Unmanned aerial vehicle adaptive decoupling method and system based on fuzzy logic under full duplex, and storage medium

Abstract

The invention provides an unmanned aerial vehicle adaptive decoupling method and system based on fuzzy logic under full duplex, and a storage medium. The self-adaptive decoupling method comprises the following steps: step 1, inputting fuzzy logic to the receiving power # imgabs0 # of an unmanned aerial vehicle k in an uplink and a candidate base station (bc), the angle # imgabs2 # between the associated # imgabs1 # in a downlink and the candidate base station (bc) in UL, and the load # imgabs3 # of the unmanned aerial vehicle k in UL and the candidate base station (bc); step 2, obtaining all BS scores # imgabs4 # corresponding to the unmanned aerial vehicle k by using fuzzy logic; step 3, determining a decoupling decision according to the BS scores # imgabs5 # in the step 2; and selecting the BS with the highest score, namely # imgabs6 #, as the UL associated BS. The method has the advantages that compared with a traditional UL-DL decoupling scheme, the method can remarkably reduce self-interference of the unmanned aerial vehicle end, and therefore the DL transmission rate is increased.

Description

Technical Field

[0001] The present invention relates to the field of communication technology, and in particular to a method, system and storage medium for adaptive decoupling of unmanned aerial vehicles based on fuzzy logic in full-duplex mode. Background Art

[0002] Traditional drone applications rely mainly on point-to-point communications, which not only limits the control range of drones, but also cannot support emerging applications such as remote surveillance that require beyond-line-of-sight operations. To address these limitations, 3GPP began to explore the capabilities and feasibility of cellular connections in drone communications and introduced relevant specifications. Through cellular networks, drones, as aerial user devices, can obtain ubiquitous and reliable connections during flight. Introducing full-duplex technology into cellular-connected drone communications can achieve simultaneous transmission and reception of data in the same frequency band, thereby greatly improving spectrum efficiency. This is particularly important for drone applications that require real-time two-way communication, such as emergency rescue. However, due to the close distance between the transmitter and the receiver, full-duplex operation may cause severe self-interference (SI), which affects system performance. Although technologies such as antenna isolation and analog and digital suppression can reduce SI to a certain extent, residual SI remains a major challenge that needs to be addressed.

[0003] Cellular connected drone communications use cellular networks to provide ubiquitous and reliable services for drones. Introducing full-duplex technology into cellular connected drone communications allows data to be sent and received simultaneously on the same frequency band, doubling the spectrum efficiency. In cellular connected drone communications under full-duplex, both the base station (BS) and the drone may operate in full-duplex mode. In this case, if uplink-downlink (UL-DL) coupled association is used, the transmit and receive beam directions of the BS / drone end overlap, and the SI mainly comes from the main lobe and becomes difficult to manage. Based on the above challenges, it is crucial to design an effective association scheme for cellular connected drone communications.

[0004] An effective solution to the above problems is to use UL-DL decoupling association, that is, the UAV performs UL and DL association with different BSs to achieve spatial separation of the transmit and receive beams on the UAV side, thereby reducing SI. In addition, the UL association strategy based on minimum path loss can enhance UL transmission and can be used for applications with UL-dominated traffic in cellular-connected UAVs (for example, emergency rescue and video surveillance). However, the current UL-DL decoupling solutions are mainly based on minimum path loss and cannot be used to solve the SI problem of UAV communications with cellular connections in full-duplex. Summary of the invention

[0005] In order to solve the problem of serious SI caused by uplink-downlink coupling correlation in UAV communication with cellular connection under full-duplex in the prior art, the present invention provides a UAV adaptive decoupling method based on fuzzy logic under full-duplex, comprising the following steps:

[0006] Step 1: Place UAV k in UL with candidate BS (b c ) received power UAV k is associated with DL and the candidate BS in UL (b c ) UAV k is in UL with candidate BS (b c ) Enter fuzzy logic;

[0007] Step 2: Use fuzzy logic to obtain all BS scores corresponding to drone k

[0008] Step 3: Based on the BS score from step 2 Determine the decoupling decision; the BS with the highest score, i.e. Selected as UL associated BS.

