SDR sensor scheduling optimization method based on Bayesian reasoning

Abstract

The present invention discloses an SDR sensor scheduling optimization method based on Bayesian reasoning, which solves the problem that the existing technology cannot effectively handle sensor scheduling problems with time window constraints. The method includes: determining model parameters of a sensor scheduling optimization model during sensor scheduling; wherein the model parameters include: the state of each sensor in the time period, the state of each static target in the time period, the state of each dynamic target in the time period, and the working state sequence of all sensors; constructing the objective function and constraint conditions of the sensor scheduling optimization model according to the model parameters; solving the objective function according to the constraint conditions based on Bayesian reasoning to obtain the optimal state sequence of sensor scheduling optimization; achieving rapid convergence when effectively handling sensor scheduling problems with time window constraints, and preventing the generation of solutions that violate time constraints during cross-mutation.

Description

Technical Field

[0001] The present invention relates to the field of wireless communication technology, and in particular to an SDR sensor scheduling optimization method based on Bayesian reasoning. Background Art

[0002] In modern wireless communications and IoT systems, the introduction of software-defined radio (SDR) technology offers a new approach to optimizing target detection and tracking. SDR dynamically configures sensor functions through software, enabling real-time optimization and flexible scheduling, thereby improving system adaptability and resource efficiency. This technology offers significant advantages in addressing the resource constraints and complex environment adaptation challenges inherent in traditional approaches.

[0003] In target detection and tracking tasks, SDR sensors adjust their on-time and frequency to accurately monitor the target's location and status, maximizing detection time and coverage. However, this process involves complex issues such as multi-target matching and resource allocation. It requires coordinating sensor resources within multi-dimensional constraints such as time, space, and frequency to ensure efficient detection mission execution. Summary of the Invention

[0004] The present invention solves the problem that the existing technology cannot effectively handle the sensor scheduling problem with time window constraints by providing an SDR sensor scheduling optimization method based on Bayesian reasoning. It achieves rapid convergence when effectively handling the sensor scheduling problem with time window constraints and can prevent the generation of solutions that violate time constraints during cross-mutation.

[0005] The present invention provides an SDR sensor scheduling optimization method based on Bayesian reasoning, the method comprising:

[0006] Determine the model parameters of the sensor scheduling optimization model for sensor scheduling; wherein the model parameters include: each sensor in the first The state of each time period, each static target in the The status of each time period, each dynamic target in the The state of the time period and the sequence of working states of all sensors;

[0007] Constructing the objective function and constraint conditions corresponding to the sensor scheduling optimization model according to the model parameters;

[0008] Under the constraints, the objective function is solved based on Bayesian reasoning to obtain an optimal state sequence for sensor scheduling optimization.

[0009] One or more technical solutions provided in the present invention have at least the following technical effects or advantages:

[0010] The present invention uses Bayesian reasoning to predict the activation state of a sensor in the next generation based on historical and new data, thereby implementing an efficient repair strategy and a mechanism for generating legitimate offspring solutions. This method not only helps improve the effective daily utilization rate of SDR sensors and obtain the optimal state sequence, but also optimizes resource utilization and reduces power loss caused by ineffective activation.

[0011] The present invention uses Bayesian reasoning to not only achieve rapid convergence but also prevent solutions that violate time constraints during cross-mutation, which can effectively handle sensor scheduling problems with time window constraints. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] Figure 1 Flowchart of the SDR sensor scheduling optimization method based on Bayesian reasoning provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0013] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making any creative efforts are within the scope of protection of the present invention.

[0014] The present invention provides an SDR sensor scheduling optimization method based on Bayesian reasoning, such as Figure 1 As shown, the method includes the following steps S101 to S103.

[0015] S101, determining the model parameters of the sensor scheduling optimization model for sensor scheduling; wherein the model parameters include: each sensor in the first The state of each time period, each static target in the The status of each time period, each dynamic target in the The state of the time period and the sequence of working states of all sensors;

[0016] Here, 24 hours a day is divided into time granularity Divided into time period, i.e. , the time set is represented as ; Set in sensor collection , , is the total number of sensors; static target set , , is the total number of static targets; the dynamic target set , , is the total number of dynamic targets.

[0017] Specifically, in step S101, each sensor The status of a time period is represented as: ;in, Indicates sensor longitude; Indicates sensor Latitude; Indicates sensor Detection distance; Indicates sensor In the The frequency set that can be detected in each time period; Indicates sensor In the The first Frequency working status of each frequency point; Indicates sensor In the The number of available channels per time period; Indicates sensor In the The status of a time period.

