WO2016000366A1 - 一种资源调度方法、装置和计算机可读存储介质 - Google Patents

一种资源调度方法、装置和计算机可读存储介质 Download PDF

Info

Publication number
WO2016000366A1
WO2016000366A1 PCT/CN2014/089822 CN2014089822W WO2016000366A1 WO 2016000366 A1 WO2016000366 A1 WO 2016000366A1 CN 2014089822 W CN2014089822 W CN 2014089822W WO 2016000366 A1 WO2016000366 A1 WO 2016000366A1
Authority
WO
WIPO (PCT)
Prior art keywords
scheduling
user terminal
network
video
indicator
Prior art date
Application number
PCT/CN2014/089822
Other languages
English (en)
French (fr)
Inventor
左雯
鲁晓牧
王宁
陈雨辰
刘贵忠
Original Assignee
中兴通讯股份有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 中兴通讯股份有限公司 filed Critical 中兴通讯股份有限公司
Publication of WO2016000366A1 publication Critical patent/WO2016000366A1/zh

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W72/00—Local resource management
    • H04W72/04—Wireless resource allocation

Definitions

  • the present invention relates to the field of wireless communication technologies, and in particular, to a resource scheduling method, apparatus, and computer readable storage medium.
  • LTE Long Term Evolution
  • MIMO multiple-input multiple-output
  • the channel quality fluctuates with time. This fluctuation will eventually affect the effective data transfer rate of the application layer.
  • the application layer data transmission rate cannot meet the required code rate of video continuous playback, the playback of the video stream may be interrupted. Frequent or prolonged video playback interruptions can seriously affect the service experience (QoE, Quality of Experience) that video streaming users receive.
  • QoE Quality of Experience
  • embodiments of the present invention are expected to provide a resource scheduling method, apparatus, and computer readable storage medium.
  • the user terminal corresponding to the maximum scheduling indicator value is selected in the scheduling matrix, and the resource block RB corresponding to the maximum scheduling indicator value is allocated to the selected user terminal.
  • the method further includes:
  • the video playback buffer information of the selected user terminal is updated, and the selected user terminal is reclassified according to the updated video playback buffer information.
  • the method further includes:
  • the classifying the user terminal according to the video play cache information of the user terminal including:
  • the corresponding user terminal Comparing the number of video frames of the video playback buffer of the user terminal with the preset first threshold and the second threshold. If the number of video frames of the video playback buffer of the user terminal is less than the first threshold, the corresponding user terminal is divided into the first a set; if the number of video frames of the video playback buffer of the user terminal is greater than the first threshold and less than the second threshold, the corresponding user terminal is divided into the second set; if the number of video frames of the video playback buffer of the user terminal is greater than the first The second threshold divides the corresponding user terminal into the third set.
  • the determining, according to the classification result of the user terminal, determining the working state of the network including:
  • determining the corresponding scheduling set according to the working state of the network includes:
  • the scheduling indicator corresponding to the scheduling set is determined by:
  • r(k,n) represents the transmission capability of the kth user terminal in the scheduling set on the nth RB
  • r min (k) represents the minimum average transmission rate required by the user terminal to ensure continuous playback.
  • RS(k) represents the size of the video information required for the correct decoding of the frame F(k)
  • F(k) represents the most urgent frame among all the video frames that the kth user terminal has not decoded in the scheduling set.
  • the continuous playing time that the video data in the cache of the user terminal k can support is displayed, and B(k) represents the video playing cache information of the kth user terminal in the scheduling set.
  • the value of the scheduling indicator of the user terminal in the scheduling set is calculated by:
  • M(k,n) FI(k)*m(k,n),k ⁇ S,n ⁇ S ava
  • S represents a scheduling set, indicating a set of RBs that can be currently used
  • M(k, n) represents a value of a scheduling indicator of the kth user terminal in the scheduling set on the nth RB
  • FI(k) represents a fairness parameter.
  • IT(k) represents the past total interruption time of the kth user terminal
  • * indicates multiplication.
  • the user terminal corresponding to the maximum scheduling indicator value is selected in the scheduling matrix, and the RB corresponding to the maximum scheduling indicator value is allocated to the selected user terminal, including:
  • n * th RBs in the scheduling matrix that satisfy the following conditions are assigned to the k * th user:
  • the classification unit is configured to classify the user terminal according to the video play cache information of the user terminal;
  • a state determining unit configured to determine a working state of the network according to a classification result of the user terminal
  • the scheduling set and the scheduling indicator determining unit are configured to determine a corresponding scheduling set and a scheduling indicator according to the working state of the network;
  • the scheduling indicator value calculation unit is configured to separately calculate the value of the scheduling indicator of all user terminals in the scheduling set, and calculate the value of the scheduling indicator of each user terminal to form a scheduling matrix;
  • the allocation unit is configured to select a user terminal corresponding to the maximum scheduling indicator value in the scheduling matrix, and allocate the resource block RB corresponding to the maximum scheduling indicator value to the selected user terminal.
  • the device further includes an update unit configured to update the video of the selected user terminal after the allocation unit allocates the RB corresponding to the maximum scheduling indicator value to the selected user terminal.
  • the cache information is played, and the classification unit is notified to reclassify the selected user terminal according to the updated video play cache information.
  • the device further includes a determining unit, configured to: after the classifying unit reclassifies the selected user terminal, the determining unit determines whether there are any unassigned RBs in the network, and the determination result is Sometimes, the state determining unit, the scheduling set and the scheduling indicator determining unit, the scheduling indicator value calculating unit, and the allocating unit are notified;
  • the state determining unit determines the working state of the network according to the classification result of the user terminal; the scheduling set and the scheduling indicator determining unit re-determine the corresponding scheduling set and the scheduling indicator according to the working state of the network; The calculation unit recalculates the location in the scheduling set.
