A permission-free resource allocation method for coexistence of D2D and UORA mechanisms

Abstract

The present invention relates to a method for allocating unlicensed resources for the coexistence of D2D and UORA mechanisms, and belongs to the field of wireless communications. The steps include: S1: setting a coexistence scenario of D2D users, cellular users, and WiFi users; S2: designing a coexistence plan and establishing an optimization problem for maximizing the total throughput of D2D users and cellular users; S3: proposing a channel allocation algorithm based on KM matching to allocate channels to D2D users; S4: optimizing the D2D transmission power through Taylor expansion, Lagrange transformation, and quadratic transformation; S5: proposing an iterative algorithm for multivariable joint optimization to solve the optimal D2D transmission power, channel allocation, duty cycle allocation, and D2D selection results. The unlicensed spectrum coexistence solution proposed in the present invention guarantees the minimum transmission rate of WiFi users while meeting the communication quality of D2D users and cellular users, thereby maximizing the total throughput of D2D and cellular users. In addition, the solution enables the base station to offload more authorized D2D users, greatly alleviating the pressure on the authorized spectrum and significantly improving the spectrum efficiency of the unlicensed spectrum.

Description

Technical Field

[0001] The present invention belongs to the technical field of wireless communications and relates to a method for allocating authorization-free resources in which D2D and UORA mechanisms coexist. Background Art

[0002] With the rapid adoption of smart devices and advances in mobile communication technology, the contradiction between the enormous demand for data traffic and the rapid increase in IoT devices, and the scarcity of spectrum resources in traditional cellular networks, has become increasingly apparent. Over the past decade, cellular networks have carried explosively increasing data traffic, and existing cellular communication networks are no longer able to meet the growing demand for high-speed services. Therefore, there is an urgent need to introduce and implement new technologies or solutions to improve spectrum efficiency and network capacity while reducing the excessive burden on base stations in traditional cellular systems. To this end, the 3rd Generation Partnership Project (3GPP) has proposed device-to-device (D2D) communication technology. Compared to traditional cellular communication, D2D communication enables direct communication between two closely located devices without the need for a base station as an intermediary, significantly reducing the burden on base stations. Furthermore, as a proximity communication technology, D2D users only require lower transmit power to meet communication quality requirements. Furthermore, D2D communication can improve network capacity and spectrum utilization by reusing the licensed spectrum resources of the cellular system.

[0003] As a mainstream user of unlicensed spectrum, WiFi has achieved a qualitative leap in throughput and spectrum efficiency over the past 20 years, from the initial standard 802.11b to the current 802.11ax. Unlike the previous 802.11 protocol, 802.11ax introduces the uplink OFDMA-based random channel access (UORA) mechanism. OFDMA divides each WiFi channel into sub-channels and resource units to support uplink or downlink transmissions of multiple WiFi users at the same time. UORA is an uplink transmission random access mechanism based on trigger frames (TFs). Trigger frames provide time synchronization for the distributed uplink transmission of multiple WiFi users and carry resource allocation information for WiFi users. UORA is initialized and synchronized by trigger frames and uses random access as a competition mechanism for resource allocation. This process does not require any pre-scheduled information for transmission. Compared with the Carrier Sensing Multiple Access with Collision Avoidance (CSMA / CA) mechanism used in the traditional 802.11 protocol, the use of trigger frames simplifies WLAN synchronization and resource allocation for WiFi users, which helps improve the spectrum efficiency and reliability of the network.

[0004] The introduction of D2D communication technology in the unlicensed band (D2D-U) has proven to be a promising approach for expanding cellular coverage, improving spectrum efficiency, increasing network capacity, and reducing transmission delays and overload. However, most previous research has considered the coexistence of D2D with 802.11n, with little research examining the coexistence of D2D with 802.11ax, particularly in the case of WiFi uplink transmissions based on the UORA mechanism. Furthermore, existing research on D2D-U, while considering the use of duty cycle mechanisms for coexistence with WiFi, has rarely combined D2D power allocation, channel allocation, duty cycle allocation, and D2D selection. The Listen Before Talk (LBT) mechanism and the Duty Cycle (DC) mechanism are currently the two mainstream D2D technologies for using unlicensed spectrum. The D2D users and cellular users in the present invention use a duty cycle mechanism to coexist with WiFi users in the unlicensed spectrum, so it is necessary to control the time that D2D and cellular users occupy the unlicensed spectrum to meet the communication quality of WiFi users. Furthermore, when D2D users and cellular users multiplex channels, it is necessary to control the transmission power of D2D to meet the communication quality of cellular users and other D2D users, and it is also necessary to allocate channels to D2D users to improve the total throughput of D2D and cellular users. When WiFi occupies the unlicensed channel, some D2D users directly access the unlicensed spectrum and share some sub-channels with WiFi users. Therefore, it is necessary to solve the D2D selection problem so that WiFi users have enough channels to meet their uplink communication needs. Under the premise of meeting the communication quality of D2D users and cellular users, the present invention guarantees the minimum transmission rate of WiFi users, and jointly considers the transmission power of D2D users, D2D channel allocation, the duty cycle time of D2D and cellular users accessing the unlicensed spectrum, and D2D mode selection, thereby maximizing the total throughput of D2D and cellular users. Summary of the Invention

[0005] In view of this, the present invention provides a method for allocating unlicensed resources in which D2D and UORA mechanisms coexist. Under the premise of ensuring the minimum transmission rate of WiFi users and the communication quality between D2D users and cellular users, in order to maximize the total throughput of D2D users and cellular users, we propose a method for allocating unlicensed resources in which D2D and UORA mechanisms coexist. In this method, we decouple the variables of the optimization objective function, propose a channel allocation algorithm based on the maximum weight (Kuhn Munkres, KM) matching of the bipartite graph to solve the channel allocation problem of D2D users and cellular users reusing the unlicensed spectrum, and propose a power allocation algorithm based on convex optimization to solve the transmission power optimization problem of D2D users. Finally, a multivariable iterative optimization algorithm combining channel-power-duty cycle-D2D selection is proposed to solve the time proportion problem of WiFi users and the D2D selection problem under the duty cycle mode, and obtain the optimal solution of all optimization variables.

