A method for scheduling hybrid business resources in a multi-service scenario

Abstract

This invention discloses a method for scheduling mixed service resources in a multi-service scenario, applied to a cellular network system. The method includes: at the initial stage of a slot, the base station allocates resource blocks for an eMBB service slice; when URLLC service traffic arrives in a sub-time slot within the slot, resource blocks are allocated for the URLLC service slice at the beginning of the next sub-time slot, ensuring that all URLLC service traffic is transmitted within that next sub-time slot; wherein, the base station utilizes a resource allocation strategy to allocate resource blocks, using OMA (Optical Method Allocation) for resource block allocation between users belonging to the same service slice, and using NOMA (Normally Owned Method Allocation) for resource blocks to be reused between user pairs belonging to different service slices. This invention combines NOMA technology to solve the problem of scheduling mixed service resources in a multi-service scenario, improving spectrum efficiency, alleviating spectrum congestion, reducing system latency, and improving transmission reliability.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of communication, and particularly relates to a method for scheduling mixed services in a multi-service scenario. BACKGROUND

[0002] There are various services with different quality of service requirements in a highly cooperative mobile communication network, and the traditional single network architecture cannot simultaneously meet such diversified requirements.

[0003] In the prior art, in order to meet the transmission requirements of different services, many studies have proposed that resource reuse schemes are used for wireless resource sharing in an eMBB (Enhanced Mobile Broadband) and URLLC (Ultra Reliable Low Latency Communication) mixed service system. Since URLLC services have randomness, it is allowed to schedule URLLC services in sub-slots as the minimum scheduling unit while eMBB services are transmitted in slots as the scheduling unit. The suddenly arrived URLLC service can puncture the eMBB service being transmitted in the next sub-slot, that is, when the URLLC service data packet arrives, the data transmission of the eMBB service is suspended first, the URLLC service is served first to guarantee its strict latency requirement, and then the data transmission of the eMBB service is resumed.

[0004] However, allowing the URLLC service to preempt the transmission resources of the eMBB service makes the eMBB service interrupted, affecting the reliability of the eMBB service, and cannot simultaneously meet the requirements of different services. Secondly, the spectrum resources are limited, and since the URLLC data packet is generally small, interrupting the eMBB service to transmit the URLLC service will reduce the spectrum utilization of the system. SUMMARY

[0005] In order to solve the above problems existing in the prior art, the present application provides a method for scheduling mixed services in a multi-service scenario, which is applied to a cellular network system, and the system supports two types of services, eMBB and URLLC. The method comprises the following steps:

[0006] The base station allocates resource blocks to eMBB service slices at the beginning of each slot, wherein the resource blocks allocated to a single eMBB user in the eMBB service slice include one or more.

[0007] In each slot, when there is URLLC service traffic arriving in a certain sub-slot in the slot, the base station allocates resource blocks for the URLLC service slice starting from the next sub-slot, and ensures that the allocated resource blocks are such that the URLLC service traffic is completely transmitted in the next sub-slot;

[0008] The base station uses a preset resource allocation strategy to implement the function of allocating resource blocks, and in the resource allocation strategy, the base station allocates resource blocks to users belonging to the same service slice in an OMA manner, and allocates resource blocks to users belonging to different service slices in an NOMA manner to multiplex resource blocks between the user pairs.

[0009] Optionally, the resource allocation strategy is based on a pre-constructed hybrid service resource scheduling optimization problem to implement resource allocation.

[0010] The hybrid service resource scheduling optimization problem model is an optimization problem of solving a resource allocation scheme under the constraint condition of the Qos requirement of URLLC, with the goal of maximizing the available data rate of the eMBB service slice.

[0011] Optionally, the hybrid service resource scheduling optimization problem is:

[0012]

[0013]

[0014]

[0015]

[0016]

[0017]

[0018] ∑ k∈K ∑ b∈B p kb (t)+∑ m∈M ∑ n∈N ∑ b∈B p nb (m)≤P max (4),

[0019]

[0020]

[0021]

[0022]

[0023]

[0024] where K denotes the set of eMBB users, N denotes the set of URLLC users, M denotes the number of sub-slots contained in a slot, slot is denoted by t, and sub-slot is denoted by m, denotes the set of resource blocks of a base station, T denotes the resource scheduling frame of the base station; f b denotes the bandwidth of a resource block;

[0025] denotes the actual data rate available to eMBB user k in sub-slot m, taking into account the data rate loss to eMBB users caused by the introduction of NOMA, denotes the actual data rate available to eMBB user k in t, taking into account the data rate loss to eMBB users caused by the introduction of NOMA; denotes the data rate constraint on eMBB user k;

[0026] x kb (t) is a first resource block scheduling factor, x kb (t) = 1 when resource block b is allocated to eMBB user k in t, otherwise x kb (t) = 0; x is a resource allocation vector in t composed of each first resource block scheduling factor;

[0027] is a second resource block scheduling factor, when resource block b is allocated to URLLC user n in sub-slot m of t, otherwise z is a resource allocation vector in sub-slot m composed of each second resource block scheduling factor, is the inverse of

[0028] γ kb (m, t) denotes the signal-to-noise ratio of eMBB user k in sub-slot m; denotes the signal-to-interference noise ratio of eMBB user k when eMBB user k and URLLC user n multiplex resource block b in sub-slot m;

[0029] denotes the data packet size of URLLC user n in sub-slot m; denotes the resource allocation indicator factor of URLLC user n in sub-slot m of t, when URLLC user n is allocated resources in sub-slot m, otherwise

[0030] ζ ndenotes the maximum latency satisfying the Qos requirement of URLLC; denotes the data rate available to URLLC user n in sub-slot m at t; denotes the arrival process of URLLC traffic packets in sub-slot m at t, which is a Poisson process with mean λ; denotes the URLLC reliability probability; denotes the probability of

[0031] p kb denotes the power obtained by eMBB user k on resource block b at t; p nb denotes the power obtained by URLLC user n on resource block b in sub-slot m; p max denotes the total power constraint of the base station; p denotes the power allocation scheme of eMBB users and URLLC users at t;

[0032] denotes the strong user identification in a pair of eMBB user k and URLLC user n sharing resource block b, if eMBB user k is the strong user then if URLLC user n is the strong user then where the strong user is defined as the one with higher channel gain in the pair.

