An unmanned aerial vehicle multi-connection switching method and system based on fuzzy logic and a storage medium

Abstract

The application provides a kind of unmanned aerial vehicle multi-connection switching method, system and storage medium based on fuzzy logic, which comprises the following steps: step one: use fuzzy logic to combine multiple switching criteria to obtain the score of each base station;Step two: update the active set AS according to the score of the base station obtained in step one.The beneficial effects of the present application are: compared with the traditional switching scheme, the present application can significantly reduce the switching rate of unmanned aerial vehicle, service interruption rate, link failure rate and decoding error probability.

Description

Technical Field

[0001] This invention relates to the field of communication technology, and in particular to a method, system, and storage medium for switching multiple connections of unmanned aerial vehicles (UAVs) based on fuzzy logic. Background Technology

[0002] Traditional drone applications are primarily based on point-to-point communication. This paradigm limits the communication range and cannot support some emerging drone applications, such as remote drone rescue and cargo delivery, which require beyond-line-of-sight (BLOS) control. To achieve safe BLOS control of drones, ensuring high reliability and low latency in control and command information transmission is crucial. Therefore, research has begun on the capability and feasibility of using cellular networks to support drone communication, and cellular-connected drone communication has been proposed. With the support of cellular networks, drones, as aerial users, can access ubiquitous and more reliable services.

[0003] However, in existing cellular systems, base stations are typically equipped with downtilted antennas and utilize the main lobe to serve ground users rather than drones. In this scenario, flying drones are only served by the low-power sidelobes of the base station's antennas. Furthermore, as flight altitude increases, drones gain access to better line-of-sight (LoS) transmission channels. When drones fly at high altitudes, they may detect multiple cells with similar reference signal received power (RSRP). If an association criterion based on RSRP is used, this will result in frequent handovers for drones. In addition, drones possess three-dimensional (3D) movement capabilities unlike ground users, such as vertical takeoff. When drones fly across the null space of the base station antenna, it triggers numerous unnecessary handovers, leading to significant communication latency and reduced link reliability. Given these challenges, designing an effective handover scheme for cellular drone communication is crucial.

[0004] Traditional Long Term Evolution (LTE) handover schemes are typically based on A3 event triggering, where the RSRP of the target base station is higher than the RSRP of the serving base station by a handover difference. Furthermore, trigger time is introduced to mitigate the ping-pong effect. However, current work employing single-connectivity technology results in hard handovers that cause severe service disruptions for drones. To reduce service disruptions for mobile users, much current research utilizes multi-connectivity technology. With multi-connectivity, drones can simultaneously connect to multiple base stations, and the set of currently serving base stations is called the active set (AS). Under multi-connectivity, base station handover includes adding, deleting, and replacing base stations. The designed multi-connectivity handover scheme needs to ensure link reliability. However, much current research primarily manages ASs based on RSRP criteria. This criterion, under actual antenna configurations and the 3D movement of drones, leads to severe handover rates and service disruptions. Furthermore, AS management also needs to consider the available resources of connected base stations. Summary of the Invention

[0005] To address the problems in the prior art, this invention provides a method, system, and storage medium for switching multiple connections of unmanned aerial vehicles (UAVs) based on fuzzy logic.

[0006] This invention provides a method for multi-connection handover of unmanned aerial vehicles (UAVs) based on fuzzy logic, comprising the following steps:

[0007] Step 1: Use fuzzy logic combined with the multi-handover criterion to obtain the score for each base station;

[0008] Step 2: Update the activation set based on the base station scores obtained in Step 1.

[0009] As a further improvement of the present invention, step one includes:

[0010] Step 1: Add the trigger difference Deletion trigger difference Replacement trigger difference Trigger time Base stations The minimum number of base stations in the activation set. Activate the maximum number of base stations in the set Added counters C add 、 Deletion counter C rel 、 Replacement counter C rep Input fuzzy logic;

[0011] Step 2: Use fuzzy logic to obtain the scores of all base stations, i.e. .

[0012] As a further improvement of the present invention, in step 2, the fuzzy logic includes a fuzzification module, an inference module, and a defuzzification module, and its operation process is as follows:

[0013] Step S1: The fuzzification module receives clear input values, converts the clear input values ​​into fuzzy sets through a membership function, and then sends them to the inference module. The clear input values ​​include average received power, the rate of change of average received power, and average base station load.