[0006] In order to achieve the above object, the present invention provides the following technical solutions:

[0007] A method for allocating unlicensed resources for coexistence of D2D and UORA mechanisms includes the following steps:

[0008] S1: Set up the coexistence scenario of D2D users, cellular users and WiFi users;

[0009] S2: Design a coexistence scheme and formulate an optimization problem to maximize the total throughput of D2D users and cellular users;

[0010] S3: A KM matching-based channel allocation algorithm is proposed to allocate channels to D2D users;

[0011] S4: Optimize D2D transmission power through Taylor expansion, Lagrange transform, and quadratic transform;

[0012] S5: A multivariable joint optimization iterative algorithm is proposed to solve the optimal D2D transmission power, channel allocation, duty cycle allocation and D2D selection results.

[0013] Furthermore, in step S1, the present invention considers a cellular network covered by a single base station, which includes D pairs of D2D users and N cellular users, both of which are randomly distributed. i , Cellular users are denoted as CU j , In addition, there is a WiFi access point (WAP) within the coverage area, and its service range includes W randomly distributed WiFi users WU. WiFi users are denoted as WU k , WiFi devices operate in the 5 GHz frequency band and use the 802.11ax protocol for communication. Specifically, WiFi users communicate with WAPs based on the UORA mechanism. In the UORA mechanism, WAP divides the unlicensed channel into different sub-channels for WiFi users to compete for. u For the channel bandwidth of WiFi users, the present invention sets the number of sub-channels to M.

[0014] Furthermore, in step S2, the present invention combines the duty cycle mechanism and the direct access method to achieve the coexistence of D2D users, cellular users and WiFi users in the unlicensed spectrum. Define a period T, T is divided into T off and T on Two time periods.

[0015] In T off During this time period, WiFi does not use the unlicensed spectrum, and D2D users and cellular users reuse the unlicensed sub-channel B allocated by the base station. n , the number of sub-channels is equal to the number of cellular users, B n is the channel bandwidth of the cellular user. Therefore, in T off During this period, DU i The throughput can be expressed as:

[0016]

[0017] In the formula, For DU i With CU j Signal-to-interference-and-noise ratio when multiplexing channels; ρ i is the D2D selection factor; P i d is the transmission power of D2D transmitter i; P c For CU j The transmission power; For DU i Channel gain from transmitter to receiver; |g j,i | 2 For CU j to DU i Channel gain at the receiving end; |g i′,i | 2 is the transmission end to DU of other D2D pairs i′ in the same channel i Channel gain at the receiving end; σ 2 is the noise power spectral density; α i,j is the channel allocation factor, α i,j ∈{0,1}. In T off CU in the time period j The throughput can be expressed as:

[0018]

[0019] In the formula, For CU j With DU i Signal-to-interference-and-noise ratio when multiplexing channels; |g j,BS | 2 For CU j Channel gain to the base station; |g i,j | 2 For DU i Transmitter to CU j channel gain.

[0020] In T on During this time period, cellular users do not use the unlicensed spectrum, WiFi users transmit data through the sub-channels they compete for, and some D2D users directly access the sub-channels to use the unlicensed spectrum. A sub-channel is occupied by at most one pair of D2D users. i The throughput is:

[0021]

[0022] In the formula, For DU i Signal-to-interference-and-noise ratio when multiplexing channels with WiFi users; P max is the maximum transmit power of D2D; P w is the transmission power of the Wi-Fi user, |g k，i | 2 For WiFi users to DU i The channel gain at the receiving end. The average throughput of each WiFi user is:

[0023]

[0024] In the formula, P tr is the probability that at least one WiFi user transmits in a certain subchannel; P s is the probability of successful transmission in a subchannel; P idle is the probability that all subchannels are idle; E[P] is the expected packet size (in bits); T tr is the duration of channel occupancy when at least one subchannel can successfully deliver a data packet; T idle The duration for WAP to transmit a trigger frame (TF); T c is the time when the channel is busy due to collisions among all WiFi users, λ is the duty cycle factor, λ = T off / T.