[0033] Optionally, the manner of allocating resources based on the mixed traffic resource scheduling optimization problem comprises:

[0034] dividing the mixed traffic resource scheduling optimization problem into a first sub-problem and a second sub-problem; wherein the first sub-problem is an optimization problem with x and p as solving parameters and with the objective of maximizing the available data rate of eMBB traffic slice under the constraint of the Qos requirement of URLLC; the second sub-problem is an optimization problem with z and p as solving parameters and with the objective of maximizing the available data rate of eMBB traffic slice under the constraint of the Qos requirement of URLLC;

[0035] allocating resource blocks for eMBB traffic slice by solving the first sub-problem at the beginning of each slot; when URLLC traffic arrives in a certain sub-slot in the slot, allocating resource blocks for the URLLC traffic slice by solving the second sub-problem.

[0036] Optionally, the first sub-problem is:

[0037]

[0038]

[0039] ​

[0040]

[0041]

[0042]

[0043] wherein, represents the data rate available to eMBB user k on resource block b in t.

[0044] Optionally, the way of solving the first sub-problem comprises:

[0045] obtaining a Lagrangian function of the first problem model;

[0046] solving optimal x and p based on the Lagrangian function.

[0047] Optionally, the second sub-problem is:

[0048]

[0049]

[0050]

[0051]

[0052]

[0053]

[0054]

[0055] wherein, represents the data rate available to URLLC user n in sub-slot m of t; p′ kb (m) represents the power obtained by eMBB user k on resource block b in sub-slot m of t.

[0056] Optionally, the way of solving the second sub-problem comprises:

[0057] determining an initial feasible solution of z and p by using Vogel method;

[0058] solving optimal solution of z and p based on the initial feasible solution by using potential method.

[0059] Optionally, the way of realizing resource allocation based on the mixed traffic resource scheduling optimization problem comprises:

[0060] A first sub-problem for eMBB slice resource allocation is divided from the mixed service resource scheduling optimization problem; the first sub-problem is an optimization problem with x and p as solving parameters and with the target of maximizing the available data rate of the eMBB service slice under the constraint condition of the Qos requirement of URLLC;

[0061] At the beginning of each slot, resource blocks are allocated for the eMBB service slice by solving the first sub-problem;

[0062] When URLLC service traffic arrives in a certain sub-slot in the slot, a resource allocation scheme for the URLLC service slice is determined based on the solving result of the first sub-problem with the target of minimizing the number of multiplexed resource blocks.

[0063] Optionally, the manner of implementing resource allocation based on the mixed service resource scheduling optimization problem comprises:

[0064] A first sub-problem for eMBB slice resource allocation is divided from the mixed service resource scheduling optimization problem; the first sub-problem is an optimization problem with x and p as solving parameters and with the target of maximizing the available data rate of the eMBB service slice under the constraint condition of the Qos requirement of URLLC;

[0065] At the beginning of each slot, resource blocks are allocated for the eMBB service slice by solving the first sub-problem;

[0066] When URLLC service traffic arrives in a certain sub-slot in the slot, a resource allocation scheme for the URLLC service slice is determined based on the solving result of the first sub-problem with the target of minimizing the number of multiplexed resource blocks.

[0067] In the method for mixed service resource scheduling in a multi-service scenario provided by the application, the base station uses a resource allocation strategy to realize the function of allocating resource blocks, and uses the OMA manner to allocate resource blocks between users belonging to the same service slice and uses the NOMA manner to multiplex resource blocks between user pairs belonging to different service slices. Thus, the application solves the problem of mixed service resource scheduling in a multi-service scenario by combining the NOMA technology, can improve the spectrum efficiency, relieve spectrum congestion, reduce system latency, and improve transmission reliability.

[0068] Wherein, the application formulates a mixed service cooperative scheduling optimization problem based on NOMA technology for eMBB and URLLC mixed service resource scheduling in a 5G cellular network, which can maximize the system throughput by optimizing the allocation strategy of resource blocks and power on a double time scale (slot and sub-slot) on the basis of guaranteeing the quality of service demand of each type of service, improve the transmission rate of eMBB service as much as possible, minimize the influence of URLLC service multiplexing resources on eMBB, and improve the spectral efficiency and user fairness index of the system.

[0069] The application will be further described in detail below with reference to the accompanying drawings and the application. BRIEF DESCRIPTION OF DRAWINGS

[0070] Figure 1 is a flowchart of the method for mixed service resource scheduling in a multi-service scenario provided by the embodiment of the application;

[0071] Figure 2 The model of resource allocation and scheduling in the prior art in the eMBB and URLLC service coexistence scenario is shown in FIG.

[0072] Figure 3 The model of resource superposition multiplexing in the eMBB and URLLC service coexistence scenario in the embodiment of the application is shown in FIG.

[0073] Figure 4 It is a vehicle networking heterogeneous service scenario model;

[0074] Figure 5 The influence of URLLC data packet multiplexing resources on the eMBB slice rate under several different resource allocation schemes is compared in FIG.

[0075] Figure 6 The influence of several different resource allocation schemes on the eMBB slice user fairness index under different URLLC data packet arrival rates is compared in FIG.

[0076] Figures 7(a) to 7(c) The ECDF (Empirical Cumulative Distribution Function) value of the eMBB slice user average rate is shown in FIG.

[0077] Figure 8 The influence of the URLLC service arrival rate on the eMBB slice user data rate in the prior art and the embodiment of the application is compared in FIG. DETAILED DESCRIPTION

[0078] The application will be further described in detail below with reference to the accompanying drawings and the application.