[0018] Specifically, in step S101, each static target The status of a time period is represented as: ;in, Represents a static target longitude; Represents a static target Latitude; Represents a static target Radiation radius; Represents a static target In the The set of radiation frequencies in a time period; Represents a static target In the Working status of each period; Represents a static target In the The state of each period, static target set , , is the total number of static targets.

[0019] Specifically, in step S101, each dynamic target The status of a time period is represented as: ;in Represents a dynamic target In the The longitude of the starting time of each period; Represents a dynamic target In the The latitude of the starting time of each period; Represents a dynamic target In the The set of radiation frequencies in a time period; Represents a dynamic target In the Working status of each period; Represents a dynamic target In the Status of each period, dynamic target set , , is the total number of dynamic targets.

[0020] Specifically, in step S101, each dynamic target The state of the time period and the working state sequence of all sensors are expressed as: ;in, Indicates sensor The working state sequence, and , sensor collection , , is the total number of sensors.

[0021] S102, constructing the objective function and constraint conditions of the sensor scheduling optimization model according to the model parameters;

[0022] Here, the objective function includes: a first objective function and a second objective function; wherein,

[0023] The first objective function is expressed as:

[0024] ;

[0025] The second objective function is expressed as:

[0026] ;

[0027] in:

[0028] ;

[0029] ;

[0030] ;

[0031] in, Indicates the total number of static targets; Indicates the total number of sensors; Indicates the total number of time periods, ; Indicates sensor Detection range and static targets The area of ​​the intersection of the radiation range; Represents a static target Radiation range; Indicates sensor In the The first frequency points; Represents a static target In the The set of radiation frequencies in a time period; Indicates sensor In the The first Frequency working status of each frequency point; Represents a static target In the Working status of each period; Indicates sensor In the Time slots open Frequency point and static target degree of matching; Represents a dynamic target In the The set of radiation frequencies in a time period; Used to represent dynamic targets In the Is the sensor in the time period within the detection range; Represents a dynamic target In the The set of radiation frequencies in a time period; Indicates sensor In the Time slots open Frequency point and dynamic target degree of matching; Represents a dynamic target In the Working status of each period; Indicates sensor The union of the sets of detectable frequency points in each time period.

[0032] When , it indicates a dynamic target In the The working status of each period is that all frequencies are turned on. When , it indicates a dynamic target In the The working status of each period is that all frequencies are closed;

[0033] When , it indicates a dynamic target In the The sensor Within the detection range, When , it indicates a dynamic target In the The sensor is not present during the period Within detection range;

[0034] When , it indicates a static target In the The working status of each period is that all frequencies are turned on. When , it indicates a static target In the The working status of this period is that all frequencies are closed.

[0035] Here, the constraints include a first constraint, a second constraint, and a third constraint. The first constraint restricts the activation time of each sensor. The second constraint restricts the number of activation times for each frequency point within each time period. The third constraint restricts the activation time of each sensor per day.

[0036] Specifically, the first constraint is expressed as:

[0037] ;

[0038] The second constraint is expressed as:

[0039] ;

[0040] The third constraint is expressed as:

[0041] ;

[0042] in, Indicates sensor In the The first Frequency working status of each frequency point; Indicates sensor The start time of the available opening period; Indicates sensor The end time of the available opening period; Indicates the total number of time periods, ; Indicates sensor The union of the sets of detectable frequency points in each time period; Indicates sensor In the The number of available channels per time period; Indicates sensor The longest opening time in a single day; Indicates the time granularity.

[0043] S103 , under the constraints, solving the objective function based on Bayesian reasoning to obtain the optimal state sequence for sensor scheduling optimization.

[0044] Specifically, in step S103, the objective function is solved based on Bayesian reasoning to obtain the optimal state sequence for sensor scheduling optimization, including: determining the maximum number of iterations , initialize the first frequency working state, and obtain the scheduling scheme set under each iteration according to the cyclic update method; and determine the optimal state sequence of sensor scheduling optimization based on the scheduling scheme set corresponding to the maximum number of iterations.

[0045] According to the cyclic update method, the scheduling scheme set for each iteration is obtained, including:

[0046] (1) According to The set of scheduling solutions after iterations , use the fitness value calculation formula to calculate the fitness corresponding to each state sequence Among them, fitness is positively correlated with the objective function; Represents a set of scheduling schemes Middle A scheduling plan, , Represents a set of scheduling schemes size;

[0047] (2) According to the fitness corresponding to each state sequence , using a binary tournament selection strategy to select The state sequence is put into the mating pool and the The first scheduling solution set after iterations ;

[0048] (3) According to The first scheduling solution set after iterations The state sequence in , using the non-zero rate predicted by Bayesian inference to generate the The second scheduling solution set under iterations ;

[0049] The specific implementation process of (3) is as follows:

[0050] (3.1) Define the size as Unit vector and size Unit vector , and determine the initial value of the weight; among them, the size of the unit vector Expressed as: ; and define a size of The empty matrix ;in, Indicates the total number of sensors; Indicates the total number of time periods; Indicates sensor The number of elements in the union of the sets of detectable frequency points in each time period.

