WO2023087746A1 - 一种用于地下空间灾后应急场景的无人机中继选择方法 - Google Patents

一种用于地下空间灾后应急场景的无人机中继选择方法 Download PDF

Info

Publication number
WO2023087746A1
WO2023087746A1 PCT/CN2022/104900 CN2022104900W WO2023087746A1 WO 2023087746 A1 WO2023087746 A1 WO 2023087746A1 CN 2022104900 W CN2022104900 W CN 2022104900W WO 2023087746 A1 WO2023087746 A1 WO 2023087746A1
Authority
WO
WIPO (PCT)
Prior art keywords
matching
preference
uncertain
pair
user
Prior art date
Application number
PCT/CN2022/104900
Other languages
English (en)
French (fr)
Inventor
王博文
孙彦景
张贝贝
云霄
王婷婷
徐永刚
郝之楹
马占国
Original Assignee
中国矿业大学
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 中国矿业大学 filed Critical 中国矿业大学
Priority to US18/113,258 priority Critical patent/US20230224021A1/en
Publication of WO2023087746A1 publication Critical patent/WO2023087746A1/zh

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
    • H04W4/70—Services for machine-to-machine communication [M2M] or machine type communication [MTC]
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04B—TRANSMISSION
    • H04B7/00—Radio transmission systems, i.e. using radiation field
    • H04B7/14—Relay systems
    • H04B7/15—Active relay systems
    • H04B7/185—Space-based or airborne stations; Stations for satellite systems
    • H04B7/18502—Airborne stations
    • H04B7/18504—Aircraft used as relay or high altitude atmospheric platform
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W24/00—Supervisory, monitoring or testing arrangements
    • H04W24/02—Arrangements for optimising operational condition
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
    • H04W4/90—Services for handling of emergency or hazardous situations, e.g. earthquake and tsunami warning systems [ETWS]
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W40/00—Communication routing or communication path finding
    • H04W40/02—Communication route or path selection, e.g. power-based or shortest path routing
    • H04W40/22—Communication route or path selection, e.g. power-based or shortest path routing using selective relaying for reaching a BTS [Base Transceiver Station] or an access point