[0079] To solve the problems proposed in the background art, embodiments of the present application provide a method for mixed service resource scheduling in a multi-service scenario, which is applied to a cellular network system supporting two types of services, eMBB and URLLC, and using network slicing technology to divide network functions and system resources for different service requirements, thereby constructing logical networks with different functions on a common infrastructure.

[0080] Referring to Figure 1 The method for mixed service resource scheduling in a multi-service scenario provided by the embodiments of the present application includes the following steps:

[0081] S10: The base station allocates resource blocks to the eMBB service slice at the beginning of each time slot slot, wherein the resource blocks allocated to a single eMBB user in the eMBB service slice include one or more.

[0082] S20: Within each slot, when URLLC service traffic arrives in a certain sub-slot in the slot, the base station allocates resource blocks to the URLLC service slice at the beginning of the next sub-slot, and ensures that the allocated resource blocks enable the URLLC service traffic to be completely transmitted within the next sub-slot.

[0083] The base station uses a preset resource allocation strategy to achieve the function of allocating resource blocks, in which the base station allocates resource blocks to users belonging to the same service slice in an OMA (Orthogonal Multiple Access) manner, and the base station allocates resource blocks to user pairs belonging to different service slices in a NOMA (Non-Orthogonal Multiple Access) manner to multiplex resource blocks between the user pairs.

[0084] It can be understood that NOMA technology allows multiple users to access the same channel non-orthogonally. Therefore, by combining NOMA technology to solve the problem of mixed service resource scheduling in a multi-service scenario in the embodiments of the present application, the spectral efficiency can be improved, the spectrum congestion can be alleviated, the system delay can be reduced, and the transmission reliability can be improved.

[0085] The resource allocation strategy used by the base station is based on a pre-constructed mixed service resource scheduling optimization problem to achieve resource allocation; the mixed service resource scheduling optimization problem model is an optimization problem of solving a resource allocation scheme with the maximum available data rate of the eMBB service slice as the target under the constraint condition of the Qos requirement of the URLLC.

[0086] The following detailed description of the specific implementation of the resource allocation based on the mixed service resource scheduling optimization problem in the embodiments of the application is described in detail.

[0087] First, the mixed service resource scheduling optimization problem model is described in detail.

[0088] In a cellular network system, two types of services, eMBB (Enhanced Mobile Broadband) and URLLC (Ultra Reliable Low Latency Communication), are supported. Let the set of eMBB users be K = {1, 2, … K}, and the set of URLLC users be N = {1, 2, … N}. The resource block (RB) set of the base station is represented as B = {1, 2, … B}, and each RB occupies 12 subcarriers in the frequency domain and 1 slot in the time domain. In the global 5G standard based on the new air interface design of 5G NR (5G New Radio) OFDM (Orthogonal Frequency Division Multiplexing), each slot can be divided into M mini-slots, represented as M = {1, 2, … M}, where the mini-slot is referred to as a sub-slot.

[0089] Referring to Figure 2 As shown, the base station allocates resources to eMBB service slices at the beginning of each slot, and one eMBB user can use multiple RBs. Since the URLLC service slice has a low latency requirement, the base station needs to allocate resources to it at the beginning of the next mini-slot when it arrives, and must ensure that the service can be completely transmitted in one mini-slot.

[0090] In the embodiments of the application, in order to improve the utilization rate of spectrum resources, the base station allocates resource blocks between users belonging to the same service slice using the OMA method, and the base station allocates resource blocks between user pairs belonging to different service slices using the NOMA method to multiplex the resource blocks between the user pairs, as shown in Figure 3 .

[0091] Specifically, for the eMBB service slice, each RB can only be allocated to one eMBB user, and the scheduling period of the base station is one slot, so the eMBB user The data rate available on the resource block is:

[0092]

[0093] where, denotes the signal-to-noise ratio SNR, f b denotes the bandwidth of each RB, h kb (t) denotes the channel gain of the Rayleigh fading of the transmission channel of eMBB user k on resource block b in time slot t, p kb (t) denotes the downlink transmission power allocated by the base station to eMBB user k on resource block b in time slot t, σ 2 denotes the noise power, and the superscript "e" is a shorthand for eMBB.

[0094] In order to meet the data rate of eMBB users, the base station can need to allocate multiple RBs to them, so the data rate available to eMBB user k in time slot t is:

[0095]

[0096] where x kb (t) is a first resource block scheduling factor, x kb (t) = 1 when resource block b is allocated to eMBB user k in t, otherwise x kb (t) = 0.

[0097] Since the URLLC traffic can arrive during the data transmission of eMBB users, in order to meet the strict QoS requirement of URLLC, the arrived URLLC data packet needs to be placed on the next mini-slot immediately. Since the data packet of URLLC user is usually small, Shannon formula cannot be directly used to calculate the rate available to URLLC user n on the mini-slot, so the finite block length channel coding theory is used to calculate the rate available to URLLC user n on the mini-slot:

[0098]

[0099] where γ nb (m, t) denotes the signal-to-noise ratio of URLLC user n on sub-slot m of t, Q -1 (·) denotes the inverse of the Gaussian Q function, V u = 1 - (1 + γ nb (m, t)) -2 denotes the channel dispersion, · u denotes the coding error rate, N u denotes the number of symbols in the sub-slot; is a second resource block scheduling factor, when resource block b is allocated to URLLC user n in sub-slot m of t, otherwise The resource allocation vector z in sub-slot m is composed of each second resource block scheduling factor.