[0025] Establish the objective function to be optimized: To obtain the maximum system throughput of cellular users and D2D users, we have:

[0026]

[0027] st:

[0028] C1:

[0029] C2:0≤λ<1

[0030] C3:P min ≤P i d ≤P max ,

[0031] C4:

[0032] C5:

[0033] C6:α i,j ∈{0,1},

[0034] C7:

[0035] C8:

[0036] C9:

[0037] C10:ρ i ∈{0,1},

[0038] C11:

[0039] Where ρ is the variable ρ i The set of α is the variable α i,j The collection of P d is the variable P i d The collection of R C is the total throughput of cellular users; R Df For T off Total throughput of D2D users in the time period; R Do For T on The total throughput of D2D users in the time period; when C1 cellular users, D2D users and WiFi coexist, WiFi users should meet the minimum throughput requirement, R min is the minimum throughput of WiFi; C2 represents the value range of the duty cycle factor λ, and its value is T off / T; C3 represents the D2D transmission power limit, P min is the minimum transmit power of D2D, P max is the maximum transmit power of D2D; C4 means that each pair of D2D users can only multiplex channels with one cellular user; C5 means that a cellular user can multiplex channels with at most D-N+1 pairs of D2D users and at least one pair of D2D users; C6 is the channel allocation factor α i,j The value range of α i,j =1, indicating DU i With CU j C7 indicates that when cellular users and D2D users are multiplexing channels, the signal-to-interference-and-noise ratio of the cellular user should meet its minimum signal-to-interference-and-noise ratio requirement. The minimum signal-to-noise ratio requirement that cellular users need to meet; C8 and off During the time period, when D2D users and cellular users reuse channels, the signal-to-interference-and-noise ratio of D2D users should meet the minimum signal-to-interference-and-noise ratio requirement. is the minimum signal-to-noise ratio requirement that D2D users need to meet; C10 represents the minimum signal-to-noise ratio requirement that D2D users need to meet at T on During the time period, when D2D users and WiFi users reuse channels, the signal-to-interference-and-noise ratio of D2D users should meet their minimum signal-to-interference-and-noise ratio requirements; C10 represents the D2D selection factor ρ i The value range of ρ i =1, represents DU i In T on Communication is carried out during the time period, when ρ i = 0, it means that D2D pair i is in T off C11 indicates that at most M-1 pairs of D2D are allowed to communicate in T on Direct access to unlicensed spectrum during the time period.

[0040] Further, in step S3, to ensure that off During the time period, each cellular user multiplexes the channel with at least one pair of D2D users. The number of D2D pairs should satisfy D≥N+M-1. Assuming that there are D′ pairs of D2D users that can multiplex the channel with cellular users, is the set of D′ pairs of D2D users. KM matching is performed once in each iteration, and the number of iterations is l. If D′ / N is an integer, l∈(1,…,D′ / N); if D′ / N is a decimal, l∈(1,…,D′ / N+1). For the first iteration, to meet the KM matching requirements, the base station randomly selects Randomly select N pairs of D2D users from the data set. Assuming that the base station can know the location and channel status information of all D2D users and cellular users, the base station can calculate DU i Separately with all CU jThe throughput during multiplexing is summed up and recorded as Will As a DU i The weight of CU is 0. j The weight of DU i With CU j When multiplexing channels, the sum of the throughput of the two is taken as the edge (DU i ,CU j ) edge weight. After KM matching, the optimal matching Ψ1 of the first iteration is obtained, and each pair of matching {DU i′ ,CU j′}Replace the vertex CU matched by KM in the next iteration j Repeat the first iteration process and update the lth matching Ψ l Until the D′ / Nth iteration, if D′ / N is an integer, output Ψ l , as the result of channel allocation and mapped to the optimization variable α, the channel allocation problem is solved; if D′ / N is a small number, at the last iteration, the number of unmatched D2D pairs is less than N, which does not meet the KM matching conditions. Since the base station knows the channel state information and location of all D2D users and cellular users, it can be matched by exhaustive method at the last iteration and output Ψ l , as the result of channel allocation and mapped to the optimization variable α, so far, the channel allocation problem is solved.

[0041] Furthermore, in step S4, since the optimization problem is a mixed integer nonlinear convex optimization problem, the present invention uses a convex optimization method using Taylor expansion, Lagrange transform, and quadratic transform to first solve the D2D transmit power optimization problem. During this process, the variables λ, α, and ρ are always fixed, and the process is as follows:

[0042] First, since the throughput of cellular users in the optimization objective function is represented by the term R C It is about the variable P i d is a convex function, so in the feasible solution The first-order Taylor expansion of this becomes is a linear function, Right now:

[0043]

[0044] Secondly, since the throughput of D2D users in the optimization objective function is represented by the term R Df About variable P i d Non-strict concave and convex, according to the Lagrange transformation, the approximate variable Υ can be introduced i To replace the complex fraction in the function, so RDf can be rewritten as:

[0045]

[0046] R Df About Y i To find the derivative, get The solution of time

[0047]

[0048] Finally, introduce the variable R Df Perform a secondary transformation and fix the variable Υ i , so R Df can be rewritten as:

[0049]

[0050] R Df about Seek the derivation, get The solution of time

[0051]

[0052] Will Substitute R Df , we can get:

[0053]

[0054] At this point, the optimization problem can be rewritten as:

[0055]

[0056] st: C1, C3, C7, C8, C9

[0057] The optimization function is continuously iterated through the convex optimization tool CVX until the problem converges, and the D2D transmission power optimization problem is solved.

