HK1171581B

HK1171581B - Traffic scheduling method of guaranteed bit rate based on quality of service

Info

Publication number: HK1171581B
Application number: HK12112152.5A
Authority: HK
Inventors: 石娴文; 于泳; 陆涛
Original assignee: 中兴通讯股份有限公司
Filing date: 2010-05-20
Publication date: 2015-06-19

Abstract

The present invention discloses a method for scheduling guaranteed bit rate services based on service quality, comprising: determining the scheduling priority of each user based on the average rate of guaranteed bit rate GBR services for all online users within the current transmission time interval TTI; Schedule users in order according to the determined scheduling priority, and allocate resource blocks RB resources to users.The present invention also discloses a service quality based guaranteed bit rate service scheduling device for implementing the above method.By using the scheduling method according to the present invention, it is possible to fully utilize RB resources, quickly increase the rate of users who have not reached GBR, and enable as many users as possible to reach GBR rate, thereby increasing the overall number of satisfied users in the system.For cases where MBR is greater than GBR, it is possible to further increase user speed and increase the number of high-speed users while meeting the GBR rate requirements for as many users as possible.

Description

Guaranteed bit rate service scheduling method based on service quality

Technical Field

The present invention relates to a Guaranteed Bit Rate (GBR) Service scheduling technique, and more particularly, to a Guaranteed Bit Rate Service scheduling method based on Quality of Service (QoS).

Background

Data with strict QoS guarantee, such as voice, image, video and other multimedia service data, is one of the major challenges faced by current mobile communication systems. In order to make the service available to the end user with expected effect, for example, when the user watches Video on line, the picture should be continuous, the third Generation Partnership Project (3 GPP) clearly defines the end-to-end QoS structure in the umts, and introduces a variety of bearer and processing mechanisms to ensure that the umts can fully exert its technical advantages and provide various differentiated services to the user.

QoS is the quality of service that a system (server) provides to a user. QoS is End-to-End (Endto End), i.e., the quality of service is evaluated from the originating End (Source) to the receiving End (Target). The bearer level QoS parameters include QoS Class Identifier (QCI), Allocation and Retention Priority (ARP), GBR, Maximum Bit Rate (MBR), and Aggregate Maximum Bit Rate (AMBR). Where QCI is a quantity class used to indicate the access point parameters that control the packet transmission process at the bearer level, ARP is primarily intended to decide whether to accept or reject a bearer establishment or modification request in case of resource limitations. The parameter GBR represents a bit rate expected to be provided by a GBR bearer, and when the transmission rate of the traffic is equal to or greater than GBR, the quality of service of the traffic is satisfactory, and when the transmission rate of the traffic is less than GBR, the quality of service of the traffic is unacceptable. The parameter MBR limits the bit rate that the GBR bearer can provide, which represents an upper bound for the GBR bearer to provide the desired data rate.

The GBR bearer is mainly used for services such as voice, video, and real-time games, and QoS needs to be guaranteed on both the access network side and the core network side to guarantee QoS of the services. The QoS of the core network side traffic is guaranteed by the transmission priority, while the access network side is guaranteed by allocating sufficient radio resources within the Base Station (BS) to guarantee the QoS of the user traffic.

The QoS of core network side traffic is easily implemented, and for the QoS of access network GBR traffic, because the base station determines the service priority of each user terminal through scheduling, the selection of a suitable scheduling algorithm is very important for whether the user terminal can obtain the GBR rate.

At present, there are three commonly used scheduling algorithms at the base station side: round Robin (RR) algorithm, Max C/I (Maximum Carrier to Interference) algorithm, Proportional Fair (PF) algorithm.

The basic idea of the RR algorithm is to ensure that the user terminals in the cell cyclically occupy radio resources of equal time according to a certain sequence for communication. Although the algorithm has the best fairness in scheduling opportunity, the difference of the user channel quality cannot be fully utilized, so that the utilization rate of system resources and the system throughput are low, and the method does not consider the requirement of guaranteed GBR of a user, so that the satisfaction degree of the user is low.

The basic idea of the Max C/I algorithm is to sort all user Channel Quality Indicator (CQI) predictors and schedule users in descending order. Although the algorithm can obtain the maximum system throughput rate, the service obtained by the user is very unfair, the central user with good channel condition can always receive the service, the rate is greater than the GBR, and the edge user with poor channel condition can not obtain scheduling, the rate is less than the GBR, and the user satisfaction is low.