  • the user equipment has a scheduling indicator value, and the allocation unit re-selects the user terminal corresponding to the maximum scheduling indicator value in the scheduling matrix formed by the value of the scheduling indicator, and allocates the RB corresponding to the maximum scheduling indicator value to the RB. Selected user terminal;
  • the classification unit is configured to compare the number of video frames of the video playback buffer of the user terminal with the preset first threshold and the second threshold, if the number of video frames of the video playback buffer of the user terminal is smaller than the first
  • the threshold is used to divide the corresponding user terminal into the first set; if the number of video frames of the video playback buffer of the user terminal is greater than the first threshold and less than the second threshold, the corresponding user terminal is divided into the second set; If the number of video frames of the video play buffer of the terminal is greater than the second threshold, the corresponding user terminal is divided into the third set.
  • the state determining unit is configured to: when the first set is not empty, determine that the network is in the first state; when the first set is empty, and the second set is not empty, determine the network. In a second state; determining that the network is in a third state when both the first set and the second set are empty.
  • the scheduling set and the scheduling indicator determining unit are configured to: when the network is in the first state, determine that the corresponding scheduling set is the first set; and when the network is in the second state, determine that the corresponding scheduling set is The second set; when the network is in the third state, determining that the corresponding scheduling set is the third set;
  • r(k,n) represents the transmission capability of the kth user terminal in the scheduling set on the nth RB
  • r min (k) represents the minimum average transmission rate required by the user terminal to ensure continuous playback.
  • RS(k) represents the size of the video information required for the correct decoding of the frame F(k)
  • F(k) represents the most urgent frame among all the video frames that the kth user terminal has not decoded in the scheduling set.
  • the continuous playing time that the video data in the cache of the user terminal k can support is displayed, and B(k) represents the video playing cache information of the kth user terminal in the scheduling set.
  • the scheduling indicator value calculation unit is configured to calculate the value of the scheduling indicator of the user terminal in the scheduling set by:
  • M(k,n) FI(k)*m(k,n),k ⁇ S,n ⁇ S ava
  • S represents a scheduling set, indicating a set of RBs that can be currently used
  • M(k, n) represents a value of a scheduling indicator of the kth user terminal in the scheduling set on the nth RB
  • FI(k) represents a fairness parameter.
  • IT(k) represents the past total interruption time of the kth user terminal
  • * indicates multiplication.
  • the allocation unit is configured to allocate the n * th RBs in the scheduling matrix that satisfy the following conditions to the k * th user:
  • the embodiment of the present invention further provides a computer readable storage medium, the storage medium comprising a set of computer executable instructions, the instructions being used to execute the resource scheduling method according to the embodiment of the present invention.
  • a resource scheduling method, apparatus, and computer readable storage medium comprehensively consider the channel quality of each user terminal on the RB, the number of video frames in the user terminal playing buffer, and the fairness between users. Factors such as sex can effectively avoid the interruption of video playback or shorten the duration of the interruption, and also ensure that the spectral efficiency of the system does not drop significantly.
  • FIG. 1 is a flowchart of a resource scheduling method according to an embodiment of the present invention
  • FIG. 2 is a schematic structural diagram of a resource scheduling apparatus according to an embodiment of the present invention.
  • FIG. 3 is a comparison diagram of average interruption durations of six algorithms in a case where a user requests video code rate difference is large according to an embodiment of the present invention
  • FIG. 4 is a comparison diagram of average interruption durations of six algorithms in a case where a user requests a video code rate difference is small according to an embodiment of the present invention
  • FIG. 5 is a diagram showing the cumulative interrupt duration of each user with the number of playback frames under the six algorithms in the embodiment of the present invention
  • FIG. 6 is a diagram showing changes in system throughput over time in six algorithms according to an embodiment of the present invention.
  • FIG. 7 is a performance diagram of an algorithm using different cache information feedback periods according to an embodiment of the present invention.
  • the number of RBs that the base station can allocate is N.
  • the user terminal requests the video stream service from the base network through the base station. After the video data arrives at the base station, the video data is first stored in the corresponding transmission buffer of the user, and then sent to the user terminal according to the result of the base station scheduling. In order to facilitate transmission, the video data is divided into a plurality of small data packets for transmission in the network. In the embodiment of the present invention, when all data packets of only one video frame are correctly received, the video frame can be successfully decoded and placed in the play buffer for playback.
  • TTI Transmission Time Interval
  • IT(k, t) represents the cumulative interruption duration of the user terminal k at time t
  • IT(k, t-1) represents the cumulative interruption duration of the user terminal k at time t-1
  • B(k, t) Indicates that the user terminal reports at time t The number of video frames to be played in the cache. Since only the buffer state of the current time is considered in the algorithm, B(k) is substituted for B(k, t).
  • the purpose of implementing the embodiments of the present invention is to minimize IT k , even if the cumulative interrupt duration of the user terminal k is minimized as much as possible, thereby effectively avoiding the occurrence of video playback interruption or shortening the duration of the interruption.
  • the embodiment of the present invention provides a resource scheduling method, as shown in FIG. 1 , the method mainly includes:
  • Step 101 Classify the user terminal according to the video play cache information of the user terminal.
  • the video playback buffer information of the user terminal mainly refers to the number of video frames in the video playback buffer of the user terminal; the user terminal can periodically report the respective video playback buffer information.
  • the embodiment of the present invention proposes two thresholds B DownTH and B UpTH , wherein B DownTH represents a set lower limit of the play buffer, referred to as a first threshold, and B UpTH represents a set play buffer upper limit, referred to as a second threshold.
