A D2D Low-Latency User Pairing Method Based on Joint Attributes

Abstract

The present invention relates to the field of communication technologies, and discloses a D2D low-latency user pairing method based on joint attributes. The method includes: establishing a one-to-one bidirectional matching model based on D2D connections, and obtaining a satisfaction function of the supplier user and a satisfaction function of the demander user according to the joint attributes; determining the pairing latency of the demander user according to the number of requests for the demander user to establish a D2D connection once and the unit request time; determining the pairing efficiency of the demander user and the supplier user in the D2D connection according to the satisfaction function of the supplier user, the satisfaction function of the demander user, and the pairing latency of the demander user; obtaining a matching algorithm with low pairing latency, and performing D2D connection user pairing through the matching algorithm with low pairing latency. The present invention sets up a bidirectional consideration joint attribute evaluation framework according to the demand differences between the supply and demand sides, proposes a matching strategy with low pairing latency, and obtains a higher matching satisfaction gain with a low time cost.

Description

Technical Field

[0001] The present invention relates to the field of communication technologies, specifically to D2D communication technologies, and particularly to a D2D low-latency user pairing method based on joint attributes. Background Art

[0002] In recent years, the rapid development of wireless networks has brought an explosive growth in mobile communication data volume. To relieve the burden on cellular networks and increase network capacity, Device-to-Device (D2D) communication technology has been proposed and has become a key technology for the latest generation (5G) mobile communication systems. The first problem to be solved in D2D communication is the user pairing problem. During the pairing process, extending the detection time can, with a certain probability, enable a device to find a more suitable pairing user, but long-term detection inevitably leads to a large consumption of the service life and energy of mobile devices. Therefore, how to achieve low-detection latency and high-performance user pairing has become a problem widely studied by scholars.

[0003] Although most current studies take into account social networks, enriching the influencing factors of system performance, the common problem still exists that the influencing factors of user willingness are not considered comprehensively enough. Some studies idealize and default that users receiving requests will definitely agree to establish D2D communication, without separately considering the demand differences between the supply and demand sides. In the setting of some optimization goals, besides common metrics such as energy efficiency and throughput, the satisfaction of users during the pairing process and the interruption probability of device connections should also be considered. Summary of the Invention

[0004] The purpose of the present invention is to overcome at least one of the above-mentioned prior art problems, and provides a D2D low-latency user pairing method based on joint attributes. A two-way consideration joint attribute evaluation framework is set according to the demand differences between the supply and demand sides, a matching strategy with low pairing latency is proposed, the problem of delayed pairing of the GS algorithm is improved, and a higher matching satisfaction gain is obtained with low time cost.

[0005] The present invention provides a D2D low-latency user pairing method based on joint attributes, including:

[0006] Establish a one-to-one two-way matching model based on D2D connections, and obtain a satisfaction function for supply-side users and a satisfaction function for demand-side users according to joint attributes;

[0007] Determine the pairing latency of demand-side users according to the number of requests for a demand-side user to establish a D2D connection once and the unit request time, and the pairing latency of the demand-side users satisfies wherein, is the pairing latency of demand-side users, The number of requests to establish a D2D connection for the demand-side user, and Δt is the unit request time, which is the time for the demand-side user to send a request each time.

[0008] Determine the pairing efficiency of the demand-side user and the supply-side user in the D2D connection according to the supply-side user satisfaction function, the demand-side user satisfaction function, and the pairing delay of the demand-side user. The pairing efficiency is:

[0009]

[0010] Among them, Pvalue i，j is the supply-side user satisfaction function, and Dvalue i，j is the demand-side user satisfaction function;

[0011] Based on the supply-side user satisfaction function and the demand-side user satisfaction function, combined with the optimal stopping theory and the GS algorithm, obtain a matching algorithm with low pairing delay, and perform D2D connection user pairing through the matching algorithm with low pairing delay.

[0012] Furthermore, the one-to-one two-way matching model based on D2D connection is:

[0013]

[0014] s.t.C1:

[0015] C2:

[0016] Among them, represents the binary pairing user selection strategy set, and C1 and C2 represent channel reuse constraints, which are used to ensure that a demand-side user d i is paired with at most one supply-side user p j in a pairing.

[0017] Furthermore, the obtaining of the supply-side user satisfaction function and the demand-side user satisfaction function according to the joint attributes specifically includes:

[0018] Obtain the remaining battery power E of the supply-side user i，j , the size Fs of the transmission file required by the demand-side user i，j and the physical distance D between the supply-side user and the demand-side user i，j , the transmission rate R i，j and the social relationship So i，j ;

[0019] Obtain the weight of the remaining battery power of the supply-side user The weight of the size of the transmission file required by the demand-side user The weight of the physical distance between the supply-side user and the demand-side user Transmission rate weight First social relationship weight and second social relationship weight

[0020] A supplier user satisfaction function is established according to the remaining power of the device, the size of the transmitted file, the social relationship and their respective weights. The supplier user satisfaction function is

[0021]

[0022] The remaining power weight of the device The transmitted file size weight and the first social relationship weight The sum is 1;

[0023] A demander user satisfaction function is established according to the physical distance, the transmission rate, the social relationship and their respective weights. The demander user satisfaction function is

[0024]

[0025] wherein, The physical distance weight The transmission rate weight and the second social relationship weight The sum is 1.