[0014] Step S2: The inference module receives the fuzzy set sent in step S1, maps the fuzzy set to the output set according to predefined rules, and then sends it to the defuzzification module;

[0015] Step S3: The deblurring module receives the output set sent in step S2, and then converts the output set into clear output values, i.e., the base station. m The score.

[0016] As a further improvement of the present invention, the predefined sets of input and output linguistic variables in the fuzzification module are as follows: , , , .

[0017] As a further improvement of the present invention, the inference module uses a maximum-minimum inference method to obtain the membership degree of the fuzzy output set, i.e. ,in Indicates the average RSRP membership degree. The membership degree represents the rate of change of the average RSRP. The membership degree represents the average base station load. For each rule, the minimum value of the input membership degree is used to obtain the output membership degree. In order to combine different rules, the maximum value of the input membership degree values ​​in each rule is used to obtain the final output membership degree.

[0018] As a further improvement of the present invention, in the deblurring module, the centroid is calculated using the center method. During the centroid calculation process, to calculate the final result, the output membership function is truncated, and then the disjunction principle is used to merge shapes and form a shadow shape. Finally, the coordinates of the shadow shape's center on the normalized output coordinate axis are calculated using the following formula. ,

[0019]

[0020] in, Let be the upper boundary value of the shadow shape, and s represent the normalized output, with a value range of [0,1]. Therefore, the base station can be obtained. m The fraction, that is .

[0021] As a further improvement of the present invention, step two further includes:

[0022] Step 1: If , ,as well as At the trigger time During this period, the following conditions can be met simultaneously: Indicates base station The score, Indicates base station The score will then be the base station Add a counter to the active set. C add Increment the counter by 1, then proceed to the next step; otherwise, increment the counter. C add Set the counter to 0 and return to step 2.

[0023] Step 2: Determine the counter added in Step 1. C add Is it equal to the trigger time? If yes, proceed to the next step; otherwise, return to step 2.

[0024] Step 3: Install the base station Add to the active set, along with the counter. C add Set the count to 0, that is ;

[0025] Step 4: If , ,as well as At the trigger time If the conditions can be met simultaneously during this period, then the base station will be... Remove the counter from the active set. C rel Increment the counter by 1, then proceed to the next step; otherwise, delete the counter. C rel Set to 0, then return to step 2.

[0026] Step 5: Check the deletion counter from step 4. C rel Is it equal to the trigger time? If yes, proceed to the next step; otherwise, return to step 2.

[0027] Step 6: Install the base station Remove from the active set, and also delete the counter. C rel Set the count to 0;

[0028] Step 7: If , At the trigger time If the conditions can be met during the period, then the base station with the minimum score is... Base station Replacement, Replacement Counter C rep Increment the counter by 1, then proceed to the next step; otherwise, replace the counter. C rep Set to 0, then return to step 2.

[0029] Step 8: Determine the counter that was replaced in step 7. C rep Is it equal to the trigger time? If yes, proceed to the next step; otherwise, return to step 2.

[0030] Step 9: Base Station Replace the base station with the lowest score Replacement counter C rep Set to 0 and return to step 2.

[0031] The present invention also discloses a UAV multi-connection switching system based on fuzzy logic, a memory, a processor, and a computer program stored in the memory, wherein the computer program is configured to implement the steps of the UAV multi-connection switching method described in the present invention when called by the processor.

[0032] The present invention also discloses a computer-readable storage medium storing a computer program configured to implement the steps of the UAV multi-connection switching method described in the present invention when invoked by a processor.