The PF algorithm is where the base station schedules one or more terminals with the largest fairness Factor (Fair Factor). The main idea is to assign a corresponding priority to the users in the cell, and the user with the highest priority in the cell receives the service. The sector throughput rate and service fairness of the algorithm are between RR and MAX-C/I.Wherein the FairFactor_i(t) is UE_iFairness factor, tbSize, at time t_i(t) is UE_iThe amount of data that can be transmitted at time t, Throughput_i(t) is UE_iThroughput in the time window ending at time t. For a User Equipment (UE) with better CQI, as its throughput increases, its priority will decrease to achieve fairness. The algorithm also does not take into account the user GBR requirements.

If the traffic established in the communication system is of the stream type, the traffic requires a GBR, and if non-GBR traffic is established, the GBR is not configured for the user, and the user is not satisfied if the service is not provided when the system resources are tight. Therefore, the appropriate GBR can be configured for the user under both GBR and non-GBR services, so that the user obtains a basic service quality, and a richer operation strategy can be provided.

Disclosure of Invention

In view of the above, the main objective of the present invention is to provide a method and an apparatus for scheduling guaranteed bit rate services based on qos, which can make full use of Resource Block (RB) resources to quickly increase the user rate that does not reach GBR, so that as many users as possible reach the GBR rate.

In order to achieve the purpose, the technical scheme of the invention is realized as follows:

a guaranteed bit rate service scheduling method based on service quality comprises the following steps:

determining the scheduling priority of each user according to the average rate of guaranteed bit rate GBR services of all online users in the current Transmission Time Interval (TTI);

and scheduling the users in sequence according to the determined scheduling priority, and allocating RB resources for the users.

Preferably, the determining the scheduling priority of each user according to the average rate of the GBR services of all online users in the current TTI is as follows:

and determining the scheduling priority of each user according to a proportional fair PF scheduling algorithm and in combination with the QoS (quality of service) of the GBR service of each user.

Preferably, the scheduling priority of each user is determined according to a PF scheduling algorithm and in combination with the QoS of GBR traffic of each user, specifically, the scheduling priority FF of a user is determined according to the following formula:

FF＝FF_PF·FF_GBR

wherein the content of the first and second substances,". represents a dot product operation, Th_HisFor the average rate of GBR service in selected N TTIs before the current TTI, TB (i) is a transport block TB successfully transmitted in the N TTIs, and ThresholdGBR is the GBR reserved proportion set by the system; GBR in the exponent bit numerator represents the guaranteed bit rate of the system configuration.

Preferably, the Th_HisDetermined as follows:

n is the selected window length; m is the number of GBR service flows of the user in N TTIs; TB _ l to TB _ m are the size of the TB once scheduled by each GBR service flow 1 to m scheduled in the TTI; the values of ACK _ l, ACK _2 and ACK _ m are 1 when the corresponding TB transmission succeeds, and 0 when the transmission fails.

Preferably, the sequentially scheduling the users according to the determined scheduling priority, and allocating RB resources to the users, are:

scheduling retransmission users in sequence according to the priority order, wherein the RB number of the retransmission users is determined by the size of the retransmission TB and is not limited by GBR factors;

and when the RB resources are still left after the retransmission user is scheduled, scheduling the newly transmitted users in sequence according to the priority order, and allocating the RB number to the newly transmitted users according to the service quality determined by the GBR service of the newly transmitted users.

Preferably, the allocating the number of RBs to the newly transmitted user according to the service quality determined by the GBR service of the newly transmitted user is:

when the average rate of the newly transmitted user is determined to be greater than the MBR, the newly transmitted user is not scheduled;

when the average rate of newly transmitted users is determined to be greater than or equal to GBR and less than or equal to MBR, and when other users with the rate less than GBR exist in the cell, the RB number is determined for the users according to the GBR and the smaller data volume in the BSR; when the average rates of other users in the cell are determined to be more than or equal to GBR, determining the number of RBs for the user according to the data volume in the BSR of the user; the other users are all online users of a service cell to which the newly transmitted user belongs, and the newly transmitted user, the retransmission user and the scheduled user before the user is scheduled in the current TTI are removed, and the users left after the user is removed;

and when the average rate of the newly transmitted user is determined to be less than the GBR, determining the RB number for the user according to the data volume in the BSR of the user.