  • B DownTH represents a set lower limit of the play buffer
  • B UpTH represents a set play buffer upper limit, referred to as a second threshold.
  • the number of video frames in the video play buffer reported by each user terminal is compared with the B DownTH and B UpTH , and the user terminal is classified according to the comparison result.
  • the user terminal can be divided into the following three sets according to the comparison result:
  • B(k) represents the video play buffer information of the user terminal k, that is, the video of the user terminal k The number of video frames in the playback buffer.
  • a first set S D to B (k) set ⁇ B DownTH the user terminal i.e. B (k) is less than the user terminal B DownTH divided into a first set S D, the user terminal first in the set S D B (k) is less than B DownTH, video frames represented by the video playback buffer of the user terminal in the first set S D are rarely, a user of the first terminal set S D there is a big possibility of interruption of playback;
  • the second set of S M is B DownTH ⁇ B set of user terminals (k) ⁇ B UpTH of, i.e. B (k) greater than or equal B DownTH, or less B UpTH user terminal S M divided into a second set, the first B (k) of the two user terminals in the set of S M is greater than or equal B DownTH, or less B UpTH, video frames represented by the video player buffer less a user terminal in the second set of S M, the second The user terminal in the set S M has a certain possibility of interruption of the playing process, but the possibility is smaller than the user terminal in the first set S D ;
  • the third set is the set S U B (k)> B UpTH user terminal, i.e. B (k) is greater than the user terminal B UpTH divided into a third set S U, B is a user terminal a third set of S U (k) is greater than the number of frames B UpTH represents more video playback buffer of the user terminal in the third set of S U, S U third set of user terminals will not be interrupted during playback of a period of time.
  • Step 102 Determine, according to the classification result of the user terminal, the working state of the network.
  • the working state of the current network is determined according to the distribution of the user terminals in the network (ie, the foregoing first set S D , the second set S M , and the third set S U ), so as to adopt different scheduling policies.
  • the specific network working status can be divided into:
  • Case a shows a first state, when the first set S D is not empty, determine that the network is in a first state, a first state representing the video frame buffer video playback in the user terminal S D in the small, S D
  • the scheduling optimization goal in this case is to avoid the occurrence of the playback interruption or reduce the duration of the playback interruption;
  • Case b represents a second state, when the first set S D is empty, and the second set S M is not empty, it is determined that the network is in the second state, and the second state represents the video play buffer of the user terminal in the S M the fewer the number of frames of video, the user terminal S M tends to occur interruption of playback, the scheduling optimization goal is an aspect of this case to avoid playback disruption occurs, on the other hand to ensure single spectrum efficiency will not Serious decline in considering the user experience;
  • Case c represents a third state.
  • the third state indicates that all user terminals in the network belong to the third set.
  • the S U that is, the number of video frames in the video play buffer of the user terminal in the third set S U is large, and the playback process is not interrupted for a period of time.
  • the scheduling optimization goal is to improve the system spectrum efficiency. .
  • Step 103 Determine a corresponding scheduling set and scheduling indicator according to the working state of the network.
  • the scheduling set when the network is in different working states is:
  • the determined corresponding scheduling set is the first set S D ; when the network is in the second state Case b, the determined corresponding scheduling set is the second set S M ; When the network is in the third state Case c, the determined corresponding scheduling set is the third set S U .
  • the scheduling indicator corresponding to the scheduling set is:
  • the scheduling indicator is r min (k), r min (k) is for the user terminal in the first set S D ; when the network is in the second state Case b
  • the scheduling indicator is For the user terminal in the second set S M; in the third state when the network scheduling indicators Case c r (k, n), r (k, n) for the third user terminal in the set S U.
  • r(k,n) represents the transmission capability of the kth user terminal on the nth RB (ie, the transport code stream);
  • r min (k) represents the minimum average transmission code required by the user terminal to ensure continuous playback.
  • Rate which can be expressed as It is assumed that F(k) is the most urgent frame among all video frames that the user terminal k has not decoded, and RS(k) indicates the size of the video information required for the correct decoding of the frame F(k) (the unit is the number of bits or bytes) ), for the experience value, Indicates the continuous play time that the video data in the cache of the user terminal k can support, and FR denotes the frame rate, which is a known variable at the time of system initialization.
  • Step 104 Calculate the value of the scheduling indicator of all user terminals in the scheduling set, and calculate the value of the scheduling indicator of each user terminal to form a scheduling matrix.
  • the current RB set is represented by S ava , and the scheduling index is calculated for k ⁇ S, n ⁇ S ava .
  • the calculation method of the scheduling indicator is:
  • M(k,n) FI(k)*m(k,n),k ⁇ S,n ⁇ S ava
  • S represents a scheduling set
  • M(k, n) represents a value of a scheduling indicator of the kth user terminal in the scheduling set on the nth RB
  • FI(k) represents a fairness parameter
  • IT(k) represents the past total interruption duration of the kth user terminal
  • Step 105 Select a user terminal corresponding to the maximum scheduling indicator value in the scheduling matrix, and allocate the RB corresponding to the maximum scheduling indicator value to the selected user terminal.
  • n * th RBs in the scheduling matrix that satisfy the following conditions are allocated to the k * th user:
  • IT(k) represents the past total interruption duration of the kth user terminal.
  • Step 106 Update the video play cache information of the scheduled user terminal (ie, the user terminal selected in step 105) according to the resource allocation result, and reclassify the user terminal according to the updated video play cache information.
  • the video play buffer information of the corresponding user terminal needs to be updated according to RS(k) and R(k), wherein R(k) is available for all current RBs allocated to the user terminal k.