Definitions

  • the present application relates to the technical field of unmanned aerial vehicle relay matching, and in particular, relates to a method for selecting an unmanned aerial vehicle relay for post-disaster emergency scenarios in underground spaces.
  • UAVs With the characteristics of strong mobility, high flexibility, convenient operation and strong reconfigurability, UAVs can provide all-round and three-dimensional coverage of disaster-prone and frequent areas such as underground spaces through UAVs and other space-based network resources. , Disaster monitoring and communication coverage without blind spots, can quickly guarantee the normal communication of users in disaster areas with damaged communication infrastructure, and complete emergency communication tasks more efficiently. Therefore, it is necessary to consider UAV relay in emergency communication scenarios to improve emergency communication support capabilities.
  • Matching theory as a mathematical tool to analyze the mutual benefit relationship between users, is widely used in the design of distributed wireless resource allocation and relay selection algorithms.
  • post-disaster emergency communication scenarios in order to ensure timely and effective communication among as many rescuers as possible, multiple D2D pairs are allowed to reuse the same relay UAV, so the one-to-one matching method is not applicable.
  • the D2D pairs that choose the same relay drone are often not independent, and their decisions are often affected by the decisions of other D2D pairs, and stable matching results cannot be obtained, which brings the same group effect The problem.
  • exchange matching is used to eliminate the influence of cohort effect, but some existing methods are centralized and not suitable for dynamic distributed scenarios.
  • the matching process usually requires accurate information such as the connection weights between the two subjects.
  • uncertain information is prone to appear due to the ambiguity and complexity of the decision-making environment.
  • Inter-sequence preference information that is, the preference sequence.
  • the dynamic mobility of drones and D2D users may lead to the inability to obtain accurate preference lists, and thus the ineffective matching between D2D users and relay drones.
  • the purpose of the present invention is to propose an efficient post-disaster UAV uncertain preference sequence bilateral stable matching relay selection method, first generate the uncertain preference sequence of D2D users and relay UAVs, and then use multiple pairs
  • a two-sided matching theory completes the matching between D2D users and relay drones, and then for the same group effect in the D2D user cluster that reuses the same drone relay, a stable matching result is further obtained through exchange matching.
  • the present invention provides a UAV relay selection method for post-disaster emergency scene in underground space, comprising the following steps:
  • the maximum rate, minimum rate and average rate of data transmission by D2D users through UAVs are collected, and the first uncertain preference sequence of D2D users for UAVs is obtained, as well as the UAVs.
  • the second uncertain preference order of the D2D user is obtained, as well as the UAVs.
  • the D2D user's preference strategy for the drone is obtained, and the many-to-one bilateral matching model is constructed according to the many-to-one bilateral matching algorithm;
  • the prediction range of the D2D user for the UAV in the target time slot is obtained.
  • the first uncertain preference order is constructed according to the maximum rate and the minimum rate, and the first preference expression of the first uncertain preference order is obtained, wherein the first preference The expression is:
  • Y ki is the upper limit of the first uncertain preference order
  • Y ki is the individual set that ranks the i-th individual in the UAV set U for D2D pair k
  • Z(U m ) is the The length of the first interval of the two-determined preference order.
  • the second uncertain preference order is constructed according to the maximum rate and the minimum rate, and the second preference expression of the second uncertain preference order is obtained, and the preference expression is:
  • Y mj is the set of individuals in which the drone U m ranks the individuals in the D2D pair set K at the jth position
  • Y ki the number of individuals in the set
  • Z(k) is the number of individuals in the set The length of the second interval of an indeterminate preference order.
  • the preference strategy and the preference list corresponding to the preference strategy are obtained, and a many-to-one bilateral matching model is constructed, wherein the constraints of the many-to-one bilateral matching model are:
  • the matching process of the many-to-one bilateral matching model includes the following steps:
  • the D2D user calculates the transmission performance by predicting the position of the relay UAV, obtains the first uncertain preference order, generates the corresponding preference list according to the first preference expression, and the D2D user sorts the relay UAV according to the preference list and select;
  • the relay UAV After receiving the request from the D2D user, the relay UAV accepts the matching request of the best candidate according to the constraints and rejects other D2D users;
  • Accepted D2D users stop their matching process, and rejected D2D users send matching requests to suboptimal relay drones, until there is no better relay link than the current match, then the matching process is terminated.
  • the process of exchanging matching includes the following steps:
  • step S3 According to the matching result of step S2, judge whether to perform the operation of step S2, wherein the judgment process includes:
  • step S2 When the transmission rate of the first D2D pair increases and the transmission rate of the second D2D pair remains unchanged, the matching process of step S2 is maintained;
  • step S2 When the transmission rate of the second D2D pair increases and the transmission rate of the first D2D pair remains unchanged, the matching process of step S2 is maintained;
  • step S2 When the transmission rates of the first D2D pair and the second D2D pair increase, the matching process of step S2 is maintained;
  • step S2 the matching process of step S2 will not be performed.
  • the constraint condition of exchange matching is: when the first D2D pair and the second D2D pair meet the following two conditions at the same time, an exchange restriction pair is formed, and the process of exchange matching is performed:
  • the steps of the process of exchanging matches include:
  • Step 3 Until there is no exchange restriction pair in the current matching, return the updated matching result.
  • the UAV relay selection system for implementing the UAV relay selection method includes,
  • the data collection and processing module is used to collect the maximum rate, minimum rate and average rate of D2D users' assistance in transmitting data through drones based on the prediction range of D2D users for drones, and obtain the first uncertainty of D2D users for drones preference order, and the second uncertain preference order of the UAV for the D2D user;
  • the preference matching module is used to obtain the preference strategy of the D2D user for the drone based on the first uncertain preference sequence and the second uncertain preference sequence, and construct a many-to-one bilateral matching model according to the many-to-one bilateral matching algorithm;
  • the exchange matching module is used for exchanging and matching the UAVs that have been matched by any two D2D users based on the matching result of the preference matching module, so as to obtain a stable matching result of bilateral exchange.
  • the computational complexity of the algorithm of the present invention is greatly reduced compared with the exhaustive algorithm; in addition, the present invention allows the same relay UAV to communicate with multiple pairs of D2D users in the same time slot, and the matching result is stable, It can ensure timely and effective communication among as many rescuers as possible; the performance of the algorithm of the present invention is superior to other comparison methods, and it is more suitable for emergency dynamic distributed scenarios.
  • Fig. 1 is the underground space emergency communication scene assisted by the UAV described in the present invention
  • Fig. 2 is the unmanned aerial vehicle relay mode and transmission process described in the present invention
  • FIG. 3 shows the relationship between the D2D user transmission success rate and the expected value of transmission data according to the present invention
  • Fig. 4 is the relationship between the D2D user transmission success rate and the number of relay drones according to the present invention.
  • FIG. 5 shows the relationship between the D2D user transmission success rate and the number of D2D pairs according to the present invention.
  • the present invention provides a UAV relay selection method for post-disaster emergency scenarios in underground space, including the following steps:
  • the maximum rate, minimum rate and average rate of data transmission by D2D users through UAVs are collected, and the first uncertain preference sequence of D2D users for UAVs is obtained, as well as the UAVs.
  • the second uncertain preference order of the D2D user is obtained, as well as the UAVs.
  • the D2D user's preference strategy for the drone is obtained, and the many-to-one bilateral matching model is constructed according to the many-to-one bilateral matching algorithm;
  • the prediction range of the D2D user for the UAV in the target time slot is obtained.
  • the process of obtaining the first uncertain preference order construct the first uncertain preference order according to the maximum rate and the minimum rate, and obtain the first preference expression of the first uncertain preference order, where the first preference The expression is:
  • Y ki is the upper limit of the first uncertain preference order
  • Y ki is the individual set that ranks the i-th individual in the UAV set U for D2D pair k
  • Z(U m ) is the The length of the first interval of the two-determined preference order.
  • the preference expression is:
  • Y mj is the set of individuals in which the drone U m ranks the individuals in the D2D pair set K at the jth position
  • Y ki the number of individuals in the set
  • Z(k) is the number of individuals in the set The length of the second interval of an indeterminate preference order.
  • the preference strategy and the preference list corresponding to the preference strategy are obtained, and a many-to-one bilateral matching model is constructed, wherein the constraints of the many-to-one bilateral matching model are:
  • matching ⁇ is defined as mapping from the set K ⁇ U to , for any D2D user k belongs to the D2D user set K, the matching object ⁇ (k) of k belongs to the UAV relay set U or the empty set (indicating that k has no matching object), and k can only match at most one Human-machine relay; for any UAV relay U m belongs to the UAV relay set U, the matching objects ⁇ (U m ) of U m all belong to the D2D user set K or the empty set (indicating that U m has no matching objects) , and each UAV relay can serve at most q 0 D2D user pairs; for any D2D user k belongs to the D2D user set K, any UAV relay U m belongs to the UAV relay set U, the matching of k
  • the object ⁇ (k) being U m is equivalent to the matching object ⁇ (U m ) of U m being k.
  • the matching process of the many-to-one bilateral matching model includes the following steps:
  • the D2D user calculates the transmission performance by predicting the position of the relay UAV, obtains the first uncertain preference order, generates the corresponding preference list according to the first preference expression, and the D2D user sorts the relay UAV according to the preference list and select;
  • the relay UAV After receiving the request from the D2D user, the relay UAV accepts the matching request of the best candidate according to the constraints and rejects other D2D users;
  • Accepted D2D users stop their matching process, and rejected D2D users send matching requests to suboptimal relay drones, until there is no better relay link than the current match, then the matching process is terminated.
  • the process of exchanging matching includes the following steps:
  • step S3 According to the matching result of step S2, judge whether to perform the operation of step S2, wherein the judgment process includes:
  • step S2 When the transmission rate of the first D2D pair increases and the transmission rate of the second D2D pair remains unchanged, the matching process of step S2 is maintained;
  • step S2 When the transmission rate of the second D2D pair increases and the transmission rate of the first D2D pair remains unchanged, the matching process of step S2 is maintained;
  • step S2 When the transmission rates of the first D2D pair and the second D2D pair increase, the matching process of step S2 is maintained;
  • step S2 the matching process of step S2 will not be performed.
  • the constraint condition of exchange matching is: when the first D2D pair and the second D2D pair meet the following two conditions at the same time, an exchange restriction pair is formed, and the process of exchange matching is performed:
  • steps in the process of exchanging matches include:
  • Step 3 Until there is no exchange restriction pair in the current matching, return the updated matching result.
  • the UAV relay selection system for realizing the UAV relay selection method includes,
  • the data collection and processing module is used to collect the maximum rate, minimum rate and average rate of D2D users' assistance in transmitting data through drones based on the prediction range of D2D users for drones, and obtain the first uncertainty of D2D users for drones preference order, and the second uncertain preference order of the UAV for the D2D user;
  • the preference matching module is used to obtain the preference strategy of the D2D user for the unmanned aerial vehicle based on the first uncertain preference sequence and the second uncertain preference sequence, and construct a many-to-one bilateral matching model according to the many-to-one bilateral matching algorithm;