[0100] In the embodiment of the present application, NOMA technology is used to schedule the overlapping URLLC service and eMBB service. Specifically, the base station uses NOMA between user pairs belonging to different service slices to multiplex resource blocks between the user pairs, that is, the URLLC user and the eMBB user multiplexing the same resource block are regarded as a pair of paired users. For paired users, successive interference cancellation (SIC) is used at the user with higher channel gain, that is, assuming that the eMBB user k and the URLLC user n are paired users on the resource block b, if the channel gain of the eMBB user k is higher than that of the URLLC, then the SNRs of them can be respectively represented as:

[0101]

[0102]

[0103] On the contrary, if the channel gain of the eMBB user k is lower than that of the URLLC, then the SNRs of them are respectively represented as:

[0104]

[0105]

[0106] In order to meet the delay and reliability requirements of the URLLC user n, the URLLC user n can be allocated multiple resource blocks in the sub-slot m, and then the SINR of the URLLC user n can be represented as:

[0107]

[0108] wherein, represents the identification of the strong user in a pair of eMBB user k and URLLC user n multiplexing the resource block b, if the eMBB user k is the strong user, then if the URLLC user n is the strong user, then wherein the strong user is defined as the one with higher channel gain in the pair of users.

[0109] Therefore, the data rate that the URLLC user n can obtain in the sub-slot m can be represented as:

[0110]

[0111] For the URLLC user n with a data packet size of , in order to meet the requirement of ζ n in the Qos requirement, the following conditions need to be met:

[0112]

[0113] wherein, ζn denotes the maximum latency satisfying the Qos requirement of URLLC, denotes the data packet size of URLLC user n in sub-slot m; denotes the resource allocation indicator factor of URLLC user n in sub-slot m at t, when URLLC user n is allocated with resources in sub-slot m, otherwise

[0114] In addition, the reliability constraint of URLLC service can be expressed as follows:

[0115]

[0116] wherein, Ψ denotes the arrival process of URLLC service data packet in sub-slot m at t, Ψ is a Poisson process with mean λ; ∈ denotes the URLLC reliability probability, in the case of ∈ = 0.999, the reliability constraint ensures that the probability of the number of served URLLC users being less than the number of arrived users is less than 0.001, thereby satisfying the reliability of URLLC service. denotes the probability of

[0117] Since the URLLC user on the sub-slot and the already allocated eMBB user will cause resource block overlap, which will increase the data rate loss of eMBB user, therefore, on the superimposed resource block b, considering the data rate loss caused by NOMA superposition, the actual data rate that eMBB user k can obtain on sub-slot m is:

[0118]

[0119] wherein, is the inverse of denotes the signal-to-interference-and-noise ratio (SINR) of eMBB user k when eMBB user k and URLLC user n multiplex resource block b in sub-slot m.

[0120] The resource allocation of eMBB user is carried out at the beginning of each time slot t, and the actual data rate that it can obtain on t is as follows:

[0121]

[0122] Based on the above mathematical modeling results, in order to simultaneously satisfy the data transmission requirements of eMBB service slice and the delay and reliability requirements of URLLC service slice, the embodiment of the application constructs an optimization problem of solving resource allocation scheme under the constraint of Qos requirement of URLLC and with the goal of maximizing the available data rate of eMBB service slice, and the expression is as follows: ​​

[0123]

[0124]

[0125]

[0126]

[0127]

[0128]

[0129] ∑ k∈K ∑ b∈B p kb (t)+∑ m∈M ∑ n∈N ∑ b∈B p nb (m)≤P max ,

[0130]

[0131]

[0132]

[0133]

[0134]

[0135] where K denotes the set of eMBB users, N denotes the set of URLLC users, M denotes the number of sub-slots contained in a slot, slot is denoted by t, sub-slot is denoted by m, denotes the set of resource blocks of the base station, T denotes the resource scheduling frame of the base station; f b denotes the bandwidth of the resource block;

[0136] denotes the actual data rate that the eMBB user k can obtain in the sub-slot m, considering the data rate loss brought to the eMBB user by the introduction of NOMA, denotes the actual data rate that the eMBB user k can obtain in t, considering the data rate loss brought to the eMBB user by the introduction of NOMA; denotes the data rate constraint on the eMBB user k;

[0137] x kb (t) is a first resource block scheduling factor, x kb (t) = 1 when the resource block b is allocated to the eMBB user k in t, otherwise xkb (t) = 0; x is the resource allocation vector in t consisting of individual first resource block scheduling factors;

[0138] For the second resource block scheduling factor, when resource block b is allocated to URLLC user n in sub-slot m of t, Otherwise z is the resource allocation vector in sub-slot m consisting of individual second resource block scheduling factors, For the inverse of

[0139] γ kb (m, t) denotes the signal-to-noise ratio of eMBB user k in sub-slot m; denotes the signal-to-interference-plus-noise ratio of eMBB user k when eMBB user k and URLLC user n multiplex resource block b in sub-slot m;

[0140] denotes the data packet size of URLLC user n in sub-slot m; denotes the resource allocation indicator of URLLC user n in sub-slot m of t, when URLLC user n is allocated resource in sub-slot m, Otherwise

[0141] ζ n denotes the maximum latency satisfying URLLC Qos requirement; denotes the achievable data rate of URLLC user n in sub-slot m of t; Ψ denotes the arrival process of URLLC traffic data packet in sub-slot m of t, Ψ is a Poisson process with mean λ; ∈ denotes the URLLC reliability probability; denotes the probability of

[0142] p kb (t) denotes the power obtained by eMBB user k on resource block b in t; p nb (m) denotes the power obtained by URLLC user n on resource block b in sub-slot m; P max denotes the total power constraint of the base station; p denotes the power configuration scheme of individual eMBB users and URLLC users in t;

[0143] denotes the strong user identification in a pair of eMBB user k and URLLC user n multiplexing resource block b, if eMBB user k is the strong user then if URLLC user n is the strong user then where the strong user is defined as the one with higher channel gain in the pair.

[0144] The constraint conditions (1) and (2) represent the delay and reliability constraints of the URLLC slice users respectively; the constraint condition (3) represents the data rate constraint of the eMBB users; the constraint condition (4) represents the total power constraint of the base station; the constraint conditions (5) and (6) represent the orthogonality constraints of the users in the eMBB slice and the URLLC slice respectively, that is, the users in the eMBB slice cannot reuse the same resource block, and the users in the URLLC slice cannot reuse the same resource block; the constraint conditions (7)-(9) represent that x, z and a, are binary variables.