  • B (k) indicates the video playback buffer information before the update (ie, the number of video frames in the video playback buffer before the update). It is assumed that F(k) is the most urgent frame among all video frames that the user terminal k has not decoded, and RS(k) represents the size of the video information required for the correct decoding of the frame F(k).
  • step 107 it is determined whether there are any unallocated RBs in the network, and if yes, return to step 102; otherwise, the process ends.
  • the next RB is allocated according to the same operations in steps 102-106 until all the RBs in the network are allocated.
  • the resource scheduling method in the embodiment of the present invention may be applied to a base station, such as may be applied to a home base station.
  • the embodiment of the present invention further provides a resource scheduling apparatus.
  • the apparatus includes:
  • the classification unit 10 is configured to classify the user terminal according to the video play cache information of the user terminal;
  • the state determining unit 20 is configured to determine an operating state of the network according to the classification result of the user terminal;
  • the scheduling set and scheduling indicator determining unit 30 is configured to determine a corresponding scheduling set and scheduling indicator according to the working state of the network
  • the scheduling indicator value calculation unit 40 is configured to separately calculate the value of the scheduling indicator of all the user terminals in the scheduling set, and calculate the value of the scheduling indicator of each user terminal to form a scheduling matrix;
  • the allocating unit 50 is configured to select a user terminal corresponding to the maximum scheduling indicator value in the scheduling matrix, and allocate the RB corresponding to the maximum scheduling indicator value to the selected user terminal.
  • the apparatus further includes an updating unit 60 configured to update the selected user after the allocation unit 50 assigns the RB corresponding to the maximum scheduling indicator value to the selected user terminal.
  • the video of the terminal plays the cache information, and the notification classification unit 10 reclassifies the selected user terminal according to the updated video play cache information.
  • the apparatus further includes a determining unit 70 configured to, after the classifying unit 10 reclassifies the selected user terminal, the determining unit 70 determines whether there are any unassigned RBs in the network, and determines the result. For example, the notification state determining unit 20, the scheduling set and scheduling indicator determining unit 30, the scheduling index value calculating unit 40, the allocating unit 50, and the updating unit 60; the state determining unit 20 re-determines the network based on the classification result of the user terminal.
  • the working state; the scheduling set and scheduling indicator determining unit 30 will re-determine the corresponding scheduling set and the scheduling indicator according to the working state of the network determined by the state determining unit 20; the scheduling indicator value calculating unit 40 will re-determine according to the scheduling set and the scheduling indicator.
  • the scheduling set and the scheduling indicator determined by the unit 30 are used to calculate the value of the scheduling indicator of all the user terminals in the scheduling set; the allocating unit 50 selects the user terminal corresponding to the maximum scheduling indicator value in the scheduling matrix formed by the scheduling indicator value.
  • updating Element 60 will again update the selected user terminal cache information video playback, and notifies the video classification unit 10 in accordance with the updated Playing cache information to reclassify selected user terminals;
  • Each of the above modules is executed until all RBs in the network are allocated.
  • the classification unit 10 is configured to compare the number of video frames of the video playback buffer of the user terminal with the preset first threshold and the second threshold, if the number of video frames of the video playback buffer of the user terminal is smaller than the first If the threshold is greater than the first threshold and less than the second threshold, the corresponding user terminal is divided into the second set; If the number of video frames of the video playback buffer of the user terminal is greater than the second threshold, the corresponding user terminal is divided into the third set.
  • the state determining unit 20 is configured to determine that the network is in the first state when the first set is not empty; and to determine when the first set is empty and the second set is not empty The network is in the second state; when both the first set and the second set are empty, it is determined that the network is in the third state.
  • the scheduling set and scheduling indicator determining unit 30 is configured to determine that the corresponding scheduling set is the first set when the network is in the first state, and determine the corresponding scheduling set when the network is in the second state. The second set; when the network is in the third state, determining that the corresponding scheduling set is the third set;
  • r(k,n) represents the transmission capability of the kth user terminal in the scheduling set on the nth RB
  • r min (k) represents the minimum average transmission rate required by the user terminal to ensure continuous playback.
  • RS(k) represents the size of the video information required for the correct decoding of the frame F(k)
  • F(k) represents the most urgent frame among all the video frames that the kth user terminal has not decoded in the scheduling set.
  • the continuous play time that the video data in the cache of the user terminal k can support is displayed, and B(k) represents the video play cache information of the kth user terminal in the dispatch set.
  • the scheduling indicator value calculation unit 40 is configured to calculate the value of the scheduling indicator of the user terminal in the scheduling set by:
  • M(k,n) FI(k)*m(k,n),k ⁇ S,n ⁇ S ava
  • S represents a scheduling set, indicating a set of RBs that can be currently used
  • M(k, n) represents a value of a scheduling indicator of the kth user terminal in the scheduling set on the nth RB
  • FI(k) represents a fairness parameter.
  • IT(k) represents the past total interruption time of the kth user terminal
  • * indicates multiplication.
  • the allocating unit 50 is configured to allocate the n * th RBs in the scheduling matrix that satisfy the following conditions to the k * th user:
  • the foregoing classification unit 10, the state determining unit 20, the scheduling set and scheduling index determining unit 30, the scheduling index value calculating unit 40, the allocating unit 50, the updating unit 60, and the determining unit 70 may be centrally located by the resource scheduling device.
  • a CPU Central Processing Unit
  • MPU Microprocessor
  • DSP Digital Signal Processor
  • FPGA Field-Programmable Gate Array
  • the resource scheduling apparatus of the embodiment of the present invention may be applied to a base station, such as may be applied to a home base station. Therefore, the embodiment of the present invention further provides a base station that includes the resource scheduling apparatus.
