Farmland cognitive sensor network transmission method, device, electronic equipment and medium

Abstract

The present invention provides a farmland cognitive sensor network transmission method, device, electronic device and medium, which relates to the field of data transmission, including: constructing a first target variable strategy set and a second target variable strategy set according to the first Pareto optimal solution of each sensor node in the first objective function and the second Pareto optimal solution in the second objective function; repeatedly performing the following steps: obtaining the current first-level strategy set according to the first current strategy and the second current strategy; iteratively processing the current first-level strategy set according to the secondary game to determine the first-level strategy set after iteration; until the absolute value of the difference between the current first-level strategy set and the first-level strategy set after iteration is less than or equal to the first preset value, the sensor network transmission is performed according to the first-level strategy set after iteration. The present invention can reduce the complexity of solving the problem, adjust the transmission power and the transmission channel, improve the transmission quality and network capacity, and reduce the network energy consumption.

Description

Technical Field

[0001] The present invention relates to the field of data transmission, and in particular to a farmland cognitive sensor network transmission method, device, electronic equipment and medium. Background Art

[0002] The efficient and precise agricultural decision-making system places high demands on the data transmission integrity, network delay, network capacity and transmission signal-to-noise ratio of the farmland sensor network. However, due to limited resources, there are conflicts between some network performance indicators. The existing multi-objective optimization usually adopts the method of empirical weighting and custom benefit function. The optimized transmission power and transmission channel are sensitive to the weight coefficient and benefit function form, and the empirical weighting lacks objectivity and cannot take into account the improvement of the overall network performance. Summary of the invention

[0003] The present invention provides a farmland cognitive sensor network transmission method, device, electronic device and medium to solve the technical problem of lack of objectivity in the selection of wireless network transmission power and transmission channel in the prior art. The present invention provides a technical solution for converting the multi-objective network performance optimization problem into a non-cooperative game model to jointly optimize and control the transmission power and transmission channel of the farm cognitive wireless sensor network.

[0004] In a first aspect, the present invention provides a farmland cognitive sensor network transmission method, comprising:

[0005] According to a first Pareto optimal solution of each sensor node in the first objective function and a second Pareto optimal solution of each sensor node in the second objective function, construct a first objective variable strategy set corresponding to the first objective function and a second objective variable strategy set corresponding to the second objective function;

[0006] Repeat the following steps:

[0007] Randomly generate a first current strategy according to the first target variable strategy set, randomly generate a second current strategy according to the second target variable strategy set, and obtain a current primary strategy set according to the first current strategy and the second current strategy;

[0008] Iteratively process the current primary strategy set according to the secondary game to determine a primary strategy set after iteration;

[0009] Until the absolute value of the difference between the current first-level strategy set and the iterated first-level strategy set is less than or equal to a first preset value, the sensor network transmission is performed according to the iterated first-level strategy set;

[0010] The first objective function is determined according to the signal to interference noise ratio of each sensor node in the sensor network, and the second objective function is constructed according to the transmission power of each sensor node and the number of sensors in the same channel;

[0011] The current first-level strategy set is iteratively processed according to the secondary game, and is used to determine the first final transmission channel corresponding to each sensor node using the first objective function as the benefit function, and then obtain the first final transmission power after taking the first-order derivative of the first final transmission channel with the transmission power as the variable; determine the second final transmission channel corresponding to each sensor node using the second objective function as the benefit function, and then obtain the second final transmission power after taking the first-order derivative of the second final transmission channel with the transmission power as the variable.

[0012] According to the farmland cognitive sensor network transmission method provided by the present invention, before constructing a first objective variable strategy set corresponding to the first objective function and a second objective variable strategy set corresponding to the second objective function according to a first Pareto optimal solution of each sensor node in the first objective function and a second Pareto optimal solution of each sensor node in the second objective function, the method further includes:

[0013] For each sensor node in the sensor network, the transmission time is determined according to the transmission bandwidth and the transmission rate;

[0014] Determine the total power of all sensor nodes based on the path gain from the current sensor node to the next hop node and the transmission power of the next hop node;

[0015] Determine a first power value according to the total power of all sensor nodes and the background noise power;

[0016] Determine a second power value according to the transmit power of the current sensor node and the next hop path gain of the current node;

[0017] Determine a power ratio according to the first power value and the second power value;

[0018] Determine the signal to interference and noise ratio of the current sensor node according to the transmission time and the power ratio;

[0019] A first objective function is constructed according to the signal to interference plus noise ratio (SINR) of each sensor node in the sensor network and a SINR threshold.

[0020] According to the farmland cognitive sensor network transmission method provided by the present invention, the first objective variable strategy set corresponding to the first objective function and the second objective variable strategy set corresponding to the second objective function are constructed according to the first Pareto optimal solution of each sensor node in the first objective function and the second Pareto optimal solution of each sensor node in the second objective function, including:

[0021] Calculate a first Euclidean distance between a first Pareto optimal solution of each sensor node and a first objective function, and calculate a second Euclidean distance between a second Pareto optimal solution of each sensor node and a second objective function;

[0022] Normalizing all first Euclidean distances to obtain a first normalized result of each sensor node, normalizing all second Euclidean distances to obtain a second normalized result of each sensor node, and constructing a fuzzy relationship matrix according to the first normalized result and the second normalized result;

[0023] The fuzzy relationship matrix is ​​processed by a square method to obtain a fuzzy equivalence matrix, the fuzzy equivalence matrix is ​​clustered, and a first objective variable strategy set corresponding to the first objective function and a second objective variable strategy set corresponding to the second objective function are obtained.

[0024] According to the farmland cognitive sensor network transmission method provided by the present invention, the iterative processing of the current primary strategy set according to the secondary game to determine the iterative primary strategy set includes:

[0025] Processing the current primary policy set based on the first iteration, obtaining a first final transmission channel and a second final transmission channel, and determining a post-iteration secondary policy set according to the first final transmission channel and the second final transmission channel;

[0026] The iterated secondary policy set is processed based on the second iteration to obtain a first final transmit power and a second final transmit power, and a post-iteration primary policy set is determined according to the first final transmit power and the second final transmit power.

[0027] According to the farmland cognitive sensor network transmission method provided by the present invention, the current primary strategy set is processed based on the first iteration, a first final transmission channel and a second final transmission channel are obtained, and a secondary strategy set after iteration is determined according to the first final transmission channel and the second final transmission channel, including:

[0028] Repeat the following steps:

[0029] Under the condition that the transmission power of each sensor node in the current secondary strategy set remains unchanged, the first target transmission channel corresponding to each sensor node in the first target function strategy set is determined by taking the first target function as the benefit function, and the second target transmission channel corresponding to each sensor node in the second target function strategy set is determined by taking the second target function as the benefit function, wherein the current secondary strategy set is determined according to the current primary strategy set;

[0030] updating the current secondary strategy set according to the first target transmission channel and the second target transmission channel to obtain an iterated secondary strategy set;

[0031] Until the absolute value of the difference between the current secondary strategy set and the iterated secondary strategy set is less than or equal to a second preset value, a first final transmission channel and a second final transmission channel are obtained.