[0009] As a further improvement of the present invention, in step 1, the fuzzy logic includes a fuzzification module, a reasoning module and a defuzzification module, wherein:

[0010] The fuzzification module is used to convert the input of step 1 into a fuzzy set through a fuzzy membership function;

[0011] The reasoning module: fuzzy input is mapped to fuzzy output through a set of 27 IF / THEN rules;

[0012] The defuzzification module: after combining the rules, the aggregated output set is converted into a numerical value, namely BSb c For the score of drone k.

[0013] As a further improvement of the present invention, in the fuzzification module, the input and output language term sets are defined as: O = {very bad, bad, fair, good, very good};

[0014] The fuzzy output set represents the suitability of BSb to become the UL associated BS of UAV k when considering the three inputs.

[0015] As a further improvement of the present invention, in the fuzzification module, the membership functions of the input and output are generated using trapezoidal and trigonometric functions.

[0016] As a further improvement of the present invention, in the reasoning module, the Min-Max reasoning principle is used to determine the membership of the fuzzy logic output set, and its expression is: O =max(min(μ X ,μ Y ,μ Z )), where μ X , μ Y and μ Z They represent the reference received power in UL, the angle between the UAV and the BS, and the membership degree of the BS load, respectively.

[0017] As a further improvement of the present invention, in the reasoning module, for each rule, the output membership is obtained by taking the minimum value of the input membership. In order to combine different rules, the final output membership is obtained by taking the maximum value of the output membership of each rule.

[0018] As a further improvement of the present invention, in the defuzzification module, a centroid method is used for defuzzification, which calculates the centroid of the aggregated output set.

[0019] As a further improvement of the present invention, in the defuzzification module, in order to calculate the final score, the output membership function is truncated and the shapes are merged using the disjunction principle to form a shadow shape, and the centroid coordinates of the shadow shape are calculated on the s-axis:

[0020]

[0021] where f O (·) is the upper edge of the shaded shape, so BSb c The score for drone k is

[0022]

[0023] The present invention also discloses a system for adaptive decoupling of unmanned aerial vehicles based on fuzzy logic in full-duplex mode, comprising: a memory, a processor, and a computer program stored in the memory, wherein the computer program is configured to implement the steps of the method of the present invention when called by the processor.

[0024] The present invention also discloses a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and the computer program is configured to implement the steps of the method of the present invention when called by a processor.

[0025] The beneficial effect of the present invention is that compared with the traditional UL-DL decoupling solution, the method of the present invention can significantly reduce the SI of the drone end, thereby improving the DL transmission rate. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 It is a system model of the UAV adaptive decoupling method of the present invention;

[0027] Figure 2 It is a flow chart of the adaptive decoupling method of the unmanned aerial vehicle of the present invention;

[0028] Figure 3 It is a fuzzy logic architecture diagram of the UAV adaptive decoupling method of the present invention;

[0029] Figure 4a -4d is a membership function diagram of the input and output of the UAV adaptive decoupling method of the present invention;

[0030] Figure 5 It is the output membership function and COG defuzzification mechanism of the UAV adaptive decoupling method of the present invention. DETAILED DESCRIPTION

[0031] Self-interference: self-interference, SI;

[0032] Uplink-downlink: uplink-downlink, UL-DL;

[0033] Macro base station: macro base station, MBS;

[0034] Small base stations: small base stations, SBSs;

[0035] Line of sight: line-of-sight, LoS;

[0036] Non-line-of-sight: non-line-of-sight, NloS;

[0037] Angle of arrival: angle of arrival, AoA;

[0038] Angle of departure: angle of departure, AoD;

[0039] Received power: reference signal received power, RSRP;

[0040] The present invention provides a UAV adaptive decoupling method based on fuzzy logic in full-duplex mode to reduce SI at the UAV end. First, the present invention considers multiple criteria: including the received power on UL, the angle between the UAV and the BS, and the load of the BS. Then, fuzzy logic is used to combine these criteria and obtain the score of each BS. Finally, the decoupling decision is determined based on the obtained score. Compared with the traditional UL-DL decoupling scheme, the present invention can significantly reduce SI at the UAV end, thereby improving the DL transmission rate.