[0026] Furthermore, the remaining power E of the supplier user's device i，j and the remaining power weight of the supplier user's device The acquisition steps include:

[0027] The remaining power of the supplier user's device is set in a stepped manner and normalized to obtain the remaining power E of the supplier user's device i，j =[0, 0.25, 0.5, 0.75, 1];

[0028] The remaining power weight is obtained according to the remaining power of the device. The remaining power weight satisfies

[0029] wherein, is the remaining power weight.

[0030] Furthermore, the size Fs of the transmitted file required by the demander user i，j and the size weight of the transmitted file required by the demander user The acquisition steps include:

[0031] Obtain the size of the transmission file required by the demand-side user according to the Gaussian random distribution, where the size of the transmission file is Fs i，j ∈[0, 1];

[0032] Obtain the weight value of the transmission file size according to the size of the transmission file, and the weight value of the transmission file size satisfies

[0033]

[0034] where is the weight value of the transmission file size.

[0035] Furthermore, the physical distance D between the supply-side user and the demand-side user i，j and the weight value of the physical distance between the supply-side user and the demand-side user The obtaining steps include:

[0036] Use the Euclidean distance to determine the physical distance between the supply-side user and the demand-side user, and the physical distance is

[0037]

[0038] Obtain the weight value of the physical distance according to the physical distance, and the weight value of the physical distance satisfies

[0039]

[0040] where is the weight value of the physical distance.

[0041] Furthermore, the transmission rate R between the supply-side user and the demand-side user i，j and the weight value of the transmission rate between the supply-side user and the demand-side user The obtaining steps include:

[0042] Use the fading channel model to determine the transmission rate between the supply-side user and the demand-side user, and the transmission rate R i，j is

[0043]

[0044] where P b and P d are the transmission powers of the base station and a group of D2D communication pairs, D j，i is the transmission distance between a group of D2D communication pairs, D B，j，i is the transmission distance between the base station and a group of D2D communication pairs, α is the path loss exponent corresponding to the large-scale fading of the transmission channel, h0 is a Rayleigh channel coefficient obeying the complex Gaussian distribution and N0 is the variance of the additive white Gaussian noise (AWGN);

[0045] Obtain a transmission rate weight value according to the said transmission rate, and the said transmission rate weight value satisfies

[0046]

[0047] wherein, is the transmission rate weight value.

[0048] Furthermore, the social relationship So between the said supplier user and the demander user i，j and the first social relationship weight value between the said supplier user and the demander user The second social relationship weight value The obtaining steps include:

[0049] Obtain the social relationship between the supplier user and the demander user according to the Pareto distribution, and the said social relationship So i，j ∈[0, 1];

[0050] Obtain the first social relationship weight value and the second social relationship weight value according to the said social relationship, and the said first social relationship weight value satisfies wherein, is the first social relationship weight value, and the said second social relationship weight value is equal to the said first social relationship weight value

[0051] Further, based on the said supplier user satisfaction function and the demander user satisfaction function, combine the optimal stopping theory and the GS algorithm to obtain a matching algorithm with low pairing delay, and perform D2D connection user pairing through the said matching algorithm with low pairing delay, specifically including:

[0052] Step 1: Create a first sequence and a second sequence for each demander user. The first sequence is used to represent the number of candidate pairing objects of each demander user, and the second sequence is used to store the candidate pairing objects of each demander user, and the second sequence is sorted from high to low according to the demander user satisfaction function;

[0053] Step 2: Create a third sequence for each supplier user. The third sequence is used to store the demander users who send pairing requests to each supplier user, and the third sequence is sorted from high to low according to the supplier user satisfaction function;

[0054] Step 3: Based on the GS algorithm, create a first matrix and a second matrix. The first matrix is used to represent the pairing status of the supply and demand sides, and the pairing status includes a paired status and an unpaired status. The second matrix is used to store the supplier users in the fully paired status. The supplier users in the fully paired status are the supplier users who have been paired with the highest-ranked demand-side user in the third sequence and will no longer accept other pairing requests.

[0055] Step 4: Based on the optimal stopping theory, determine whether the cumulative number of pairing requests sent by each demand-side user is greater than the product of 0.37 and the first sequence. If so, stop sending; otherwise, approve the sending.

[0056] Step 5: Based on the second matrix, determine the pairing status of all supplier users in the second sequence of each demand-side user. Based on the judgment result, obtain a supplier user with the highest ranking in the non-fully paired status of the second sequence of each demand-side user and send a pairing request.

[0057] Step 6: Determine that the first supplier user accepts the request of the highest-ranked demand-side user in the third sequence of the first supplier user, and add the first supplier user to the second matrix.

[0058] Step 7: Determine that the second supplier user changes the paired object of the second supplier user. The pairing status of the paired object of the second supplier user changes from the paired status to the unpaired status, and update the first matrix.

[0059] Step 8: Determine whether the pairing status of each demand-side user is the unpaired status and the request is still approved to be sent. If so, return to Step 4.

[0060] Compared with the prior art, the present invention has at least one of the following technical effects:

[0061] 1. The evaluation framework of the combined attributes has certain advantages over the traditional evaluation framework in terms of throughput and outage probability. The matching algorithm with low pairing delay is significantly superior to the GS algorithm in terms of pairing delay, while maintaining substantially the same throughput and global utility performance as the latter.

[0062] 2. Considering the needs of D2D providers and demanders from multiple aspects such as social relationships, physical distances, device status, and communication performance, stable D2D user pairs are formed, obtaining higher user satisfaction.

[0063] 3. Reasonably set different attributes according to the respective needs of the supply and demand sides and establish an evaluation framework with two-way considerations. An evaluation framework of two-way combined attributes and related performance indicators are proposed. The satisfaction of the supply and demand sides is obtained using the combined attributes, and the pairing efficiency is obtained by combining with the detection time as a performance indicator for target optimization.