  • the exchange matching module is used for exchanging and matching the UAVs that have been matched by any two D2D users based on the matching result of the preference matching module, so as to obtain a stable matching result of bilateral exchange.
  • Embodiment 1 the following content is set forth in conjunction with accompanying drawing to the specific embodiment of the present invention:
  • Figure 1 is a schematic diagram of an emergency communication scene in underground space assisted by drones taking a mine as an example.
  • affected by mine accidents such as rock bursts, gas explosions, and coal fires, underground fixed ) partially or completely fail and it is difficult to restore communication services in a short time.
  • Rescuers cannot quickly perceive the disaster situation in the disaster-stricken area and transmit emergency information.
  • Personnel complete the search and rescue task, based on this, the present invention adopts the decoding and forwarding protocol for data transmission.
  • the UAV uses time-division multiple access technology to cooperate with these D2D pairs.
  • the D2D pair needs to transmit data with a size of , and the data packets are transmitted frame by frame, and the length of each frame is equal.
  • a frame is divided into two phases. In phase 1, the transmitting end of the D2D pair transmits data to the relay drone, and in phase 2, the relay drone transmits the data to the receiving end of the D2D pair.
  • the transmission model is shown in Figure 2, assuming In the case of relay UAV cooperative D2D pair 1 and D2D pair 2 transmission, since the transmission data size and transmission rate of D2D users are different, the relay UAV can allocate different time resources to different D2D pairs during the transmission time. . In the time slot, the data rate transmitted from the transmitting end of the D2D pair to the receiving end through the UAV is:
  • B is the channel bandwidth, and are the signal-to-noise ratios from the transmitting end S k of the D2D pair k to the UAV U m , U m to the receiving end D k of the D2D pair k, and S k to D k respectively, q(U m ) represents the U m Number of D2D pairs assisted. If the ratio of the data size of the D2D pair k to the rate, that is, the transmission duration is smaller than the threshold ⁇ th , it is considered that the data of the D2D pair k can be successfully transmitted.
  • the dynamic characteristics of the network cause the data rate may be different in different time slots, and thus the number of D2D pairs successfully transmitted may be different. Therefore, by optimizing the real-time relay UAV allocation to maximize the average total transmission success rate of D2D users, that is, to maximize the overall
  • the average ratio of the total number of successfully transmitted D2D pairs to the total number of D2D pairs in the network transmission phase can be expressed by the following formula
  • the optimization problem is NP-hard.
  • the optimal UAV relay assignment for D2D users is time-varying.
  • the flight trajectories of UAVs are determined by their missions and are unknown to ground users, and the dynamic positions of D2D users are also unknown to UAVs. Therefore, an offline planning method is not desirable and an online method is required.
  • the objective optimization problem is actually to maximize the number of successful transmissions by D2D users, and the choice of relay drone not only affects the transmission of the D2D pair for the current decision, but also affects the transmission of other D2D pairs that multiplex the same relay drone, Therefore, the optimization problem will be converted to the optimization of relay UAV selection for solution.
  • this invention proposes a multi-to-one bilateral stable matching method under an uncertain preference order, which is suitable for the uncertainty of the matching strategy caused by the dynamic location of UAVs and D2D users
  • the scene has good stability and effectiveness.
  • the method consists of two steps: 1) D2D users and UAVs generate uncertain preference sequences based on uncertain information, and comprehensively evaluate the uncertain preference sequences to generate corresponding preference lists, and establish a many-to-one bilateral matching model to obtain matching results ; 2)
  • the matching result at this time is unstable, and each exchange operation needs to be further iterated to eliminate the influence of the peer effect to obtain a stable matching result.
  • the present invention first proposes a hierarchical strategy, a many-to-one bilateral matching model based on an uncertain preference order.
  • the flight trajectory of the drone is unknown to D2D users. It is difficult for D2D users to directly generate a specific preference list based on uncertain information. It is necessary to consider the case where the preference information is an ordinal value.
  • the ground user has a prediction range l(U m ,t n ) for the UAV U m at t n-1 time slot position, and in the prediction range l( U m , t n )
  • the maximum rate, minimum rate and average rate of D2D pair k assisting in data transmission through the UAV U m are obtained, and they are expressed as and And according to these rates, the uncertain preference order of individual k in K to individual U m in U is obtained and sequence interval
  • the upper and lower bounds of , respectively, are given by and The ranking is determined. In particular, if but degenerates to an exact ordinal value.
  • the UAV calculates the maximum speed of the D2D pair within the moving range l(k,t n ) minimum rate and the average rate And according to these rates, the uncertain preference order of individual U m in U to individual k in K is obtained and sequence interval The upper and lower bounds of , respectively, are given by and The ranking is determined. In particular, if but degenerates to an exact ordinal value.
  • the smaller the value of the uncertain preference order the higher the position of k in the evaluation ranking of U m , and U m will give priority to providing relay transmission services. Based on this, the comprehensive score of individual k in D2D pair set K to individual U m in relay UAV set U and the comprehensive score of individual U m in U to individual k in K can be expressed by the following formulas:
  • Y ki is the set of individuals in which k ranks individuals in U at the i position
  • Y mj is the set of individuals in which U m ranks the individuals in K at the j position
  • Z(U m ), Z(k) are the lengths of the uncertain preference sequence intervals of U m and k respectively. In particular, if There is no need to subtract the corresponding points.
  • the comprehensive score obtained based on the uncertain preference order of the participants not only takes into account the degree of competition among matching individuals in the same rank, but also reflects the difference in preferences of different individuals, which largely reflects the true nature of the subject. will.
  • the idea of the many-to-one matching algorithm based on the uncertain preference order proposed in the present invention is: in each time slot, the D2D user calculates the transmission performance by predicting the position of the relay UAV, and obtains the uncertain preference order , based on which the comprehensive evaluation generates the corresponding preference list. D2D users sort and select relay drones according to the preference list. After the relay UAV receives the request from the D2D user, it will accept the matching request of the best candidate and reject other D2D users under the quota constraint requirements. Accepted D2D users stop their matching process, while rejected D2D users continue to issue matching requests to suboptimal relay drones. The matching process terminates until there is no better relay link than the current match.
  • the many-to-one matching algorithm under uncertain preference order is summarized as follows:
  • Step 1 The algorithm progresses to time slot t n .
  • D2D establishes a preference list for k ⁇ K and relay drone Um ⁇ U according to their respective uncertain preference order comprehensive evaluation.
  • Step 2 If
  • 0, and the preference list of k is not empty, then perform the following steps for k ⁇ K and U m ⁇ U respectively until all D2D pairs are matched to the relay UAVs in the preference list or are relayed Drone refused.
  • the numbers of D2D pairs and UAVs participating in the matching are K and M respectively.
  • the candidate relay set of any D2D pair contains all UAVs. machine, all relay drones’ preferences for D2D users do not meet the requirements of relay drones.
  • D2D users participating in matching need to continuously send requests to other relay drones and are rejected , so the worst time complexity of the algorithm is O(MK), and the worst time complexity of the algorithm in the whole task transmission cycle of the system is O(NMK).
  • the matching result of the above-mentioned many-to-one matching algorithm based on uncertain preference order is unstable because of the peer effect, that is, a D2D user may constantly change his preference order according to the preference list formed by other D2D users, so The final trunk selection result is never reached. Therefore, in the second stage, the present invention introduces how to exchange their matched relay UAVs between any two D2D pairs through an exchange operation, so as to avoid the peer effect and achieve a stable matching state.
  • D2D pairs k and k' exchange matching objects with each other.
  • one of the participating D2D pairs is allowed to be empty, so a single D2D pair moves to the available space corresponding to the drone.
  • exchange matching is performed between exchange-restricted pairs.
  • the transmission rate of all the D2D pairs participating in the exchange will not decrease, but the transmission rate of at least one of the D2D pairs will increase, which also avoids cycling between equivalent matches.
  • the exchange operation is performed on the basis of utility value, based on the rate increase within the forecast range as the exchange condition to reflect the uncertainty in the transmission process.
  • stability is usually used to measure the quality of the matching scheme. For a many-to-one bilateral matching ⁇ , if there are no exchange-restricted pairs, the matching ⁇ is said to be bilaterally exchange-stable.
  • the present invention iteratively completes each exchange operation on the exchange restriction pair based on the matching result ⁇ (t n ) under the many-to-one matching algorithm under the uncertain preference order. Therefore, this method can dynamically obtain the stable matching results of bilateral exchanges.
  • the relay unmanned aerial vehicle selection method that avoids group effect proposed by the present invention can be summarized as follows:
  • Step 3 Until there is no exchange restriction pair in the current matching, return the updated matching result ⁇ (t n ).
  • C kk' is expressed as the number of times D2D pairs k send exchange requests to k', and k can be exchanged with k' at most twice, which avoids the ping-pong effect and ensures the convergence of the algorithm.
  • the exchange matching process ends, and the matching result ⁇ (t n ) is updated.
  • the method proposed in the second stage has more steps of exchange and matching, and the upper limit of the number of exchanges is Therefore, the worst time complexity of the relay UAV selection method to avoid the peer effect in the whole mission transmission cycle of the system is
  • the matching ⁇ finally obtained by the method proposed by the present invention is bilateral exchange stable.
  • Embodiment 2 A specific embodiment of the present invention is described as follows: the system simulation adopts Matlab2015a, and the parameter setting does not affect the generality of the present invention. D2D pairs are randomly distributed in a 3km ⁇ 3km area, the distance between the transmitter and the corresponding receiver is randomly selected within (100m, 200m), and each time slot between the transmitter and the receiver is randomly within (1m, 2m) ) to move within the range.
  • the UAV randomly selects the flight direction and does not fly out of the specified area.
  • the flight altitude is [100m, 200m]
  • the flight speed is 10m/s
  • the battery capacity is 2 ⁇ 10 5 J.
  • the energy consumption model of the UAV refers to the multi-rotor UAV energy consumption model (reference: Wang B, Sun Y, Liu D, et al. Social-aware UAV-assisted Mobile Crowd Sensing in Stochastic and Dynamic Environments for Disaster Relief Networks. IEEE Transactions on Vehicular Technology, 2020, 69( 1): 1070-1074), the comparison algorithm of the present invention is exhaustive search algorithm (optimum solution can be obtained but time complexity is extremely high), random relay selection algorithm (time complexity is low but performance is unstable) and one-to-one Matching algorithm (references: Liu D, Yang Y, Xu Y, et al. Uncertain Preference Matching-Based Relay Selection and Position Adjustment in Dynamic UAV Systems//2020 International Conference on Wireless Communications and Signal Processing (WCSP), 2020:1170 -1175).
  • the comparison algorithm of the present invention is exhaustive search algorithm (optimum solution can be obtained but time complexity is extremely high), random relay selection algorithm (time complexity is low but performance is unstable) and one-to-
  • Fig. 3, Fig. 4 and Fig. 5 respectively show the method proposed by the present invention (based on the many-to-one matching algorithm under the uncertain preference order and the relay UAV selection method to avoid the same group effect) and the existing The method compares the success rate of D2D user transmission in the three cases of the change of D2D user transmission data expectation value, the change of the number of relay drones and the change of the number of D2D pairs. It can be seen from these three graphs that although the transmission success rate of D2D users obtained by the algorithm proposed in this paper is lower than that of the exhaustive algorithm, and the result is suboptimal, the computational complexity of the proposed algorithm is higher than that of the exhaustive algorithm. significantly reduced.
  • the proposed method allows the same relay UAV to communicate with multiple pairs of D2D users in the same time slot, and the matching result is stable, which can ensure timely and effective communication with as many rescuers as possible. Therefore, the performance of the proposed algorithm is superior to other comparison methods, and it is more suitable for emergency dynamic distributed scenarios.