[0041] The present invention considers a heterogeneous network, the system model is as follows Figure 1 As shown in Figure 1, it includes one MBS, S SBSs and K UAVs. The set of BSs is represented as Where 0 represents MBS, and {1,2,...,B} represents the SBS set. In addition, the drone set is represented as The present invention focuses on real-time two-way communication applications, such as emergency rescue operations. In these scenarios, video content is transmitted on the UL, while control data for the drone is transmitted on the DL. To support these applications, each drone is assumed to operate in full-duplex mode and is equipped with receiving antennas and In addition, each M is equipped with receiving antennas and To reduce SI in full-duplex operation, the UAV adopts an adaptive UL-DL decoupling association strategy to achieve spatial separation of transmit and receive beams at the UAV and BS ends.

[0042] The path loss between BSb and UAV k can be modeled as:

[0043]

[0044] in represents the distance between BSb and UAV k, represents the horizontal distance between BSb and UAV k, H k is the height of UAV k, H BS is the height of BS. v and β v denote the path loss exponent and the path loss constant at 1 m, respectively, where v∈{los,nlos} denotes line-of-sight or non-line-of-sight communication.

[0045] The LoS probability of the link between BSb and UAV k is expressed as:

[0046]

[0047] Where a and c are constants related to the propagation environment, and the elevation angle And θ b,k ∈[0,π / 2].

[0048] The channel between BSb and UAV k is modeled as the Rician channel model:

[0049]

[0050] where τ∈{d,u} represents DL or UL, and x∈{MBS,SBS}, denotes the channel between BSb and UAV k in DL and UL, respectively. κ denotes the Rician factor. represents the LoS component in DL, which can be expressed as

[0051]

[0052] in as well as and φ b,k are the arrival angle and departure angle in the vertical direction respectively. λ is the wavelength, D a Denotes the antenna spacing, set to D a =λ / 2. Similarly, the LoS component in UL And expressed as

[0053]

[0054] in and Represents the NLoS component, whose elements are In addition, the SI fading channels at the BS and the UAV are denoted by h k,k and h b,b , where the elements are An iid complex Gaussian random variable.

[0055] The conventional UL-DL decoupled association strategy allows a UAV to associate with the BS providing the maximum reference signal received power in DL, while connecting with the BS with the minimum path loss in UL. The present invention considers beamforming gain on this basis and adopts a modified association strategy for BS association in DL and UL. Therefore, the RSRP received by UAV k from BSb in DL is expressed as where x∈{MBS,SBS}, v∈{los,nlos}, P x represents the transmit power of BS. Similarly, the RSRP received by BSb from UAV k in UL is Where P UAV Indicates the transmitting power of the drone.

[0056] To obtain long-term parameters, the fast fading effects should be averaged, i.e. Where h represents the channel between the transmitter and the receiver, N r and N t denotes the number of antennas at the receiving end and the transmitting end, respectively. Therefore, the RSRP in DL and UL can be expressed as and The BS association results of UAV k in DL and UL can be expressed as and Assume the decoupling instruction of drone k is o k , where o k ∈{0,1}. If Then the UL association is decoupled, that is, o k =1; otherwise, o k = 0. Since beamforming is used to reduce the SI of the UAV, the traditional decoupling scheme based on RSRP is not enough to reduce SI. In addition, the available wireless resources are affected by the BS load, which also needs to be considered in the decoupling process.

[0057] In order to solve these problems, the present invention proposes a new decoupling scheme based on multiple criteria in full-duplex: a multi-criteria decoupling scheme based on fuzzy logic. The following are three criteria considered.

[0058] · UAV k is in UL with candidate BS (b c ) is expressed as This has been introduced above.