[0064] 4. The algorithm using the optimal stopping theory enables users to complete the optimal pairing with a high probability without going through a long global detection, reducing the detection time and the system latency.

[0065] 5. The GS algorithm is improved and combined with the improved GS algorithm, enabling some users to complete the optimal selection and pairing in advance without waiting until the global user detection stops, reducing the detection time of some users, thereby reducing the service latency and improving the communication quality. Brief Description of the Drawings

[0066] To more clearly illustrate the technical solutions in the embodiments of the present application, the following will briefly introduce the drawings required in the embodiments. Obviously, the drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0067] Figure 1 It is a schematic diagram of the scenario of a D2D low-latency user pairing method based on joint attributes provided by an embodiment of the present invention;

[0068] Figure 2 It is a schematic diagram of the simulation parameter settings of each evaluation framework;

[0069] Figure 3 It is a schematic diagram of the D2D establishment ratio when there are different supplier users in each evaluation framework;

[0070] Figure 4 It is a schematic diagram of the transmission interruption probability when there are different supplier users in each evaluation framework;

[0071] Figure 5 It is a schematic diagram of the system throughput when there are different supplier users in each evaluation framework;

[0072] Figure 6 It is a schematic diagram of the number of pairing detection requests when there are different supplier users in each evaluation framework;

[0073] Figure 7 It is a schematic diagram of the comparison of connection establishment between the low pairing latency matching algorithm and the GS algorithm;

[0074] Figure 8 It is a schematic diagram of the CDF graph of the time when the demander users of each algorithm successfully establish a pairing relationship;

[0075] Figure 9 It is a schematic diagram of the number of supplier users and the total D2D pairing efficiency of each algorithm;

[0076] Figure 10 It is a schematic diagram of the number of supplier users and the total user satisfaction of each algorithm;

[0077] Figure 11 It is a schematic diagram of the number of supplier users and the number of requests for each algorithm;

[0078] Figure 12 It is a schematic diagram of the number of supplier users and the system throughput of each algorithm;

[0079] Figure 13 It is a schematic diagram of the number of supplier users and the D2D establishment ratio of each algorithm. Detailed implementation manners

[0080] In the following description, for the purpose of illustration rather than limitation, specific details such as specific system architectures and technologies are presented to provide a thorough understanding of the embodiments of the present application. However, those skilled in the art should clearly understand that the present application can also be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to avoid unnecessary details from interfering with the description of the present application.

[0081] It should be understood that when used in the specification of the present application and the appended claims, the term "comprising" indicates the presence of the described features, wholes, steps, operations, elements, and / or components, but does not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components, and / or their combinations.

[0082] It should also be understood that the term "and / or" used in the specification of the present application and the appended claims refers to any combination and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0083] As used in the specification of the present application and the appended claims, the term "if" can be interpreted as "when", "once", "in response to determining", or "in response to detecting" according to the context. Similarly, the phrase "if determined" or "if [the described condition or event] is detected" can be interpreted as meaning "once determined", "in response to determining", "once [the described condition or event] is detected", or "in response to detecting [the described condition or event]" according to the context.

[0084] In addition, in the description of the specification of the present application and the appended claims, the terms "first", "second", "third", etc. are only used for distinguishing descriptions and cannot be understood as indicating or implying relative importance.

[0085] References to "one embodiment" or "some embodiments" etc. described in the specification of this application mean that specific features, structures, or characteristics described in connection with that embodiment are included in one or more embodiments of this application. Thus, statements such as "in one embodiment", "in some embodiments", "in other some embodiments", "in still other embodiments", etc. that appear in different places in this specification do not necessarily all refer to the same embodiment, but mean "one or more but not all embodiments", unless otherwise specifically emphasized. The terms "comprising", "including", "having" and their variants all mean "including but not limited to", unless otherwise specifically emphasized.

[0086] Referring to Figure 1 , Figure 1 FIG. is a schematic diagram of a scenario of a D2D low-latency user pairing method based on joint attributes provided by an embodiment of the present invention. It includes a base station and several D2D user devices. A supplier user device and a demander user device form a group of D2D communication pairs. The supplier user device and the demander user device are connected through D2D. At the same time, each user device can also be connected to the base station through a cellular network, and the D2D communication and the cellular network do not affect each other. The user device is not limited to any device with wireless communication functions, including smart phones, computers, etc.; the base station is not limited to any form of wireless base station, including macro base stations, micro base stations, pico base stations, femto base stations, etc.

[0087] An embodiment of the present invention provides a D2D low-latency user pairing method based on joint attributes, including:

[0088] Establish a one-to-one bidirectional matching model based on D2D connections, and obtain a supplier user satisfaction function and a demander user satisfaction function according to the joint attributes;

[0089] Determine the demander user pairing delay according to the number of requests for the demander user to establish a D2D connection once and the unit request time. The demander user pairing delay satisfies Wherein, is the demander user pairing delay, is the number of requests for the demander user to establish a D2D connection once, and Δt is the unit request time, and the unit request time is the time for the demander user to send a request each time.

[0090] Determine the pairing efficiency of the demander user and the supplier user in the D2D connection according to the supplier user satisfaction function, the demander user satisfaction function, and the demander user pairing delay. The pairing efficiency is:

[0091]

[0092] Wherein, Pbaluei，j is the satisfaction function of the supplier user, Dvalue i,j is the satisfaction function of the demander user;

[0093] Based on the satisfaction function of the supplier user and the satisfaction function of the demander user, combined with the optimal stopping theory and the GS algorithm, a matching algorithm with low pairing delay is obtained, and D2D connection user pairing is performed through the matching algorithm with low pairing delay.