[0033] The beneficial effects of this invention are: compared with traditional switching schemes, this invention can significantly reduce the switching rate, service interruption rate, link failure rate and decoding error probability of UAVs. Attached Figure Description

[0034] Figure 1 This is a model diagram of the cellular-connected unmanned aerial vehicle communication system of the present invention;

[0035] Figure 2 This is a flowchart of the UAV multi-connection switching method of the present invention;

[0036] Figure 3This is a block diagram of the fuzzy logic principle of the present invention;

[0037] Figure 4a ~4d is the membership function graph of the input and output of this invention;

[0038] Figure 5 This is a diagram of the output membership function and COG defuzzification mechanism of this invention. Detailed Implementation

[0039] In situations beyond normal line-of-sight, cellular networks are used to provide communication services for drones, enabling remote control—that is, drone communication via cellular connections. However, because most current cellular systems are equipped with downtilted base station antennas, drones may connect to the base station through sidelobes and suffer severe handover interruptions. Therefore, it is difficult to guarantee seamless and low-latency service for drones. To address these technical challenges, existing solutions employ multi-connectivity technology to resolve service interruptions for mobile users. However, existing multi-connectivity handover schemes are primarily based on received power, which cannot guarantee the reliability of drone communication connections.

[0040] This invention provides a fuzzy logic-based method for UAV multi-connection handover to reduce handover rate and provide more reliable communication. First, the invention considers multiple handover criteria, including average received power, the rate of change of average received power, and average base station load. Then, fuzzy logic is used to combine these criteria and obtain a score for each base station. Finally, based on the obtained scores, the invention updates the serving base station set according to traditional multi-connection handover schemes. Compared to traditional handover schemes, this invention can significantly reduce the UAV handover rate, service interruption rate, link failure rate, and decoding error probability.

[0041] This invention considers a cellular-connected unmanned aerial vehicle (UAV) communication system, the system model of which is as follows: Figure 1 As shown, this includes one drone, J ground users, and M base stations. The set of base stations can be represented as... Assume the drone can receive downlink control information from ground base stations. Furthermore, assume the drone is equipped with a single antenna, while each base station is equipped with one... A uniform linear array of array elements, vertically mounted. To achieve more reliable connectivity, the drone can connect to multiple base stations simultaneously, and the active set AS can be represented as... In a multi-connectivity scenario, the UAV maintains data and control plane connections with the primary base station and data plane connections with the secondary base station. All ground users connect to the base station using a single-connectivity association criterion based on maximum RSRP. In time slot t, the number of users connected to base station m can be expressed as... Furthermore, all base stations have the same transmission power. At the same height All base stations are interconnected via wired links.

[0042] Because most current cellular base stations are equipped with base station antennas that have a fixed radiation pattern and downtilt angle. The base station antenna is omnidirectional in the horizontal direction, while the power radiation pattern in the vertical direction is equal to the array power gain multiplied by the element power gain. Therefore, in the vertical direction, the element power gain of the m-th base station in time slot t can be modeled as:

[0043]

[0044] in, For half-power beamwidth and The maximum antenna gain. The elevation angle between the UAV and the m-th base station in time slot t can be expressed as: , ,in, and Let be the flight altitude of the UAV in time slot t and the horizontal distance between the UAV and the m-th base station, respectively. The array factor of the m-th base station in time slot t can be expressed as:

[0045]

[0046] make Let represent the array power gain. In time slot t, the total antenna gain of the m-th base station can be expressed as:

[0047]

[0048] In time slot t, the path loss between m users at the m-th base station is related to the Loss-of-Stake (LoS) link and can be expressed as:

[0049]

[0050] in, Let be the distance between the UAV and the m-th base station in time slot t. and These are the path loss exponent and path loss constant, respectively, when the reference distance is 1m. and These represent LoS and NLoS communication, respectively.

[0051] In time slot t, the Loss Probability (LoS) of the link between base station m and the drone can be expressed as:

[0052]

[0053] in, In this model, the urban area is viewed as a set of buildings on a square grid, where 'a' represents the proportion of the total land area occupied by buildings, 'b' represents the average number of buildings per square kilometer, and the height of the buildings is modeled by a Rayleigh probability distribution function with a proportionality parameter 'c'.

[0054] Based on the above model, in time slot t, the RSRP from base station m can be expressed as:

[0055]

[0056] Therefore, in time slot t, the signal-to-interference-plus-noise ratio (SINR) from the k-th link to the UAV can be calculated by the following formula:

[0057]

[0058] in, Represents the set of interfering base stations of base station k and Let represent the noise power. Using the maximum proportional combination scheme, the total received SINR at the UAV in time slot t is expressed as:

[0059] .