[0145] The above optimization problem is a combined optimization problem with opportunity constraints, and it is very difficult to find the global optimal solution. The inventors consider that the eMBB users need a large amount of data transmission in a period of time, and therefore the base station allocates resources to the eMBB users at the initial time of the slot t and keeps the allocation result unchanged in this period of time. On this basis, in order to meet the QoS requirements of the sporadic URLLC slice user data packets that arrive in the sub-slot m, the base station needs to allocate resources to the URLLC slice users in the sub-slot m+1 and transmit, as shown in Figure 3

[0146] In the embodiments of the present application, there are various ways to allocate resources based on the above mixed service resource scheduling optimization problem.

[0147] For example, in the first implementation mode, in order to effectively solve the above mixed service resource scheduling optimization problem, the divide-and-conquer strategy can be adopted to divide the above mixed service resource scheduling optimization problem into two resource allocation sub-problems.

[0148] Specifically, the way to allocate resources based on the above mixed service resource scheduling optimization problem includes:

[0149] (a) dividing the mixed service resource scheduling optimization problem into a first sub-problem and a second sub-problem; wherein the first sub-problem is an optimization problem with x and p as solving parameters and with the maximum available data rate of the eMBB service slice as the target under the constraint condition of the QoS requirement of the URLLC; the second sub-problem is an optimization problem with z and p as solving parameters and with the maximum available data rate of the eMBB service slice as the target under the constraint condition of the QoS requirement of the URLLC;

[0150] (b) at the beginning of each slot, allocating resource blocks to the eMBB service slice by solving the first sub-problem; when the URLLC service traffic arrives in a certain sub-slot in the slot, allocating resource blocks to the URLLC service slice by solving the second sub-problem.

[0151] ​wherein the first sub-problem is:

[0152]

[0153]

[0154]

[0155]

[0156]

[0157]

[0158] wherein, represents a data rate available to the eMBB user k on the resource block b within t.

[0159] The second sub-problem is:

[0160]

[0161]

[0162]

[0163]

[0164]

[0165]

[0166]

[0167] wherein, represents a data rate available to the URLLC user n in the sub-slot m of t; p' kb (m) represents a power obtained by the eMBB user k on the resource block b in the sub-slot m of t.

[0168] As to the above first sub-problem, a manner of solving the first sub-problem includes: constructing a Lagrangian function of the first problem model; and solving optimal x and p based on the Lagrangian function.

[0169] Specifically, the Lagrangian function of the above first sub-problem can be written as:

[0170]

[0171] wherein λ and μ are dual variables. In addition, the Lagrangian dual function of the first sub-problem can be written as:

[0172]

[0173] Here, by Lagrangian relaxation, the connection between RBs can be eliminated and can be divided into B sub-problems at each RB, each of which can be solved independently and has minimal complexity. At the same time, each sub-problem can be further split into sub-problems of power allocation and resource block allocation by two-level decomposition. When any one sub-problem is determined, the other sub-problem will be easy to solve. The specific resource allocation process can be referred to the following algorithm:

[0174]

[0175]

[0176] Wherein, δ and ε are both preset error parameters for judging whether the algorithm converges, and l is the iteration number of the algorithm.

[0177] Regarding the above second sub-problem, the resource allocation for URLLC service slice includes the initial pairing of URLLC users and eMBB users and the power allocation required by URLLC users. Since the resource allocation for eMBB users has been completed at the beginning of time slot t, the RBs are fully occupied. Therefore, it is necessary to select resource blocks for URLLC users for NOMA superposition multiplexing.

[0178] As a prerequisite for NOMA technology, it is necessary to allocate the required power to each user to achieve its reliability constraint. Then, the problem is transformed and modeled as an unbalanced transmission problem model, and is converted into a balanced transmission problem, and the initial feasible solution of the balanced problem is determined by using the Vogel method, and then the optimal solution is obtained according to the initial feasible solution and the potential method. The specific process is as follows:

[0179] (1) The initial feasible solution of z and p is determined by using the Vogel method:

[0180] a: Determine the row penalty value and column penalty value of each row and each column in the cost matrix C, respectively called the lowest loss and the second lowest loss. The cost matrix is a BxN matrix, and the element in the matrix is the loss of eMBB user k when the resource block already allocated to eMBB user k is superimposed to URLLC user n.

[0181] b: Select the cell with the smallest loss in the row or column with the largest loss, and then delete the row or column.

[0182] c: In each row or column that has not been deleted, repeat the above steps until the last cell is also allocated, and the initial basic feasible solution of the problem is obtained.

[0183] (2) Based on the initial feasible solution, the optimal solution of z and p is solved by using the potential method:

[0184] a: Using the initial feasible solution obtained by the Vogel method as the basic variable, add a potential column l and a potential row m to the table of basic variables;

[0185] b: Constructing the potential equation system: l n +m b =c nb Calculate l separately n m b ;

[0186] c: According to σ nb =c nb -l n -m b Calculate the test number σ nb The test result table is obtained;

[0187] d: Determine whether σ exists in the test number table. nb If the value is less than 0, and it exists, then the minimum test number σ is found using the closed-loop adjustment method. nb Perform closed-loop testing to obtain a new basic variable table, and repeat steps b through d until all test numbers σ are obtained. nb If the value is greater than or equal to 0, then the solution in the basic variable table is the optimal solution.

[0188] It should be noted that there are already many examples describing the specific implementation methods of solving problems using the Vogel method or potential method in related technologies, so they will not be described in detail in the embodiments of this invention.