[0051] (3.2) The first scheduling scheme is set The scheduling sequence in is saved in Matrix and initialize the command ;

[0052] (3.3) Statistical matrix Sensors in each scheduling scheme In the Period The frequency point is The number of times the work status is open in the scheduling plan ;

[0053] (3.4) According to the probability calculation formula and the number of times , determine the sensor In the Period The frequency point working state is the frequency of opening;

[0054] (3.5) Calculate the frequency in the matrix The index in , put the frequency into the matrix according to the index middle;

[0055] (3.6) Select the matrix Any column , judge the column Whether the probability corresponding to each element in satisfies the first judgment condition, according to the judgment result, the matrix Modify the elements in to obtain a first modified matrix;

[0056] (3.7) According to the rows and columns of the elements in the first modified matrix, determine the corresponding matrix Elements in and judge the matrix whether the elements in the first modification matrix meet the second judgment condition, and modify the elements in the first modification matrix according to the judgment result to obtain the second modification matrix;

[0057] (3.8) According to the second modified matrix, we get The second scheduling solution set under iterations .

[0058] (4) The second scheduling scheme is combined and a set of scheduling schemes Merge to get the third scheduling solution set ;

[0059] (5) The third scheduling scheme set Perform non-dominated sorting, retain the first W frontiers after each sorting, and delete the scheduling plan with the smallest fitness from the Wth frontier until the third scheduling plan set The size of , and obtain the scheduling plan set for the next iteration.

[0060] For example, in a specific use embodiment provided by the present invention:

[0061] Step 1: Define the current number of iterations as , and initialize ; The maximum number of iterations is set to ;

[0062] Randomly initialize the The sequence of on-states produces the The number of iterations is The set of scheduling schemes ;in, Indicates the The next iteration The state sequence corresponding to the set of scheduling schemes, and ; Indicates the The next iteration The sensors in the scheduling scheme set State sequence , ;

[0063] Step 2: Define two sizes Unit vector: and ; Initialize weights ;in, ;

[0064] Step 3: Define a size of The empty matrix ;

[0065] Step 4: Calculate the scheduling plan set according to the fitness value calculation formula Each state sequence in The fitness value of ;in, express The fitness value is expressed as follows:

[0066] ;

[0067] in, Indicates the Scheduling plan for the next iteration The first target value ; Indicates the The next iteration The second target value ; Wherein, the target value is calculated based on the first objective function and the second objective function.

[0068] Step 5: Use the binary tournament selection strategy to select The scheduling plans are put into the mating pool and called the first scheduling plan set , where the scheduling scheme with greater fitness has a higher probability of being selected.

[0069] Step 6: The non-zero rate predicted by Bayesian inference is obtained from the first scheduling solution set The scheduling scheme in The second scheduling solution set under iterations ;

[0070] (6.1) The first scheduling scheme is set The scheduling plan in Matrix middle;

[0071] (6.2) Initialization ;

[0072] (6.3) Statistics Sensors in the scheduling scheme In the Period The number of frequency points in different working states ;

[0073] (6.4) Calculate the first The sensor in the next iteration In the Period The probability of different working states at each frequency point:

[0074] ;

[0075] in, Indicates the The matrix under the iteration No. Rank elements, ; Represents a unit vector Middle elements; Represents a unit vector Middle elements.

[0076] (6.5) Order , Represents the matrix No. Column; judge separately The corresponding probability of each element Is it satisfied , if satisfied, then The corresponding scheduling scheme in If you are not satisfied, but ,in, ;

[0077] (6.6) Order , ;

[0078] (6.7) If ,make , execute step (6.3);

[0079] like , ,make , , execute step (6.3);

[0080] like , , ,make , , , execute step (6.3);

[0081] like , execute step (6.8);

[0082] (6.8) ;

[0083] (6.9) Selection matrix A random column of elements in the row Go to step (6.10); if Perform step (6.11); Indicates the The matrix under the iteration No. Rank elements;

[0084] (6.10) If ,make Otherwise, let ; Indicates the The matrix under the iteration No. Rank elements.