[0058] Further, in step S5, since on During the time period, WiFi users will cause co-channel interference to D2D users. Therefore, when allocating D2D devices to T on Before the time period, each D2D pair performs channel energy detection on each subchannel assigned by the WAP to the WiFi user and feeds it back to the base station. The base station forms a D×M interference matrix I based on the feedback information. D×M , I i,m ∈I D×M , Ii,m For DU i Interference from channel m is detected, m∈{1,2,…,M}. The base station selects M-1 interference items from the interference matrix from small to large, and each interference item is in a different row and column in the interference matrix. i,m When selected, ρ i =1, when I i,m When not selected, ρ i = 0. From the constraint C10 in S2, we know that when ρ i =1, represents DU i In T on Communication is carried out during the time period, when ρ i =0, represents DU i In T off Communicate within the time period, and thus we can preliminarily obtain the D2D selection factor variable set ρ. From M-1 to 0, gradually discard the largest I i,m ,ρ is updated gradually, and judge Is it satisfied? If not, continue to discard I i,m If satisfied, the optimal channel matching α and optimal transmit power P for each pair of D2D can be obtained from S3 and S4. d , and calculate (R C +R Df ) and R Do If (R C +R Df )>R Do ,λ=1-WR min / R W , R W is the total throughput of WiFi users when λ=0; if (R C +R Df )≤R Do ,λ=0. At this point, the total throughput R of D2D users and cellular users in this iteration is recalculated total , R total =λ(R C +R Df )+(1-λ)R Do . Record R for each iteration total After the iteration is completed, the maximum R total and their corresponding λ, α, ρ and P d , the output R total is the maximum total throughput of D2D and cellular users in the optimization problem, λ, α, ρ and P d is the optimal solution corresponding to the maximum throughput.

[0059] Advantages and beneficial effects of the present invention

[0060] The advantages of the present invention are as follows: Most previous studies have considered the coexistence scenario of D2D and 802.11n, and there are few studies on the coexistence of D2D and 802.11ax, especially the WiFi uplink transmission scenario based on the UORA mechanism. Furthermore, existing research on D2D-U considers the use of a duty cycle mechanism to coexist with WiFi, but there is little work on combining D2D power allocation, channel allocation, duty cycle allocation, and D2D selection. Therefore, the present invention takes advantage of D2D's low transmission power, short communication distance, and high spectrum utilization, allowing it to coexist with cellular users and WiFi users in the unlicensed spectrum, significantly improving spectrum efficiency and maximizing the total throughput of D2D and cellular users.

[0061] The beneficial effects of the present invention are as follows: the present invention jointly considers the transmit power of D2D users, D2D channel allocation, the duty cycle of D2D and cellular users accessing unlicensed spectrum, and D2D mode selection. To address the optimization problem of maximizing the total throughput of D2D and cellular users, a method for unlicensed resource allocation that allows coexistence of D2D and the UORA mechanism is proposed. In this method, the variables of the optimization objective function are decoupled, and a channel allocation algorithm based on KM matching is proposed to address the channel allocation problem for D2D and cellular users reusing unlicensed spectrum. A power allocation algorithm based on convex optimization is proposed to address the transmit power optimization problem for D2D users. Finally, a multivariable iterative optimization algorithm combining channel, power, duty cycle, and D2D selection is proposed to address the time share of WiFi users and the D2D selection problem in the duty cycle mode, and to obtain the optimal solution for all optimization variables. The unlicensed spectrum coexistence scheme proposed in this invention guarantees the minimum transmission rate for WiFi users while ensuring the communication quality of D2D and cellular users, thereby maximizing the total throughput of D2D and cellular users. In addition, this solution enables the base station to offload more authorized D2D users, greatly alleviating the pressure on the licensed spectrum and significantly improving the spectrum efficiency of the unlicensed spectrum. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] In order to make the purpose, technical solutions and beneficial effects of the present invention more clear, the present invention provides the following drawings for illustration:

[0063] Figure 1 A schematic diagram of a coexistence network according to an embodiment of the present invention;

[0064] Figure 2 This is a schematic diagram of unlicensed spectrum allocation according to an embodiment of the present invention;

[0065] Figure 3 Schematic diagram of a flow chart of an embodiment of the present invention. DETAILED DESCRIPTION

[0066] The preferred embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0067] In order to solve the problem of spectrum resource shortage in the licensed spectrum and the low spectrum utilization rate of the unlicensed spectrum, this paper proposes a method for allocating unlicensed resources for the coexistence of D2D and UORA mechanisms. In order to enable D2D devices to coexist with Wi-Fi networks, this paper designs a multivariable joint optimization algorithm based on convex optimization and KM matching. Figure 1 As shown, the present invention considers a cellular network covered by a single base station, which includes D pairs of D2D users (DU) and N randomly distributed cellular users (CU). Furthermore, a WAP exists within the coverage area, serving W randomly distributed WiFi users (WU). WiFi devices operate in the 5 GHz band and communicate using the 802.11ax protocol, with an unlicensed bandwidth of 20 MHz. WiFi users communicate with the WAP via the UORA uplink transmission mechanism.

[0068] The diagram of unlicensed spectrum allocation is as follows: Figure 2 As shown, in the time domain, the unlicensed spectrum is divided into T on and T off In the frequency domain, the unlicensed matching is divided into the D2D and cellular user multiplexing part, the D2D and WiFi multiplexing part, and the WiFi usage part in the absence of interference. on During this time, the unlicensed spectrum is divided by WAP into M bandwidths of B u The sub-channels are used by WiFi users to compete, and D2D users multiplex some of the sub-channels with direct access. off During this time, the unlicensed spectrum is divided into N bands with bandwidth B by the base station. n The sub-channels are multiplexed by D2D users and cellular users. Each D2D pair can multiplex the channel with at most one cellular user, and one cellular user can multiplex the channel with at most D-N+1 pairs of D2D users.