[0189] In the second implementation, to reduce the impact of URLLC slice service resource reuse on eMBB slice user rates, resource allocation based on the hybrid service resource scheduling optimization problem can include:

[0190] (1) The first subproblem for eMBB slice resource allocation is divided from the mixed service resource scheduling optimization problem. The first subproblem is an optimization problem with x and p as solution parameters and the objective of maximizing the available data rate of eMBB service slices under the constraint of URLLC QoS requirements.

[0191] (2) At the beginning of each slot, resource blocks are allocated to the eMBB service slice by solving the first subproblem;

[0192] (3) When URLLC service traffic arrives at a sub-slot in this slot, based on the solution of the first sub-problem, determine the URLLC service slice resource allocation scheme with the goal of minimizing the number of reused resource blocks.

[0193] It can be understood that the number of multiplexing resource blocks is minimized as the goal, that is, the URLLC slice user service is transmitted on one resource block as much as possible.

[0194] In the second implementation, the first sub-problem and its solving method are the same as the first implementation, which will not be repeated here.

[0195] Wherein, based on the solution of the first sub-problem, the specific algorithm process of determining the URLLC service slice resource allocation scheme with the goal of minimizing the number of multiplexing resource blocks is as follows:

[0196]

[0197]

[0198] Wherein, resource pools B1 and B2 are resource blocks allocated to eMBB service slices in the solution of the first sub-problem, and τ is the length of the sub-slot. Indicates that the allocated resource block can meet the QoS requirement of the URLLC user n, that is, the obtained data rate is greater than the required data rate.

[0199] In the third implementation, in order to reduce the influence of URLLC slice service resource multiplexing on the rate of eMBB slice users, the resource allocation method based on the mixed service resource scheduling optimization problem can include:

[0200] (1) The first sub-problem for eMBB slice resource allocation is divided from the mixed service resource scheduling optimization problem; the first sub-problem is an optimization problem with x and p as solving parameters and with the goal of maximizing the available data rate of eMBB service slice under the constraint condition of the Qos requirement of URLLC;

[0201] (2) At the beginning of each slot, the resource blocks are allocated to the eMBB service slice by solving the first sub-problem;

[0202] (3) When URLLC service traffic arrives in a certain sub-slot in the slot, based on the solution of the first sub-problem, the resource scheduling weight matrix is calculated to allocate resources to the URLLC service slice under the consideration of the QoS requirement of the URLLC service slice and the influence of multiplexing resource blocks on the eMBB service slice, so as to maximize the data rate of the eMBB service slice.

[0203] In this implementation, the QoS requirements of the URLLC slice service and the influence of the multiplexing resource block on the eMBB slice service are comprehensively considered, the resource scheduling weight matrix is calculated, and the URLLC resource is allocated according to the weight matrix. Among them, the first sub-problem and its solving method are the same as the above implementation, which will not be repeated here. Then, based on the solving result of the first sub-problem, the specific algorithm process of determining the URLLC service slice resource allocation scheme is to minimize the number of multiplexing resource blocks. Referring to the following:

[0204]

[0205] Among them, step 5 is to allocate the resource block corresponding to the maximum element in the weight matrix W to user n according to the weight matrix W.

[0206] The above completes the detailed description of the way of implementing resource allocation based on mixed service resource scheduling optimization in the embodiment of the application.

[0207] In summary, the method of mixed service resource scheduling in a multi-service scenario provided by the embodiment of the application formulates a mixed service cooperative scheduling optimization problem based on NOMA technology for eMBB and URLLC mixed service resource scheduling in a 5G cellular network multi-service scenario, aiming to maximize the system throughput by optimizing the allocation strategy of resource blocks and power on the double time scale (slot and sub-slot) on the basis of guaranteeing the quality of service of each type of business, and to minimize the influence of URLLC service multiplexing resources on eMBB, improve the system spectral efficiency and user fairness index.

[0208] In a preferred implementation, the method of mixed service resource scheduling in a multi-service scenario provided by the embodiment of the application can be applied to a vehicle network, for the following reasons:

[0209] With the development of vehicle intelligence and the rapid evolution of communication technology, the Internet of Vehicles gradually develops from supporting vehicle information services to supporting intelligent and networked applications such as assisted driving, autonomous driving and intelligent transportation, so there are various services with different quality of service requirements in the highly cooperative Internet of Vehicles scenario, including information service type services with high data rate requirements, and safety type services with strict delay and reliability requirements, etc. Figure 4 As shown in the following figure.

[0210] In a dense mobile environment, resource conflicts can occur between vehicle users due to occupying the same time-frequency resources, and it is difficult to guarantee the access rate of users through the way of Orthogonal Multiple Access (OMA). Especially for dense topology, this serious resource collision can cause a large number of packet losses, seriously affecting the transmission reliability of the system, and cannot meet the needs of high data rate services and low latency and high reliability services in the vehicle networking service.

[0211] In addition, with the continuous development of vehicle networking technology, the frequency spectrum solely divided for vehicle networking is very limited, and how to meet the needs of different services under limited wireless resources and improve the spectrum utilization rate is a problem to be solved in vehicle networking.

[0212] As can be seen, mixed service resource scheduling in a multi-service scenario is particularly prominent in vehicle networking, so the method for mixed service resource scheduling in a multi-service scenario provided by the embodiment of the application can be applied to a vehicle networking system to solve the above problems existing in vehicle networking.

[0213] Next, taking vehicle networking as an example, the system performance of the embodiment of the application is evaluated through simulation. In the simulation process, a vehicle networking system is constructed, the base station of which is deployed in the center of a coverage range with a radius of 500 meters, and eMBB users and URLLC users are randomly distributed in the coverage range of the base station. The duration of one time slot is 1ms, and each time slot is further divided into 7 equal time slots. The bandwidth of each resource block is 180kHz, and each resource block is composed of 12 subcarriers with a subcarrier spacing of 15kHz, and the total bandwidth of the system is 10MHz. The arrival of URLLC data packets obeys a Poisson distribution with a rate of λ, and the size of each data packet is 32 bytes. The simulation parameters are shown in the following table:

[0214]

[0215]

[0216] The simulation results are as follows:

[0217] Figure 5 In the comparison of several different resource allocation schemes, the influence of URLLC packet multiplexing resources on eMBB slice rate is compared. The horizontal axis is the arrival rate of URLLC data packets, and the vertical axis is the eMBB slice rate; TM corresponds to the "first implementation mode" in the above, Strategy I corresponds to the "second implementation mode" in the above, Strategy II corresponds to the "third implementation mode" in the above, and SINRmax corresponds to the resource allocation method based on maximum SINR in the prior art.