Preferably, when it is determined that the average rate of the newly transmitted user is greater than or equal to GBR and less than or equal to MBR, the method further includes:

when the BSR of a user is larger than the GBR, adding the user into a user list reaching the GBR and limited in the number of RBs;

and when the RB resources of the other users are still remained after the allocation is finished, allocating the RB resources for the users in the user list reaching the GBR.

A guaranteed bit rate service scheduling device based on service quality comprises a determining unit and a resource scheduling unit; wherein the content of the first and second substances,

a determining unit, configured to determine a scheduling priority of each user according to an average rate of GBR services of all online users in a current TTI;

and the resource scheduling unit is used for sequentially scheduling the users according to the determined scheduling priority and allocating RB resources to the users.

Preferably, the determining unit further determines the scheduling priority of each user according to a PF scheduling algorithm and in combination with the QoS of the GBR service of each user.

Preferably, the determining unit further determines the scheduling priority FF of the user according to the following formula:

FF＝FF_PF·FF_GBR

wherein the content of the first and second substances,". represents a dot product operation, Th_HisFor the average rate of GBR service in selected N TTIs before the current TTI, TB (i) is the TB successfully transmitted in the N TTIs, and ThresholdGBR is the GBR reserved proportion set by the system; GBR in the exponent bit numerator represents the guaranteed bit rate of the system configuration.

Preferably, the determination unit further determines Th as follows_His：

Wherein N is the selected window length; m is the number of GBR service flows of the user in N TTIs; here, m is generally 1 or 2; TB _ l to TB _ m are the size of the TB once scheduled by each GBR service flow l to m scheduled in the TTI; the values of ACK _1, ACK _2 and ACK _ m are 1 when the corresponding TB transmission succeeds, and 0 when the transmission fails.

Preferably, the resource scheduling unit further,

when the average rate of the newly transmitted user is determined to be greater than the maximum bit rate MBR, the newly transmitted user is not scheduled;

Preferably, the resource scheduling unit further determines that when the average rate of the newly transmitted user is greater than or equal to GBR and less than or equal to MBR, and when the BSR of the user is greater than GBR, the resource scheduling unit adds the user to the GBR-reached user list of the limited number of RBs; and when the RB resources of the other users are still remained after the allocation is finished, allocating the RB resources for the users in the user list reaching the GBR.

In the invention, the scheduling priority of each user is determined according to a PF scheduling algorithm and in combination with the QoS of GBR service of each user, so that the service can meet the GBR rate as far as possible on the premise of ensuring the scheduling fairness. The scheduling method can fully utilize RB resources, quickly improve the user rate which does not reach GBR, and ensure that the users as many as possible reach the GBR rate, thereby improving the overall satisfied user number in the system. For the condition that MBR is larger than GBR, the speed of the user can be further improved on the basis of meeting the requirement that the maximum number of users reach the GBR speed, and the number of high-speed users is increased.

Drawings

FIG. 1 is a flowchart of a guaranteed bit rate QoS-based service scheduling method according to a first embodiment of the present invention;

FIG. 2 is a flowchart of a second embodiment of a QoS-based guaranteed bit rate service scheduling method according to the present invention;

FIG. 3 is a schematic diagram of a simulation of an MBR with an equal GBR of 200Kbps, in accordance with an embodiment of the present invention;

FIG. 4 is a schematic diagram of a simulation of an MBR with an equal GBR of 300Kbps, in accordance with an embodiment of the present invention;

FIG. 5 is a schematic diagram of a simulation of an example of the present invention in which the MBR is 350Kbps and the GBR is 200 Kbps;

FIG. 6 is a schematic diagram of a simulation of an example of the present invention in which the MBR is 400Kbps and the GBR is 300 Kbps;

fig. 7 is a schematic diagram of a configuration of a guaranteed bit rate service scheduler based on qos according to the present invention.

Detailed Description

In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is further described in detail below with reference to the accompanying drawings by way of examples.

The following two embodiments are described by taking an LTE downlink system, a 10MHz bandwidth, and a macro base station as examples. The basic parameters and parameters used for system simulation are shown in table 1 below:

TABLE 1

Based on a service model for promoting the next generation mobile network technology development organization (NGMN), a scene with larger load is selected, the size of a user data packet is set to be 0.16M, new data is injected into a user data Buffer (BSR) according to the size of MBR in each TTI, so that the phenomenon that the user rate is greater than MBR cannot occur, and all users in the system are supposed to have the same GBR service requirements. The essence of the technical solution of the present invention is further illustrated below with reference to specific examples.