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Aviation & Aerospace Engineering (AREA)
  • General Physics & Mathematics (AREA)
  • Physics & Mathematics (AREA)
  • Astronomy & Astrophysics (AREA)
  • Business, Economics & Management (AREA)
  • Health & Medical Sciences (AREA)
  • Emergency Management (AREA)
  • Environmental & Geological Engineering (AREA)
  • Public Health (AREA)
  • Mobile Radio Communication Systems (AREA)

Abstract

本发明公开了一种用于地下空间灾后应急场景的无人机中继选择方法，包括以下步骤：基于D2D用户对于无人机的预测范围，采集D2D用户通过无人机协助传输数据的最大速率、最小速率和平均速率，获取D2D用户对于无人机的第一不确定偏好序，以及无人机对于D2D用户的第二不确定偏好序；基于第一不确定偏好序和第二不确定偏好序，获取D2D用户对于无人机的偏好策略，根据多对一双边匹配算法进行匹配后，将任意两个D2D用户已匹配的无人机进行交换匹配，获取双边交换稳定的匹配结果；本发明的算法计算复杂度低，允许同一中继无人机在同一时隙协作多对D2D用户进行通信，且匹配结果稳定，能保障尽可能多的救援人员及时有效沟通。

Description

一种用于地下空间灾后应急场景的无人机中继选择方法 技术领域
本申请涉及无人机中继匹配技术领域，具体而言，涉及一种用于地下空间灾后应急场景的无人机中继选择方法。
背景技术
随着经济的高速发展与基建智能化进程的加快，包含矿井、隧道、地铁、人防工程、地下停车场等的地下空间的开发规模不断扩大，同时给安全性方面带来了严峻挑战。地下空间一旦发生灾害，基础传感通信设施部分或全部损毁，受困群众和救援人员无法实时传递信息，救援行动难以有效实施，容易造成人员伤亡和经济损失。因此，需要一个弹性的应急通信网络，快速实现断点续传。无人机凭借移动性强、灵活度高、操控方便以及可重构性强等特性，通过无人机等空基网络资源，对地下空间等灾害事故易发多发频发区域全方位、立体化、无盲区的灾情监测与通信覆盖，能迅速保障通信基础设施受损的灾害地区用户的正常通信，更为高效地完成应急通信任务。因此，在应急通信场景中考虑无人机中继提高应急通信保障能力十分必要。
匹配理论作为分析用户互利关系的数学工具，被广泛应用于分布式无线资源分配与中继选择算法设计中。在灾后应急通信场景中，为了保证尽可能多的救援人员及时有效沟通，允许多个D2D对复用同一中继无人机，因此一对一匹配方法不适用。但在多对一匹配中，选择同一中继无人机的D2D对间往往不是独立的，他们的决定经常受到其他D2D对决定的影响，无法得到稳定的匹配结果，这带来了同群效应 的问题。一般通过交换匹配消除同群效应的影响，但现有的一些方法是集中式的，不适用于动态分布式场景中。而且通常匹配过程需要双边主体间的连接权重等准确信息，而在现实情况中，由于决策环境的模糊性和复杂性容易出现不确定信息，匹配双边中一边更容易获得另一边个体的不精确的序区间偏好信息，即偏好序列。在灾后环境中，无人机和D2D用户的动态移动性可能导致各自不能获得精确的偏好列表，进而D2D用户和中继无人机之间无法进行有效的匹配。
发明内容
针对以上问题，本发明的目的是提出一种高效的灾后无人机不确定偏好序双边稳定匹配中继选择方法，首先生成D2D用户和中继无人机的不确定偏好序，再利用多对一双边匹配理论完成D2D用户和中继无人机间的匹配，然后针对复用同一无人机中继的D2D用户簇内存在同群效应，通过交换匹配进一步得到稳定的匹配结果。
为了实现上述目的，本发明提供了一种用于地下空间灾后应急场景的无人机中继选择方法，包括以下步骤：
基于D2D用户对于无人机的预测范围，采集D2D用户通过无人机协助传输数据的最大速率、最小速率和平均速率，获取D2D用户对于无人机的第一不确定偏好序，以及无人机对于D2D用户的第二不确定偏好序；
基于第一不确定偏好序和第二不确定偏好序，获取D2D用户对于无人机的偏好策略，根据多对一双边匹配算法，构建多对一双边匹配模型；
基于多对一双边匹配模型，将任意两个D2D用户已匹配的无人机进行交换匹配，获取双边交换稳定的匹配结果。
优选地，在获取第一不确定偏好序和第二不确定偏好序的过程中，通过选择目标时隙，获取D2D用户对于无人机在目标时隙的预测范围。
优选地，在获取第一不确定偏好序的过程中，根据最大速率、最小速率，构建第一不确定偏好序，并获取第一不确定偏好序的第一偏好表达式，其中，第一偏好表达式为：
Figure PCTCN2022104900-appb-000001
其中，
Figure PCTCN2022104900-appb-000002
为第一不确定偏好序的下限，
Figure PCTCN2022104900-appb-000003
为第一不确定偏好序的上限，Y ki为D2D对k将无人机集合U中个体排在第i位的个体集合，|Y ki|集合中个体的数量，Z(U m)为第二不确定偏好序的第一区间长度。
优选地，在获取第二不确定偏好序的过程中，根据最大速率、最小速率，构建第二不确定偏好序，并获取第二不确定偏好序的第二偏好表达式，偏好表达式为：
Figure PCTCN2022104900-appb-000004
其中，
Figure PCTCN2022104900-appb-000005
为第二不确定偏好序的下限，
Figure PCTCN2022104900-appb-000006
为第二不确定偏好序的上限，Y mj为无人机U m将D2D对集合K中个体排在第j位的个体集合，|Y ki|集合中个体的数量，Z(k)为第一不确定偏好序的第二区间长度。
优选地，根据第一偏好表达式和第二偏好表达式，获取偏好策略 以及偏好策略对应的偏好列表，构建多对一双边匹配模型，其中，多对一双边匹配模型的约束条件为：
Figure PCTCN2022104900-appb-000007
其中，
Figure PCTCN2022104900-appb-000008
Figure PCTCN2022104900-appb-000009
Figure PCTCN2022104900-appb-000010
优选地，在构建多对一双边匹配模型的过程中，多对一双边匹配模型的匹配过程包括以下步骤：
D2D用户通过预测中继无人机的位置来计算传输性能，得到第一不确定偏好序，根据第一偏好表达式生成对应的偏好列表，D2D用户根据偏好列表，对中继无人机进行排序和选择；
中继无人机收到D2D用户的请求后，根据约束条件，接受最佳候选者的匹配请求，并拒绝其他的D2D用户；
被接受的D2D用户停止其匹配过程，被拒绝的D2D用户向次优中继无人机发出匹配请求，直到没有比当前匹配项更好的中继链路，则匹配过程终止。
优选地，在匹配过程终止的过程后，交换匹配的过程包括以下步骤：
S1.选取D2D用户的第一D2D对和第二D2D对，以及第一D2D对的第一匹配对象、第二D2D对的第二匹配对象；
S2.将第一D2D对与第二匹配对象进行匹配，同时将第二D2D对与第一匹配对象进行匹配；
S3.根据步骤S2的匹配结果，判断是否执行步骤S2的操作，其中，判断过程包括：
当第一D2D对的传输速率增加，第二D2D对的传输速率不变时，则保持步骤S2的匹配过程；
当第二D2D对的传输速率增加，第一D2D对的传输速率不变时，则保持步骤S2的匹配过程；
当第一D2D对和第二D2D对的传输速率增加时，则保持步骤S2的匹配过程；
否则，将不执行步骤S2的匹配过程。
优选地，交换匹配的约束条件为：当第一D2D对和第二D2D对同时满足以下两个条件时，形成交换限制对，执行交换匹配的过程：
条件一：
Figure PCTCN2022104900-appb-000011
条件二：
Figure PCTCN2022104900-appb-000012
优选地，交换匹配的过程的步骤包括：
步骤1：初始阶段，初始化D2D对k向D2D对k'发送交换请求的次数，即C kk'＝0；
步骤2：每个D2D对k∈K搜索另一个D2D对k'∈{K\k}以形成交换限制对，如果(k,k')形成交换限制对，并且满足C kk'+C k'k≤2，则根据基于不确定偏好序下的多对一匹配算法更新匹配结果，且C kk'＝C kk'+1；否则保持当前的匹配状态；
步骤3：直到当前匹配中不存在任何交换限制对，则返回更新后的匹配结果。
优选地，用于实现无人机中继选择方法的无人机中继选择系统包括，
数据采集处理模块，用于基于D2D用户对于无人机的预测范围，采集D2D用户通过无人机协助传输数据的最大速率、最小速率和平均速率，获取D2D用户对于无人机的第一不确定偏好序，以及无人机对于D2D用户的第二不确定偏好序；
偏好匹配模块，用于基于第一不确定偏好序和第二不确定偏好序，获取D2D用户对于无人机的偏好策略，根据多对一双边匹配算法，构建多对一双边匹配模型；
交换匹配模块，用于基于偏好匹配模块的匹配结果，将任意两个D2D用户已匹配的无人机进行交换匹配，获取双边交换稳定的匹配结果。
与现有技术相比，本发明的算法计算复杂度较穷举算法大幅度降低；另外，本发明允许同一中继无人机在同一时隙协作多对D2D用户进行通信，且匹配结果稳定，能保障尽可能多的救援人员及时有效沟通；本发明算法性能优于其他对比方法，更适用于应急动态分布式场景。
附图说明
为了更清楚地说明本发明实施例或现有技术中的技术方案，下面将对实施例中所需要使用的附图作简单地介绍，显而易见地，下面描述中的附图仅仅是本发明的一些实施例，对于本领域普通技术人员来讲，在不付出创造性劳动性的前提下，还可以根据这些附图获得其他 的附图。
图1为本发明所述的无人机辅助的地下空间应急通信场景；
图2为本发明所述的无人机中继模式和传输过程；
图3为本发明所述的D2D用户传输成功率随传输数据期望值变化的关系；
图4为本发明所述的D2D用户传输成功率随中继无人机数量变化的关系；
图5为本发明所述的D2D用户传输成功率随D2D对数量变化的关系。
具体实施方式
下为使本申请实施例的目的、技术方案和优点更加清楚，下面将结合本申请实施例中附图，对本申请实施例中的技术方案进行清楚、完整地描述，显然，所描述的实施例仅仅是本申请一部分实施例，而不是全部的实施例。通常在此处附图中描述和示出的本申请实施例的组件可以以各种不同的配置来布置和设计。因此，以下对在附图中提供的本申请的实施例的详细描述并非旨在限制要求保护的本申请的范围，而是仅仅表示本申请的选定实施例。基于本申请的实施例，本领域技术人员在没有做出创造性劳动的前提下所获得的所有其他实施例，都属于本申请保护的范围。
如图1-5所示，本发明提供了一种用于地下空间灾后应急场景的无人机中继选择方法，包括以下步骤：
基于D2D用户对于无人机的预测范围，采集D2D用户通过无人机协助传输数据的最大速率、最小速率和平均速率，获取D2D用户对于无人机的第一不确定偏好序，以及无人机对于D2D用户的第二不确定偏好序；