[0069] like Figure 3 As shown, a method for allocating unlicensed resources for coexistence of D2D and UORA mechanisms includes the following steps:

[0070] S1: Set up the coexistence scenario of D2D users, cellular users and WiFi users;

[0071] S2: Design a coexistence scheme and formulate an optimization problem to maximize the total throughput of D2D users and cellular users;

[0072] S3: A KM matching-based channel allocation algorithm is proposed to allocate channels to D2D users;

[0073] S4: Optimize D2D transmission power through Taylor expansion, Lagrange transform, and quadratic transform;

[0074] S5: A multivariable joint optimization iterative algorithm is proposed to solve the optimal D2D transmission power, channel allocation, duty cycle allocation and D2D selection results.

[0075] The present invention considers a cellular network covered by a single base station, which includes D pairs of D2D users and N cellular users, both of which are randomly distributed. D2D pair i is denoted as DU i , Cellular users are denoted as CU j , In addition, there is a WiFi access point (WAP) within the coverage area, and its service range includes W randomly distributed WiFi users WU. WiFi users are denoted as WU k , WiFi devices operate in the 5 GHz frequency band and use the 802.11ax protocol for communication. WiFi users communicate with WAP based on the UORA mechanism. In the UORA mechanism, WAP divides the unlicensed channel into different sub-channels for WiFi users to compete for. u For the channel bandwidth of WiFi users, the present invention sets the number of sub-channels to M.

[0076] The invention combines the duty cycle mechanism and the direct access method to achieve the coexistence of D2D users, cellular users and WiFi users in the unlicensed spectrum. Define a period T, T is divided into T off and T on Two time periods.

[0077] In T off During this time period, WiFi does not use unlicensed spectrum. D2D users and cellular users reuse unlicensed sub-channels divided by the base station. The number of sub-channels is equal to the number of cellular users. n is the channel bandwidth of the cellular user. Therefore, in T off During this period, DU i The throughput can be expressed as:

[0078]

[0079] In the formula, For DU i With CU j Signal-to-interference-and-noise ratio when multiplexing channels; ρ i is the D2D selection factor; P i d is the transmission power of D2D transmitter i; P c For CU j The transmission power; For DU i Channel gain from transmitter to receiver; |g j,i | 2 For CU j to DU i Channel gain at the receiving end; |g i′,i | 2 is the transmission end to DU of other D2D pairs i′ in the same channel i Channel gain at the receiving end; σ 2 is the noise power spectral density; α i,j is the channel allocation factor, α i,j ∈{0,1}. In T off CU in the time period j The throughput can be expressed as:

[0080]

[0081] In the formula, For CU j With DU i Signal-to-interference-and-noise ratio when multiplexing channels; |g j,BS | 2 For CU j Channel gain to the base station; |g i,j | 2 For DU i Transmitter to CU j channel gain.

[0082] In T on During this time period, cellular users do not use the unlicensed spectrum, WiFi users transmit data through the sub-channels they compete for, and some D2D users directly access the sub-channels to use the unlicensed spectrum. A sub-channel is occupied by at most one pair of D2D users. i The throughput is:

[0083]

[0084] In the formula, For DU i Signal-to-interference-and-noise ratio when multiplexing channels with WiFi users; P max is the maximum transmit power of D2D; P w is the transmission power of the Wi-Fi user, |g k,i | 2 For WiFi users to DU i The channel gain at the receiving end. The average throughput of each WiFi user is:

[0085]

[0086] In the formula, Ptr is the probability that at least one WiFi user transmits in a certain subchannel; P s is the probability of successful transmission in a subchannel; P idle is the probability that all subchannels are idle; E[P] is the expected packet size (in bits); T tr is the duration of channel occupancy when at least one subchannel can successfully deliver a data packet; T idle The duration for WAP to transmit a trigger frame (TF); T c is the time when the channel is busy due to collisions among all WiFi users, λ is the duty cycle factor, λ = T off / T.

[0087] Establish the objective function to be optimized: To obtain the maximum system throughput of cellular users and D2D users, we have:

[0088]

[0089] st:

[0090] C1:

[0091] C2:0≤λ<1

[0092] C3:P min ≤P i d ≤P max ,

[0093] C4:

[0094] C5:

[0095] C6:α i,j ∈{0,1},

[0096] C7:

[0097] C8:

[0098] C9:

[0099] C10:ρ i ∈{0,1},

[0100] C11: Where ρ is the variable ρ i The set of α is the variable αi,j The collection of P d is the variable P i d The collection of R C is the total throughput of cellular users; R Df For T off Total throughput of D2D users in the time period; R Do For T on The total throughput of D2D users in the time period; when C1 cellular users, D2D users and WiFi coexist, WiFi users should meet the minimum throughput requirement, R min is the minimum throughput of WiFi; C2 represents the value range of the duty cycle factor λ, and its value is T off / T; C3 represents the D2D transmission power limit, P min is the minimum transmit power of D2D, P max is the maximum transmit power of D2D; C4 means that each pair of D2D users can only multiplex channels with one cellular user; C5 means that a cellular user can multiplex channels with at most D-N+1 pairs of D2D users and at least one pair of D2D users; C6 is the channel allocation factor α i,j The value range of α i,j =1, indicating DU i With CU j C7 indicates that when cellular users and D2D users are multiplexing channels, the signal-to-interference-and-noise ratio of the cellular user should meet its minimum signal-to-interference-and-noise ratio requirement. The minimum signal-to-noise ratio requirement that cellular users need to meet; C8 and off During the time period, when D2D users and cellular users reuse channels, the signal-to-interference-and-noise ratio of D2D users should meet the minimum signal-to-interference-and-noise ratio requirement. is the minimum signal-to-noise ratio requirement that D2D users need to meet; C10 represents the minimum signal-to-noise ratio requirement that D2D users need to meet at T on During the time period, when D2D users and WiFi users reuse channels, the signal-to-interference-and-noise ratio of D2D users should meet their minimum signal-to-interference-and-noise ratio requirements; C10 represents the D2D selection factor ρ i The value range of ρ i =1, represents DU i In T on Communication is carried out during the time period, when ρ i = 0, it means that D2D pair i is in T off C11 indicates that at most M-1 pairs of D2D are allowed to communicate in T on Direct access to unlicensed spectrum during the time period.