Example one

This example is for the case of the MBR ═ GBR of the user, i.e. the way in which new data is injected into the user data Buffer (BSR) per TTI at the GBR size.

When new BSR data is injected in GBR size every TTI, once the user retransmits, the rate is smaller than GBR, so in case of MBR ═ GBR, instantaneously, there are more users smaller than GBR rate, and then take quick response: for users less than the GBR rate, the number of RBs is determined according to the data size in a user Buffer (BSR), and for users more than or equal to the GBR rate, the number of RBs is determined according to the smaller value of the GBR and the BSR as the data size.

Fig. 1 is a flowchart of a first embodiment of a method for scheduling guaranteed bit rate services based on qos according to the present invention, and as shown in fig. 1, the method for scheduling guaranteed bit rate services based on qos according to the present invention includes the following steps:

step 11, calculate the average historical throughput rate (i.e. the average rate of users) of all online users in the system over the current Transmission Time Interval (TTI). The calculation formula is as follows:

in the above equation, N is a set window length, and is usually 1024, but may be set according to the system requirements and the actual scheduling scenario. In this example, it is described that a user has two GBR traffic flows at the same time, and if there are multiple GBR traffic flows in the user, the calculation method can still determine the average rate by the above formula. TB _1 (stream 1) and TB _2 (stream 2) are TB sizes scheduled once within a window time, and take 0 if not scheduled. ACK _1 and ACK _2 are Hybrid Automatic Repeat Request (HARQ) acknowledgement information received by the base station (for downlink) for whether TB _1 and TB _2 are successfully transmitted, where ACK is 1 when transmission is successful, and ACK is 0 when transmission is failed.

And step 12, the base station queues the priority of each user terminal according to the scheduling algorithm.

The scheduling algorithm in step 12 is an algorithm that integrates two factors, namely, the proportional fairness and the quality of service of the GBR service, and the comprehensive priority (fairness factor) FF expression of the user is as follows:

FF＝FF_PF·FF_GBR

in the formula FF_PFIs a fairness factor calculated by a (general) proportional fairness algorithm (G-PF),the numerator represents the user's TB Size for N TTI (selected window length) intra-couple; the Historythreughput in the denominator is the user average historical throughput rate obtained in step 11); "·" denotes a dot product operation.

FF_GBRIs a fairness factor for the GBR scheduling,GBR in the exponent bit numerator represents the guaranteed bit rate of the system configuration, threshold GBR is the GBR reservation ratio, corresponding to the upper and lower amplitude thresholds of GBR, e.g., threshold GBR ═ 10%; the HistoryThroughput in the exponent denominator is the average historical throughput of the user obtained in step 11.

And according to the calculation result in the step, carrying out priority ranking on the users, wherein the users with higher priorities are ranked at the top of the queue.

And step 13, scheduling the retransmission users according to the user priorities obtained in the step 12, wherein the RB number of the retransmission users is determined by the retransmission TB Size and is not limited by GBR factors.

And step 14, scheduling the newly transmitted users according to the user priority obtained in the step 12.

Wherein the step 14 may comprise the steps of:

step 141, determining whether the rate of the user obtained in step 11 is less than GBR, if so, entering step 142, otherwise, entering step 143.

Step 142, the average rate of the user is less than GBR, according to the number of RBs needed by the data amount in the user Buffer (BSR)And determining the RB quantity, and taking a smaller value with the current residual RB quantity.

Namely, it is

If the user is a single stream transmission:

if the user is in dual-stream transmission:

in the above formula: head is the number of bits of the MAC header, and is 16 bits in the LTE system. The CRC is a length of a Cyclic Redundancy Check (CRC), which is 24 bits in the LTE system.For the number of subcarriers included in each RB, 12 subcarriers are included at 15kHz intervals, and 24 subcarriers are included at 7.5kHz intervals.For the number of available symbols per TTI, 10 are dropped for the normal CP and 8 are dropped for the extended CP. SE₁Is the spectrum efficiency (bit/Symbol) corresponding to the single stream CQI or the CQI of the first stream, SE₂The spectral efficiency corresponding to the CQI of the second stream for two streams. The CQI is the wideband CQI reported by the user.

