal-to-noise ratio model; the power intelligent sensor network comprises a main sensor network and a secondary user network, the main sensor network comprises a base station and M main sensors, and the secondary user network is composed of N pairs of secondary sensors for device-to-device communication; the energy collection model is used to characterize the energy collected at the nth pair of secondary sensors; the signal-to-noise ratio model is used to characterize the signal-to-noise ratio of the nth pair of secondary sensors; according to the energy collection model and the signal-to-noise ratio model, a target function is constructed with the goal of maximizing network energy efficiency, and constraint conditions of the target function are set; a polyhedral model is constructed, the polyhedral model is learned using sample data of independent and identically distributed imperfect channel states, and channel gain information is output; according to the channel gain information, the target function is solved based on the constraint conditions of the target function, the allocation power of each pair of secondary sensors on different sub-channels and the binary selection information of each pair of secondary sensors on the sub-channels are determined, and a resource allocation scheme is obtained.

[0150] In one embodiment, the processor further implements the following steps when executing the computer program: dividing the sample data of independent and identically distributed imperfect channel states into two parts to obtain first sample data and second sample data; using the first sample data to approximate the shape of the high probability region, and setting the center of the polyhedron through the sample mean of the first sample data to determine a parameter value and a parameter value; using the second sample data to calibrate the size of the polyhedron to determine a parameter value; determining and outputting the channel gain information in the polyhedral model with the determined parameter values and

[0151] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When executed by a processor, the computer program performs the following steps: Based on a power smart sensor network, an energy harvesting model and a signal-to-noise ratio (SNR) model are constructed; the power smart sensor network includes a main sensor network and a secondary user network, the main sensor network includes a base station and M main sensors, and the secondary user network consists of N pairs of secondary sensors for device-to-device communication; the energy harvesting model characterizes the energy harvested at the nth pair of secondary sensors; the SNR model characterizes the SNR of the nth pair of secondary sensors; based on the energy harvesting model and the SNR model, an objective function is constructed with the goal of maximizing network energy efficiency, and constraints on the objective function are set; a polyhedral model is constructed, and the polyhedral model is learned using sample data of independent and identically distributed imperfect channel states to output channel gain information; based on the channel gain information and the constraints of the objective function, the objective function is solved to determine the power allocation of each pair of secondary sensors on different sub-channels and the binary selection information of each pair of secondary sensors for the sub-channels, thereby obtaining a resource allocation scheme.

[0152] In one embodiment, when the computer program is executed by a processor, it further performs the following steps: dividing sample data of independent and identically distributed imperfect channel states into two parts to obtain first sample data and second sample data; using the first sample data to approximate the shape of a high-probability region, and using the first sample data...

[0153] The sample mean is set at the center of the polyhedron to determine... Parameter values ​​and Parameter values; using the second sample data to calibrate the size of the polyhedron and determine the λ parameter value; using the determined parameter values ​​in the polyhedron model, determine and output the channel gain information. and

[0154] In one embodiment, a computer program product is provided, comprising a computer program which, when executed by a processor, implements the following steps: based on a power smart sensor network, constructing an energy collection model and a signal-to-noise ratio model; the power smart sensor network comprises a primary sensor network and a secondary user network, the primary sensor network comprises a base station and M primary sensors, and the secondary user network is composed of N pairs of secondary sensors for device-to-device communication; the energy collection model is used to characterize the energy collected at the nth pair of secondary sensors; the signal-to-noise ratio model is used to characterize the signal-to-noise ratio of the nth pair of secondary sensors; according to the energy collection model and the signal-to-noise ratio model, a target function is constructed with the goal of maximizing network energy efficiency, and constraint conditions of the target function are set; a polyhedral model is constructed, the polyhedral model is learned by using sample data of independent and identically distributed imperfect channel states, and channel gain information is output; according to the channel gain information, based on the constraint conditions of the target function, the target function is solved to determine the allocation power of each pair of secondary sensors on different sub-channels and the binary selection information of each pair of secondary sensors on the sub-channels, and a resource allocation scheme is obtained.

[0155] In one embodiment, the computer program, when executed by the processor, further implements the following steps: dividing the sample data of independent and identically distributed imperfect channel states into two parts to obtain first sample data and second sample data; using the first sample data to approximate the shape of the high probability region, and setting the center of the polyhedron by the sample mean of the first sample data to determine a parameter value and a parameter value; using the second sample data to calibrate the size of the polyhedron to determine a parameter value; determining and outputting the channel gain information and

[0156] It should be noted that the user information (including but not limited to user equipment information, user personal information, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) involved in the present application are all information and data authorized by the user or authorized by all parties, and the collection, use and processing of related data need to comply with relevant regulations.

[0157] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer readable storage medium, and when executed, can include the processes of the above-mentioned embodiment methods. Any reference to memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile memory and volatile memory. The non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical storage, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. The volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration but not limitation, the RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc. The database involved in the embodiments provided in the present application can include at least one of a relational database and a non-relational database. The non-relational database can include a distributed database based on a block chain, etc., without being limited thereto. The processor involved in the embodiments provided in the present application can be a general-purpose processor, a central processing unit, a graphics processing unit, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, an artificial intelligence (AI) processor, etc., without being limited thereto.

[0158] The technical features of the above embodiments can be combined arbitrarily. In order to make the description simple, all possible combinations of the technical features in the above embodiments are not described, however, as long as the combinations of the technical features do not exist contradictory, they should be considered as the scope of the present application.