[0060] Furthermore, the motion model of the drone can be modeled as follows:

[0061]

[0062] in, and Represent the 3D position of the UAV in time slot t and in time slot t, respectively. The direction of movement of the time slot and These represent the drone's flight speed and the duration of a single time slot, respectively.

[0063] This invention provides a multi-criteria switching method based on fuzzy logic for updating the activation set (AS). For example... Figure 1As shown, assume the drone is currently flying along the marked green dot and connected to base stations 1 and 2. Flying forward along the antenna sidelobe of base station 1 may result in a longer connection duration. However, flying sideways along the antenna sidelobe of base station 2 may cause the drone to quickly fly out of the sidelobe and only achieve a shorter connection duration. It's important to note that the rate of change of the similar received power (RSRP) is relatively small when the drone flies forward along a sharp beam, but relatively large when it flies sideways. Therefore, base stations with a smaller rate of change of RSRP in the active set (AS) will be able to provide longer connection durations for the drone. Furthermore, the available spectrum resources in the serving base station depend on the base station's load. When a drone connects to a base station with more available spectrum resources, this results in a higher transmission rate. Therefore, considering multiple handover criteria becomes more meaningful: namely, the average similar received power (RSRP), the average RSRP change rate, and the average base station load, which are respectively denoted as… , ,as well as ,in This refers to the average window size. To balance the impact of RSRP, link connection duration, and available spectrum resources, this invention employs fuzzy logic tools to combine these multiple criteria. The multiple criteria are used as input to a fuzzy logic program, which then yields a score for each base station. The fuzzy logic part will be introduced in a later section.

[0064] Let the trigger differences for adding, deleting, and replacing base stations be represented as follows: , as well as The trigger time is determined by This is represented as follows. Furthermore, let the minimum and maximum values ​​of the number of base stations in the AS be represented as follows: as well as The switching process proposed in this invention is as follows:

[0065] Step 1: Obtain the scores of all base stations using fuzzy logic, i.e. .

[0066] Step 2: Update the active set (AS) based on the base station scores obtained in Step 1.

[0067] Specifically, if , ,as well as exist During this period, the following conditions can be met simultaneously: and Representing base stations With base station The score, then the base station It will be added to AS; if , ,as well as exist If the conditions can be met simultaneously during this period, then the base station It will be removed from the active set (AS); if , At the trigger time If the conditions can be met during the period, then the base station with the minimum score is... (Right now, ) will be base station Replacement. The specific switching process is as follows: Figure 2 As shown, the detailed steps are as follows:

[0068] Step 1: If , ,as well as At the trigger time During this period, the following conditions can be met simultaneously: Indicates base station The score, Indicates base station The score will then be the base station Add the counter to the active set (AS). C add Count +1, that is C add= C add+1 Then proceed to the next step; if not, add a counter. C add Set the count to 0, that is C add =0, return to step 2;

[0069] Step 2: Determine the counter added in Step 1. C add Is it equal to the trigger time? If yes, proceed to the next step; otherwise, return to step 2.

[0070] Step 3: Install the base station Add a counter to the active set (AS). C add Set the count to 0, that is ;

[0071] Step 4: If , ,as well as At the trigger time If the conditions can be met simultaneously during this period, then the base station will be... Remove the counter from the active set (AS). C rel Count +1, that is C rel = C rel +1, then proceed to the next step; otherwise, delete the counter. C rel Set to 0, that is C rel =0, return to step 2;

[0072] Step 5: Check the deletion counter from step 4. C rel Is it equal to the trigger time? If yes, proceed to the next step; otherwise, return to step 2.

[0073] Step 6: Install the base station Remove from the active set (AS), and also remove the counter. C rel Set the count to 0, that is ;

[0074] Step 7: If , At the trigger time If the conditions can be met during the period, then the base station with the minimum score is... ,Right now, Base station Replacement, Replacement Counter C rep Count +1, that is C rep = C rep Increment the counter by 1, then proceed to the next step; otherwise, replace the counter. C rep Set to 0, that is C rep =0, return to step 2;

[0075] Step 8: Determine the counter that was replaced in step 7. C rep Is it equal to the trigger time? If yes, proceed to the next step; otherwise, return to step 2.