[0218] It can be seen from the comparison simulation results that, when the arrival rate of the URLLC data packet is small, the scheduling scheme using the TM can make the system obtain better performance; as the arrival rate continuously increases, the performance of the TM gradually decreases, but the overall performance is still better than the existing SINRmax. In addition, the overall performance of the two other implementation manners proposed in the embodiment of the application is better than SINRmax, and the performance of Strategy II is improved compared with Strategy I, especially for the case of large arrival rate of the URLLC data packet.

[0219] Figure 6 The influence of several different resource allocation schemes on the fairness index of the eMBB slice user under different arrival rates of the URLLC data packet is compared in FIG. 3. Figure 6 As shown in FIG. 3, the fairness index of Strategy II is better than TM and Strategy I, and the performance of the three schemes proposed in the embodiment of the application is better than the existing SINRmax.

[0220] Figures 7(a) to 7(c) The ECDF (Empirical Cumulative Distribution Function) values of the average rate of the eMBB slice user under several different resource allocation schemes with the increase of the arrival rate of the URLLC data packet are shown in FIG. 4. It can be seen by comparison that the probability of the eMBB slice user rate less than 10 Mbps is 0.2 for TM, the probability of the eMBB slice user rate less than 10 Mbps is about 0.168 for Strategy II, the probability of the eMBB slice user rate less than 10 Mbps is about 0.3 for Strategy I, and the performance of the three schemes is better than the existing SINRmax. In addition, it can be seen that Strategy II has a higher achievable rate of the eMBB slice user, because Strategy II considers the influence of the URLLC slice user on the rate of the eMBB slice user when allocating resources to the URLLC slice user, so that the achievable rate of the eMBB slice user is higher under the same condition.

[0221] In the prior art, the commonly used multiplexing mode for the wireless resource allocation problem in the eMBB service and URLLC service coexistence scenario is puncturing, that is, when the URLLC service data packet arrives, the data transmission of the eMBB service is suspended first, the URLLC service is served first to guarantee its strict delay requirement, and then the data transmission of the eMBB service is resumed. In contrast, in the embodiment of the application, the NOMA superposition technology is used to multiplex the two slice services. Figure 8 The influence of the arrival rate of the URLLC service on the data rate of the eMBB slice user under the two modes is compared in FIG. 5, where NOMA is the TM scheme in the embodiment of the application, and puncturing is the puncturing mode in the prior art.

[0222] As can be seen from Figure 8 The system performance of the embodiment of the present application is obviously superior to the existing puncturing scheme, because the NOMA technology can multiplex the URLLC user and the eMBB user in the same RB, and obtain the signals of different users by setting different transmission power and using the serial interference cancellation technology. This means that during the transmission of the URLLC data packet, the eMBB service will be affected, but the transmission will not be completely terminated, so compared with the puncturing scheme, the embodiment of the present application reduces the influence of the URLLC service on the eMBB slice service, and improves the rate of the eMBB slice user.

[0223] It should be noted that the terms "first", "second", and the like are used to distinguish similar objects, and do not necessarily have to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged under appropriate circumstances, so that the embodiments of the disclosure described herein can be implemented in an order other than those illustrated or described herein. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present disclosure. Rather, they are merely examples of devices and methods consistent with some aspects of the present disclosure.

[0224] In the description of the present specification, the description of the terms "one embodiment", "some embodiments", "an example", "a specific example", or "some examples" and the like means that the specific features or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In the present specification, the illustrative description of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features or characteristics described can be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art can combine and combine different embodiments or examples described in the present specification.

[0225] Although the present application is described herein in conjunction with various embodiments, other variations of the disclosed embodiments can be understood and implemented by those skilled in the art with reference to the drawings and the disclosure. In the description of the present application, the word "comprising" does not exclude other components or steps, "one" or "an" does not exclude a plurality, and "plurality" means two or more, unless otherwise explicitly specified. In addition, some measures are described in different embodiments, but this does not mean that these measures cannot be combined to produce good results.

[0226] The above is further detailed description of the present application in combination with specific preferred embodiments, and cannot be deemed as limitation of the specific implementation of the present application to these descriptions. For those skilled in the art to which the present application belongs, without departing from the concept of the present application, a number of simple deductions or substitutions can be made, and all should be deemed as falling within the protection scope of the present application.