[0101] To ensure that T offDuring the time period, each cellular user multiplexes the channel with at least one pair of D2D users. The number of D2D pairs should satisfy D≥N+M-1. Assuming that there are D′ pairs of D2D users that can multiplex the channel with cellular users, is the set of D′ pairs of D2D users. KM matching is performed once in each iteration, and the number of iterations is l. If D′ / N is an integer, l∈(1,…,D′ / N); if D′ / N is a decimal, l∈(1,…,D′ / N+1). For the first iteration, to meet the KM matching requirements, the base station randomly selects Randomly select N pairs of D2D users from the data set. Assuming that the base station can know the location and channel status information of all D2D users and cellular users, the base station can calculate DU i Separately with all CU j The throughput during multiplexing is summed up and recorded as Will As a DU i The weight of CU is 0. j The weight of DU i With CU j When multiplexing channels, the sum of the throughput of the two is taken as the edge (DU i ,CU j ) edge weight. After KM matching, the optimal matching Ψ1 of the first iteration is obtained, and each pair of matching {DU i′ ,CU j′}Replace the vertex CU matched by KM in the next iteration j Repeat the first iteration process and update the lth matching Ψ l Until the D′ / Nth iteration, if D′ / N is an integer, output Ψ l , as the result of channel allocation and mapped to the optimization variable α, the channel allocation problem is solved; if D′ / N is a small number, at the last iteration, the number of unmatched D2D pairs is less than N, which does not meet the KM matching conditions. Since the base station knows the channel state information and location of all D2D users and cellular users, it can be matched by exhaustive method at the last iteration and output Ψ l , as the result of channel allocation and mapped to the optimization variable α, so far, the channel allocation problem is solved.

[0102] Since the optimization problem is a mixed integer nonlinear convex optimization problem, the present invention uses a convex optimization method using Taylor expansion, Lagrange transform, and quadratic transform to first solve the D2D transmit power optimization problem. During the process, the variables λ, α, and ρ are always fixed. The process is as follows:

[0103] First, since the throughput of cellular users in the optimization objective function is represented by the term R C It is about the variable P i dis a convex function, so in the feasible solution The first-order Taylor expansion of this becomes is a linear function, Right now:

[0104]

[0105] Secondly, since the throughput of D2D users in the optimization objective function is represented by the term R Df About variable P i d Non-strict concave and convex, according to the Lagrange transformation, the approximate variable Υ can be introduced i To replace the complex fraction in the function, so R Df can be rewritten as:

[0106]

[0107] R Df About Y i Seek the derivation, get The solution of time

[0108]

[0109] Finally, introduce the variable R Df Perform a secondary transformation and fix the variable Υ i , so R Df can be rewritten as:

[0110]

[0111] R Df about To find the derivative, get The solution of time

[0112]

[0113] Will Substitute R Df , we can get:

[0114]

[0115] At this point, the optimization problem can be rewritten as:

[0116]

[0117] st: C1, C3, C7, C8, C9

[0118] The optimization function is continuously iterated through the convex optimization tool CVX until the problem converges, and the D2D transmission power optimization problem is solved.

[0119] Because in T on During the time period, WiFi users will cause co-channel interference to D2D users. Therefore, when allocating D2D devices to T on Before the time period, each D2D pair performs channel energy detection on each subchannel assigned by the WAP to the WiFi user and feeds it back to the base station. The base station forms a D×M interference matrix I based on the feedback information. D×M , I i,m ∈I D×M , I i,m For DU i Interference from channel m is detected, m∈{1,2,…,M}. The base station selects M-1 interference items from the interference matrix from small to large, and each interference item is in a different row and column in the interference matrix. i,m When selected, ρ i =1, when I i,m When not selected, ρ i = 0. From the constraint C10 in S2, we know that when ρ i =1, represents DU i In T on Communication is carried out during the time period, when ρ i =0, represents DU i In T off Communicate within the time period, and thus we can preliminarily obtain the D2D selection factor variable set ρ. From M-1 to 0, gradually discard the largest I i,m ,ρ is updated gradually, and judge Is it satisfied? If not, continue to discard I i,m If satisfied, the optimal channel matching α and optimal transmit power P for each pair of D2D can be obtained from S3 and S4. d , and calculate (R C +R Df ) and R Do If (R C +R Df )>R Do ,λ=1-WR min / R W , R W is the total throughput of WiFi users when λ=0; if (R C +R Df )≤R Do ,λ=0. At this point, the total throughput R of D2D users and cellular users in this iteration is recalculated total , R total =λ(RC +R Df )+(1-λ)R Do . Record R for each iteration total After the iteration is completed, the maximum R total and their corresponding λ, α, ρ and P d , the output R total is the maximum total throughput of D2D and cellular users in the optimization problem, λ, α, ρ and P d is the optimal solution corresponding to the maximum throughput.