[0076] Step 9: Base Station Replace the base station with the lowest score ,Right now Replacement counter C rep Set to 0, that is Return to step 2.

[0077] This invention utilizes fuzzy logic tools to combine the proposed multiple criteria, with each module such as... Figure 3 As shown in the diagram. In the fuzzification module, the sharp input values ​​are converted into fuzzy sets using a membership function. Subsequently, the input set is mapped to the output set using predefined rules in the inference module. Finally, the defuzzification module converts the output set back into sharp output values. The specific execution process is as follows:

[0078] Step S1: The fuzzification module receives clear input values, then converts the clear input values ​​into fuzzy sets through a membership function, and then sends them to the inference module. The clear input values ​​are average received power, the rate of change of average received power, and average base station load.

[0079] Step S2: The inference module receives the fuzzy set sent in step S1, maps the fuzzy set to the output set according to predefined rules, and then sends it to the defuzzification module;

[0080] Step S3: The deblurring module receives the output set sent in step S2, and then converts the output set into clear output values, i.e., the base station. m The score.

[0081] The functions of each module are as follows:

[0082] Fuzzification Module: To perform the fuzzification process, the three inputs and outputs need to be converted into fuzzy sets. The predefined sets of linguistic variables for input and output are as follows: , , , The fuzzy output represents the rating of the base station's service to the drone. Figure 4a ~4d shows the membership functions for the three inputs and outputs, where trigonometric and trapezoidal functions are used to generate the membership functions.

[0083] Table 1. Rule Base

[0084]

[0085] Inference Module: As shown in Table 1, for three level-three inputs, this invention establishes 27 fuzzy rules to map the levels of the three input variables to the levels of the outputs. Since multiple rules may be applied simultaneously, the inference system employs a min-max inference approach to obtain the membership degree of the fuzzy output set. ,in , as well as These represent the membership degree of average RSRP, the membership degree of average RSRP change rate, and the membership degree of average base station load, respectively. For each rule, the minimum value of the input membership degree is used to obtain the output membership degree. To combine different rules, the maximum value of the input membership degree values ​​in each rule is used to obtain the final output membership degree.

[0086] Deblurring Module: After combining the rules, the output set needs to be converted into sharp output values, i.e., the scores of base station m. This invention uses the center method to calculate the center of gravity (COG). Figure 5 In this example, we illustrate the mechanism for calculating COG. Let the normalized output be s, with a value range of [0,1]. Assume that the membership degree of "Very Poor" and "Very Good" is 0, and the membership degrees of "Poor," "Medium," and "Good" are 0.4, 0.5, and 0.2, respectively. To calculate the final result, we truncate the output membership function and then use the disjunction principle to merge the shapes and form the shadow shape. Then, the coordinates of the center of the shadow shape on the normalized output coordinate axis can be calculated using the following formula (…). ):

[0087]

[0088] in, This represents the upper boundary of the shadow shape. Therefore, the fraction of base station m can be obtained, i.e. .

[0089] The present invention also discloses a UAV multi-connection switching system based on fuzzy logic, a memory, a processor, and a computer program stored in the memory, wherein the computer program is configured to implement the steps of the UAV multi-connection switching method described in the present invention when called by the processor.

[0090] The present invention also discloses a computer-readable storage medium storing a computer program configured to implement the steps of the UAV multi-connection switching method described in the present invention when invoked by a processor.

[0091] The beneficial effects of this invention are: compared with traditional switching schemes, this invention can significantly reduce the switching rate, service interruption rate, link failure rate and decoding error probability of UAVs.

[0092] The above description, in conjunction with specific preferred embodiments, provides a further detailed explanation of the present invention. It should not be construed that the specific implementation of the present invention is limited to these descriptions. For those skilled in the art, various simple deductions or substitutions can be made without departing from the concept of the present invention, and all such modifications and substitutions should be considered within the scope of protection of the present invention.