  • a base station that includes the resource scheduling apparatus.
  • the resource scheduling method and device for the video stream in the LTE network provided by the embodiment of the present invention firstly provide an optimization target type for minimizing the interruption time of all user terminals in the network, and According to this goal, the scheduling order of the user and the allocation strategy of the wireless resource are determined.
  • the playback buffer information periodically reported by the user terminal specifically, the number of video frames to be played in the playback buffer
  • the user terminals can be classified into three types: user terminals that are about to have or have been interrupted by play; The user terminal and the user terminal that does not have a playback interruption in a short period of time.
  • the distribution of three types of user terminals in the network can be used to describe the current working state of the network.
  • the adopted policies are dynamically adjusted according to different network states.
  • the network status can be divided into three categories: a.
  • the user terminal in the network that has a video playback process that is about to be interrupted or has been interrupted.
  • the network has some user terminals whose cache has fewer video frames and has a higher probability.
  • the playback process is interrupted.
  • the number of cached video frames of all user terminals in the network is large, which can effectively resist the fluctuation of the network transmission capability, and the playback process will not be interrupted for a period of time. In each case, a specific schedule set and scheduling metrics are determined.
  • each user terminal in the scheduling set calculates its scheduling indicator value on each available RB to form a scheduling matrix.
  • the user terminal having the largest scheduling indicator value and the corresponding RB are searched in the scheduling matrix, and the RB is allocated to the user terminal. After that, the playback buffer information of the user terminal needs to be updated according to the allocation situation.
  • the queue packet scheduling may be performed by using a first in first out (FIFO) manner.
  • FIFO first in first out
  • the resource scheduling method and device in the embodiment of the present invention comprehensively considers the channel quality of each user terminal on the RB, the number of video frames in the user terminal playing buffer, and the fairness between users, etc., thereby effectively avoiding video.
  • the occurrence of a playback interrupt or the duration of the interrupt is shortened, and the spectral efficiency of the system is not significantly reduced.
  • the user location, the moving speed (0, 3 km/h), and the service type used by the user are randomly generated in the network, wherein the macro base station user randomly generates voice or data services, and the home base station user randomly generates data or video services.
  • the amount of data to be downloaded is randomly generated in the (0, 10M) interval, which is accurate to 1k.
  • the dynamic factor ⁇ i is randomly selected within the set ⁇ 2, 5, 10, 15 ⁇ , and the encoding rate REQ i is randomly selected in the set ⁇ 0.8 Mbps, 0.9 Mbps, 1 Mbps, 1.1 Mbps, 1.2 Mbps ⁇ .
  • the simulation test environment of the macro cell refers to the TR 25.814 protocol implementation, and the simulation test environment of the home base station adopts the simulation environment setting for the home base station in TS 36.814.
  • the detailed simulation parameter settings are shown in Table 2, Table 3, and Table 4.
  • the existing algorithm is the MOSMax algorithm, and the algorithm corresponding to the embodiment of the present invention is the proposed algorithm.
  • Figure 3 compares the total interrupt durations of the six algorithms in the environment shown in the video stream request configuration table. These six algorithms are: RR (Polling) algorithm, MAXCI (Maximum Carrier to Interference Ratio). Algorithm, PF (proportional fairness) algorithm, RG (equalization) algorithm, PB algorithm, and PBA algorithm proposed by the embodiment of the present invention.
  • RR Policy
  • MAXCI Maximum Carrier to Interference Ratio
  • Algorithm Algorithm
  • PF proportional fairness
  • RG equalization
  • PB Packequalization
  • PBA PBA algorithm proposed by the embodiment of the present invention.
  • the RR can effectively ensure the fairness of resource allocation among users, but because the requested video sequence is different, and the channel quality of each user is also different, the quality of service obtained by the user is relatively better. Great difference. Among them, users who request large-rate video or poor channel quality will experience longer playback interruption when watching video.
  • the PF algorithm shows better fairness in the first four users (requesting video sequences with similar bit rates), but it shows a significant bias for the last two users who request a smaller bit rate video.
  • the RG algorithm can effectively avoid the defects of the PF algorithm by considering the specific transmission rate requirements of each video stream, and can obtain good performance. However, the RG guarantees the long-term transmission rate, and does not consider the real-time bit rate requirement of video smooth playback. , still cause video playback to be interrupted.
  • PB considers the real-time status of the cache, which can better meet the real-time bit rate requirements of smooth video playback. However, PB considers that the data in the cache is in units of bits, and not in units of video frames. This will cause PB and PF to have the same defect, which will cause partiality to users with small code rates.
  • the PBA algorithm proposed by the embodiment of the present invention can obtain good performance, can not only effectively reduce the interruption duration, but also ensure fairness between users.
  • Figure 4 compares the total interrupt duration for each of the six algorithms in the environment shown in the video stream request configuration table 2. Due to the small difference in the code rate requested by the user under such a configuration condition, the performance improvement is not as obvious as in the configuration 1 environment. However, as can be seen from Figure 4, the performance of PBA is still superior to other comparison algorithms.
  • FIG. 5 illustrates the cumulative interruption duration of each user's playback time when six algorithms are employed in the environment shown in the video stream request configuration table 1.
  • the interrupt time of each user is continuously increased. Long, indicating the fairness of RR resource allocation.
  • the resource allocation bias is obvious, that is, the user with poor channel conditions waits for a long time before getting the service. At the same time, you can see the bias of PF and PB for small code rate users.