[0120] Finally, it should be noted that the above preferred embodiments are only used to illustrate the technical solutions of the present invention and are not limiting. Although the present invention has been described in detail through the above preferred embodiments, those skilled in the art should understand that various changes can be made in form and details without departing from the scope defined by the claims of the present invention.

Claims

1. A method for allocating unlicensed resources for coexistence of D2D and UORA mechanisms, comprising the following steps: S1: Set up the coexistence scenario of D2D users, cellular users, and WiFi users: D pairs of D2D users and N cellular users are randomly distributed in the cellular network covered by a single base station, and D2D pair i is denoted as DU i , Cellular users are denoted as CU j , There is also a WiFi access point within the coverage area, and its service range includes W randomly distributed WiFi users WU; WiFi users are recorded as WU k , WiFi devices operate in the 5GHz frequency band and use the 802.11ax protocol for communication. Specifically, WiFi users communicate with WAP based on the UORA mechanism; in the UORA mechanism, WAP divides the unlicensed channel into different sub-channels B u For WiFi users to compete, the number of sub-channels is set to M; S2: Design a coexistence solution and establish an optimization problem to maximize the total throughput of D2D users and cellular users: Combining the duty cycle mechanism and direct access method, D2D users, cellular users and WiFi users can coexist in the unlicensed spectrum. A period T is defined, which is divided into T off and T on Two time periods, in T off During this time period, WiFi does not use the unlicensed spectrum, and D2D users and cellular users reuse the unlicensed sub-channel B allocated by the base station. n , the number of sub-channels is equal to the number of cellular users; exist T off During the time period, with cellular user CU j Multiplexed DU i The throughput can be expressed as: In the formula, For DU i With CU j Signal-to-interference-and-noise ratio when multiplexing channels; ρ i is the D2D selection factor; P i d is the transmission power of D2D transmitter i; P c For CU j The transmission power; For DU i Channel gain from transmitter to receiver; |g j,i | 2 For CU j to DU i Channel gain at the receiving end; |g i′,i | 2 is the transmission end to DU of other D2D pairs i′ in the same channel i Channel gain at the receiving end; σ 2 is the noise power spectral density; α i,j is the channel allocation factor, α i,j ∈{0,1}; in T off CU in the time period j The throughput can be expressed as: In the formula, For CU j With DU i Signal-to-interference-and-noise ratio when multiplexing channels; |g j,BS | 2 For CU j Channel gain to the base station; |g i,j | 2 For DU i Transmitter to CU j The channel gain of In T on During this time period, cellular users do not use the unlicensed spectrum, WiFi users transmit data through the sub-channels they compete for, and some D2D users directly access the sub-channels to use the unlicensed spectrum. A sub-channel is occupied by at most one pair of D2D users. At this time, DU i The throughput is: In the formula, For DU i Signal-to-interference-and-noise ratio when multiplexing channels with WiFi users; P max is the maximum transmit power of D2D; P w is the transmission power of the Wi-Fi user, |g k,i | 2 For WiFi users to DU i Channel gain at the receiving end; The average throughput per WiFi user is: In the formula, P tr is the probability that at least one WiFi user transmits in a certain subchannel, P s is the probability of successful transmission in a subchannel, P idle is the probability that all subchannels are idle; E[P] is the expected size of the data packet in bits, T tr is the duration of channel occupancy when at least one subchannel can successfully deliver a data packet, T idle The duration of a trigger frame (TF) transmitted by WAP, T c is the time when the channel is busy due to collisions among all WiFi users; Establish the objective function to be optimized: To obtain the maximum system throughput of cellular users and D2D users, we have: st: C2:0≤λ<1 Where ρ is the variable ρ i The set of α is the variable α i,j The collection of P d is the variable P i d The collection of R C is the total throughput of cellular users, R Df For T off The total throughput of D2D users in the time period, R Do For T on The total throughput of D2D users in the time period, C1 When cellular users and D2D users coexist with WiFi, WiFi users should meet the minimum throughput requirement, R min is the minimum throughput of WiFi, C2 represents the value range of duty cycle factor λ, and its value is T off / T, C3 represents the D2D transmission power limit, P min is the minimum transmit power of D2D, P max is the maximum transmission power of D2D, C4 means that each pair of D2D users can only multiplex channels with one cellular user, C5 means that a cellular user can multiplex channels with at most D-N+1 pairs of D2D users, and can multiplex channels with at least one pair of D2D users, and C6 is the channel allocation factor α i,j The value range of α i,j =1, indicating DU i With CU j Multiplexing channel, C7 indicates that when cellular users and D2D users are multiplexing channels, the signal-to-interference-and-noise ratio of the cellular user should meet its minimum signal-to-interference-and-noise ratio requirement. The minimum signal-to-noise ratio requirement that needs to be met for cellular users is C8 and is expressed in T off During the time period, when D2D users and cellular users reuse channels, the signal-to-interference-and-noise ratio of D2D users should meet the minimum signal-to-interference-and-noise ratio requirement. is the minimum signal-to-noise ratio requirement that D2D users need to meet, and C10 represents the minimum signal-to-noise ratio requirement that D2D users need to meet. on During the time period, when D2D users and WiFi users reuse channels, the signal-to-interference-and-noise ratio of D2D users should meet their minimum signal-to-interference-and-noise ratio requirements. C10 represents the D2D selection factor ρ. i The value range of ρ i =1, represents DU i In T on Communication is carried out during the time period, when ρ i = 0, it means that D2D pair i is in T off C11 indicates that at most M-1 pairs of D2D are allowed to communicate in T on Direct access to unlicensed spectrum within the time period; S3: A channel allocation algorithm based on KM matching is proposed to allocate channels to D2D users: To ensure that T off During the time period, each cellular user multiplexes the channel with at least one pair of D2D users. The number of D2D pairs should satisfy D≥N+M-1. Assuming that there are D′ pairs of D2D users that can multiplex the channel with cellular users, For the set of D2D users D′, a KM matching is performed in each iteration, and the number of iterations is l. If D′ / N is an integer, l∈(1,…,D′ / N); if D′ / N is a decimal, l∈(1,…,D′ / N+1). For the first iteration, in order to meet the requirements of KM matching, the base station randomly selects Randomly select N pairs of D2D users from the data set. Assuming that the base station can know the location and channel status information of all D2D users and cellular users, the base station can calculate DU i Separately with all CU j The throughput during multiplexing is summed up and recorded as Will As a DU i The weight of CU is 0. j The weight of DU i With CU j When multiplexing channels, the sum of the throughput of the two is taken as the edge (DU i ,CU j ) edge weight; after KM matching, the optimal matching Ψ1 of the first iteration is obtained, and each pair of matching {DU i′ ,CU j′ }Replace the vertex CU matched by KM in the next iteration j Repeat the first iteration process and update the lth matching Ψ l Until the D′ / Nth iteration, if D′ / N is an integer, output Ψ l , as the result of channel allocation and mapped to the optimization variable α, so far, the channel allocation problem is solved; If D′ / N is a decimal, in the last iteration, the number of unmatched D2D pairs is less than N, which does not meet the KM matching condition. Since the base station knows the channel state information and location of all D2D users and cellular users, it can be matched by exhaustive method in the last iteration and output Ψ l , as the result of channel allocation and mapped to the optimization variable α, so far, the channel allocation problem is solved; S4: Optimize D2D transmission power through Taylor expansion, Lagrange transform, and quadratic transform; S5: Calculate the optimal D2D transmission power, channel allocation, duty cycle allocation, and D2D selection results based on the iterative algorithm of multi-variable joint optimization.