[0085] like , ;otherwise, ;

[0086] (6.11) , ;

[0087] (6.12) If , execute step (6.13); otherwise, let Execute step (6.9);

[0088] (6.13) Let the second scheduling solution set middle, ;

[0089] (6.14) .

[0090] Step 7: Assemble the scheduling plans and the second scheduling solution set Merge to get the third combined scheduling solution set .

[0091] Step 8: Set the third scheduling solution Perform non-dominated sorting and retain the first W frontier surfaces, where W represents the sum of the number of all scheduling solutions that meet the W frontier surfaces is greater than or equal to The minimum value of

[0092] Step 9: Delete the scheduling solution with the minimum fitness from the Wth frontier surface;

[0093] Step 10: Repeat the process of step 9 until the number of all state sequences in the W frontiers is So far, we get The number of iterations is The set of scheduling solutions for the next iteration .

[0094] The introduction of a Bayesian strategy effectively enhances the algorithm's global search capabilities. By dynamically updating the posterior probability, the genetic operator is guided to generate high-quality solutions, reducing the risk of falling into local optima, thereby enabling the rapid discovery of near-optimal scheduling solutions. Secondly, the algorithm employs only targeted crossover and mutation operations during the evolutionary process. Adjustments are made only to dimensions that do not violate constraints, avoiding the generation of offspring that violate these constraints and maintaining the feasibility and stability of the solution. This strategy not only maintains population diversity but also ensures the efficiency and reliability of the evolutionary process, resulting in a set of near-optimal scheduling solutions.

[0095] The various embodiments in this specification are described in a progressive manner. References to the same or similar parts between the various embodiments are sufficient. Each embodiment focuses on the differences from other embodiments. All or part of the present invention can be used in a variety of general or specialized computer system environments or configurations. For example, personal computers, server computers, handheld or portable devices, tablet devices, mobile communication terminals, multiprocessor systems, microprocessor-based systems, programmable electronic devices, network PCs, minicomputers, mainframe computers, and distributed computing environments that include any of the above systems or devices.

[0096] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, it should be understood by those skilled in the art that the technical solutions described in the aforementioned embodiments may still be modified, or some or all of the technical features thereof may be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the scope of the technical solutions of the present invention.

Claims

1. A method for optimizing SDR sensor scheduling based on Bayesian reasoning, characterized in that: include: Determine the model parameters of the sensor scheduling optimization model for sensor scheduling; wherein the model parameters include: each sensor in the first The state of each time period, each static target in the The status of each time period, each dynamic target in the The state of the time period and the sequence of working states of all sensors; The objective function and constraints corresponding to the sensor scheduling optimization model are constructed based on the model parameters; the constraints include a first constraint, a second constraint, and a third constraint; wherein: the first constraint is used to constrain the activation time of each sensor; the second constraint is used to constrain the number of frequency points activated in each time period; and the third constraint is used to constrain the daily activation time of each sensor; Under the constraints, the objective function is solved based on Bayesian reasoning to obtain an optimal state sequence for sensor scheduling optimization; under the constraints, the objective function is solved based on Bayesian reasoning to obtain an optimal state sequence for sensor scheduling optimization, including: Determine the maximum number of iterations , initialize the first frequency working state, and obtain the scheduling scheme set under each iteration according to the cyclic update method; and determine the optimal state sequence of sensor scheduling optimization based on the scheduling scheme set corresponding to the maximum number of iterations; The scheduling scheme set obtained in each iteration according to the cyclic update method includes: According to The set of scheduling solutions after iterations , use the fitness value calculation formula to calculate the fitness corresponding to each state sequence ; wherein the fitness is positively correlated with the objective function; wherein, Represents a set of scheduling schemes Middle A scheduling plan, , Represents the scheduling scheme set size; According to the fitness corresponding to each state sequence , using a binary tournament selection strategy to select The state sequence is put into the mating pool and the The first scheduling solution set after iterations ; According to the said The first scheduling solution set after iterations The state sequence in , using the non-zero rate predicted by Bayesian inference to generate the The second scheduling solution set under iterations ; The second scheduling scheme set and a set of scheduling schemes Merge to get the third scheduling solution set ; The third scheduling solution set Perform non-dominated sorting, retain the first W frontiers after each sorting, and delete the scheduling plan with the smallest fitness from the Wth frontier until the third scheduling plan set The size of , get the scheduling solution set for the next iteration .