  • the PBA algorithm proposed by the embodiment of the present invention can ensure that each user gets the fairness of the service and control the interruption time to a lower level.
  • Figure 6 illustrates the throughput of the six algorithms in the environment shown in the video stream request configuration table 1. It can be seen that although the throughput performance of PBA is not as good as MAXCI, PF and RR, it is worse than RG and PB. This result shows that PBA does not seriously reduce the throughput of the system while ensuring user QoE.
  • Figure 7 illustrates the performance variation of the PBA algorithm when different buffer state feedback cycles are used. Since the frame rate of video playback is usually chosen to be 30 frames/second or lower, the duration of one frame playback is 33 ms or longer. Compared with the CQI feedback period of 1 ms, the user terminal playback buffer state change is very slow. In this way, in order to reduce the overhead caused by the buffer status information feedback, the feedback period can be appropriately amplified. The specific experimental results can be seen in Figure 7. When the feedback period is 1 TTI, 11 TTI and 22 TTI, the performance difference is not large. However, when it exceeds 33 TTIs, it deteriorates as the feedback period becomes longer. Therefore, the buffer feedback period can indeed select a larger value to reduce the overhead of signaling feedback.
  • the embodiment of the present invention comprehensively considers the channel quality of each user terminal on the RB, the number of video frames in the user terminal playing buffer, and the fairness between users, etc., which can effectively avoid interruption of video playback.
  • the occurrence or shortening of the duration of the interruption also ensures that the spectral efficiency of the system does not decrease significantly.
  • the embodiment of the present invention can effectively reduce the interruption of the video stream user during video watching.
  • the duration can also solve the bias problem for specific users in the comparison algorithm. For example, PF and PB will favor users who request small rate video.
  • MAXCI will favor users with better channel quality.
  • the embodiment of the present invention further provides a computer readable storage medium, the storage medium comprising a set of computer executable instructions, the instructions being used to execute the resource scheduling method according to the embodiment of the present invention.
  • embodiments of the present invention can be provided as a method, system, or computer program product. Accordingly, the present invention can take the form of a hardware embodiment, a software embodiment, or a combination of software and hardware. Moreover, the invention can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage and optical storage, etc.) including computer usable program code.
  • the computer program instructions can also be stored in a computer readable memory that can direct a computer or other programmable data processing device to operate in a particular manner, such that the instructions stored in the computer readable memory produce an article of manufacture comprising the instruction device.
  • the apparatus implements the functions specified in one or more blocks of a flow or a flow and/or block diagram of the flowchart.