2. The method for allocating unlicensed resources for coexistence of D2D and UORA mechanisms according to claim 1, characterized in that: In step S4, since the optimization problem is a mixed integer nonlinear convex optimization problem, a convex optimization method using Taylor expansion, Lagrange transform, and quadratic transform is used to first solve the D2D transmit power optimization problem. During the process, the variables λ, α, and ρ are always fixed. The process is as follows: First, since the throughput of cellular users in the optimization objective function is represented by the term R C It is about the variable P i d is a convex function, so in the feasible solution The first-order Taylor expansion of this becomes is a linear function, Right now: Secondly, since the throughput of D2D users in the optimization objective function is represented by the term R Df About variable P i d Non-strict concave and convex, according to the Lagrange transformation, the approximate variable Υ can be introduced i To replace the complex fraction in the function, so R Df can be rewritten as: R Df About Y i To find the derivative, get The solution of time Finally, introduce the variable R Df Perform a secondary transformation and fix the variable Υ i , so R Df can be rewritten as: R Df about Seek the derivation, get The solution of time Will Substitute R Df , we can get: At this point, the optimization problem can be rewritten as: st: C1, C3, C7, C8, C9 The optimization function is continuously iterated through the convex optimization tool CVX until the problem converges, and the D2D transmission power optimization problem is solved.

3. The method for allocating unlicensed resources for coexistence of D2D and UORA mechanisms according to claim 1, characterized in that: In step S5, since T on During the time period, WiFi users will cause co-channel interference to D2D users. Therefore, when allocating D2D devices to T on Before the time period, each D2D pair performs channel energy detection on each subchannel allocated by the WAP to the WiFi user and feeds it back to the base station; the base station forms a D×M interference matrix I based on the feedback information D×M , I i,m ∈I D×M , I i,m For DU i Interference from channel m is detected, m∈{1,2,…,M}; The base station selects M-1 interference items from the interference matrix from small to large, and each interference item is located in a different row and column in the interference matrix; when I i,m When selected, ρ i =1, when I i,m When not selected, ρ i =0; From the constraint C10 in S2, we know that when ρ i =1, represents DU i In T on Communication is carried out during the time period, when ρ i =0, represents DU i In T off Communication is carried out within the time period, from which the D2D selection factor variable set ρ can be preliminarily obtained; From M-1 to 0, gradually discard the largest I i,m ,ρ is updated gradually, and judge Is it satisfied? If not, continue to abandon I i,m If satisfied, the optimal channel matching α and optimal transmit power P for each pair of D2D can be obtained from S3 and S4. d , and calculate (R C +R Df ) and R Do ; If (R C +R Df )>R Do ,λ=1-WR min / R W , R W is the total throughput of WiFi users when λ=0; if (R C +R Df )≤R Do ,λ=0; At this point, the total throughput R of D2D users and cellular users in this iteration is recalculated total , R total =λ(R C +R Df )+(1-λ)R Do , record the R of each iteration total After the iteration is completed, the maximum R total and their corresponding λ, α, ρ and P d , the output R total is the maximum total throughput of D2D and cellular users in the optimization problem, λ, α, ρ and P d is the optimal solution corresponding to the maximum throughput.

Images