[0094] In this embodiment, during the process of establishing a D2D connection, the demander user sends a connection establishment request to the supplier user with high satisfaction and waits for a reply. If the request is rejected by the supplier user, the demander user will send a request to the next supplier user. For the demander user, the fewer the number of requests sent, the earlier the establishment of the relationship is determined, the shorter the time for establishing the D2D connection, and the better the user experience. Therefore, this waiting time is defined as the pairing delay of the demander user: And it is assumed that the time for each request is constant, called the unit request time Δt, and the number of requests for the demander user to establish a D2D connection once is The pairing delay of the demander user is defined as the product of the number of requests and the unit request time:

[0095] Considering the satisfaction function of the supplier and the satisfaction function of the demander with joint attributes provides guidance for D2D users and can better solve the problem of poor communication quality in establishing D2D connections, such as unreliable physical transmission, malicious behavior of supplier users, etc. At the same time, it also considers shortening the time for the demander user to establish a D2D connection and improving the user experience. Therefore, based on the two-way joint attribute evaluation framework and the pairing delay, this efficiency is quantified and further regarded as a performance indicator, that is, the formula for the pairing efficiency in this embodiment.

[0096] It can be seen from the formula of the pairing efficiency that if Pvalue i，j +Dvalue i，j is larger, the individual satisfaction of the demander user and the corresponding supplier user is higher, and the D2D connection communication is stronger. If is smaller, the pairing delay of the demander user is shorter, the time cost for establishing the D2D connection is less, and the user experience is better. Therefore, the larger it is, the less time cost the users participating in D2D pair communication spend and the higher quality communication they obtain, that is, the higher the D2D pair matching efficiency.

[0097] In some embodiments, the one-to-one two-way matching model based on D2D connection is:

[0098]

[0099] s.t.C1:

[0100] C2:

[0101] wherein, represents the binary pairing user selection strategy set, and C1 and C2 represent channel multiplexing constraints for ensuring that a demand user d i is paired with at most one supply user p j in a pairing.

[0102] In this embodiment, the pairing selection matrix of D2D pairs is expressed as where the (i, j) element x i，j ∈ {0, 1} is a binary variable representing the pairing relationship of supply and demand users of the D2D pair composed of and (d i , p j ). If x i，j = 1, d i is paired with p j , and if x i，j = 0, d i is not paired with p j .

[0103] In some embodiments, obtaining the satisfaction function of the supply user and the satisfaction function of the demand user according to the joint attributes specifically includes:

[0104] Obtaining the remaining battery power E of the supply user's device i,j , the size Fs of the transmission file required by the demand user i，j , the physical distance D i，j , the transmission rate R i，j and the social relationship So i，j between the supply user and the demand user;

[0105] Obtaining the weight of the remaining battery power of the supply user's device , the weight of the size of the transmission file required by the demand user , the weight of the physical distance between the supply user and the demand user , the weight of the transmission rate , the first social relationship weight and the second social relationship weight

[0106] Establishing a satisfaction function of the supply user according to the remaining battery power, the size of the transmission file, the social relationship and their respective weights, and the satisfaction function of the supply user is

[0107]

[0108] The weight value of the remaining power of the device The weight value of the size of the transmitted file and the weight value of the first social relationship sum to 1;

[0109] According to the physical distance, the transmission rate, the social relationship and their respective weight values, a satisfaction function of the demand-side user is established. The satisfaction function of the demand-side user is

[0110]

[0111] wherein, The weight value of the physical distance The weight value of the transmission rate and the weight value of the second social relationship sum to 1.

[0112] In this embodiment, the normalized combined attribute values are linearly weighted to obtain the satisfaction function. In the establishment of D2D communication, most previous studies start from the physical domain perspective, aiming to pursue high system throughput or strong stability. However, from the social domain dimension, in reality, not all users can obtain the required files through their relatives and friends, and there is also a high probability of receiving requests from strange users to establish D2D connections. The idealized research from a single dimension obviously cannot adapt to the real scenario, and most studies will adopt fixed weight values in the process of linear weighting. This embodiment establishes a dynamic weight value, which represents the degree of importance of the user to this attribute, and establishes a corresponding functional relationship between the weight values of different attributes and the corresponding attributes according to the actual situation. In the satisfaction function of the supply-side user or the demand-side user, the weight values of different evaluation attributes are between [0, 1]. The satisfaction function of the supply-side user or the demand-side user is the sum of each weight value divided by the sum of the weight values, and it is ensured that the sum of the three dynamic weight values of the satisfaction function of the supply-side user or the demand-side user is 1.

[0113] In some embodiments, the remaining power E of the device of the supply-side user i，j and the weight value of the remaining power of the device of the supply-side user The obtaining steps include:

[0114] Perform stepped setting and normalization processing on the remaining power of the device of the supply-side user to obtain the remaining power E of the device of the supply-side user i，j = [0, 0.25, 0.5, 0.75, 1];

[0115] Obtain the weight value of the remaining power of the device according to the remaining power of the device. The weight value of the remaining power of the device satisfies

[0116] wherein, is the weight value of the remaining power of the device.