[0032] According to the farmland cognitive sensor network transmission method provided by the present invention, the iterative secondary strategy set is processed based on the second iteration to obtain the first final transmission power and the second final transmission power, and the iterative primary strategy set is determined according to the first final transmission power and the second final transmission power, including:

[0033] Repeat the following steps:

[0034] When the final transmission channel of each sensor node in the iterative secondary strategy set remains unchanged, for each secondary strategy set in the iterative secondary strategy set, taking the transmission power as a variable, first-order derivative of the final transmission channel is obtained, and when the first-order derivative is 0, a first final transmission power and a second final transmission power are obtained;

[0035] updating the iterated secondary strategy set according to the first final transmit power and the second final transmit power to obtain a target secondary strategy set;

[0036] Until the absolute value of the difference between the iterated secondary strategy set and the target secondary strategy set is less than or equal to a third preset value, the target secondary strategy set is determined to be the iterated primary strategy set.

[0037] According to the farmland cognitive sensor network transmission method provided by the present invention, the sensor network transmission is performed according to the iterated first-level strategy set, including:

[0038] The final transmission power and the final transmission channel of each sensor node are determined according to the iterated first-level strategy set, and the transmitter network transmission is performed according to the final transmission power and the final transmission channel of each sensor node.

[0039] In a second aspect, a farmland cognitive sensor network transmission device is provided, comprising:

[0040] A construction unit: configured to construct a first objective variable strategy set corresponding to the first objective function and a second objective variable strategy set corresponding to the second objective function according to a first Pareto optimal solution of each sensor node in the first objective function and a second Pareto optimal solution of each sensor node in the second objective function;

[0041] Repeating unit: used to repeatedly perform the following steps:

[0042] Randomly generate a first current strategy according to the first target variable strategy set, randomly generate a second current strategy according to the second target variable strategy set, and obtain a current primary strategy set according to the first current strategy and the second current strategy;

[0043] Iteratively process the current primary strategy set according to the secondary game to determine a primary strategy set after iteration;

[0044] A transmission unit: configured to perform sensor network transmission according to the iterative first-level policy set until the absolute value of the difference between the current first-level policy set and the iterative first-level policy set is less than or equal to a first preset value;

[0045] The first objective function is determined according to the signal to interference noise ratio of each sensor node in the sensor network, and the second objective function is constructed according to the transmission power of each sensor node and the number of sensors in the same channel;

[0046] The current first-level strategy set is iteratively processed according to the secondary game, and is used to determine the first final transmission channel corresponding to each sensor node using the first objective function as the benefit function, and then obtain the first final transmission power after taking the first-order derivative of the first final transmission channel with the transmission power as the variable; determine the second final transmission channel corresponding to each sensor node using the second objective function as the benefit function, and then obtain the second final transmission power after taking the first-order derivative of the second final transmission channel with the transmission power as the variable.

[0047] In a third aspect, an electronic device is also provided, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the farmland cognitive sensor network transmission method when executing the program.

[0048] In a fourth aspect, a non-transitory computer-readable storage medium is also provided, on which a computer program is stored, and when the computer program is executed by a processor, the farmland cognitive sensor network transmission method is implemented.

[0049] The beneficial effects of the present invention are as follows: the present invention provides a farmland cognitive sensor network transmission method, device, electronic device and medium, obtains a current first-level strategy set according to a first current strategy and a second current strategy; it iterates the current first-level strategy set according to a secondary game to determine an iterated first-level strategy set; until the difference between the current first-level strategy set and the iterated first-level strategy set is less than or equal to a first preset value, the sensor network transmission is performed according to the optimal transmission power and transmission channel determined by the iterated first-level strategy set. When optimizing and solving the game nodes, the present invention designs a two-layer game model for the non-convex problem of the benefit function, takes each sensor node as a first-level game party, and the transmission power and transmission channel of each sensor node as the strategy space of the secondary game party, and uses the best response principle to obtain the optimization results of network power control and channel allocation, thereby converting the optimization problem of multiple network performance parameters into two low-dimensional problems, reducing the complexity of solving the problem, and being able to adjust the transmission power and transmission channel according to the dynamic impact sensitivity of different nodes on different network performance, thereby achieving improvement in transmission quality and network capacity, and reducing network energy consumption. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0051] Figure 1 This is one of the flow diagrams of the farmland cognitive sensor network transmission method provided by the present invention;

[0052] Figure 2 This is the second flow chart of the farmland cognitive sensor network transmission method provided by the present invention;

[0053] Figure 3 It is a flowchart of constructing a first target variable strategy set and a second target variable strategy set provided by the present invention;

[0054] Figure 4 It is a schematic diagram of a process for determining a set of first-level strategies after iteration provided by the present invention;

[0055] Figure 5 It is a schematic diagram of a process for determining a secondary strategy set after iteration provided by the present invention;

[0056] Figure 6 It is a schematic diagram of a process for determining a set of first-level strategies after iteration provided by the present invention;

[0057] Figure 7It is a structural schematic diagram of the farmland cognitive sensor network transmission device provided by the present invention;

[0058] Figure 8 It is a structural schematic diagram of the electronic device provided by the present invention. DETAILED DESCRIPTION

[0059] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0060] At present, precision agriculture decision-making systems use modern wireless technology and computer technology to monitor and manage farmland soil, water, light, temperature and humidity, and crop growth, which helps to improve crop yield and quality and reduce labor demand. However, in actual practice, due to the vast area of ​​open fields and dense crops, as the crop growth period changes, the environmental parameters at different locations and heights of the farmland vary greatly. Sparsely distributed sensors are difficult to achieve accurate environmental perception, which may cause inaccurate agricultural operation decisions.

[0061] With the rapid development of Internet of Things technology, the frequency resources that wireless sensor network transmission relies on are becoming more and more crowded, and the interference between nodes is becoming more and more serious. The types of data monitored by wireless sensor networks in farmland have increased, including crop phenotypic data, multi-dimensional environmental parameters, and agricultural machinery status information, etc., with the characteristics of heterogeneous perception models, dense node deployment, typical wide bandwidth and high throughput, which further aggravates network interference. In order to solve the above technical problems, the present invention provides a solution for finding the optimal transmission power and optimal transmission channel of wireless sensor networks. Specifically, Figure 1 This is one of the flow diagrams of the farmland cognitive sensor network transmission method provided by the present invention, which discloses a farmland cognitive sensor network transmission method, including:

[0062] Step 101: construct a first objective variable strategy set corresponding to the first objective function and a second objective variable strategy set corresponding to the second objective function according to a first Pareto optimal solution of each sensor node in the first objective function and a second Pareto optimal solution of each sensor node in the second objective function;

[0063] Repeat the following steps:

[0064] Step 102: randomly generate a first current strategy according to the first target variable strategy set, randomly generate a second current strategy according to the second target variable strategy set, and obtain a current primary strategy set according to the first current strategy and the second current strategy;

[0065] Step 103, iteratively processing the current primary strategy set according to the secondary game to determine an iterated primary strategy set;

[0066] Step 104: until the absolute value of the difference between the current first-level policy set and the iterated first-level policy set is less than or equal to a first preset value, sensor network transmission is performed according to the iterated first-level policy set;

[0067] The first objective function is determined according to the signal to interference noise ratio of each sensor node in the sensor network, and the second objective function is constructed according to the transmission power of each sensor node and the number of sensors in the same channel;

[0068] The current first-level strategy set is iteratively processed according to the secondary game, and is used to determine the first final transmission channel corresponding to each sensor node using the first objective function as the benefit function, and then obtain the first final transmission power after taking the first-order derivative of the first final transmission channel with the transmission power as the variable; determine the second final transmission channel corresponding to each sensor node using the second objective function as the benefit function, and then obtain the second final transmission power after taking the first-order derivative of the second final transmission channel with the transmission power as the variable.