[0059] · UAV k is associated with DL and the candidate BS in UL (b c ), expressed as like Figure 1 As shown, it is assumed that the decoupling decision for UAVs 1 and 2 has been made. For UAV 3, it is assumed that BS3 provides the maximum RSRP in both DL and UL. If UAV 3 is associated with BS3 in a coupled manner, the transmit and receive beams of UAV 3 are completely aligned, that is, the main lobes of the antennas will have the same direction, resulting in significant SI and poor DL ​​performance (due to possible sidelobe leakage and reflections of the UAV's transmitted signals). Therefore, it may be necessary to decouple the UL association of UAV 3 to achieve spatial separation of the main lobe. In this case, the SI comes only from the side lobes of the antenna. The degree of spatial beam separation is related to the angle between UAV k and the associated BS in DL and UL. Note that when the angle is large, the SI is relatively low; when the angle is small, the SI is relatively high.

[0060] · UAV k is in UL with candidate BS (b c ) load, expressed as The available resources of the serving BS are determined by its load. When a UAV associates with a BS with more available bandwidth resources, it will obtain a higher transmission rate.

[0061] like Figure 2 As shown in Figure 2, the above multiple criteria are used as inputs to the fuzzy logic process. Through the fuzzy logic framework, the scores of all BSs are obtained, namely, The proposed decoupling process is as follows:

[0062] First, fuzzy logic is used to obtain all BS scores corresponding to UAV k, that is,

[0063] Then, the decoupling decision is determined based on these scores. Specifically, the BS with the highest score, i.e. will be selected as the UL associated BS. Then the UL association will be decoupled (o k =1); otherwise, keep coupling state (o k =0). Note that the proposed scheme can be executed in a distributed manner at the UAV end. Therefore, it is necessary to collect physical layer parameters at the UAV end, such as RSRP in UL, BS position for calculating the angle between the UAV and BS, and BS load. The specific decoupling process is as follows: Figure 2 shown.

[0064] The present invention uses fuzzy logic tools to combine the proposed multiple criteria. The fuzzy logic framework used is as follows: Figure 3 As shown in Figure 2. The architecture consists of three main modules: fuzzification module, reasoning module, and defuzzification module. Each module plays a key role in converting precise input into final decision. The functions of each module are described as follows:

[0065] Fuzzification module: In the fuzzification module, the input is converted into fuzzy sets through fuzzy membership functions. The linguistic term sets of these inputs and outputs are defined as: O = {very bad, bad, fair, good, very good}.

[0066] The set of fuzzy outputs represents the suitability of BSb to become the UL associated BS for UAV k when considering the three inputs. Figure 4 shows the membership functions of the inputs and outputs, which are generated using trapezoidal and trigonometric functions.

[0067] Reasoning module: In the reasoning module, fuzzy inputs are mapped to fuzzy outputs through a set of 27 IF / THEN rules, as listed in Table 1. These rules consider all possible combinations of three tertiary inputs. Since multiple rules can be applied simultaneously, the Min-Max reasoning principle is used to determine the membership of the fuzzy logic output set. Its expression is: μ O =max(min(μ X ,μ Y ,μ Z )), where μ X , μ Y and μ ZThey represent the membership of RSRP in UL, the angle between the drone and the BS, and the BS load, respectively. For each rule, the output membership is obtained by taking the minimum of the input membership. In order to combine different rules, the final output membership is obtained by taking the maximum of the output membership of each rule.

[0068] Table 1. Rule base

[0069]

[0070]

[0071] Defuzzification module: After combining the rules, the aggregated output set needs to be converted into an exact value, namely BSb c For the score of drone k, in the present invention, the centroid method is used for defuzzification, which calculates the center of gravity (COG) of the aggregated output set. Figure 5 An example of the COG mechanism is shown. Let the normalized output be s, ranging between [0,1]. Assume that the membership of "very bad" and "very good" is zero, while the membership of "bad", "medium" level and "good" is 0.2, 0.5 and 0.4 respectively. To calculate the final score, a shadow shape is formed by truncating the output membership function and merging the shapes using the disjunction principle. The coordinates of the centroid of this shadow shape are calculated on the s axis:

[0072]

[0073] where f O (·) is the upper edge of the shaded shape. Therefore, BSb c The score for drone k is

[0074] The beneficial effect of the present invention is that compared with the traditional UL-DL decoupling solution, the method of the present invention can significantly reduce the SI of the drone end, thereby improving the DL transmission rate.