[0117] In this embodiment, the remaining power of the device conforms to a rectangular distribution among idle users. For the supplier users, the intuitive feeling of the power within a certain range does not vary much. For example, when the mobile phone power is 10% or 15%, the willingness of idle users to establish D2D communication is very low. Therefore, the device power can be set in a stepped manner and normalized, that is, the power from 0 to 100 is attributed to 0 to 1 and divided into five intervals. Since the remaining power of the device of the supplier user belongs to the subjective conditions of the supplier user itself, the more the remaining power of the device of the supplier user, the lower the influence of this item of the remaining power of the device on the willingness of the supplier user to establish a D2D connection, and the smaller the weight value should be.

[0118] In some embodiments, the size Fs of the transmission file required by the demander user i，j and the weight value of the size of the transmission file required by the demander user The acquisition steps include:

[0119] According to the Gaussian random distribution, obtain the size of the transmission file required by the demander user, and the size of the transmission file Fs i，j ∈[0, 1];

[0120] Obtain the weight value of the transmission file size according to the size of the transmission file, and the weight value of the transmission file size satisfies

[0121]

[0122] where is the weight value of the transmission file size.

[0123] In this embodiment, for idle supplier users, when there are multiple demanders of different files initiating connection requests, the consideration of the size of the transmission file becomes particularly important. Simply considering the transmission rate cannot reflect the true demands of idle supplier users. For supplier users, the larger the file required by the other party, the lower their willingness to participate in D2D communication; on the contrary, the smaller the transmission file, the higher the willingness of the supplier user to participate in D2D communication; that is to say, when the file transmission time tends to two extremes, the supplier user attaches more importance to this attribute of the transmission time, and the weight value is also larger; when the size of the transmission file is around the average level, the supplier user's consideration of the size of the transmission file is lower.

[0124] In some embodiments, the physical distance Di,j between the supplier user and the demander user and the weight value of the physical distance between the supplier user and the demander user The acquisition steps include:

[0125] Use the Euclidean distance to determine the physical distance between the supplier user and the demander user, and the physical distance is

[0126]

[0127] Obtain a physical distance weight value according to the physical distance, and the physical distance weight value satisfies

[0128]

[0129] wherein, is the physical distance weight value.

[0130] In this embodiment, when the D2D user pair is at a relatively long distance, it will affect the communication quality or even make communication impossible, and its signal-to-noise ratio is inversely proportional to the transmission distance. Therefore, the demanding user is more inclined to select a node with a close geographical location for communication. Place the positions of each node into a coordinate system, and use the Euclidean distance as the physical distance between the supplying user and the demanding user.

[0131] In some embodiments, the transmission rate R between the supplying user and the demanding user i，j and the transmission rate weight value between the supplying user and the demanding user The obtaining steps include:

[0132] Use the fading channel model to determine the transmission rate between the supplying user and the demanding user, and the transmission rate R i，j is

[0133]

[0134] wherein, P b and P d are the transmission powers of the base station and a group of D2D communication pairs, D j,i is the transmission distance between a group of D2D communication pairs, D B,j,i is the transmission distance between the base station and a group of D2D communication pairs, α is the path loss exponent corresponding to the large-scale fading of the transmission channel, h0 is a Rayleigh channel coefficient subject to a complex Gaussian distribution and N0 is the variance of the additive white Gaussian noise (AWGN);

[0135] Obtain a transmission rate weight value according to the transmission rate, and the transmission rate weight value satisfies

[0136]

[0137] wherein, is the transmission rate weight value.

[0138] In this embodiment, the larger the transmission rate of the communication established with the supplying user, the more in line with the evaluation criteria of the demanding user. For the demanding user, too large or too small communication transmission rate will strongly affect its selection willingness.

[0139] In some embodiments, the social relationship So between the supplier user and the demander user i，j and the first social relationship weight between the supplier user and the demander user The second social relationship weight The obtaining steps include:

[0140] According to the Pareto distribution, obtain the social relationship between the supplier user and the demander user, and the social relationship So i，j ∈[0, 1]; obtain the first social relationship weight and the second social relationship weight according to the social relationship, and the first social relationship weight satisfies Wherein is the first social relationship weight, and the second social relationship weight is equal to the first social relationship weight

[0141] In this embodiment, the social relationships between social relationship users conform to the Pareto distribution, that is, users with strong social relationships with D2D users (such as relatives, friends, etc.) only exist in a small part around them, and more are strangers with weak social relationships. Both the social relationship and the size of the transmitted file belong to objective conditions, and the influence relationship of their attribute values on the user's willingness to establish a connection is the same. However, due to human selfishness, the influence of social relationships, compared with evaluation attributes such as file size, increases or decreases with the increase or decrease of social relationship intimacy towards both poles, and the communication willingness increases slowly. And since the social relationship is mutual for the demander user and the supplier user, this weight formula in the evaluation criteria of the demander user is also the same as the weight formula of the supplier user.

[0142] In some embodiments, based on the supplier user satisfaction function and the demander user satisfaction function, combined with the optimal stopping theory and the GS algorithm, obtain a matching algorithm with low pairing delay, and perform D2D connection user pairing through the matching algorithm with low pairing delay, specifically including:

[0143] Step 1: Create a first sequence and a second sequence for each demander user. The first sequence is used to represent the number of candidate pairing objects of each demander user, and the second sequence is used to store the candidate pairing objects of each demander user, and the second sequence is sorted from high to low according to the demander user satisfaction function;

[0144] Step 2: Create a third sequence for each supplier user. The third sequence is used to store the demander users who send pairing requests to each supplier user, and the third sequence is sorted from high to low according to the supplier user satisfaction function;

[0145] Step 3: Based on the GS algorithm, create a first matrix and a second matrix. The first matrix is used to represent the pairing status between the supply and demand sides, where the pairing status includes a paired status and an unpaired status. The second matrix is used to store the supplier users in a fully paired status. The supplier users in a fully paired status are those who have been paired with the highest-ranked demand user in the third sequence and will no longer accept other pairing requests.