[0069] In step 101, the first objective function is determined according to the signal-to-interference-to-noise ratio of each sensor node in the sensor network, and the second objective function is constructed according to the transmission power of each sensor node and the number of sensors under the same channel. The farmland cognitive wireless sensor network control is aimed at densely deployed sensor networks, balancing the network's multiple energy efficiency indicators and extending the network life cycle. However, the optimization of individual node energy efficiency is to ensure its own energy efficiency, while the optimization of the overall network life is to balance the energy efficiency of different nodes to improve the overall performance of the network. However, without considering the needs of the nodes themselves, only optimizing the overall network life each time is likely to cause uneven node energy consumption, and still cannot guarantee the network survival time.

[0070] According to existing research results, improving the signal-to-noise ratio can optimize network performance such as network capacity and network transmission quality, but unlimited improvement of the signal-to-noise ratio will cause the network energy consumption to be too fast. The first objective function is used to limit the signal-to-noise ratio increase space in combination with the signal-to-noise ratio threshold. When the network has a high signal-to-noise ratio, the node transmission power and channel allocation results have limited impact on this goal. However, crops are densely distributed in farmland, and the land area is vast, making it difficult to charge wireless nodes. Therefore, optimizing link transmission energy efficiency and extending network life are important contents of farm wireless monitoring networks. Node energy consumption is proportional to its transmission power, and the more nodes working on the same channel at the same time, the greater the network interference, which will further aggravate network energy consumption. For this reason, the second objective function is designed, and the farm cognitive network control optimization under multi-objective constraints is designed based on the first objective function and the second objective function. For practical problems, it is almost impossible to obtain a solution that maximizes all objective vector elements in the farm cognitive network control optimization under multi-objective constraints, so it is considered a utopian problem. The solution to this type of problem is usually expressed as Pareto optimality, that is, if there are no other design vectors and only the target solution is solved, which can improve some objective functions without causing at least one other objective function to decrease at the same time, then the target solution is considered to be Pareto optimal while satisfying the constraints.

[0071] After obtaining the first Pareto optimal solution of each sensor node in the first objective function and the second Pareto optimal solution of each sensor node in the second objective function, a first objective variable strategy set corresponding to the first objective function and a second objective variable strategy set corresponding to the second objective function are constructed according to the first Pareto optimal solution and the second Pareto optimal solution.

[0072] Step 102 and step 103 are repeatedly performed until the absolute value of the difference between the current first-level policy set and the iterative first-level policy set is less than or equal to a first preset value, and sensor network transmission is performed according to the iterative first-level policy set.

[0073] In step 102, a first current strategy is randomly generated according to the first target variable strategy set, a second current strategy is randomly generated according to the second target variable strategy set, and a current primary strategy set is obtained according to the first current strategy and the second current strategy. The basic game model consists of three basic elements, namely, participants, strategy sets and utility functions. The present invention uses two target functions as players and divides the set S into the strategy space {S1, S2} owned by the corresponding players, where S1 = {s v ,…,s e},S2={s t ,…,s Q}. Among them, S1∪S2=S, S a ∪S b=Φ(a, b=1,2; a≠b).

[0074] Therefore, after completing fuzzy clustering and set mapping, the present invention maps each single objective function into a benefit function of the game parties in the game process, and the constraint conditions are used to constrain the variable iteration in the game process.

[0075] The multi-parameter optimization game problem of cognitive wireless network in farmland can be expressed as:

[0076] G={S1,S2;u1,u2} (1)

[0077] In formula (1), G is the multi-parameter optimization of cognitive wireless network in farmland. According to the target variable subsets S1 and S2 determined by fuzzy clustering, u1 and u2 are the benefit functions of the game parties. A joint optimization algorithm for the parameters of the transmission power and transmission channel of cognitive wireless network in farmland is designed.

[0078] The first current strategy is randomly generated according to the first target variable strategy set, the second current strategy is randomly generated according to the second target variable strategy set, and the initial feasible strategy is randomly generated in the strategy set of each participant to form a strategy set S (0) =S (0)1 ∪S (0)2 , let the number of iterations T = 0.

[0079] In step 103, the current first-level strategy set is iteratively processed according to the secondary game to determine the iterative first-level strategy set. The present invention uses the first objective function as the benefit function to determine the first final transmission channel corresponding to each sensor node, and then takes the transmission power as the variable to obtain the first final transmission power after taking the first-order derivative of the first final transmission channel; uses the second objective function as the benefit function to determine the second final transmission channel corresponding to each sensor node, and then takes the transmission power as the variable to obtain the second final transmission power after taking the first-order derivative of the second final transmission channel, and finally determines the iterative first-level strategy set according to the first final transmission power and the second final transmission power.

[0080] In step 104, until the absolute value of the difference between the current first-level policy set and the iterated first-level policy set is less than or equal to a first preset value, sensor network transmission is performed according to the iterated first-level policy set. After determining the iterated first-level policy set, the present invention determines whether the iterated first-level policy set meets the convergence condition, that is, the absolute value of the difference between the current first-level policy set and the iterated first-level policy set is less than or equal to a first preset value. If so, the game ends. If not, the above steps 102 and 103 are repeated until the algorithm meets the convergence condition and the iteration ends, that is, the transmission power and transmission channel of each node after optimization are obtained, and sensor network transmission is performed according to the transmission power and transmission channel of each node after optimization.

[0081] Optionally, performing sensor network transmission according to the iterated first-level strategy set includes: determining a final transmission power and a final transmission channel of each sensor node according to the iterated first-level strategy set, and performing transmitter network transmission according to the final transmission power and the final transmission channel of each sensor node.

[0082] Those skilled in the art understand that in order to achieve intensive and accurate perception of crop growth and farmland environment, the present invention constructs a farmland cognitive network model and a multi-objective network performance optimization method based on non-cooperative game theory. The densely distributed nodes in the network are used as variables for optimizing the cognitive network, and the node transmission power and transmission channel are used to construct the variable tuple of each node. The optimal solution of each objective function is solved separately to obtain the transmission power and transmission channel of each node for the purpose of optimizing different network performance parameters. Then, in the feasible solution space, the influence of each node on different objective functions is explored; then, the fuzzy clustering theory is used to divide the variable tuples of different nodes into strategy spaces of different objective functions; finally, different objective functions are used as participants in the non-cooperative game. In the divided strategy subspace, different game parties use network signal-to-noise ratio, network energy consumption, etc. as benefits to further optimize the sensor node transmission power and transmission channel.

[0083] The efficient and precise agricultural decision-making system has high requirements for the data transmission integrity, network delay, network capacity and transmission signal-to-noise ratio of farmland sensor networks. However, due to limited resources, there are conflicts between some network performance indicators. For example, when optimizing network energy consumption, it is necessary to reduce the transmission power as much as possible. However, it is difficult to ensure the communication quality and network capacity of the network if the transmission power is too low. Increasing the transmission power of any node in the farmland wireless network is likely to cause interference to its neighboring nodes. The neighboring nodes increase the transmission power to counter the interference, which leads to the deterioration of network performance. Therefore, it is difficult to meet the decision-making needs of agricultural production operations with a single network performance optimization target, and it is not suitable for farmland wireless monitoring networks with mutually coupled network performance indicators. Existing multi-objective optimization methods usually adopt empirical weighting and custom benefit functions. The optimization results are sensitive to the weight coefficient and the form of the benefit function, and the empirical weighting lacks objectivity and cannot take into account the improvement of the overall network performance. In addition, the farm network state is affected by crop growth and open environment, and the topology changes frequently. The empirical weighting and custom benefit function are not suitable for complex agricultural wireless sensor networks. In response to the above problems, the present invention intends to combine fuzzy clustering and non-cooperative game methods to study a farm cognitive network model for intensive crop information perception, and transform the multi-objective network performance optimization problem into a non-cooperative game model to jointly optimize the transmission power and transmission channel of the farm cognitive wireless sensor network.