Claims

1. A method for multi-connection handover of unmanned aerial vehicles based on fuzzy logic, characterized in that... This includes the following steps: Step 1: Use fuzzy logic combined with the multi-handover criterion to obtain the score for each base station; Step 2: Update the activation set based on the base station scores obtained in Step 1; Step one includes: Step 1: Add the trigger difference Deletion trigger difference Replacement trigger difference Trigger time Base stations The minimum number of base stations in the activation set. Activate the maximum number of base stations in the set Added counters C add 、 Deletion counter C rel 、 Replacement counter C rep Input fuzzy logic; Step 2: Use fuzzy logic to obtain the scores of all base stations, i.e. ; Step two also includes: Step 1: If , ,as well as At the trigger time During this period, the conditions can be met simultaneously, among which Indicates base station The score, Indicates base station The score, If the base station set is represented, then the base stations will be... Add a counter to the active set. C add Increment the counter by 1, then proceed to the next step; otherwise, increment the counter. C add Set the counter to 0 and return to step 2. Step 2: Determine the counter added in Step 1. C add Is it equal to the trigger time? If yes, proceed to the next step; otherwise, return to step 2. Step 3: Install the base station Add to the active set, along with the counter. C add Set the count to 0, that is ; Step 4: If , ,as well as At the trigger time If the conditions can be met simultaneously during this period, then the base station will be... Remove the counter from the active set. C rel Increment the counter by 1, then proceed to the next step; otherwise, delete the counter. C rel Set to 0, then return to step 2. Step 5: Determine the deletion counter from Step 4. C rel Is it equal to the trigger time? If yes, proceed to the next step; otherwise, return to step 2. Step 6: Install the base station Remove from the active set, and also delete the counter. C rel Set the count to 0, that is ; Step 7: If , At the trigger time If the conditions can be met during the period, then the base station with the minimum score is... Base station Replacement, Replacement Counter C rep Increment the counter by 1, then proceed to the next step; otherwise, replace the counter. C rep Set to 0, then return to step 2. Step 8: Determine the counter that was replaced in step 7. C rep Is it equal to the trigger time? If yes, proceed to the next step; otherwise, return to step 2. Step 9: Base Station Replace the base station with the lowest score Replacement counter C rep Set to 0, that is Then return to step 2.

2. The UAV multi-connection switching method based on fuzzy logic according to claim 1, characterized in that, In step 2, the fuzzy logic includes a fuzzification module, an inference module, and a defuzzification module, and its operation process is as follows: Step S1: The fuzzification module receives clear input values, converts the clear input values ​​into fuzzy sets through a membership function, and then sends them to the inference module. The clear input values ​​include average received power, the rate of change of average received power, and average base station load. Step S2: The inference module receives the fuzzy set sent in step S1, maps the fuzzy set to the output set according to predefined rules, and then sends it to the defuzzification module; Step S3: The deblurring module receives the output set sent in step S2, and then converts the output set into clear output values, i.e., the base station. m The score.

3. The UAV multi-connection switching method based on fuzzy logic according to claim 2, characterized in that, In the fuzzification module, the predefined sets of input and output linguistic variables are as follows: , , , .

4. The UAV multi-connection switching method based on fuzzy logic according to claim 2, characterized in that, In the inference module, the membership degree of the fuzzy output set is obtained using a maximum-minimum inference method, i.e. ,in Indicates the average RSRP membership degree. The membership degree represents the rate of change of the average RSRP. The membership degree represents the average base station load. For each rule, the minimum value of the input membership degree is used to obtain the output membership degree. In order to combine different rules, the maximum value of the input membership degree values ​​in each rule is used to obtain the final output membership degree.

5. The UAV multi-connection switching method based on fuzzy logic according to claim 2, characterized in that, In the deblurring module, the centroid is calculated using the center method. During this calculation, to arrive at the final result, the output membership function is truncated. Then, the disjunction principle is used to merge shapes and form a shadow shape. Finally, the coordinates of the shadow shape's center on the normalized output coordinate axes are calculated using the following formula. , in, Let be the upper boundary value of the shadow shape, and s represent the normalized output, with a value range of [0,1]. Therefore, the base station can be obtained. m The fraction, that is .

6. A multi-connection switching system for unmanned aerial vehicles based on fuzzy logic, characterized in that, include: A memory, a processor, and a computer program stored on the memory, the computer program being configured to implement the steps of the UAV multi-connection switching method according to any one of claims 1-5 when invoked by the processor.

7. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program configured to implement the steps of the UAV multi-connection switching method according to any one of claims 1-5 when invoked by a processor.

Images