[0146] Step 4: Based on the optimal stopping theory, determine whether the cumulative number of pairing requests sent by each demand user is greater than the product of 0.37 and the first sequence. If so, stop sending; otherwise, approve the sending.

[0147] Step 5: Based on the second matrix, determine the pairing status of all supplier users in the second sequence of each demand user. Based on the judgment result, obtain a supplier user with the highest rank among the non-fully paired status in the second sequence of each demand user and send a pairing request.

[0148] Step 6: Determine that the first supplier user accepts the request from the highest-ranked demand user in the third sequence of the first supplier user, and add the first supplier user to the second matrix.

[0149] Step 7: Determine that the second supplier user changes the paired object of the second supplier user. The pairing status of the paired object of the second supplier user changes from the paired status to the unpaired status, and update the first matrix.

[0150] Step 8: Determine whether the pairing status of each demand user is the unpaired status and the request is still approved for sending. If so, return to Step 4.

[0151] In this embodiment, the optimal stopping theory is used to optimize the matching technology to achieve the most suitable supplier user with the highest probability within a short detection time. During pairing, the demand user sends requests to the supplier users that meet the D2D communication conditions one by one. Assume the number of supplier users that meet the D2D communication conditions is N. According to the optimal stopping theory, after the demand user sends 0.37N (rounded up) requests, it stops sending requests. At this time, the probability of pairing with the most suitable supplier user is the highest, that is, the probability that the utility value of the D2D communication pair is the largest is the highest. Using the algorithm of the optimal stopping theory, users can complete the optimal pairing with a high probability without going through a long global detection, reducing the detection time and the system latency.

[0152] The GS algorithm is a two-sided matching algorithm for stable matching, which achieves the global optimum of weak Pareto balance through deferred matching. However, in the process of pursuing the optimum, the deferred matching of this matching algorithm will inevitably bring a slow D2D connection establishment service experience to the demand-side users, which cannot be ignored in large-scale networks. Therefore, in this embodiment, the GS algorithm is improved. Before the pairing starts, the satisfaction preference rankings of both sides of the pairing (the supply-side user and the demand-side user) are known, which are the second sequence and the third sequence respectively. Suppose the supply-side user a receives a request from a certain demand-side user b and forms a pairing with it. If the user b ranks first in the third sequence of the supply-side user a, the supply-side user a will no longer receive pairing requests from other demand-side users, and the demand-side users of the system will not send pairing applications to it either, which is called the full-pairing state of the supply-side user. The supply-side users entering the full-pairing state are incorporated into the second matrix. At the beginning of the matching, the demand-side users are the initiators of the pairing and send pairing requests to the supply-side users according to their respective second sequences. According to the second matrix, judge the status of the supply-side user to whom the request is about to be sent. If the requested target user is in the full-pairing state, the request will not be sent, and the request will be sent to the next supply-side target user in sequence. Such an improved GS algorithm enables some users to complete the optimal selection and pairing in advance, without waiting until the global user detection stops to complete the user pairing. It reduces the detection time of some users, thereby reducing the service delay and improving the communication quality.

[0153] The two-way joint attribute reference disclosed in the embodiment of the present invention Figures 2 to 6 , Figure 2 is a schematic diagram of the simulation parameter settings of each evaluation framework. When the simulation system is modeled, it is assumed that all users are randomly distributed in a single-cell environment. Therefore, it is considered that the influence of the user's moving speed on the path loss can be ignored. Through simulation analysis using the MATLAB simulation software, the advantages and disadvantages of the evaluation framework 1 based on joint attributes and other evaluation frameworks are compared in four dimensions: communication establishment ratio, outage probability, system throughput, and pairing detection request times. The other evaluation frameworks are as follows: the one-way distance and social relationship evaluation framework 2 that only considers distance and social relationship from the demand-side perspective and defaults that the supply-side agrees to establish D2D communication, the two-way social evaluation framework 3 where both the supply-side and the demand-side only use social relationship as the evaluation attribute, and the evaluation framework 4 (or 5) where the supply-side considers device power (or transmission time) and social relationship, and the demand-side uses distance as the evaluation attribute.

[0154] Figure 3 and Figure 4Comparisons are made from two dimensions: the communication establishment ratio and the communication interruption probability. It can be seen that the two-way evaluation framework based on joint attributes has a higher communication establishment ratio and can ensure a lower interrupted transmission probability. Although Evaluation Framework 5 has a slightly higher establishment ratio, its forwarding interruption probability is the highest, which means that even if a connection is established, the transmission is likely to be interrupted due to insufficient power or poor channel quality. At the same time, as the number of supplier users increases from 100 to 300, the system's successful establishment ratio shows an upward trend to stability. This is because as the number of supplier users increases, the set of users that the demander can choose from becomes larger. When the number of supplier users is much larger than that of the demander, the establishment ratio stabilizes close to 1. From Figure 5 it can be seen that when the number of supplier users changes from 100 to 300, the performance of the system throughput of Evaluation Framework 1 based on joint attributes far exceeds that of other evaluation models. From Figure 6 it can be seen that one of the optimization objectives of the evaluation framework based on joint attributes is to maximize user satisfaction. Although traversing the set inevitably increases the number of requests to a certain extent, the gap with other evaluation frameworks is acceptable.