[0084] Farmland is vast, the terrain is complex, and the crops are diverse, so scene perception needs to be achieved through dense deployment of sensors. However, the increase in the number of nodes will lead to a proportional increase in the amount of data, exacerbating the current problem of tight spectrum resources in wireless sensor networks, resulting in unsatisfactory transmission quality and network energy consumption. Cognitive radio technology can not only improve network capacity and reduce network overhead, but also alleviate the current situation of tight network resources. Using cognitive wireless networks to control transmission power and channel allocation can optimize network performance, extend network life cycle, and thus improve the monitoring quality of farmland environment. The present invention studies the farm cognitive wireless sensor network transmission technology suitable for intensive perception of crop information, and adopts cognitive radio (CR) to enable authorized users and unauthorized users to better utilize available channel resources on the same bandwidth, which can greatly improve the performance of wireless sensor networks and alleviate the problem of spectrum scarcity.

[0085] The distribution of the environment in different areas of farmland is quite different, and sensors need to be densely deployed to achieve accurate monitoring. The rapid promotion of the Internet of Things also affects the resource utilization of agricultural wireless sensor networks. In response to this problem, the present invention constructs a multi-objective cognitive network performance optimization model that optimizes network signal-to-noise interference and reduces network energy consumption. It combines the node residual energy and the neighbor node information of each node to explore the influence of parameters such as the transmission power and transmission channel of the cognitive wireless network on different network performances on the basis of ensuring network connectivity. Based on the multi-objective optimization model, a non-cooperative game method is disclosed to obtain a farm cognitive wireless network parameter optimization scheme under multi-objective compromise. The farm cognitive wireless network parameter optimization scheme studies the strategy space partitioning method based on fuzzy transfer packet clustering, which can effectively divide the power and channel of different network nodes into the strategy subspace of each game node, avoiding the subjective influence caused by methods such as experience weighting in the traditional multi-objective optimization process. Based on the strategy space partitioning results, a non-cooperative game network performance optimization model is constructed, which can not only guarantee the collective interests of all nodes, but also consider the individual interests of each node.

[0086] The present invention provides a farmland cognitive sensor network transmission method, device, electronic device and medium. The method comprises the following steps: obtaining a current first-level strategy set according to a first current strategy and a second current strategy; iteratively processing the current first-level strategy set according to a secondary game to determine a first-level strategy set after iteration; until the difference between the current first-level strategy set and the first-level strategy set after iteration is less than or equal to a first preset value, performing sensor network transmission according to the optimal transmission power and transmission channel determined by the first-level strategy set after iteration. When optimizing and solving the game nodes, the present invention designs a double-layer game model for the non-convex problem of the benefit function, takes each sensor node as a first-level game party, and uses the transmission power and transmission channel of each sensor node as the strategy space of the secondary game party, and uses the best response principle to obtain the optimization results of network power control and channel allocation, thereby converting the optimization problem of multiple network performance parameters into two low-dimensional problems, reducing the complexity of solving the problem, and being able to adjust the transmission power and transmission channel according to the dynamic influence sensitivity of different nodes on different network performances, thereby achieving improvement in transmission quality and network capacity, and reducing network energy consumption.

[0087] Figure 2 The second flow diagram of the farmland cognitive sensor network transmission method provided by the present invention includes: before constructing a first objective variable strategy set corresponding to the first objective function and a second objective variable strategy set corresponding to the second objective function according to a first Pareto optimal solution of each sensor node in the first objective function and a second Pareto optimal solution of each sensor node in the second objective function, the first objective variable strategy set corresponding to the first objective function and the second objective variable strategy set corresponding to the second objective function are constructed,

[0088] Step 201: for each sensor node in the sensor network, determine the transmission time according to the transmission bandwidth and the transmission rate;

[0089] Step 202: determining the total power of all sensor nodes according to the path gain from the current sensor node to the next hop node and the transmission power of the next hop node;

[0090] Step 203: determining a first power value according to the total power of all sensor nodes and the background noise power;

[0091] Step 204: determining a second power value according to the transmit power of the current sensor node and the next hop path gain of the current node;

[0092] Step 205: determining a power ratio according to the first power value and the second power value;

[0093] Step 206: determining the signal to interference and noise ratio of the current sensor node according to the transmission time and the power ratio;

[0094] Step 207: construct a first objective function according to the signal to interference plus noise ratio (SINR) and the SINR threshold of each sensor node in the sensor network.

[0095] In step 201, the present invention assumes that there is a primary user in the farmland cognitive wireless network, that is, an authorized user, which refers to multiple types of mobile devices, and N secondary users, that is, densely distributed sensor nodes in the farmland, work together to form a sensor set S = {s1, s2, ..., s N}. The spectrum resources are divided into M mutually orthogonal wireless channels, represented as U = {channel1, channel2, ..., channelM}. Each node can only select one channel at any time, and the channel selection of node i is marked as c i , c i ∈U, for each sensor node in the sensor network, the transmission time is determined according to the quotient of the transmission bandwidth and the transmission rate.

[0096] In step 202, the total power of all sensor nodes is determined according to the path gain from the current sensor node to the next hop node and the transmission power of the next hop node. Specifically, the total power of all sensor nodes can be determined by It indicates that, is the path gain from the current sensor node to the next hop node, p j is the transmit power of the next hop node.

[0097] In step 203, optionally, a first power value is determined according to the sum of the total power of all sensor nodes and the background noise power.

[0098] In step 204, optionally, a second power value is determined according to the product of the transmit power of the current sensor node and the next-hop path gain of the current node.

[0099] In step 205, a power ratio is determined according to a quotient of the second power value and the first power value.

[0100] In step 206, the signal to interference and noise ratio of the current sensor node is determined according to the product of the transmission time and the power ratio. Specifically, according to the physical interference model, the signal to interference and noise ratio of the sensor node i can be expressed as:

[0101]

[0102] In formula (2), W represents the transmission bandwidth, R represents the transmission rate, and c i represents the channel selected by node i, p i represents the transmit power of node i, represents the path gain from node i to the next hop, represents the path gain from node i to node j, N σ is the background noise power, and I is the set of neighbor nodes working in the same channel as node i.

[0103] In step 207, a first objective function is constructed according to the signal to interference plus noise ratio (SINR) and the SINR threshold of each sensor node in the sensor network. The construction of the first objective function may refer to the following formula:

[0104]

[0105] In formula (3), F1 is the first objective function, γ i is the signal-to-interference-noise ratio, γ th is the signal-to-interference-noise ratio threshold.

[0106] More specifically, for the second objective function, the following formula can be used:

[0107]

[0108] In formula (4), q i is the number of nodes working on the same channel as node i, p i is the transmit power of node i.