Step 143, if the average rate of the user is greater than or equal to GBR, then the minimum value of GBR and BSR is usedAnd determining the RB quantity, and taking a smaller value with the current residual RB quantity.

Namely, it is

If the user is a single stream transmissionIf the user is in dual-stream transmission:

where Ceil (·) represents the rounding operation, min (·) is the operation of the smaller of 2, BSR is the number of bits in the DTCH buffer, GBR is the guaranteed bit rate (in kbps) of GBR traffic, EP is the rate enhancement ratio (to prevent the service rate from decreasing due to block error in advance), Header is the ratio to be enhanced due to the overhead of RLC/PDCP Header/Padding, etc., TTI is the length of TTI (here, 1ms), SE₂Is the spectral efficiency of stream 2 (for the CQI of the second GBR traffic stream).

And step 15, repeating the steps 11 to 14 when entering the next transmission time interval.

FIG. 3 is a schematic diagram of a simulation of an MBR with an equal GBR of 200Kbps, in accordance with an embodiment of the present invention; FIG. 4 is a schematic diagram of a simulation of an MBR with an equal GBR of 300Kbps, in accordance with an embodiment of the present invention; referring to fig. 3 and 4, in a scenario assumed in the first embodiment, simulation shows that, by using the scheduling scheme of GBR service of the present invention, the GBR rate of a user, especially an edge user, can be fully guaranteed. The simulation system adopted by the invention is an LTE (Long term evolution) commercial simulation system.

Example two

This example is for the case where the MBR of the user is larger than the GBR, i.e. the way new data is injected into the user data Buffer (BSR) per TTI by the MBR size.

Compared with the first embodiment, the basic idea is the same, namely, the number of RBs of users who reach the GBR rate is limited, and the saved RB resources are allocated to users who do not reach the GBR rate, so that users who do not reach the GBR rate can have more sufficient RB resources to transmit more data, the transmission rate is improved, and further the GBR is reached. The difference is that the users whose limited RB number reaches the GBR rate are refined, so that the RB resources can be fully utilized on the basis of ensuring the GBR rate of the edge users, the user rate is increased more, and more users can enjoy the service quality with higher rate on the basis of meeting the basic GBR rate.

Fig. 2 is a flowchart of a second embodiment of a guaranteed bit rate service scheduling method based on qos according to the present invention, and as shown in fig. 2, the guaranteed bit rate service scheduling method based on qos according to the present embodiment includes the following steps:

step 21, calculate the average historical throughput rate of all online users in the system (i.e. the average rate of users) over the current transmission time interval. The calculation formula is the same as that in step 11 of the previous embodiment, and is not described herein again.

And step 22, the base station queues the priority of each user terminal according to the scheduling algorithm. The calculation formula of the user priority is the same as that of step 12 in the first embodiment, and is not described here again.

And step 23, scheduling the retransmission users according to the user priorities obtained in the step 22, wherein the RB number of the retransmission users is determined by the retransmission TB Size and is not limited by GBR factors.

And step 24, scheduling the newly transmitted users according to the user priority obtained in the step 22.

Wherein, the step 24 specifically comprises the following steps:

step 241, queuing the newly transmitted user according to the user priority obtained in step 22.

In step 242, for a scheduled newly transmitted user Uei, if the average rate is less than GBR, the number of RBs is determined according to the data amount in the Buffer (BSR) of the user UeiAnd takes a smaller value with the current number of remaining RBs.The calculation formula is the same as that of step 142 in the first embodiment. If the average rate is equal to or greater than GBR, proceed to step 243.

Step 243, for a certain scheduled new transmission user Uei, a list of other online users in the cell is established, and the list excludes the retransmission user, the user on the last TTI, and the new transmission user that has been scheduled before Uei. If Uei the list of other users is not empty and the average rate of the users in the list is less than GBR, step 244 is entered, otherwise step 245 is entered.

Step 244, according to the smaller of the GBR requirement of the user Uei and the other BSRThe number of RBs allocated to Uei is determined and is taken to be a small value compared to the current number of remaining RBs.The calculation formula is the same as that in step 143 of the example. Step 246 is entered.