  • These computer program instructions can also be loaded onto a computer or other programmable data processing device such that a series of operational steps are performed on a computer or other programmable device to produce computer-implemented processing for execution on a computer or other programmable device. Instructions are provided for implementation in the stream The steps of a function specified in one or more processes and/or block diagrams in one or more blocks.
  • the embodiment of the present invention comprehensively considers the channel quality of each user terminal on the RB, the number of video frames in the user terminal playing buffer, and the fairness between users, etc., which can effectively avoid the occurrence of video playback interruption or shorten the interruption.
  • the duration of the system also ensures that the spectral efficiency of the system does not decrease significantly.
  • the embodiment of the present invention can effectively reduce the interruption duration of the video stream user watching the video, and can also solve the specific algorithm in the comparison algorithm. Hemiplegia problems, such as PF and PB will favor users who request small bit rate video, MAXCI will favor users with better channel quality.

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Two-Way Televisions, Distribution Of Moving Picture Or The Like (AREA)
  • Data Exchanges In Wide-Area Networks (AREA)

Abstract

本发明公开了一种资源调度方法，包括：根据用户终端的视频播放缓存信息，对用户终端进行分类（101）；根据用户终端的分类结果，确定网络的工作状态（102）；根据网络的工作状态，确定相应的调度集和调度指标（103）；分别计算调度集中的所有用户终端的调度指标取值，计算所得各用户终端的调度指标取值形成调度矩阵（104）；在调度矩阵中选择最大调度指标取值所对应的用户终端，并将最大调度指标取值所对应的资源块(RB)分配给所选的用户终端（105）。本发明还公开了一种资源调度装置和计算机可读存储介质。

Description

一种资源调度方法、装置和计算机可读存储介质 技术领域
本发明涉及无线通信技术领域，尤其涉及一种资源调度方法、装置和计算机可读存储介质。
背景技术
持续增高的网络传输速率和日益提高的无线终端性能为提供高质量视频服务提供了可能。在当今无线网络中，视频数据已经成为一种主要的数据来源。为了满足这种需求，第三代合作伙伴计划(3GPP，3rd Generation Partnership Project)提出了长期演进(LTE，Long Term Evolution)标准。这种标准的物理层基于正交频分复用(OFDM，Orthogonal Frequency Division Multiplexing)传输技术实现，并采用了多输入多输出(MIMO，Multiple-Input Multiple-Output)等多种传输增强技术，可以有效的提高物理层的传输速率。具体来说，LTE中上行传输速率可以达到50Mbit/s，下行可以达到100Mbit/s。
然而，由于无线信道的固有特性，信道质量会随时间波动。这种波动最终会影响到应用层的有效数据传输速率。当应用层数据传输速率不能满足视频连续播放的所需码率时，视频流的播放会出现中断。频繁或长时间的视频播放中断会严重影响视频流用户得到的服务体验(QoE，Quality of Experience)。
发明内容
为解决现有存在的技术问题，本发明实施例期望提供一种资源调度方法、装置和计算机可读存储介质。
本发明实施例提供的一种资源调度方法，包括：
根据用户终端的视频播放缓存信息，对用户终端进行分类；
根据用户终端的分类结果，确定网络的工作状态；
根据网络的工作状态，确定相应的调度集和调度指标；
分别计算调度集中的所有用户终端的调度指标取值，计算所得各用户终端的调度指标取值形成调度矩阵；
在调度矩阵中选择最大调度指标取值所对应的用户终端，并将最大调度指标取值所对应的资源块RB分配给所选的用户终端。
上述方案中，在将最大调度指标取值所对应的RB分配给所选的用户终端后，所述方法还包括：
更新所选的用户终端的视频播放缓存信息，并根据更新后的视频播放缓存信息，对所选的用户终端进行重新分类。
上述方案中，在对所选的用户终端进行重新分类后，所述方法还包括：
判断网络中是否还有未分配的RB，如果有，则重新执行所述根据用户终端的分类结果确定网络的工作状态、以及后续流程；如果无，则结束流程。
上述方案中，所述根据用户终端的视频播放缓存信息，对用户终端进行分类，包括：
将用户终端的视频播放缓存的视频帧数与预设的第一门限和第二门限进行比较，如果用户终端的视频播放缓存的视频帧数小于第一门限，则将对应的用户终端分入第一集合；如果用户终端的视频播放缓存的视频帧数大于第一门限、且小于第二门限，则将对应的用户终端分入第二集合；如果用户终端的视频播放缓存的视频帧数大于第二门限，则将对应的用户终端分入第三集合。
上述方案中，所述根据用户终端的分类结果，确定网络的工作状态，包括：
当所述第一集合不为空时，确定网络处于第一状态；
当所述第一集合为空、且第二集合不为空时，确定网络处于第二状态；
当所述第一集合和第二集合都为空时，确定网络处于第三状态。
上述方案中，所述根据网络的工作状态，确定相应的调度集包括：
网络处于第一状态时，确定相应的调度集为所述第一集合；网络处于第二状态时，确定相应的调度集为所述第二集合；网络处于第三状态时，确定相应的调度集为所述第三集合。
上述方案中，通过以下方式确定调度集相对应的调度指标：
Figure PCTCN2014089822-appb-000001
其中，r(k,n)表示调度集中第k个用户终端在第n个RB上的传输能力，rmin(k)表示用户终端需要保证播放过程连续所需的最低平均传输码率，
Figure PCTCN2014089822-appb-000002
RS(k)表示帧F(k)正确解码所需要的视频信息的大小，F(k)表示调度集中第k个用户终端还未解码的所有视频帧中最紧急的帧，
Figure PCTCN2014089822-appb-000003
示用户终端k的缓存中视频数据能够支撑的连续播放时间，B(k)表示调度集中第k个用户终端的视频播放缓存信息。
上述方案中，通过以下方式计算调度集中的用户终端的调度指标取值：
M(k,n)＝FI(k)*m(k,n),k∈S,n∈Sava