[0109] Combining formula (3) and formula (4), the optimal design of farm cognitive network control under multi-objective constraints in the present invention is as follows:

[0110]

[0111] In formula (5), F is the optimal target vector of the objective function, X = {(c1, p1), (c2, p2), ..., (c N , p N )} is a variable vector, which includes the transmission power of each node and the selected channel. The present invention constructs a first objective variable strategy set corresponding to the first objective function and a second objective variable strategy set corresponding to the second objective function according to the first Pareto optimal solution of each sensor node in the first objective function and the second Pareto optimal solution of each sensor node in the second objective function, wherein the set composed of the first Pareto optimal solution in the first objective function and the second Pareto optimal solution in the second objective function is the Pareto optimal in this application. The present invention designs the result of farm cognitive network control optimization under multi-objective constraints according to the first objective function and the second objective function, and adopts Pareto optimality to express it, that is, if there are no other design vectors, only solving the target solution can improve some objective functions without causing at least one other objective function to decrease at the same time, then it is considered that the target solution is Pareto optimal under the condition of satisfying the constraints.

[0112] In view of the limited spectrum resources, long monitoring cycle, and limited energy of dense wireless sensors in farms, the present invention proposes a farm cognitive network model for the purpose of improving the network signal-to-interference-noise ratio and reducing network energy consumption. The model improves the network signal-to-noise-noise ratio to a limited extent based on the signal-to-noise-noise ratio threshold on the basis of the S-type function, thereby improving the transmission quality and network capacity. Based on the positive correlation between the transmit power and energy consumption of cognitive wireless nodes, the present invention constructs an objective function for reducing network energy consumption by minimizing the transmit power. In addition, considering that the more nodes share a channel, the greater the network interference, which easily leads to an increase in the overall network transmit power. The network energy consumption optimization function is modified using the number of nodes contained in each channel, and the network energy consumption is optimized from the two aspects of transmit power control and channel allocation.

[0113] Figure 3 The present invention provides a flowchart of constructing a first target variable strategy set and a second target variable strategy set, wherein the first target variable strategy set corresponding to the first target function and the second target variable strategy set corresponding to the second target function are constructed according to the first Pareto optimal solution of each sensor node in the first target function and the second Pareto optimal solution of each sensor node in the second target function, including:

[0114] Step 1011, calculating a first Euclidean distance of a first Pareto optimal solution of each sensor node to a first objective function, and calculating a second Euclidean distance of a second Pareto optimal solution of each sensor node to a second objective function;

[0115] Step 1012: normalize all first Euclidean distances to obtain a first normalized result of each sensor node, normalize all second Euclidean distances to obtain a second normalized result of each sensor node, and construct a fuzzy relationship matrix according to the first normalized result and the second normalized result;

[0116] Step 1013: Process the fuzzy relationship matrix using a square method to obtain a fuzzy equivalence matrix, cluster the fuzzy equivalence matrix, and obtain a first objective variable strategy set corresponding to the first objective function and a second objective variable strategy set corresponding to the second objective function.

[0117] In step 1011, the farmland cognitive wireless network contains two objective functions. In order to obtain the Pareto optimal solution of the multi-objective function, the present invention first optimizes each objective function separately, and combines formula (5) to obtain the optimal solution F1(X*1), F2(X*2). j ={x* 1j , x* 2j , …, , x* Nj}, (j=1, 2), where x* qj ={p* qj , c* qj}, the solution of each node consists of optimizing the transmission power and optimizing the transmission channel selection, and then each design variable x ji Divide into L segments with a step size of Δ. Calculate x based on the Euclidean distance ij To target F j Impact:

[0118]

[0119] In step 1012, the normalization method of the impact factor can refer to the following formula:

[0120]

[0121] Based on the above influencing factors, the fuzzy relationship matrix Re between the design variables and the objective function is established. kl}N×N. Among them r kl It represents the correlation similarity between the kth design variable and the lth design variable. The Euclidean distance method can be used to obtain r kl :

[0122] r kl =1-w·d(x k ,x l ) (8)

[0123] In formula (8), d(x k ,x l) can refer to the following formula:

[0124]

[0125] In formula (9), w is the value that all r kl ∈[0,1], in order to effectively analyze the impact of different nodes on network performance, the relationship between different objective functions needs to be equivalently corrected, and the transitive closure of the fuzzy similarity matrix is ​​obtained using the square method.

[0126] In step 1013, combined with step 1012, in the transitive closure method, there exists a condition satisfying t(Re)=Re h The smallest natural number h (h ≤ N), for all natural numbers z greater than h, there is Re z =Re h , the present invention uses the square method to obtain t(Re), and squares the original similarity matrix Re one by one: Re→Re 2 →Re 4 →…→Re 2tr →…, by squaring the matrix, let Re h ×Re h =Re h Re h is the fuzzy equivalence matrix t(Re). According to the fuzzy equivalence matrix, fuzzy clustering of network nodes can be realized, and the set S is divided into strategy spaces S1 and S2 belonging to objective functions F1 and F2.

[0127] The present invention provides a method for dividing farmland cognitive network nodes based on fuzzy clustering. In view of the problem that traditional optimization methods are not applicable to farm cognitive wireless networks with limited energy and frequent state changes, this paper uses non-cooperative game theory to take the network signal-to-noise ratio and network energy consumption objective functions that affect the transmission quality of farmland networks as game participants, and takes the transmission power and transmission channel parameters of each node in the network as design variable tuples. By optimizing each objective function separately, the transmission power and transmission power of different groups of nodes are obtained, and then in the feasible space of the solution, the influence factor of the design variable tuple on the objective function is defined, and a fuzzy similarity matrix is ​​obtained based on the influence factor. The transitive closure method is used to divide multiple node variables into non-intersecting strategy spaces belonging to different objective function game parties. Among them, in the process of calculating the influence factors of different variables on each optimization target, considering the different effects of power and channel in each node variable tuple on network performance, the present invention quantifies the influence of each node on different networks by comprehensively considering the objective function changes caused by individual power changes and individual channel changes.

[0128] Figure 4It is a flow chart of determining the first-level strategy set after iteration provided by the present invention, wherein the iterative processing of the current first-level strategy set according to the secondary game to determine the first-level strategy set after iteration includes:

[0129] Step 1031: Based on the first iteration, process the current primary policy set to obtain a first final transmission channel and a second final transmission channel, and determine an iterative secondary policy set according to the first final transmission channel and the second final transmission channel;

[0130] Step 1032: Process the iterated secondary policy set based on the second iteration to obtain a first final transmit power and a second final transmit power, and determine the iterated primary policy set according to the first final transmit power and the second final transmit power.

[0131] In step 1031, in the process of iteratively processing the current primary strategy set according to the secondary game, it is necessary to undergo a first iteration and a second iteration. During the first iteration, it is necessary to ensure that the transmission power of each sensor node in the current secondary strategy set remains unchanged, and determine the first final transmission channel and the second final transmission channel during the first iteration. The first final transmission channel and the second final transmission channel are the optimal channels, and then the secondary strategy set after iteration is determined based on the first final transmission channel and the second final transmission channel.

[0132] In step 1032, the iterative secondary strategy set is processed based on the second iteration to obtain the first final transmission power and the second final transmission power, and the iterative primary strategy set is determined based on the first final transmission power and the second final transmission power. In the second iteration, it is necessary to ensure that the final transmission channel of each sensor node in the iterative secondary strategy set remains unchanged, and the optimal transmission power, i.e., the first final transmission power and the second final transmission power, is determined during the continuous second iteration process, and the iterative primary strategy set is determined based on the first final transmission power and the second final transmission power.