[0159] The above-described embodiments are merely illustrative of several embodiments of the present application, and the description is relatively specific and detailed, but should not be understood as a limitation on the scope of the patent. It should be noted that for those skilled in the art, without departing from the concept of the present application, a number of modifications and improvements can be made, which are all within the scope of the present application. Therefore, the scope of protection of the present application should be subject to the appended claims.

Claims

1. A method for energy efficiency optimization of low power consumption power ad hoc network, characterized in that, The method comprises: Based on the power intelligent sensor network, an energy collection model and a signal-to-noise ratio model are constructed; the power intelligent sensor network comprises a main sensor network and a secondary user network, the main sensor network comprises a base station and M main sensors, and the secondary user network is composed of N pairs of secondary sensors for device-to-device communication; the energy collection model is used to represent the energy collected at the nth pair of secondary sensors; and the signal-to-noise ratio model is used to represent the signal-to-noise ratio of the nth pair of secondary sensors; According to the energy collection model and the signal-to-noise ratio model, a target function is constructed with the goal of maximizing network energy efficiency, and a constraint condition of the target function is set; A polyhedral model is constructed, sample data of independent and identically distributed imperfect channel states are used to learn the polyhedral model, and channel gain information is outputted; According to the channel gain information, the target function is solved based on the constraint condition of the target function, the allocation power of each pair of secondary sensors on different sub-channels and the binary selection information of each pair of secondary sensors on the sub-channels are determined, and a resource allocation scheme is obtained.

2. The method of claim 1, wherein, The energy collection model is represented by the following formula (1): In Equation (1), P M , α and β are constants, P M represents the maximum harvested power when the actual energy harvesting circuit reaches saturation, and α and β are determined by the characteristics of the actual energy harvesting circuit. represents the input power of the nth pair of secondary sensors, and is expressed by Equation (2) below. In Equation (2), θ denotes a power allocation factor, denotes the transmission power of the nth pair of secondary sensors on the subchannel occupied by the mth primary sensor, denotes the channel gain of the mth primary sensor on the subchannel, denotes the channel gain of the nth pair of secondary sensors to the mth primary sensor, denotes the channel gain of the primary sensor.

3. The method of claim 2, wherein, The signal-to-noise ratio model is represented by the following formula (3): In Equation (3), denotes the transmission power of the nth secondary sensor on the subchannel occupied by the mth primary sensor, denotes the channel gain of the mth primary sensor on the subchannel, denotes the channel gain of the nth secondary sensor to the mth primary sensor, denotes the channel gain of the primary sensor; a i，m is the user association factor between the ith secondary sensor and the mth primary sensor, when a i，m = 1, it means that the ith secondary sensor occupies the channel of the mth primary sensor, when a i,m = 0, it means that the ith secondary sensor does not occupy the channel of the mth primary sensor; σ 2 denotes the noise power.

4. The method of claim 3, wherein, The target function is represented by the following formula (4): In Equation (4), denotes the transmission power of the nth pair of secondary sensors on the subchannel occupied by the mth primary sensor; denotes the data rate in the secondary user network, which is represented by the following Equation (5): In equation (5), Signal-to-noise ratio of the nth pair of secondary sensors; The constraint condition of the target function comprises a signal-to-noise ratio constraint condition, an energy collection restriction condition, a secondary sensor power restriction condition, and a sub-channel constraint condition; wherein: The signal-to-noise ratio constraint condition is represented by the following formula (6): In Equation (6), γ th is a set threshold value; The energy collection restriction condition is represented by the following formula (7): E min ≤E eh (7); In Equation (7), E min is the minimum energy required for the secondary sensor; The secondary sensor power restriction condition is represented by the following formula (8): In Equation (8), P max is the power limit value of the secondary sensor; The sub-channel constraint condition is represented by the following formula (9): In Equation (9), α n，m is a user correlation factor between the nth pair of secondary sensors and the mth primary sensor.

5. The method of claim 1, wherein, The polyhedral model is represented by the following formula (10): In Equation (10), represents the center of the polyhedron; λ > 0 represents the size of the polyhedron, represents the uncertain channel state region learned from sample data of independent and identically distributed imperfect channel states; the channel gain information output by the polyhedron model is and represents the channel gain from the nth pair of secondary sensors to the mth primary sensor; represents the channel gain of the primary sensor.

6. The method of claim 5, wherein, The learning of the polyhedral model by using sample data of independent and identically distributed imperfect channel states to output channel gain information comprises: The sample data of independent and identically distributed imperfect channel states are divided into two parts to obtain first sample data and second sample data; approximating a shape of a high-probability region using the first sample data, and setting a center of a polyhedron by a sample mean value of the first sample data, determining a parameter value and a parameter value; The size of the polyhedron is calibrated by using the second sample data to determine the value of the λ parameter; determining and outputting channel gain information in the polyhedral model with the determined parameter values and 7. A power system, characterized by The power system is provided with a power intelligent sensor network, the power intelligent sensor network comprises a main sensor network and a secondary user network, the main sensor network comprises a base station and M main sensors, and the secondary user network is composed of N pairs of secondary sensors for device-to-device communication; The secondary sensor is configured with a resource allocation scheme determined by the method according to any one of claims 1 to 6.

8. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that, The processor executes the computer program to realize the steps of the method according to any one of claims 1 to 6.

9. A computer-readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by the processor to realize the steps of the method according to any one of claims 1 to 6.

10. A computer program product comprising a computer program, characterized in that, The computer program is executed by the processor to realize the steps of the method according to any one of claims 1 to 6.

Images