Claims

1. A method for mixed traffic resource scheduling in a multi-service scenario, characterized in that, The application is applied to a cellular network system supporting eMBB and URLLC services; The method comprises: allocating resource blocks to the eMBB service slice at the beginning of each time slot slot, wherein the resource blocks allocated to a single eMBB user in the eMBB service slice comprise one or more; within each slot, when URLLC service traffic arrives in a certain sub-slot in the slot, the base station allocates resource blocks to the URLLC service slice at the beginning of the next sub-slot, and ensures that the allocated resource blocks enable the URLLC service traffic to be completely transmitted within the next sub-slot; The base station uses a preset resource allocation strategy to achieve the function of allocating resource blocks, wherein the base station allocates resource blocks to users belonging to the same service slice in an OMA manner, and allocates resource blocks to users belonging to different service slices in an NOMA manner to multiplex resource blocks between the user pair. The resource allocation strategy is based on a pre-constructed mixed service resource scheduling optimization problem to achieve resource allocation; wherein the mixed service resource scheduling optimization problem model is an optimization problem of solving a resource allocation scheme under the constraint condition of the Qos requirement of URLLC, with the goal of maximizing the available data rate of the eMBB service slice; the mixed service resource scheduling optimization problem is: ， ， s.t. (1)， (2)， (3)， (4)， (5)， (6)， (7)， (8)， (9)； wherein, denotes a set of eMBB users, denotes a set of URLLC users, denotes the number of sub-slots contained in a slot, slot is used t denotes, sub-slot is used m denotes denotes a set of resource blocks of a base station, denotes a resource scheduling frame of a base station; denotes the bandwidth of a resource block; denotes the actual data rate achievable by the eMBB user k in the sub-slot m after considering the data rate loss to the eMBB user due to the introduction of NOMA denotes the actual data rate achievable by the eMBB user k in the sub-slot t after considering the data rate loss to the eMBB user due to the introduction of NOMA denotes the data rate constraint on the eMBB user k ; is a first resource block scheduling factor when the resource block b is a second resource block scheduling factor when the resource block t is allocated to an eMBB user k , , otherwise ; is a resource allocation vector within t consisting of individual first resource block scheduling factors; is the second resource block scheduling factor when the resource block b is the first resource block scheduling factor when the resource block t is allocated to the URLLC user n , , otherwise ; is the resource allocation vector in the sub-slot m consisting of the individual second resource block scheduling factors is the inverse of . representing eMBB users k in sub-slots on a signal-to-noise ratio; representing eMBB users m in sub-slots k with URLLC users n multiplexing resource blocks b of eMBB users k on a signal-to-interference-plus-noise ratio; Indicates URLLC user n In the sub-time slot The size of the data packets within; Indicates URLLC user n exist t Sub-slot m Resource allocation indicator factor in URLLC users n In the sub-time slot m When resources are allocated, ,otherwise ; denotes the maximum latency satisfying the Qos requirement of URLLC; denotes a URLLC user n In t a sub-slot m of the data rate available; denotes the arrival process of URLLC traffic data packets within t a sub-slot m of the data rate available; is a Poisson process with mean denotes the probability of denotes the URLLC reliability probability; denotes the probability of denotes the power obtained by the eMBB user t in k resource blocks b ; denotes the power obtained by the URLLC user m in n resource blocks b ; denotes the total power constraint of the base station; denotes the power configuration scheme of the eMBB users and URLLC users t in represents multiplexing resource blocks b a pair of eMBB users k and URLLC users n strong user identification among k if the eMBB user , if the URLLC user n strong user where a strong user is defined as the one with higher channel gain among the pair.

2. The method of claim 1, wherein, The way of achieving resource allocation based on the mixed service resource scheduling optimization problem comprises: The mixed service resource scheduling optimization problem is divided into a first sub-problem and a second sub-problem; wherein the first sub-problem is to solve parameters, and the optimization problem is to maximize the available data rate of the eMBB service slice under the constraint condition of the Qos requirement of the URLLC; and The second sub-problem is to solve parameters, and the optimization problem is to maximize the available data rate of the eMBB service slice under the constraint condition of the Qos requirement of the URLLC; and The second sub-problem is to solve parameters, and the optimization problem is to maximize the available data rate of the eMBB service slice under the constraint condition of the Qos requirement of the URLLC; at the beginning of each slot, allocate resource blocks to the eMBB service slice by solving the first sub-problem; when URLLC service traffic arrives in a certain sub-slot in the slot, allocate resource blocks to the URLLC service slice by solving the second sub-problem.

3. The method of claim 2, wherein, The first sub-problem is: wherein, indicates the data rate available to eMBB users t k on the resource blocks b available to the eMBB user.​ 4. The method of claim 3, wherein, The way of solving the first sub-problem comprises: obtaining the Lagrangian function of the first problem model; solving an optimal based on the Lagrangian function and .

5. The method of claim 2, wherein, The second sub-problem is: wherein represents a URLLC user n In t a sub-slot m of a slot represents a data rate available in a sub-slot t of a slot m , eMBB user k obtained power on resource blocks b .

6. The method of claim 5, wherein, The way of solving the second sub-problem comprises: The initial feasible solution of the problem is determined using the Vogel's method and the initial feasible solution of the problem is determined using the Vogel's method Based on the initial feasible solution, the potential method is used to solve and optimal solution.

7. The method of claim 1, wherein, The way of achieving resource allocation based on the mixed service resource scheduling optimization problem comprises: a first sub-problem for eMBB slice resource allocation is divided from the mixed service resource scheduling optimization problem; the first sub-problem is to maximize the available data rate of the eMBB service slice under the constraint of the Qos requirement of the URLLC and an optimization problem for solving the parameters and taking the maximum available data rate of the eMBB service slice as the target at the beginning of each slot, allocate resource blocks to the eMBB service slice by solving the first sub-problem; when URLLC service traffic arrives in a certain sub-slot in the slot, determine the resource allocation scheme of the URLLC service slice based on the solving result of the first sub-problem, with the goal of minimizing the number of multiplexed resource blocks.

8. The method for mixed service resource scheduling in a multi-service scenario according to claim 1, characterized in that, The way of achieving resource allocation based on the mixed service resource scheduling optimization problem comprises: a first sub-problem for eMBB slice resource allocation is divided from the mixed service resource scheduling optimization problem; the first sub-problem is to maximize the available data rate of the eMBB service slice under the constraint of the Qos requirement of the URLLC and an optimization problem for solving parameters and taking the maximum available data rate of the eMBB service slice as the target at the beginning of each slot, allocate resource blocks to the eMBB service slice by solving the first sub-problem; when URLLC service traffic arrives in a certain sub-slot in the slot, consider the QoS requirement of the URLLC service slice and the impact of multiplexed resource blocks on the eMBB service slice, and based on the solving result of the first sub-problem, calculate a resource scheduling weight matrix to allocate resource blocks to the URLLC service slice, to maximize the data rate of the eMBB service slice.

Images