[0133] The present invention is based on a joint optimization mechanism of farmland cognitive network parameters based on a double-layer non-cooperative game mechanism. The design variables composed of farmland cognitive network nodes are divided into game party strategies belonging to different objective functions by using fuzzy clustering theory. On this basis, a joint optimization mechanism of farmland cognitive network parameters based on a double-layer non-cooperative game is proposed. Specifically, according to the strategy space, each node optimizes its own transmission power and channel selection as a game party according to different objectives in the divided parameter strategy space. During the game process, the objective function of each objective function is mapped to the benefit function of a single node, and each node in the network is used as a secondary game party, and the corresponding transmission power and channel selection are used as the strategy space of the secondary game party. The nodes in the network exchange information with each other to determine the transmission power and channel selection that maximize their own benefits. After the game iteration is completed, the optimal transmission power and optimal transmission channel assigned to each node can be obtained.

[0134] Figure 5 The present invention provides a flow chart of determining a secondary strategy set after iteration, wherein the current primary strategy set is processed based on the first iteration, a first final transmission channel and a second final transmission channel are obtained, and the secondary strategy set after iteration is determined according to the first final transmission channel and the second final transmission channel, including:

[0135] Repeat the following steps:

[0136] Step 10311: when the transmission power of each sensor node in the current secondary strategy set remains unchanged, a first target transmission channel corresponding to each sensor node in the first target function strategy set is determined using the first target function as the benefit function, and a second target transmission channel corresponding to each sensor node in the second target function strategy set is determined using the second target function as the benefit function, wherein the current secondary strategy set is determined based on the current primary strategy set;

[0137] Step 10312: Update the current secondary strategy set according to the first target transmission channel and the second target transmission channel to obtain an iterated secondary strategy set;

[0138] Step 10313: until the absolute value of the difference between the current secondary strategy set and the iterated secondary strategy set is less than or equal to a second preset value, obtain the first final transmission channel and the second final transmission channel.

[0139] In step 10311, the current secondary strategy set is first determined based on the current primary strategy set, and then the transmission power of each sensor node in the current secondary strategy set is kept unchanged, the transmission channel is changed, and the node transmission channel that optimizes its benefit function is selected.

[0140] In step 10312, the current secondary strategy set is updated according to the first target transmission channel and the second target transmission channel to obtain the iterated secondary strategy set. After determining the first target transmission channel and the second target transmission channel, the present invention can update the current secondary strategy set according to the changed target transmission channel. For the convenience of description, the updated current secondary strategy set is the iterated secondary strategy set. At this time, the iterated secondary strategy set is not the final optimized strategy set and further iteration is required.

[0141] In the present invention, let S (T) -1 ,S (T) -2 The initial strategy set S is (T) S (T) 1, S (T) 2. The strategy combination selected by other players, for any game goal, maintain S (T) -j Unchanged, let its benefit function F j (S* j , S* -j )→max / min, solve S*. Since the benefit function Fj is a non-convex function that is difficult to solve, the present invention converts the optimization solution into two target optimization functions as the primary game of the game player and the secondary game of each node in the strategy set as the game player. The benefit function of each secondary game player is the corresponding F1 or F2. Let the number of iterations in the channel iteration t0=0, and let the number of iterations in the power game t1=0.

[0142] In step 10313, in the first iteration of the secondary game, each player first keeps the transmission power unchanged, selects the node transmission channel that makes its benefit function optimal, and determines whether the strategy set meets the convergence condition || S (t0+1) -S (t0) ||≤δ, if it does not meet, let S (t0) =S (t0+1) , t0=t0+1, that is, if it is not satisfied, repeat step 10313 until the absolute value of the difference between the current secondary strategy set and the iterated secondary strategy set is less than or equal to the second preset value, and obtain the first final transmission channel and the second final transmission channel. At this time, the first final transmission channel and the second final transmission channel are the optimal transmission channels.

[0143] Figure 6It is a flow chart of determining a primary strategy set after iteration provided by the present invention, wherein the secondary strategy set after iteration is processed based on the second iteration, a first final transmit power and a second final transmit power are obtained, and the primary strategy set after iteration is determined according to the first final transmit power and the second final transmit power, including:

[0144] Repeat the following steps:

[0145] Step 10321: When the final transmission channel of each sensor node in the iterative secondary strategy set remains unchanged, for each secondary strategy set in the iterative secondary strategy set, take the transmission power as a variable to obtain the first-order derivative of the final transmission channel, and when the first-order derivative is 0, obtain the first final transmission power and the second final transmission power;

[0146] Step 10322: Update the iterated secondary strategy set according to the first final transmit power and the second final transmit power to obtain a target secondary strategy set;

[0147] Step 10323: until the absolute value of the difference between the iterated secondary policy set and the target secondary policy set is less than or equal to a third preset value, determine that the target secondary policy set is the iterated primary policy set.

[0148] In step 10321, after determining the final transmission channel, that is, determining the first final transmission power and the second final transmission power, the present invention ensures that the final transmission channel of each sensor node in the secondary strategy set after the iteration remains unchanged, takes the transmission power as a variable, performs a first-order derivative on u1, and obtains the node transmission power in the F1 strategy set by setting the first-order derivative to 0; solves the transmission power that minimizes u2 according to the signal-to-interference-noise ratio threshold, and obtains the node transmission power in the F2 strategy set.

[0149] In step 10322, the iterated secondary strategy set is updated according to the first final transmission power and the second final transmission power to obtain a target secondary strategy set. After determining the first final transmission power and the second final transmission power, the present invention can update the iterated secondary strategy set according to the changed transmission power. For the convenience of description, the updated iterated secondary strategy set is the target secondary strategy set. At this time, the iterated secondary strategy set is not the final optimized strategy set and further iteration is required.

[0150] In step 10323, the second iteration process is continuously performed to determine whether the strategy set meets the convergence condition || S (t1 +1) -S (t1) ||≤δ, if it does not meet the requirement, let S(t1) =S (t1+1) , t1=t1+1, that is, until the absolute value of the difference between the iterative secondary strategy set and the target secondary strategy set is less than or equal to the third preset value, until the algorithm meets the convergence condition, it is determined that the target secondary strategy set is the iterative primary strategy set.

[0151] After executing the iteration of step 10323, all secondary games are completed. However, for the primary game, only one corresponding iteration of the primary game is completed. Specifically, in the present invention, the secondary games of two primary games are executed in sequence, and S*1 and S*2 are obtained according to the output transmission power and the transmission channel. Let S (T+1) =S*1∪S*2, judgment strategy set S (T) Whether the convergence condition is met||S (T+1) -S (T) ||≤δ, if it is satisfied, the game ends, otherwise, let S (T) =S (T +1) , return to step 102, and repeat the above steps until the algorithm meets the convergence condition, and the iteration ends. The final iteration result includes the optimized transmission power and transmission channel of each node.

[0152] Figure 7 It is a structural schematic diagram of the farmland cognitive sensor network transmission device provided by the present invention. The present invention provides a farmland cognitive sensor network transmission device, including a construction unit 1: used to construct a first objective variable strategy set corresponding to the first objective function and a second objective variable strategy set corresponding to the second objective function according to a first Pareto optimal solution of each sensor node in the first objective function and a second Pareto optimal solution of each sensor node in the second objective function. The working principle of the construction unit 1 can refer to the aforementioned step 101 and will not be repeated here.

[0153] The farmland cognitive sensor network transmission device further includes a repeating unit 2: for repeatedly executing the following steps:

[0154] Randomly generate a first current strategy according to the first target variable strategy set, randomly generate a second current strategy according to the second target variable strategy set, and obtain a current primary strategy set according to the first current strategy and the second current strategy;

[0155] The current first-level strategy set is iteratively processed according to the secondary game to determine the first-level strategy set after iteration. The working principle of the repetition unit 2 can refer to the aforementioned steps 102 and 103 and will not be described in detail here.