基于第一不确定偏好序和第二不确定偏好序，获取D2D用户对于无人机的偏好策略，根据多对一双边匹配算法，构建多对一双边匹配模型；
基于多对一双边匹配模型，将任意两个D2D用户已匹配的无人机进行交换匹配，获取双边交换稳定的匹配结果。
进一步地，在获取第一不确定偏好序和第二不确定偏好序的过程中，通过选择目标时隙，获取D2D用户对于无人机在目标时隙的预测范围。
进一步地，在获取第一不确定偏好序的过程中，根据最大速率、最小速率，构建第一不确定偏好序，并获取第一不确定偏好序的第一偏好表达式，其中，第一偏好表达式为：
Figure PCTCN2022104900-appb-000013
其中，
Figure PCTCN2022104900-appb-000014
为第一不确定偏好序的下限，
Figure PCTCN2022104900-appb-000015
为第一不确定偏好序的上限，Y ki为D2D对k将无人机集合U中个体排在第i位的个体集合，|Y ki|集合中个体的数量，Z(U m)为第二不确定偏好序的第一区间长度。
进一步地，在获取第二不确定偏好序的过程中，根据最大速率、最小速率，构建第二不确定偏好序，并获取第二不确定偏好序的第二 偏好表达式，偏好表达式为：
Figure PCTCN2022104900-appb-000016
其中，
Figure PCTCN2022104900-appb-000017
为第二不确定偏好序的下限，
Figure PCTCN2022104900-appb-000018
为第二不确定偏好序的上限，Y mj为无人机U m将D2D对集合K中个体排在第j位的个体集合，|Y ki|集合中个体的数量，Z(k)为第一不确定偏好序的第二区间长度。
进一步地，根据第一偏好表达式和第二偏好表达式，获取偏好策略以及偏好策略对应的偏好列表，构建多对一双边匹配模型，其中，多对一双边匹配模型的约束条件为：
Figure PCTCN2022104900-appb-000019
其中，
Figure PCTCN2022104900-appb-000020
Figure PCTCN2022104900-appb-000021
Figure PCTCN2022104900-appb-000022
上式表示为：匹配ω定义为从集合K∪U映射到
Figure PCTCN2022104900-appb-000023
的集合，对于任意D2D用户k属于D2D用户集合K，k的匹配对象ω(k)都属于无人机中继集合U或空集(表示k无匹配对象)，且k最多只能匹配一个无人机中继；对于任意无人机中继U m属于无人机中继集合U，U m的匹配对象ω(U m)都属于D2D用户集合K或空集(表示U m无匹配对象)，且每个无人机中继最多可服务q 0个D2D用户对；对于任意D2D用户k属于D2D用户集合K，任意无人机中继U m属于无人机中继集合U，k的匹配对象ω(k)为U m等价于U m的匹配对象ω(U m)为k。
进一步地，在构建多对一双边匹配模型的过程中，多对一双边匹 配模型的匹配过程包括以下步骤：
D2D用户通过预测中继无人机的位置来计算传输性能，得到第一不确定偏好序，根据第一偏好表达式生成对应的偏好列表，D2D用户根据偏好列表，对中继无人机进行排序和选择；
中继无人机收到D2D用户的请求后，根据约束条件，接受最佳候选者的匹配请求，并拒绝其他的D2D用户；
被接受的D2D用户停止其匹配过程，被拒绝的D2D用户向次优中继无人机发出匹配请求，直到没有比当前匹配项更好的中继链路，则匹配过程终止。
进一步地，在匹配过程终止的过程后，交换匹配的过程包括以下步骤：
S1.选取D2D用户的第一D2D对和第二D2D对，以及第一D2D对的第一匹配对象、第二D2D对的第二匹配对象；
S2.将第一D2D对与第二匹配对象进行匹配，同时将第二D2D对与第一匹配对象进行匹配；
S3.根据步骤S2的匹配结果，判断是否执行步骤S2的操作，其中，判断过程包括：
当第一D2D对的传输速率增加，第二D2D对的传输速率不变时，则保持步骤S2的匹配过程；
当第二D2D对的传输速率增加，第一D2D对的传输速率不变时，则保持步骤S2的匹配过程；
当第一D2D对和第二D2D对的传输速率增加时，则保持步骤S2 的匹配过程；
否则，将不执行步骤S2的匹配过程。
进一步地，交换匹配的约束条件为：当第一D2D对和第二D2D对同时满足以下两个条件时，形成交换限制对，执行交换匹配的过程：
条件一：
Figure PCTCN2022104900-appb-000024
条件二：
Figure PCTCN2022104900-appb-000025
进一步地，交换匹配的过程的步骤包括：
步骤1：初始阶段，初始化D2D对k向D2D对k'发送交换请求的次数，即C kk'＝0；
步骤2：每个D2D对k∈K搜索另一个D2D对k'∈{K\k}以形成交换限制对，如果(k,k')形成交换限制对，并且满足C kk'+C k'k≤2，则根据基于不确定偏好序下的多对一匹配算法更新匹配结果，且C kk'＝C kk'+1；否则保持当前的匹配状态；
步骤3：直到当前匹配中不存在任何交换限制对，则返回更新后的匹配结果。
进一步地，用于实现无人机中继选择方法的无人机中继选择系统包括，
数据采集处理模块，用于基于D2D用户对于无人机的预测范围，采集D2D用户通过无人机协助传输数据的最大速率、最小速率和平均速率，获取D2D用户对于无人机的第一不确定偏好序，以及无人机对于D2D用户的第二不确定偏好序；
偏好匹配模块，用于基于第一不确定偏好序和第二不确定偏好 序，获取D2D用户对于无人机的偏好策略，根据多对一双边匹配算法，构建多对一双边匹配模型；
交换匹配模块，用于基于偏好匹配模块的匹配结果，将任意两个D2D用户已匹配的无人机进行交换匹配，获取双边交换稳定的匹配结果。
实施例1：以下内容结合附图对本发明具体实施方式进行阐述：
1、场景及问题概述：
图1是无人机辅助以矿井为例的地下空间应急通信场景示意图，在该场景中，受冲击地压、瓦斯爆炸以及煤火等矿井事故的影响，井下固定的基础设施(如矿用基站)部分或全部失效且短时间内难以恢复通信服务，救援人员无法快速感知受灾区域的灾情态势及传递应急信息，因此需要无人机集群迅速进入灾区域充当地面用户的中继传输节点，协助救援人员完成搜救任务，基于此，本发明采用解码转发协议进行数据传输。当有多个D2D对选择同一无人机，无人机采用时分多址接入技术协作这些D2D对。D2D对有大小为的数据需要传输，数据包逐帧传输，每帧的长度均相等。一帧分为两个阶段，阶段1D2D对的发射端将数据传输至中继无人机，阶段2中继无人机将数据传输至D2D对的接收端，传输模型如图2所示，假设中继无人机协作D2D对1和D2D对2传输的情况，由于D2D用户的传输数据大小和传输速率不同，中继无人机在传输时间内可以将不同的时间资源分配给不同的D2D对。在时隙，D2D对的发射端通过无人机传输至接收端的数据速率为：
Figure PCTCN2022104900-appb-000026
其中，B为信道带宽，
Figure PCTCN2022104900-appb-000027
和
Figure PCTCN2022104900-appb-000028
分别为D2D对k的发射端S k到无人机U m、U m到D2D对k的接收端D k和S k到D k的信噪比，q(U m)表示无人机U m协助的D2D对数量。D2D对k的数据大小与速率的比值，即传输时长若小于阈值τ th，则视D2D对k的数据能成功传输。
网络的动态特性导致数据速率在不同时隙可能不同，从而成功传输的D2D对数量可能不同，因此，通过优化实时中继无人机分配最大化D2D用户的平均总传输成功率，即最大化整个网络传输阶段中总的传输成功的D2D对数量与D2D对总数的平均比值，可用以下公式表示
Figure PCTCN2022104900-appb-000029
其中，
Figure PCTCN2022104900-appb-000030
要使D2D用户的平均总传输成功率最大化，必须解决以下几个难点。首先，该优化问题是NP难问题。然后，由于无人机动态飞行，位置实时变化，因此对D2D用户而言的最佳无人机中继分配是随时间变化而变化的。最后，无人机的飞行轨迹是由它们的任务决定的，对于地面用户来说是未知的，且D2D用户的动态位置对无人机同样未知。因此，离线规划方法不可取，需要在线方法。目标优化问题实际上是最大化D2D用户成功传输的数量，而中继无人机的选择不仅影响 当前决策的D2D对的传输，还影响复用同一中继无人机的其他D2D对的传输，因此，下面将优化问题转为中继无人机选择的优化进行求解。
2、方法概述
为了解决上述D2D对的中继无人机选择问题，本发明提出一种不确定偏好序下多对一双边稳定匹配方法，适用于无人机和D2D用户的动态位置导致匹配策略的不确定性场景，具有良好的稳定性与有效性。该方法由两步构成：1)D2D用户和无人机根据不确定的信息生成不确定偏好序，并对不确定偏好序综合评估生成相应的偏好列表，建立多对一双边匹配模型得到匹配结果；2)此时的匹配结果是不稳定的，需要进一步迭代完成每个交换操作，消除同群效应的影响得到稳定的匹配结果。
2.1、基于不确定偏好序的多对一匹配
在第一阶段，本发明首先提出一种分层策略，基于不确定偏好序的多对一双边匹配模型。首先无人机的飞行轨迹对于D2D用户来说是未知的，D2D用户根据不确定的信息很难直接生成具体的偏好列表，需要考虑偏好信息为序值的情况。在t n时隙(n＞0)，地面用户对无人机U m在t n-1时隙位置存在预测范围l(U m,t n)，在预测范围l(U m,t n)内，求出D2D对k通过无人机U m协助传输数据的最大速率、最小速率和平均速率，且分别表示为
Figure PCTCN2022104900-appb-000031
和
Figure PCTCN2022104900-appb-000032
并根据这些速率得到K中个体k对U中个体U m的不确定偏好序
Figure PCTCN2022104900-appb-000033
和
Figure PCTCN2022104900-appb-000034
分别为序区间
Figure PCTCN2022104900-appb-000035
的上限和下限，各由
Figure PCTCN2022104900-appb-000036
和
Figure PCTCN2022104900-appb-000037
的排名情况确定。特别 的，若
Figure PCTCN2022104900-appb-000038
则
Figure PCTCN2022104900-appb-000039
退化为一个精确的序值。不确定偏好序序值越小，则说明U m在k的评价排序中位置越靠前，即更加满足k的需求。同理，无人机求出D2D对在移动范围l(k,t n)内的最大速率
Figure PCTCN2022104900-appb-000040
最小速率
Figure PCTCN2022104900-appb-000041
和平均速率
Figure PCTCN2022104900-appb-000042
并根据这些速率得到U中个体U m对K中个体k的不确定偏好序
Figure PCTCN2022104900-appb-000043
Figure PCTCN2022104900-appb-000044
和