[0075] The above contents are further detailed descriptions of the present invention in combination with specific preferred embodiments, and it cannot be determined that the specific implementation of the present invention is limited to these descriptions. For ordinary technicians in the technical field to which the present invention belongs, several simple deductions or substitutions can be made without departing from the concept of the present invention, which should be regarded as falling within the protection scope of the present invention.

[0076] The above content is a further detailed description of the present invention in combination with specific preferred embodiments, and it cannot be determined that the specific implementation of the present invention is limited to these descriptions. For ordinary technicians in the technical field to which the present invention belongs, several simple deductions or substitutions can be made without departing from the concept of the present invention, which should be regarded as falling within the scope of protection of the present invention.

Claims

1. A UAV adaptive decoupling method based on fuzzy logic in full-duplex mode, characterized in that , including performing the following steps: Step 1: Place UAV k in UL with candidate BS (b c ) received power UAV k and BS associated with it in DL and the candidate BS in UL (b c ) UAV k is in UL with candidate BS (b c ) Enter fuzzy logic; Step 2: Use fuzzy logic to obtain all BS scores corresponding to drone k Step 3: Based on the BS score from step 2 Determine the decoupling decision; the BS with the highest score, i.e. Selected as UL associated BS.

2. The UAV adaptive decoupling method according to claim 1, characterized in that: In step 1, the fuzzy logic includes a fuzzification module, an inference module and a defuzzification module, wherein the fuzzification module is used to convert the input of step 1 into a fuzzy set through a fuzzy membership function; The reasoning module: fuzzy input is mapped to fuzzy output through a set of 27 IF / THEN rules; the defuzzification module: after combining the rules, the aggregated output set is converted into a numerical value, i.e., BSb c For the score of drone k.

3. The UAV adaptive decoupling method according to claim 2, characterized in that: In the fuzzification module, the input and output language term sets are defined as: O = {very bad, bad, fair, good, very good}; The fuzzy output set represents the suitability of BSb to become the UL associated BS of UAV k when considering the three inputs.

4. The UAV adaptive decoupling method according to claim 2, characterized in that: In the fuzzification module, the membership functions of input and output are generated using trapezoidal and trigonometric functions.

5. The UAV adaptive decoupling method according to claim 2, characterized in that: In the reasoning module, the Min-Max reasoning principle is used to determine the membership of the fuzzy logic output set, which is expressed as: μ O =max(min(μ X ,μ Y ,μ Z )), where μ X , μ Y and μ Z They represent the reference received power in UL, the angle between the UAV and the BS, and the membership degree of the BS load, respectively.

6. The UAV adaptive decoupling method according to claim 5, characterized in that: In the inference module, for each rule, the output membership is obtained by taking the minimum value of the input membership. In order to combine different rules, the final output membership is obtained by taking the maximum value of the output membership of each rule.

7. The UAV adaptive decoupling method according to claim 2, characterized in that: In the defuzzification module, a centroid method is used for defuzzification, which calculates the centroid of the aggregated output set.

8. The UAV adaptive decoupling method according to claim 7, characterized in that: In the defuzzification module, in order to calculate the final score, the output membership function is truncated and the shapes are merged using the disjunction principle to form a shadow shape, and the centroid coordinates of the shadow shape are calculated on the s-axis: where f O (·) is the upper edge of the shaded shape, so BS b c The score for drone k is 9. A system for adaptive decoupling of unmanned aerial vehicles based on fuzzy logic in full duplex mode, characterized in that: include: A memory, a processor and a computer program stored in the memory, wherein the computer program is configured to implement the steps of the method according to any one of claims 1 to 8 when called by the processor.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and the computer program is configured to implement the steps of the method according to any one of claims 1 to 8 when called by a processor.

Images