[0155] The matching algorithm with low pairing delay disclosed in the embodiments of the present invention refers to Figures 7 to 13 , Figure 7 which shows the similarities and differences in establishing D2D connection relationships between the proposed matching algorithm with low pairing delay and the GS algorithm in a certain pairing. It can be seen that most of the relationships established by the matching algorithm with low pairing delay are the same as those of the latter. Figure 8 shows the time taken by the demander users to successfully establish pairing relationships under the two algorithms in a representative matching. When N d = 100 and N p = 150, due to delayed matching, all demander users need 1.46 s of detection time to reach the global optimum and establish pairing relationships under the GS algorithm; while 54% of the demander users with low pairing delay only spend 0.99 s of detection time and have already successfully established pairing relationships. The remaining users completed pairing after 1.42 s of detection time.

[0156] Figures 9 to 13 shows the impact on various indicators when changing different supplier users (N d = 100) while keeping the demander users (N p = 100:50:300). At the same time, the proposed matching algorithm with low pairing delay is compared with other heuristic algorithms, such as: the GS algorithm, the greedy algorithm, and the random algorithm, demonstrating the effectiveness of the matching algorithm with low pairing delay in maximizing the D2D pairing efficiency while ensuring comparable D2D link throughput and establishment ratio performance.

[0157] Figure 9Shows the impact of different numbers of supplier users on the total pairing efficiency of the optimization target D2D. The total pairing efficiency of D2D increases as the number of supplier users increases. Because as the number of supplier users increases, on the one hand, the number of demander users who can successfully match and establish D2D links also increases, and the total pairing satisfaction of users will also increase. On the other hand, the more users are available for selection, the shorter the pairing delay required to detect the target user will be. Therefore, the total pairing efficiency of the optimization target will also increase. Among them, compared with the other two algorithms, the total pairing efficiency of the algorithm we proposed has been improved.

[0158] Figure 10 Describes the impact of the number of supplier users on the total user satisfaction. The larger this value is, the more satisfied the users are. It can be seen that except for the random algorithm, the other three algorithms all maintain a relatively high total user satisfaction.

[0159] Figure 11 Describes the impact of the number of supplier users on the total number of requests from 100 demander users. The number of requests increases as the number of supplier users increases because the more supplier users there are, the easier and faster it is for both parties to find the users they like. Among them, compared with the greedy algorithm and the GS algorithm, the number of requests of the matching algorithm with low pairing delay is significantly reduced.

[0160] Figure 12 and Figure 13 respectively show the impact of the number of supplier users on the system throughput and the D2D link establishment ratio. It can be seen that compared with the greedy algorithm and the GS algorithm, the indicators of the matching algorithm with low pairing delay are comparable.

[0161] The embodiments of the present invention also provide a computer-readable storage medium, characterized in that a computer program is stored on the computer-readable storage medium, and when the computer program is executed by a processor, it implements the D2D low-latency user pairing method based on joint attributes as described in any one of the above methods.

[0162] In this embodiment, if the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, to implement all or part of the processes in the above embodiment methods of this application, a computer program can be used to instruct the relevant hardware to complete. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by a processor, the steps of the above various method embodiments can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in the form of source code, object code, executable file or some intermediate form, etc. The computer-readable medium can at least include: any entity or device that can carry the computer program code to the photographing device / terminal device, recording medium, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electrical carrier signal, telecommunication signal, and software distribution medium. For example, a USB flash drive, a mobile hard disk, a magnetic disk or an optical disc, etc. In some jurisdictions, according to legislation and patent practice, the computer-readable medium cannot be an electrical carrier signal and a telecommunication signal.

[0163] In the above embodiments, the descriptions of the various embodiments have their own focuses. For the parts not detailed or recorded in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0164] Those of ordinary skill in the art can realize that the units and algorithm steps of the examples described in combination with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. Professional technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of this application.

[0165] In the embodiments disclosed in this application, it should be understood that the disclosed device / terminal device and method can be implemented in other ways. For example, the device / terminal device embodiments described above are only illustrative. For example, the division of the modules or units is only a logical function division. In actual implementation, there can be other division methods. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed coupling or direct coupling or communication connection to each other can be through some interfaces. The indirect coupling or communication connection of the device or unit can be in an electrical, mechanical or other form.

[0166] The unit described as a separation component may or may not be physically separated. The component displayed as a unit may or may not be a physical unit, that is, it may be located in one place or distributed across multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0167] The above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit it; although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements on some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application, and should all be included in the protection scope of the present application.