2. The SDR sensor scheduling optimization method based on Bayesian reasoning according to claim 1 is characterized in that: The sensors are in the The status of a time period is represented as: ； in, Indicates sensor longitude; Indicates sensor Latitude; Indicates sensor Detection distance; Indicates sensor In the The frequency set that can be detected in each time period; Indicates sensor In the The first Frequency working status of each frequency point; Indicates sensor In the The number of available channels per time period; Indicates sensor In the The status of a time period.

3. The SDR sensor scheduling optimization method based on Bayesian reasoning according to claim 1 is characterized in that: The static targets are The status of a time period is represented as: ； in, Represents a static target longitude; Represents a static target Latitude; Represents a static target Radiation radius; Represents a static target In the The set of radiation frequencies in a time period; Represents a static target In the Working status of each period; Represents a static target In the The status of the time period, , is the total number of static targets.

4. The SDR sensor scheduling optimization method based on Bayesian reasoning according to claim 1 is characterized in that: The dynamic targets are The status of a time period is represented as: ； in, Represents a dynamic target In the The longitude of the starting time of each period; Represents a dynamic target In the The latitude of the starting time of each period; Represents a dynamic target In the The set of radiation frequencies in a time period; Represents a dynamic target In the Working status of each period; Represents a dynamic target In the The status of the time period, , is the total number of dynamic targets.

5. The SDR sensor scheduling optimization method based on Bayesian reasoning according to claim 1 is characterized in that: The working state sequence of all sensors is expressed as: ； in, Indicates sensor The working state sequence, and , , is the total number of sensors; Indicates sensor In the Working status of each period, , The total number of time periods.

6. The SDR sensor scheduling optimization method based on Bayesian reasoning according to claim 1 is characterized in that: The first constraint is expressed as: ； The second constraint is expressed as: ； The third constraint is expressed as: ； in, Indicates sensor In the The first Frequency working status of each frequency point; Indicates sensor The start time of the available opening period; Indicates sensor The end time of the available opening period; Indicates the total number of time periods, ; Indicates sensor The union of the sets of detectable frequency points in each time period; Indicates sensor In the The number of available channels per time period; Indicates sensor The longest opening time in a single day; Indicates the time granularity.

7. The SDR sensor scheduling optimization method based on Bayesian reasoning according to claim 1 is characterized in that: The objective function includes: a first objective function and a second objective function; wherein, The first objective function is expressed as: ； The second objective function is expressed as: ； in, Indicates the total number of static targets; Indicates the total number of sensors; Indicates the total number of time periods, ; Indicates sensor Detection range and static targets The area of ​​the intersection of the radiation range; Represents a static target Radiation range; Indicates sensor In the The first frequency points; Represents a static target In the The set of radiation frequencies in a time period; Indicates sensor In the The first Frequency working status of each frequency point; Represents a static target In the Working status of each period; Indicates sensor In the Time slots open Frequency point and static target degree of matching; Represents a dynamic target In the The set of radiation frequencies in a time period; Used to represent dynamic targets In the Is the sensor in the time period within the detection range; Represents a dynamic target In the The set of radiation frequencies in a time period; Indicates sensor In the Time slots open Frequency point and dynamic target degree of matching; Represents a dynamic target In the Working status of each period; Indicates sensor The union of the sets of detectable frequency points in each time period.

8. The SDR sensor scheduling optimization method based on Bayesian reasoning according to claim 1 is characterized in that: According to the said The first scheduling solution set after iterations The state sequence in , using the non-zero rate predicted by Bayesian inference to generate the The second scheduling solution set under iterations ,include: Define the size as Unit vector and size Unit vector , and determine the initial value of the weight; among them, , Indicates the total number of sensors; Indicates the total number of time periods; Indicates sensor The number of elements in the union of the set of detectable frequency points in each time period; and define a size of The empty matrix ; The first scheduling scheme set The scheduling sequence in is saved in Matrix and initialize the command ; Separate statistical matrix The sensors in each scheduling scheme In the Period, The frequency point is The number of times the work status is open in the scheduling plan ; According to the number of times , use the probability calculation formula to determine the sensor In the Period The frequency at which the working state of each frequency point is turned on; Calculate the frequency in the matrix The index in , put the frequency into the matrix according to the index middle; Select Matrix Any column , judge the column Whether the probability corresponding to each element in satisfies the first judgment condition, the matrix Modify the elements in to obtain a first modified matrix; According to the rows and columns of the elements in the first modified matrix, the corresponding matrix is ​​determined and determine the elements of the matrix whether the elements in the first modification matrix meet a second judgment condition, and modifying the elements in the first modification matrix according to the judgment result to obtain a second modification matrix; According to the second modified matrix, the The second scheduling solution set under iterations .

Images