Step 245, determining the number of RBs allocated to Uei according to the data amount in the user Uei Buffer (BSR):and takes a smaller value with the current number of remaining RBs.The calculation formula is the same as that of step 142 in the first embodiment. Step 246 is performed.

Step 246, next new transmission user is scheduled according to the priority of the new transmission user in step 241, and steps 242 to 245 are repeated until the RB resources are used up or the new transmission user is scheduled completely.

And step 25, repeating the steps 21 to 24 when entering the next transmission time interval.

Referring to fig. 3, 4, 5, and 6, in the scenario assumed by the two embodiments, it is found through simulation that, by using the scheduling scheme of GBR service of the present invention, the GBR rate of a user, especially an edge user, can be ensured, and when MBR > GBR, the present invention can further improve the rate of an average, central user on the basis of ensuring the GBR rate of the edge user.

FIG. 5 is a schematic diagram of a simulation of an example of the present invention in which the MBR is 350Kbps and the GBR is 200 Kbps; FIG. 6 is a schematic diagram of a simulation of an example of the present invention in which the MBR is 400Kbps and the GBR is 300 Kbps; referring to fig. 5 and fig. 6, simulation finds that, by using the scheduling scheme of GBR service of the present invention, the GBR rate of a user, especially an edge user, can be ensured, and when MBR is greater than GBR, the present invention can further improve the rate of an average, central user on the basis of ensuring the GBR rate of the edge user.

Fig. 7 is a schematic structural diagram illustrating a configuration of a guaranteed bit rate service scheduler based on qos according to the present invention, and as shown in fig. 7, the guaranteed bit rate service scheduler based on qos according to the present invention includes a determining unit 71 and a resource scheduling unit 72; wherein the content of the first and second substances,

a determining unit 71, configured to determine a scheduling priority of each user according to an average rate of GBR services of all online users in a current TTI;

and a resource scheduling unit 72, configured to sequentially schedule the users according to the determined scheduling priorities, and allocate RB resources to the users.

The determining unit 71 further determines the scheduling priority of each user according to the PF scheduling algorithm and in combination with the QoS of the GBR service of each user.

FF＝FF_PF·FF_GBR

wherein the content of the first and second substances,". represents a dot product operation, Th_HisFor the average rate of GBR traffic in selected N TTIs before the current TTI, TB (i) is the TB successfully transmitted in N TTIs, and threshold GBR is the GBR reservation ratio set by the system.

The determining unit 71 further determines Th according to the following equation_His：

Wherein N is the selected window length(ii) a TB _ l to TB _ m are the size of the TB once scheduled by each GBR service flow l to m scheduled in the TTI; the values of ACK _1, ACK _2 and ACK _ m are 1 when the corresponding TB transmission succeeds, and 0 when the transmission fails.

The above-mentioned resource scheduling unit 72 further,

when the average rate of newly transmitted users is determined to be greater than or equal to GBR and less than or equal to MBR, and when other users with the rate less than GBR exist in the cell, the RB number is determined for the users according to the GBR and the smaller data volume in the BSR; when the average rates of other users in the cell are determined to be more than or equal to GBR, determining the number of RBs for the user according to the data volume in the BSR of the user;

The resource scheduling unit 72 further determines that the average rate of the newly transmitted user is greater than or equal to GBR and less than or equal to MBR, and adds the user to the GBR-reached user list of the limited number of RBs when the BSR of the user is greater than GBR; and when the RB resources of the other users are still remained after the allocation is finished, allocating the RB resources for the users in the user list reaching the GBR.

It should be understood by those skilled in the art that the apparatus for guaranteed bit rate service based on qos shown in fig. 7 is designed to implement the aforementioned method for guaranteed bit rate service based on qos, and the functions of each processing unit in the apparatus shown in fig. 7 can be understood with reference to the descriptions in the first to second embodiments, and the functions of each processing unit can be implemented by a program running on a processor or by a specific logic circuit.

The above description is only a preferred embodiment of the present invention, and is not intended to limit the scope of the present invention.

Claims

1. A method for guaranteed bit rate service scheduling based on quality of service, the method comprising:

determining the scheduling priority of each user according to the average rate of guaranteed bit rate GBR services of all online users in the current transmission time interval TTI;

sequentially scheduling the retransmission users according to the determined scheduling priority order, wherein the RB number of the resource blocks of the retransmission users is determined by the size of a retransmission transport block TB and is not limited by GBR factors; and when the RB resources are still left after the retransmission user is scheduled, scheduling the newly transmitted users in sequence according to the determined scheduling priority order, and allocating the RB number to the newly transmitted users according to the service quality determined by the GBR service of the newly transmitted users.