Claims

1. A D2D low-latency user pairing method based on joint attributes, characterized in that Including: Establish a one-to-one two-way matching model based on D2D connections, and linearly weight the normalized joint attribute values to obtain the satisfaction function of the supplier user and the satisfaction function of the demander user respectively; Among them, the process of linearly weighting the normalized joint attribute values to obtain the satisfaction function of the supplier user and the satisfaction function of the demander user respectively specifically includes: Obtain the remaining power of the supplier user's device , the size of the transfer file required by the demander user and the physical distance between the supplier user and the demander user , transfer rate and social relationship ; Obtain the remaining power weight of the supplier user's device , the weight of the size of the transfer file required by the demander user , the weight of the physical distance between the supplier user and the demander user , the transmission rate weight , the weight of the first social relationship and the weight of the second social relationship ; Obtain the device remaining power weight according to the remaining power of the device, and the device remaining power weight satisfies ; Obtain the transmission file size weight according to the transmission file size, and the transmission file size weight satisfies ， ； Obtain the physical distance weight according to the physical distance, and the physical distance weight satisfies ， ； Obtain the transmission rate weight according to the transmission rate, and the transmission rate weight satisfies ， ； Obtain a first social relationship weight value and a second social relationship weight value according to the social relationship, and the first social relationship weight value satisfies , where the second social relationship weight value is equal to the first social relationship weight value ; Establish a satisfaction function of the supplier user according to the remaining battery power of the device, the transmission file size, the social relationship and their respective weights. The satisfaction function of the supplier user is ； The remaining battery power weight of the device 、The weight of the transmitted file size and the weight of the first social relationship sum to 1; Establish a satisfaction function of the demander user according to the physical distance, the transmission rate and the social relationship and their respective weights. The satisfaction function of the demander user is ； Among them, , the physical distance weight , the transmission rate weight and the second social relationship weight sum to 1; Determine the pairing delay of the requesting user according to the number of requests for establishing a D2D connection by the requesting user once and the unit request time, and the pairing delay of the requesting user satisfies , where is the pairing delay of the requesting user, is the number of requests for the requesting user to establish a D2D connection once, is the unit request time, and the unit request time is the time for the requesting user to send a request each time; Determine the pairing efficiency of the demander user and the supplier user in the D2D connection according to the satisfaction function of the supplier user, the satisfaction function of the demander user and the pairing delay of the demander user. The pairing efficiency is: ； Among them, is the supplier user satisfaction function, is the demander user satisfaction function; Based on the satisfaction function of the supplier user and the satisfaction function of the demander user, combined with the optimal stopping theory and the GS algorithm, obtain a matching algorithm with low pairing delay, and perform D2D connection user pairing through the matching algorithm with low pairing delay. The GS algorithm is a stable matching bilateral matching algorithm that achieves the global optimum of weak Pareto balance through delayed matching.

2. The method according to claim 1, characterized in that, The one-to-one two-way matching model based on D2D connections is: ； Among them, represents a binary pairing user selection strategy set, and C1 and C2 represent channel multiplexing constraints to ensure that a demand-side user is paired with at most one supply-side user for pairing.

3. The method according to claim 1, wherein The remaining power of the equipment of the supplier user The obtaining steps include: Perform stepped setting and normalization processing on the remaining power of the equipment of the supplier user to obtain the remaining power of the equipment of the supplier user =[0, 0.25, 0.5, 0.75, 1].

4. The method according to claim 1, characterized in that The size of the transfer file required by the demander user The acquisition steps include: Obtain the size of the transmission file required by the demander user according to the Gaussian random distribution, and the size of the transmission file .

5. The method according to claim 1, wherein The physical distance between the supplier user and the demander user The obtaining steps include: Use the Euclidean distance to determine the physical distance between the supplier user and the demander user. The physical distance is 。 6. The method according to claim 1, wherein The transmission rate between the supplier user and the demander user The obtaining steps include: Determine the transmission rate between the supplier user and the demander user using a fading channel model, the transmission rate is ； Among them, and are the transmission powers of the base station and a group of D2D communication pairs, is the transmission distance between a group of D2D communication pairs, is the transmission distance between the base station and a group of D2D communication pairs, and α is the path loss exponent corresponding to the large-scale fading of the transmission channel, obeys the complex Gaussian distribution of the Rayleigh channel coefficient, is the variance of additive white Gaussian noise (AWGN).

7. The method according to claim 1, wherein The acquisition steps of the social relationship between the supplier user and the demander user include: Obtain the social relationship between the supplier users and the demander users according to the Pareto distribution, and the social relationship .

8. The method according to claim 1, wherein Based on the satisfaction function of the supplier user and the satisfaction function of the demander user, combined with the optimal stopping theory and the GS algorithm, obtain a matching algorithm with low pairing delay, and perform D2D connection user pairing through the matching algorithm with low pairing delay. Specifically, it includes: Step 1: Create a first sequence and a second sequence for each demander user. The first sequence is used to represent the number of candidate pairing objects of each demander user, and the second sequence is used to store the candidate pairing objects of each demander user. And the second sequence is sorted from high to low according to the satisfaction function of the demander user; Step 2: Create a third sequence for each supplier user. The third sequence is used to store the demander users who send pairing requests to each supplier user, and the third sequence is sorted from high to low according to the satisfaction function of the supplier user; Step 3: Based on the GS algorithm, create a first matrix and a second matrix. The first matrix is used to represent the pairing status of the supply and demand sides. The pairing status includes the paired status and the unpaired status. The second matrix is used to store the supplier users in the full pairing status. The supplier users in the full pairing status are the supplier users who have been paired with the demander user with the highest ranking in the third sequence and no longer accept other pairing requests; Step 4: Based on the optimal stopping theory, determine whether the cumulative number of pairing requests sent by each of the demand-side users is greater than the product of 0.37 and the first sequence. If so, stop sending; otherwise, approve the sending. Step 5: Based on the second matrix, determine the pairing status of all the supply-side users in the second sequence of each of the demand-side users. Based on the determination result, obtain a supply-side user with the highest ranking among the non-full pairing statuses in the second sequence of each of the demand-side users and send a pairing request. Step 6: Determine that the first supply-side user receives the request from the demand-side user with the highest ranking in the third sequence of the first supply-side user, and add the first supply-side user to the second matrix. Step 7: Determine that the second supply-side user replaces the pairing partner of the second supply-side user, and the pairing status of the pairing partner of the second supply-side user changes from the paired status to the unpaired status, and update the first matrix. Step 8: Determine whether the pairing status of each of the demand-side users is the unpaired status and the request sending is still approved. If so, return to Step 4.

Images