Figure PCTCN2022104900-appb-000045
分别为序区间
Figure PCTCN2022104900-appb-000046
的上限和下限，各由
Figure PCTCN2022104900-appb-000047
和
Figure PCTCN2022104900-appb-000048
的排名情况确定。特别的，若
Figure PCTCN2022104900-appb-000049
则
Figure PCTCN2022104900-appb-000050
退化为一个精确的序值。不确定偏好序序值越小，则说明k在U m的评价排序中位置越靠前，U m会优先提供中继传输服务。基于此，D2D对集合K中个体k对中继无人机集合U中个体U m的综合得分与U中个体U m对K中个体k的综合得分可分别用下式表示：
Figure PCTCN2022104900-appb-000051
Figure PCTCN2022104900-appb-000052
其中，Y ki为k将U中个体排在第i位的个体集合，Y mj为U m将K中个体排在第j位的个体集合，|Y ki|和|Y mj|表示集合中个体的数量。Z(U m)、Z(k)分别为U m、k的不确定偏好序区间长度。特别地，如果
Figure PCTCN2022104900-appb-000053
Figure PCTCN2022104900-appb-000054
则不需要减去相应的分数。基于参与者的不确定偏好序得到的综合得分，既考虑了在同一位次匹配个体的竞争程度，同时减去的分数又体现出不同个体的偏好差异，在很大程度上反映出主体的真实意愿。
根据综合得分，可得到D2D对k的偏好策略：对于k偏好列表中的匹配对象U m和，1)当H km＞H km'时，则
Figure PCTCN2022104900-appb-000055
2)当H km＝H km'时，若
Figure PCTCN2022104900-appb-000056
则
Figure PCTCN2022104900-appb-000057
若
Figure PCTCN2022104900-appb-000058
则U m＝ kU m'。同理可得无人机U m的偏好策略：对于在无人机U m的偏好列表中的匹配对象k和k'，1)当H mk＞H mk'时，则
Figure PCTCN2022104900-appb-000059
2)当H mk＝H mk'时，若
Figure PCTCN2022104900-appb-000060
则
Figure PCTCN2022104900-appb-000061
若
Figure PCTCN2022104900-appb-000062
则k＝U m k'。基于偏好列表建立多对一双边匹配模型：给定两个不同的有限参与集合K和U，ω定义为多对一匹配关系，一个匹配是满足以下条件的双映射
Figure PCTCN2022104900-appb-000063
1)
Figure PCTCN2022104900-appb-000064
2)
Figure PCTCN2022104900-appb-000065
3)
Figure PCTCN2022104900-appb-000066
基于以上讨论，本发明所提基于不确定偏好序的多对一匹配算法的思想为：在每个时隙，D2D用户通过预测中继无人机的位置来计算传输性能，得到不确定偏好序，在此基础上综合评估生成相应的偏好列表。D2D用户根据偏好列表，对中继无人机进行排序和选择。中继无人机收到D2D用户的请求后，在满足配额约束要求下，将接受最佳候选者的匹配请求，拒绝其他的D2D用户。被接受的D2D用户停止其匹配过程，而被拒绝的D2D用户继续向次优中继无人机发出匹配请求。直到没有比当前匹配项更好的中继链路，匹配过程终止。不确定偏好序下的多对一匹配算法归纳如下：
步骤1：算法进程到时隙t n。D2D对k∈K和中继无人机U m∈U根据 各自的不确定偏好序综合评估建立偏好列表。
步骤2：如果
Figure PCTCN2022104900-appb-000067
|ω(k)|＝0，且k的偏好列表非空，则分别对k∈K、U m∈U执行以下步骤直至所有D2D对匹配到偏好列表中的中继无人机或者被中继无人机拒绝。
1:所有D2D对向位于其偏好列表第一位的中继无人机
Figure PCTCN2022104900-appb-000068
发送请求，将
Figure PCTCN2022104900-appb-000069
设置为1，然后把
Figure PCTCN2022104900-appb-000070
从k的偏好列表中移除。
2:对于所有中继无人机U m∈U，如果当前D2D对的请求数量大于q 0(U m)，U m根据其偏好列表，接受q 0(U m)个D2D对，拒绝其他的D2D对，且被拒绝的D2D对
Figure PCTCN2022104900-appb-000071
值设置为0；如果当前D2D对的请求数量小于等于q 0(U m)，U m接受当前所有请求者。
基于不确定偏好序的多对一匹配算法中参与匹配的D2D对和无人机的数量分别为K和M，在最坏情况下，任意D2D对的候选中继集合中都包含所有的无人机，所有的中继无人机对D2D用户的偏好都不满足中继无人机的要求，在这种情况下参与匹配的D2D用户需要不断地向其他中继无人机发送请求并被拒绝，因此算法的最坏时间复杂度为O(MK)，在系统整个任务传输周期算法的最坏时间复杂度为O(NMK)。
2.2消除同群效应的中继无人机选择
上述基于不确定偏好序下的多对一匹配算法的匹配结果是不稳定的，原因是存在同群效应，即一个D2D用户可能会根据其他D2D用户形成的偏好列表不断改变自己的偏好顺序，因此永远不会达到最终的中继选择结果。因此，在第二阶段，本发明介绍如何使任意两个 D2D对之间通过交换操作来交换它们已匹配的中继无人机，从而规避同群效应达到稳定的匹配状态。
交换匹配的概念定义为：
Figure PCTCN2022104900-appb-000072
即在保持其他D2D对和无人机相应的匹配不变的同时，D2D对k和k'交换彼此的匹配对象。特别地，在交换的过程中允许参与的一方D2D对位置为空，因此单个D2D对向无人机对应的可用空位移动。
在此基础上，对于D2D对(k,k')，当且仅当同时满足
Figure PCTCN2022104900-appb-000073
和
Figure PCTCN2022104900-appb-000074
条件时，则称(k,k')为交换限制对。
基于上述定义可知，交换匹配是在交换限制对之间进行的。交换后，所有参与交换的D2D对的传输速率不会降低，而至少其中一个D2D对的传输速率会增加，这同时也避免了在等价匹配之间循环。交换操作在效用值的基础上进行，基于预测范围内的速率增加作为交换条件体现出传输过程中的不确定性。在双边匹配中，稳定性通常用来衡量匹配方案的优劣。对于多对一双边匹配ω，如果不存在交换限制对，则称匹配ω是双边交换稳定的。为了消除匹配ω中的交换限制对，本发明对基于不确定偏好序下的多对一匹配算法下的匹配结果ω(t n)的交换限制对迭代完成每个交换操作。因此该方法可以动态得到双边交换稳定匹配结果。本发明所提出的避免同群效应的中继无人机选择方法可归纳如下：
步骤1：初始阶段，初始化D2D对k向k'发送交换请求的次数， 即C kk'＝0。
步骤2：每个D2D对k∈K搜索另一个D2D对k'∈{K\k}以形成交换限制对，如果(k,k')形成交换限制对，并且满足C kk'+C k'k≤2，则根据基于不确定偏好序下的多对一匹配算法更新匹配结果，且C kk'＝C kk'+1；否则保持当前的匹配状态。
步骤3：直到当前匹配中不存在任何交换限制对，则返回更新后的匹配结果ω(t n)。
将C kk'表示为D2D对k向k'发送交换请求的次数，且k最多可与k'交换两次，这避免了乒乓效应，保证了算法收敛。当前匹配中不存在任何交换限制对时，交换匹配过程结束，更新匹配结果ω(t n)。第二阶段提出的方法相较于第一阶段中的算法多了交换匹配的步骤，交换次数上限为
Figure PCTCN2022104900-appb-000075
因此，在系统整个任务传输周期避免同群效应的中继无人机选择方法的的最坏时间复杂度为
Figure PCTCN2022104900-appb-000076
本发明提出的方法最终得到的匹配ω是双边交换稳定的。
证明：假设最终得到的匹配结果
Figure PCTCN2022104900-appb-000077
不是双边交换稳定的，则至少存在一个交换限制对
Figure PCTCN2022104900-appb-000078
但双边交换稳定匹配中继选择算法在存在交换限制对的情况下是不会终止的，因此，该匹配结果
Figure PCTCN2022104900-appb-000079
不是最终结果，与假设条件冲突。因此，本文所提算法得到的匹配是双边交换稳定的。
实施例2：本发明的一个具体实施例如下描述：系统仿真采用Matlab2015a，参数设定不影响本发明的一般性。D2D对随机分布在3km×3km区域内，发射端与对应的接收端之间的距离随机取值于 (100m,200m)以内，且发射端与接收端用户每个时隙随机在(1m,2m)范围内移动。无人机随机选择飞行方向，且不飞出规定区域，飞行高度为[100m,200m]，飞行速度为10m/s，电池容量为2×10 5J，无人机的能耗模型参考多旋翼无人机能耗模型(参考文献：Wang B,Sun Y,Liu D,et al.Social-aware UAV-assisted Mobile Crowd Sensing in Stochastic and Dynamic Environments for Disaster Relief Networks.IEEE Transactions on Vehicular Technology,2020,69(1):1070-1074)，本发明对比算法为穷举搜索算法(可得到最优解但时间复杂度极高)、随机中继选择算法(时间复杂度低但性能不稳定)以及一对一匹配算法(参考文献：Liu D,Yang Y,Xu Y,et al.Uncertain Preference Matching-Based Relay Selection and Position Adjustment in Dynamic UAV Systems//2020 International Conference on Wireless Communications and Signal Processing(WCSP),2020:1170-1175)。
仿真结果分析：图3、图4和图5分别展示了本发明所提方法(基于不确定偏好序下的多对一匹配算法和避免同群效应的中继无人机选择方法)与现有方法随D2D用户传输数据期望值变化、中继无人机数量变化和D2D对数量变化这三种情况下的D2D用户传输成功率比较。从这三张图种可以看出，本文提出的算法得到的D2D用户传输成功率虽然低于穷举算法的传输成功率，得到次优的结果，但是所提算法的计算复杂度较穷举算法大幅度降低。另外，所提方法允许同一中继无人机在同一时隙协作多对D2D用户进行通信，且匹配结果是稳定 的，能保障尽可能多的救援人员及时有效沟通。因此，所提算法性能是优于其他对比方法的，更适用于应急动态分布式场景。