[0156] The farmland cognitive sensor network transmission device also includes a transmission unit 3: used to perform sensor network transmission according to the iterated first-level strategy set until the absolute value of the difference between the current first-level strategy set and the iterated first-level strategy set is less than or equal to a first preset value. The working principle of the transmission unit 3 can refer to the aforementioned step 104 and will not be repeated here.

[0157] The first objective function is determined according to the signal to interference noise ratio of each sensor node in the sensor network, and the second objective function is constructed according to the transmission power of each sensor node and the number of sensors in the same channel;

[0158] The current first-level strategy set is iteratively processed according to the secondary game, and is used to determine the first final transmission channel corresponding to each sensor node using the first objective function as the benefit function, and then obtain the first final transmission power after taking the first-order derivative of the first final transmission channel with the transmission power as the variable; determine the second final transmission channel corresponding to each sensor node using the second objective function as the benefit function, and then obtain the second final transmission power after taking the first-order derivative of the second final transmission channel with the transmission power as the variable.

[0159] The present invention provides a farmland cognitive sensor network transmission method, device, electronic device and medium. The method comprises the following steps: obtaining a current first-level strategy set according to a first current strategy and a second current strategy; iteratively processing the current first-level strategy set according to a secondary game to determine a first-level strategy set after iteration; until the difference between the current first-level strategy set and the first-level strategy set after iteration is less than or equal to a first preset value, performing sensor network transmission according to the optimal transmission power and transmission channel determined by the first-level strategy set after iteration. When optimizing and solving the game nodes, the present invention designs a double-layer game model for the non-convex problem of the benefit function, takes each sensor node as a first-level game party, and uses the transmission power and transmission channel of each sensor node as the strategy space of the secondary game party, and uses the best response principle to obtain the optimization results of network power control and channel allocation, thereby converting the optimization problem of multiple network performance parameters into two low-dimensional problems, reducing the complexity of solving the problem, and being able to adjust the transmission power and transmission channel according to the dynamic influence sensitivity of different nodes on different network performances, thereby achieving improvement in transmission quality and network capacity, and reducing network energy consumption.

[0160] Figure 8 Schematic diagram of the structure of the electronic device provided by the present invention. Figure 8As shown, the electronic device may include: a processor 810, a communication interface 820, a memory 830 and a communication bus 840, wherein the processor 810, the communication interface 820 and the memory 830 communicate with each other through the communication bus 840. The processor 810 may call the logic instructions in the memory 830 to execute the farmland cognitive sensor network transmission method, the method comprising: constructing a first target variable strategy set corresponding to the first target function and a second target variable strategy set corresponding to the second target function according to the first Pareto optimal solution of each sensor node in the first target function and the second Pareto optimal solution of each sensor node in the second target function; repeatedly executing the following steps: randomly generating a first current strategy according to the first target variable strategy set, randomly generating a second current strategy according to the second target variable strategy set, and obtaining a current primary strategy set according to the first current strategy and the second current strategy; iteratively processing the current primary strategy set according to the secondary game to determine the iterated primary strategy set; until the current primary strategy set and the iterated primary strategy set are consistent. The absolute value of the difference of the first-level strategy set is less than or equal to the first preset value, and the sensor network transmission is performed according to the iterated first-level strategy set; the first objective function is determined according to the signal-to-interference-noise ratio of each sensor node in the sensor network, and the second objective function is constructed according to the transmission power of each sensor node and the number of sensors under the same channel; the current first-level strategy set is iteratively processed according to the secondary game to determine the first final transmission channel corresponding to each sensor node using the first objective function as the benefit function, and then obtain the first final transmission power by taking the first order derivative of the first final transmission channel with the transmission power as the variable; the second final transmission channel corresponding to each sensor node is determined using the second objective function as the benefit function, and then obtain the second final transmission power by taking the first order derivative of the second final transmission channel with the transmission power as the variable.

[0161] In addition, the logic instructions in the above-mentioned memory 830 can be implemented in the form of software functional units and can be stored in a computer-readable storage medium when it is sold or used as a product to be analyzed. Based on such an understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art or the part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.

[0162] On the other hand, the present invention also provides a computer program product, which includes a computer program. The computer program can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute a farmland cognitive sensor network transmission method provided by the above methods. The method includes: according to the first Pareto optimal solution of each sensor node in the first objective function and the second Pareto optimal solution of each sensor node in the second objective function, construct a first objective variable strategy set corresponding to the first objective function and a second objective variable strategy set corresponding to the second objective function; repeatedly perform the following steps: randomly generate a first current strategy according to the first objective variable strategy set, randomly generate a second current strategy according to the second objective variable strategy set, and obtain a current first-level strategy set according to the first current strategy and the second current strategy; iteratively process the current first-level strategy according to the secondary game a strategy set, and determine the first-level strategy set after iteration; until the absolute value of the difference between the current first-level strategy set and the first-level strategy set after iteration is less than or equal to the first preset value, the sensor network transmission is performed according to the first-level strategy set after iteration; the first objective function is determined according to the signal to interference noise ratio of each sensor node in the sensor network, and the second objective function is constructed according to the transmission power of each sensor node and the number of sensors under the same channel; the current first-level strategy set is iteratively processed according to the secondary game, and is used to determine the first final transmission channel corresponding to each sensor node using the first objective function as the benefit function, and then obtain the first final transmission power after taking the first order derivative of the first final transmission channel with the transmission power as the variable; determine the second final transmission channel corresponding to each sensor node using the second objective function as the benefit function, and then obtain the second final transmission power after taking the first order derivative of the second final transmission channel with the transmission power as the variable.

[0163] On the other hand, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which is implemented by a processor to perform the above-mentioned methods to provide a farmland cognitive sensor network transmission method, the method comprising: constructing a first target variable strategy set corresponding to the first target function and a second target variable strategy set corresponding to the second target function according to a first Pareto optimal solution of each sensor node in the first target function and a second Pareto optimal solution of each sensor node in the second target function; repeatedly performing the following steps: randomly generating a first current strategy according to the first target variable strategy set, randomly generating a second current strategy according to the second target variable strategy set, and obtaining a current first-level strategy set according to the first current strategy and the second current strategy; iteratively processing the current first-level strategy set according to the secondary game to determine the first-level strategy set after iteration; Until the absolute value of the difference between the current first-level strategy set and the iterated first-level strategy set is less than or equal to the first preset value, the sensor network transmission is performed according to the iterated first-level strategy set; the first objective function is determined according to the signal-to-interference-noise ratio of each sensor node in the sensor network, and the second objective function is constructed according to the transmission power of each sensor node and the number of sensors under the same channel; the current first-level strategy set is iteratively processed according to the secondary game to determine the first final transmission channel corresponding to each sensor node using the first objective function as the benefit function, and then obtain the first final transmission power by taking the first order derivative of the first final transmission channel with the transmission power as the variable; the second final transmission channel corresponding to each sensor node is determined using the second objective function as the benefit function, and then obtain the second final transmission power by taking the first order derivative of the second final transmission channel with the transmission power as the variable.