2. The method of claim 1, wherein the determining the scheduling priority of each user according to the average rate of the GBR traffic of all online users in the current TTI is as follows:

3. The method according to claim 2, wherein the determining the scheduling priority of each user according to PF scheduling algorithm and in combination with the QoS of GBR traffic of each user is specifically performed by determining the scheduling priority FF of a user according to the following formula:

FF=FF_PF·FF_GBR

wherein the content of the first and second substances,". represents a dot product operation, Th_HisFor the average rate of GBR service in selected N TTIs before the current TTI, TB (i) is the size of the transport block TB successfully transmitted in the ith TTI of the N TTIs, and ThresholdGBR is the GBR reserved proportion set by the system; GBR in the exponent bit numerator represents the guaranteed bit rate of the system configuration.

4. The method of claim 3, wherein the Th is_HisDetermined as follows:

n is the selected window length; m is the number of GBR service flows of the user in N TTIs; TB _1 to TB _ m are the TB sizes of the GBR service flows scheduled in the TTI from 1 to m in one-time scheduling; the values of ACK _1, ACK _2 and ACK _ m areIf the corresponding TB transmission is successful, the value is 1, and if the transmission is failed, the value is 0.

5. The method according to any of claims 1 to 4, wherein the number of RBs allocated to the newly transmitted user according to the determined quality of service of the GBR service of the newly transmitted user is:

when the average rate of newly transmitted users is determined to be greater than or equal to GBR and less than or equal to MBR, and when other users with the rate less than GBR exist in the cell, the RB number is determined for the users according to the GBR and the smaller data volume in the BSR; when the average rates of other users in the cell are determined to be more than or equal to GBR, determining the number of RBs for the user according to the data volume in the BSR of the user; the other users are all online users of a service cell to which the newly transmitted user belongs, and the newly transmitted user, the retransmission user and the scheduled user before the user is scheduled in the current TTI are removed, and the remaining users behind the user are removed;

6. The method of claim 5, wherein when it is determined that the average rate of the newly transmitted users is greater than or equal to GBR and less than or equal to MBR, the method further comprises:

7. A guaranteed bit rate service scheduling device based on service quality is characterized in that the device comprises a determining unit and a resource scheduling unit; wherein the content of the first and second substances,

a determining unit, configured to determine a scheduling priority of each user according to an average rate of guaranteed bit rate GBR services of all online users within a current transmission time interval TTI;

the resource scheduling unit is used for sequentially scheduling the retransmission users according to the determined scheduling priority order, and the RB number of the retransmission users is determined by the size of a retransmission transport block TB and is not limited by GBR factors; and when the RB resources are still left after the retransmission user is scheduled, scheduling the newly transmitted users in sequence according to the determined scheduling priority order, and allocating the RB number to the newly transmitted users according to the service quality determined by the GBR service of the newly transmitted users.

8. The apparatus of claim 7, wherein the determining unit further determines the scheduling priority of each user according to a proportional fair PF scheduling algorithm in combination with the QoS of GBR traffic of each user.

9. The apparatus of claim 8, wherein the determining unit is further configured to determine the scheduling priority, FF, of the user according to the following equation:

FF=FF_PF·FF_GBR

10. The apparatus of claim 9, wherein the determining unit further determines Th according to the following equation_His：

Wherein N is the selected window length; m is the number of GBR service flows of the user in N TTIs; TB _1 to TB _ m are the TB sizes of the GBR service flows scheduled in the TTI from 1 to m in one-time scheduling; the values of ACK _1, ACK _2 and ACK _ m are 1 when the corresponding TB transmission succeeds, and 0 when the transmission fails.

11. The apparatus according to any of claims 7 to 10, wherein the resource scheduling unit further,

12. The apparatus of claim 11, wherein the resource scheduling unit further determines that the average rate of the newly transmitted user is greater than or equal to GBR and less than or equal to MBR, and adds the user to the GBR-reached user list of the limited number of RBs when the BSR of the user is greater than GBR; and when the RB resources of the other users are still remained after the allocation is finished, allocating the RB resources for the users in the user list reaching the GBR.