Claims (10)

  1. 一种用于地下空间灾后应急场景的无人机中继选择方法，其特征在于，包括以下步骤：
    基于D2D用户对于无人机的预测范围，采集所述D2D用户通过所述无人机协助传输数据的最大速率、最小速率和平均速率，获取所述D2D用户对于所述无人机的第一不确定偏好序，以及所述无人机对于所述D2D用户的第二不确定偏好序；
    基于所述第一不确定偏好序和所述第二不确定偏好序，获取所述D2D用户对于所述无人机的偏好策略，根据多对一双边匹配算法，构建多对一双边匹配模型；
    基于多对一双边匹配模型，将任意两个所述D2D用户已匹配的所述无人机进行交换匹配，获取双边交换稳定的匹配结果。
  2. 根据权利要求1所述一种用于地下空间灾后应急场景的无人机中继选择方法，其特征在于：
    在获取所述第一不确定偏好序和所述第二不确定偏好序的过程中，通过选择目标时隙，获取所述D2D用户对于所述无人机在所述目标时隙的所述预测范围。
  3. 根据权利要求2所述一种用于地下空间灾后应急场景的无人机中继选择方法，其特征在于：
    在获取所述第一不确定偏好序的过程中，根据所述最大速率、所述最小速率，构建所述第一不确定偏好序，并获取所述第一不确定偏好序的第一偏好表达式，其中，所述第一偏好表达式为：
    Figure PCTCN2022104900-appb-100001
    其中，
    Figure PCTCN2022104900-appb-100002
    为第一不确定偏好序的下限，
    Figure PCTCN2022104900-appb-100003
    为第一不确定偏好序的上限，Y ki为D2D对k将无人机集合U中个体排在第i位的个体集合，|Y ki|集合中个体的数量，Z(U m)为第二不确定偏好序的第一区间长度。
  4. 根据权利要求3所述一种用于地下空间灾后应急场景的无人机中继选择方法，其特征在于：
    在获取所述第二不确定偏好序的过程中，根据所述最大速率、所述最小速率，构建所述第二不确定偏好序，并获取所述第二不确定偏好序的第二偏好表达式，所述偏好表达式为：
    Figure PCTCN2022104900-appb-100004
    其中，
    Figure PCTCN2022104900-appb-100005
    为第二不确定偏好序的下限，
    Figure PCTCN2022104900-appb-100006
    为第二不确定偏好序的上限，Y mj为无人机U m将D2D对集合K中个体排在第j位的个体集合，|Y ki|集合中个体的数量，Z(k)为第一不确定偏好序的第二区间长度。
  5. 根据权利要求4所述一种用于地下空间灾后应急场景的无人机中继选择方法，其特征在于：
    根据所述第一偏好表达式和所述第二偏好表达式，获取所述偏好策略以及所述偏好策略对应的偏好列表，构建所述多对一双边匹配模型，其中，所述多对一双边匹配模型的约束条件为：
    Figure PCTCN2022104900-appb-100007
    其中，
    Figure PCTCN2022104900-appb-100008
    Figure PCTCN2022104900-appb-100009
    Figure PCTCN2022104900-appb-100010
    式中，ω表示匹配。
  6. 根据权利要求5所述一种用于地下空间灾后应急场景的无人机中继选择方法，其特征在于：
    在构建多对一双边匹配模型的过程中，所述多对一双边匹配模型的匹配过程包括以下步骤：
    所述D2D用户通过预测中继无人机的位置来计算传输性能，得到所述第一不确定偏好序，根据所述第一偏好表达式生成对应的偏好列表，所述D2D用户根据所述偏好列表，对所述中继无人机进行排序和选择；
    所述中继无人机收到所述D2D用户的请求后，根据所述约束条件，接受最佳候选者的匹配请求，并拒绝其他的所述D2D用户；
    被接受的所述D2D用户停止其匹配过程，被拒绝的所述D2D用户向次优中继无人机发出匹配请求，直到没有比当前匹配项更好的中继链路，则匹配过程终止。
  7. 根据权利要求6所述一种用于地下空间灾后应急场景的无人机中继选择方法，其特征在于：
    在匹配过程终止的过程后，所述交换匹配的过程包括以下步骤：
    S1.选取所述D2D用户的第一D2D对和第二D2D对，以及所述第一D2D对的第一匹配对象、所述第二D2D对的第二匹配对象；
    S2.将所述第一D2D对与所述第二匹配对象进行匹配，同时将所述第二D2D对与所述第一匹配对象进行匹配；
    S3.根据步骤S2的匹配结果，判断是否执行所述步骤S2的操作，其中，判断过程包括：
    当所述第一D2D对的传输速率增加，所述第二D2D对的传输速率不变时，则保持所述步骤S2的匹配过程；
    当所述第二D2D对的传输速率增加，所述第一D2D对的传输速率不变时，则保持所述步骤S2的匹配过程；
    当所述第一D2D对和所述第二D2D对的传输速率增加时，则保持所述步骤S2的匹配过程；
    否则，将不执行所述步骤S2的匹配过程。
  8. 根据权利要求7所述一种用于地下空间灾后应急场景的无人机中继选择方法，其特征在于：
    所述交换匹配的约束条件为：当所述第一D2D对和所述第二D2D对同时满足以下两个条件时，形成交换限制对，执行所述交换匹配的过程：
    条件一：
    Figure PCTCN2022104900-appb-100011
    条件二：
    Figure PCTCN2022104900-appb-100012
  9. 根据权利要求8所述一种用于地下空间灾后应急场景的无人机中继选择方法，其特征在于：
    所述交换匹配的过程的步骤包括：
    步骤1：初始阶段，初始化D2D对k向D2D对k'发送交换请求的 次数，即C kk'＝0；
    步骤2：每个所述D2D对k∈K搜索另一个D2D对k'∈{K\k}以形成交换限制对，如果(k,k')形成所述交换限制对，并且满足C kk'+C k'k≤2，则根据基于不确定偏好序下的多对一匹配算法更新匹配结果，且C kk'＝C kk'+1；否则保持当前的匹配状态；
    步骤3：直到当前匹配中不存在任何所述交换限制对，则返回更新后的匹配结果。
  10. 根据权利要求1所述一种用于地下空间灾后应急场景的无人机中继选择方法，其特征在于：用于实现所述无人机中继选择方法的无人机中继选择系统包括，
    数据采集处理模块，用于基于所述D2D用户对于所述无人机的所述预测范围，采集所述D2D用户通过所述无人机协助传输数据的最大速率、最小速率和平均速率，获取所述D2D用户对于所述无人机的第一不确定偏好序，以及所述无人机对于所述D2D用户的第二不确定偏好序；
    偏好匹配模块，用于基于所述第一不确定偏好序和所述第二不确定偏好序，获取所述D2D用户对于所述无人机的偏好策略，根据多对一双边匹配算法，构建多对一双边匹配模型；
    交换匹配模块，用于基于所述偏好匹配模块的匹配结果，将任意两个所述D2D用户已匹配的所述无人机进行交换匹配，获取双边交换稳定的匹配结果。
PCT/CN2022/104900 2021-11-18 2022-07-11 一种用于地下空间灾后应急场景的无人机中继选择方法 WO2023087746A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
US18/113,258 US20230224021A1 (en) 2021-11-18 2023-02-23 Relay selection method for unmanned aerial vehicles in post disaster emergency scenarios in underground space