[0164] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0165] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0166] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A farmland cognitive sensor network transmission method, characterized in that: include: According to a first Pareto optimal solution of each sensor node in the first objective function and a second Pareto optimal solution of each sensor node in the second objective function, construct a first objective variable strategy set corresponding to the first objective function and a second objective variable strategy set corresponding to the second objective function; Repeat the following steps: Randomly generate a first current strategy according to the first target variable strategy set, randomly generate a second current strategy according to the second target variable strategy set, and obtain a current primary strategy set according to the first current strategy and the second current strategy; Iteratively process the current primary strategy set according to the secondary game to determine a primary strategy set after iteration; Until the absolute value of the difference between the current first-level strategy set and the iterated first-level strategy set is less than or equal to a first preset value, the sensor network transmission is performed according to the iterated first-level strategy set; The first objective function is determined according to the signal to interference noise ratio of each sensor node in the sensor network, and the second objective function is constructed according to the transmission power of each sensor node and the number of sensors in the same channel; The current primary strategy set is iteratively processed according to the secondary game, for determining the first final transmission channel corresponding to each sensor node using the first objective function as the benefit function, and then taking the transmission power as a variable to obtain the first final transmission power after taking the first order derivative of the first final transmission channel; The second final transmission channel corresponding to each sensor node is determined by taking the second objective function as the benefit function, and the second final transmission power is obtained by taking the first-order derivative of the second final transmission channel with the transmission power as a variable.

2. The farmland cognitive sensor network transmission method according to claim 1, characterized in that: Before constructing a first objective variable strategy set corresponding to the first objective function and a second objective variable strategy set corresponding to the second objective function according to a first Pareto optimal solution of each sensor node in the first objective function and a second Pareto optimal solution of each sensor node in the second objective function, the method further includes: For each sensor node in the sensor network, the transmission time is determined according to the transmission bandwidth and the transmission rate; Determine the total power of all sensor nodes based on the path gain from the current sensor node to the next hop node and the transmission power of the next hop node; Determine a first power value according to the total power of all sensor nodes and the background noise power; Determine a second power value according to the transmit power of the current sensor node and the next hop path gain of the current node; Determine a power ratio according to the first power value and the second power value; Determine the signal to interference and noise ratio of the current sensor node according to the transmission time and the power ratio; A first objective function is constructed according to the signal to interference plus noise ratio (SINR) of each sensor node in the sensor network and a SINR threshold.

3. The farmland cognitive sensor network transmission method according to claim 1, characterized in that: The method of constructing a first objective variable strategy set corresponding to the first objective function and a second objective variable strategy set corresponding to the second objective function according to a first Pareto optimal solution of each sensor node in the first objective function and a second Pareto optimal solution of each sensor node in the second objective function includes: Calculate a first Euclidean distance between a first Pareto optimal solution of each sensor node and a first objective function, and calculate a second Euclidean distance between a second Pareto optimal solution of each sensor node and a second objective function; Normalizing all first Euclidean distances to obtain a first normalized result of each sensor node, normalizing all second Euclidean distances to obtain a second normalized result of each sensor node, and constructing a fuzzy relationship matrix according to the first normalized result and the second normalized result; The fuzzy relationship matrix is ​​processed by a square method to obtain a fuzzy equivalence matrix, the fuzzy equivalence matrix is ​​clustered, and a first objective variable strategy set corresponding to the first objective function and a second objective variable strategy set corresponding to the second objective function are obtained.

4. The farmland cognitive sensor network transmission method according to claim 1, characterized in that: The iterative processing of the current primary strategy set according to the secondary game to determine the iterated primary strategy set includes: Processing the current primary policy set based on the first iteration, obtaining a first final transmission channel and a second final transmission channel, and determining a post-iteration secondary policy set according to the first final transmission channel and the second final transmission channel; The iterated secondary policy set is processed based on the second iteration to obtain a first final transmit power and a second final transmit power, and a post-iteration primary policy set is determined according to the first final transmit power and the second final transmit power.

5. The farmland cognitive sensor network transmission method according to claim 4 is characterized in that: The first iterative processing of the current primary policy set, obtaining a first final transmission channel and a second final transmission channel, and determining an iterative secondary policy set according to the first final transmission channel and the second final transmission channel, includes: Repeat the following steps: Under the condition that the transmission power of each sensor node in the current secondary strategy set remains unchanged, the first target transmission channel corresponding to each sensor node in the first target function strategy set is determined by taking the first target function as the benefit function, and the second target transmission channel corresponding to each sensor node in the second target function strategy set is determined by taking the second target function as the benefit function, wherein the current secondary strategy set is determined according to the current primary strategy set; updating the current secondary strategy set according to the first target transmission channel and the second target transmission channel to obtain an iterated secondary strategy set; Until the absolute value of the difference between the current secondary strategy set and the iterated secondary strategy set is less than or equal to a second preset value, a first final transmission channel and a second final transmission channel are obtained.

6. The farmland cognitive sensor network transmission method according to claim 5, characterized in that: The step of processing the iterated secondary strategy set based on the second iteration, obtaining the first final transmit power and the second final transmit power, and determining the iterated primary strategy set according to the first final transmit power and the second final transmit power, includes: Repeat the following steps: When the final transmission channel of each sensor node in the iterative secondary strategy set remains unchanged, for each secondary strategy set in the iterative secondary strategy set, taking the transmission power as a variable, first-order derivative of the final transmission channel is obtained, and when the first-order derivative is 0, a first final transmission power and a second final transmission power are obtained; updating the iterated secondary strategy set according to the first final transmit power and the second final transmit power to obtain a target secondary strategy set; Until the absolute value of the difference between the iterated secondary strategy set and the target secondary strategy set is less than or equal to a third preset value, the target secondary strategy set is determined to be the iterated primary strategy set.

7. The farmland cognitive sensor network transmission method according to claim 1, characterized in that: The performing sensor network transmission according to the iterated primary strategy set includes: The final transmission power and the final transmission channel of each sensor node are determined according to the iterated first-level strategy set, and the transmitter network transmission is performed according to the final transmission power and the final transmission channel of each sensor node.

8. A farmland cognitive sensor network transmission device, characterized in that: include: A construction unit: configured to construct a first objective variable strategy set corresponding to the first objective function and a second objective variable strategy set corresponding to the second objective function according to a first Pareto optimal solution of each sensor node in the first objective function and a second Pareto optimal solution of each sensor node in the second objective function; Repeating unit: used to repeatedly perform the following steps: Randomly generate a first current strategy according to the first target variable strategy set, randomly generate a second current strategy according to the second target variable strategy set, and obtain a current primary strategy set according to the first current strategy and the second current strategy; Iteratively process the current primary strategy set according to the secondary game to determine a primary strategy set after iteration; A transmission unit: configured to perform sensor network transmission according to the iterative first-level policy set until the absolute value of the difference between the current first-level policy set and the iterative first-level policy set is less than or equal to a first preset value; The first objective function is determined according to the signal to interference noise ratio of each sensor node in the sensor network, and the second objective function is constructed according to the transmission power of each sensor node and the number of sensors in the same channel; The current primary strategy set is iteratively processed according to the secondary game, for determining the first final transmission channel corresponding to each sensor node using the first objective function as the benefit function, and then taking the transmission power as a variable to obtain the first final transmission power after taking the first order derivative of the first final transmission channel; The second final transmission channel corresponding to each sensor node is determined by taking the second objective function as the benefit function, and the second final transmission power is obtained by taking the first-order derivative of the second final transmission channel with the transmission power as a variable.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the farmland cognitive sensor network transmission method according to any one of claims 1 to 7 is implemented.

10. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the farmland cognitive sensor network transmission method according to any one of claims 1 to 7 is implemented.

Images