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202111371557.2A CN114071432B (zh) 2021-11-18 2021-11-18 一种用于地下空间灾后应急场景的无人机中继选择方法
CN202111371557.2 2021-11-18

Related Child Applications (1)

Application Number Title Priority Date Filing Date
US18/113,258 Continuation US20230224021A1 (en) 2021-11-18 2023-02-23 Relay selection method for unmanned aerial vehicles in post disaster emergency scenarios in underground space

Publications (1)

Publication Number Publication Date
WO2023087746A1 true WO2023087746A1 (zh) 2023-05-25

Family

ID=80278119

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2022/104900 WO2023087746A1 (zh) 2021-11-18 2022-07-11 一种用于地下空间灾后应急场景的无人机中继选择方法

Country Status (3)

Country Link
US (1) US20230224021A1 (zh)
CN (1) CN114071432B (zh)
WO (1) WO2023087746A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN118075728A (zh) * 2024-04-24 2024-05-24 北京语言大学 一种面向应急通信场景的无人机响应决策方法及装置

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114071432B (zh) * 2021-11-18 2024-04-26 中国矿业大学 一种用于地下空间灾后应急场景的无人机中继选择方法

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110417847A (zh) * 2019-01-09 2019-11-05 北京邮电大学 无人机通信网络用户接入和内容缓存的方法及装置
CN110736478A (zh) * 2018-07-20 2020-01-31 华北电力大学 一种无人机辅助移动云感知的路径规划与任务分配方案
CN111615200A (zh) * 2020-04-10 2020-09-01 洛阳理工学院 混合Hybrid NOMA网络的无人机辅助通信资源分配方法
CN112737837A (zh) * 2020-12-28 2021-04-30 北京邮电大学 一种高动态网络拓扑下无人机群带宽资源分配方法
CN114071432A (zh) * 2021-11-18 2022-02-18 中国矿业大学 一种用于地下空间灾后应急场景的无人机中继选择方法

Family Cites Families (16)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR100297289B1 (ko) * 1998-08-24 2001-08-07 박태진 교환시스템의통합가입자정합장치및그제어방법
US7864752B1 (en) * 2006-08-09 2011-01-04 Nortel Networks Limited Bearer path resource matching in a wireless communication network
WO2010113261A1 (ja) * 2009-03-31 2010-10-07 富士通株式会社 無線通信ネットワークにおける中継局、基地局、移動局及び中継方法
US10225844B2 (en) * 2013-07-05 2019-03-05 Lg Electronics Inc. Method for selecting or reselecting relay for proximity service
US10873895B2 (en) * 2014-12-19 2020-12-22 Samsung Electronics Co., Ltd. Apparatus and method for providing relay selection in device-to-device communication system
US20180092017A1 (en) * 2015-04-08 2018-03-29 Interdigital Patent Holdings, Inc. Realizing Mobile Relays For Device-to-Device (D2D) Communications
CN105792370B (zh) * 2016-03-02 2019-03-12 南京邮电大学 一种双向多中继ofdm系统中的资源联合分配方法
US9713180B1 (en) * 2016-03-15 2017-07-18 Qualcomm Incorporated Relay centric mobility management in a mesh network
CN106658606B (zh) * 2016-11-01 2020-09-08 洛阳理工学院 一种基于QoE的分布式分层异构网络用户基站匹配方法
CN107302801B (zh) * 2017-05-19 2020-11-06 南京邮电大学 一种5G混合场景下面向QoE的双层匹配博弈方法
CN107809785B (zh) * 2017-11-07 2019-06-25 中国人民解放军陆军工程大学 一种基于互利匹配博弈的分布式中继选择方法
CN108900265B (zh) * 2018-06-15 2021-02-05 南京航空航天大学 基于匹配博弈的全双工认知无线电网络信道分配方法
US20220248239A1 (en) * 2019-05-21 2022-08-04 Trustees Of Tufts College Network User Association
CN111866901B (zh) * 2020-06-10 2022-02-08 西北大学 中继选择和资源信息优化方法、系统、计算机设备及应用
CN112616189B (zh) * 2020-12-10 2022-09-09 北京邮电大学 一种静态和动态相结合的毫米波波束资源分配与优化方法
CN112838887B (zh) * 2021-01-06 2022-02-08 中国矿业大学 一种灾后应急通信地下飞行自组网拓扑控制方法

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110736478A (zh) * 2018-07-20 2020-01-31 华北电力大学 一种无人机辅助移动云感知的路径规划与任务分配方案
CN110417847A (zh) * 2019-01-09 2019-11-05 北京邮电大学 无人机通信网络用户接入和内容缓存的方法及装置
CN111615200A (zh) * 2020-04-10 2020-09-01 洛阳理工学院 混合Hybrid NOMA网络的无人机辅助通信资源分配方法
CN112737837A (zh) * 2020-12-28 2021-04-30 北京邮电大学 一种高动态网络拓扑下无人机群带宽资源分配方法
CN114071432A (zh) * 2021-11-18 2022-02-18 中国矿业大学 一种用于地下空间灾后应急场景的无人机中继选择方法

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN118075728A (zh) * 2024-04-24 2024-05-24 北京语言大学 一种面向应急通信场景的无人机响应决策方法及装置

Also Published As

Publication number Publication date
US20230224021A1 (en) 2023-07-13
CN114071432B (zh) 2024-04-26
CN114071432A (zh) 2022-02-18

Similar Documents

Publication Publication Date Title
WO2023087746A1 (zh) 一种用于地下空间灾后应急场景的无人机中继选择方法
Chen et al. A multihop task offloading decision model in MEC-enabled internet of vehicles
Luo et al. Optimization of task scheduling and dynamic service strategy for multi-UAV-enabled mobile-edge computing system
Alioua et al. UAVs for traffic monitoring: A sequential game-based computation offloading/sharing approach
Ateya et al. Energy-and latency-aware hybrid offloading algorithm for UAVs
Cheng et al. AI for UAV-assisted IoT applications: A comprehensive review
Hawbani et al. Fuzzy-based distributed protocol for vehicle-to-vehicle communication
Dao et al. Adaptive resource balancing for serviceability maximization in fog radio access networks
Deng et al. Actions at the edge: Jointly optimizing the resources in multi-access edge computing
Liu et al. Opportunistic data collection in cognitive wireless sensor networks: Air–ground collaborative online planning
CN114827198B (zh) 一种应用于车联网的多层中心异步联邦学习方法
KR102508504B1 (ko) 기계 학습 기반의 무인항공기 모바일 엣지 서버 간 협업 태스크 매칭 및 오프로딩 방법
Tang et al. Integration of UAV and fog-enabled vehicle: application in post-disaster relief
Wu et al. Distributed federated deep reinforcement learning based trajectory optimization for air-ground cooperative emergency networks
CN111988792A (zh) 无人机联盟网络卸载模型及决策计算方法
Wang et al. Edge intelligence for mission cognitive wireless emergency networks
Zheng et al. A multi-channel load awareness-based MAC protocol for flying ad hoc networks
Hussain et al. Co-DLSA: Cooperative delay and link stability aware with relay strategy routing protocol for flying Ad-hoc network
Chen et al. Delay optimization with FCFS queuing model in mobile edge computing-assisted UAV swarms: A game-theoretic learning approach
de Souza et al. A task offloading scheme for wave vehicular clouds and 5g mobile edge computing
Liu et al. Opportunistic data ferrying in UAV-assisted D2D networks: A dynamic hierarchical game
Gharehchopogh et al. A survey and taxonomy of leader election algorithms in distributed systems
Tshilongamulenzhe et al. Traffic-based Congestion Management algorithm for Wireless Sensor networks
Wang et al. Learning based edge computing in air-to-air communication network
CN116133082A (zh) 一种提高航空自组网拓扑持续时间的多跳分簇方法

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 22894289

Country of ref document: EP

Kind code of ref document: A1