其中，S表示调度集，表示当前可以使用的RB集合；M(k,n)表示调度集中的第k个用户终端在第n个RB上的调度指标取值；FI(k)表示公平性参数，
Figure PCTCN2014089822-appb-000004
用以表征用户终端之间的公平性，IT(k)表示第k个用户终端的过去总中断时长；*表示相乘。
上述方案中，所述在调度矩阵中选择最大调度指标取值所对应的用户终端，并将最大调度指标取值所对应的RB分配给所选的用户终端，包括：
将调度矩阵中满足以下条件的第n*个RB分配给第k*个用户：
Figure PCTCN2014089822-appb-000005
其中，*表示相乘。
本发明实施例提供的一种资源调度装置，包括：
分类单元，配置为根据用户终端的视频播放缓存信息，对用户终端进行分类；
状态确定单元，配置为根据用户终端的分类结果，确定网络的工作状态；
调度集和调度指标确定单元，配置为根据网络的工作状态，确定相应的调度集和调度指标；
调度指标取值计算单元，配置为分别计算调度集中的所有用户终端的调度指标取值，计算所得各用户终端的调度指标取值形成调度矩阵；
分配单元，配置为在调度矩阵中选择最大调度指标取值所对应的用户终端，并将最大调度指标取值所对应的资源块RB分配给所选的用户终端。
上述方案中，所述装置还包括更新单元，配置为在所述分配单元将最大调度指标取值所对应的RB分配给所选的用户终端后，所述更新单元更新所选的用户终端的视频播放缓存信息，并通知所述分类单元根据更新后的视频播放缓存信息，对所选的用户终端进行重新分类。
上述方案中，所述装置还包括判断单元，配置为在所述分类单元对所选的用户终端进行重新分类后，所述判断单元判断网络中是否还有未分配的RB，并在判断结果为有时，通知所述状态确定单元、调度集和调度指标确定单元、调度指标取值计算单元、分配单元；
所述状态确定单元重新根据用户终端的分类结果，确定网络的工作状态；所述调度集和调度指标确定单元重新根据网络的工作状态，确定相应的调度集和调度指标；所述调度指标取值计算单元重新计算调度集中的所 有用户终端的调度指标取值；所述分配单元重新在调度指标取值形成的调度矩阵中选择最大调度指标取值所对应的用户终端，并将最大调度指标取值所对应的RB分配给所选的用户终端；
直至网络中的所有RB都分配完毕。
上述方案中，所述分类单元配置为，将用户终端的视频播放缓存的视频帧数与预设的第一门限和第二门限进行比较，如果用户终端的视频播放缓存的视频帧数小于第一门限，则将对应的用户终端分入第一集合；如果用户终端的视频播放缓存的视频帧数大于第一门限、且小于第二门限，则将对应的用户终端分入第二集合；如果用户终端的视频播放缓存的视频帧数大于第二门限，则将对应的用户终端分入第三集合。
上述方案中，所述状态确定单元配置为，在所述第一集合不为空时，确定网络处于第一状态；在所述第一集合为空、且第二集合不为空时，确定网络处于第二状态；在所述第一集合和第二集合都为空时，确定网络处于第三状态。
上述方案中，所述调度集和调度指标确定单元配置为，在网络处于第一状态时，确定相应的调度集为所述第一集合；在网络处于第二状态时，确定相应的调度集为所述第二集合；在网络处于第三状态时，确定相应的调度集为所述第三集合；
并通过以下方式确定调度集相对应的调度指标：
Figure PCTCN2014089822-appb-000006
其中，r(k,n)表示调度集中第k个用户终端在第n个RB上的传输能力，rmin(k)表示用户终端需要保证播放过程连续所需的最低平均传输码率，
Figure PCTCN2014089822-appb-000007
RS(k)表示帧F(k)正确解码所需要的视频信息的大小，F(k)表 示调度集中第k个用户终端还未解码的所有视频帧中最紧急的帧，
Figure PCTCN2014089822-appb-000008
示用户终端k的缓存中视频数据能够支撑的连续播放时间，B(k)表示调度集中第k个用户终端的视频播放缓存信息。
上述方案中，调度指标取值计算单元配置为，通过以下方式计算调度集中的用户终端的调度指标取值：
M(k,n)＝FI(k)*m(k,n),k∈S,n∈Sava
其中，S表示调度集，表示当前可以使用的RB集合；M(k,n)表示调度集中的第k个用户终端在第n个RB上的调度指标取值；FI(k)表示公平性参数，
Figure PCTCN2014089822-appb-000009
用以表征用户终端之间的公平性，IT(k)表示第k个用户终端的过去总中断时长；*表示相乘。
上述方案中，所述分配单元配置为，将调度矩阵中满足以下条件的第n*个RB分配给第k*个用户：
Figure PCTCN2014089822-appb-000010
其中，*表示相乘。
本发明实施例还提供一种计算机可读存储介质，所述存储介质包括一组计算机可执行指令，所述指令用于执行本发明实施例所述的资源调度方法。
本发明实施例所提供的一种资源调度方法、装置和计算机可读存储介质，综合考虑了每个用户终端在RB上的信道质量、用户终端播放缓存中的视频帧数和用户之间的公平性等因素，既可以有效地避免视频播放中断的发生或缩短中断的持续时间，还能保证系统的频谱效率不发生明显下降。
附图说明
图1为本发明实施例的一种资源调度方法的流程图；
图2为本发明实施例的一种资源调度装置的结构示意图；
图3为本发明实施例中六种算法在用户请求视频码率差异较大的情况下的平均中断时长对比图；
图4为本发明实施例中六种算法在用户请求视频码率差异较小的情况下的平均中断时长对比图；
图5为本发明实施例中六种算法下每个用户随播放帧数的累积中断时长图；
图6为本发明实施例中六种算法下系统吞吐量随时间的变化图；
图7为本发明实施例的算法采用不同的缓存信息反馈周期的性能图。
具体实施方式
下面结合附图和具体实施例对本发明的技术方案进一步详细阐述。
作为一个具体的LTE网络，其中包括一个基站和K个视频流用户，基站可以分配的RB数目为N。用户终端通过基站向核心网请求视频流服务，视频数据到达基站后会首先被存入用户对应的发送缓存，然后根据基站调度的结果，被发送给用户终端。为了方便传输，视频数据在网络中是被切分成多个小数据包进行传输的。本发明实施例假设只有一个视频帧的所有数据包都被正确接收时，该视频帧才能被成功解码并放入播放缓存等待播放。
假设网络中用户都是在时间0发出视频流服务请求。本发明实施例中的时间是以传输时间间隔(TTI，Transmission Time Interval)为单位计算的，那么可以定义用户终端k在时刻t时的累积中断时长表示为：
Figure PCTCN2014089822-appb-000011
其中，IT(k,t)表示用户终端k在时刻t时的累积中断时长，IT(k,t-1)表示用户终端k在时刻t-1时的累积中断时长，B(k,t)表示用户终端在时刻t上报 的缓存中待播放的视频帧数。由于算法中只考虑当前时刻的缓存状态，所以后文以B(k)代替B(k,t)。当B(k,t)＝0时，说明用户终端在时刻t上报的缓存中待播放的视频帧数为0，待播放的视频帧为空，那么在时刻t视频流的播放会中断，此时累积的IT(k,t)需要在IT(k,t-1)的基础上加1；当B(k,t)≠0时，说明用户终端在时刻t上报的缓存中待播放的视频帧数不为0，待播放的视频帧不为空，那么在时刻t视频流的播放不会中断，此时累积的IT(k,t)不需在IT(k,t-1)的基础上加1，IT(k,t)与IT(k,t-1)相等。
那么，本发明实施例的优化目标即可以表示为：
Figure PCTCN2014089822-appb-000012
也就是说，实施本发明实施例的目的是为了使ITk尽量达到最小化，即使用户终端k的累积中断时长尽量最小化，从而有效地避免视频播放中断的发生或缩短中断的持续时间。
为达上述目的，本发明实施例提供了一种资源调度方法，如图1所示，该方法主要包括：
步骤101，根据用户终端的视频播放缓存信息，对用户终端进行分类。
用户终端的视频播放缓存信息，主要是指用户终端的视频播放缓存中的视频帧数；用户终端可以周期性的上报各自的视频播放缓存信息。
本发明实施例提出两个门限BDownTH和BUpTH，其中，BDownTH表示设定的播放缓存下限，简称第一门限，BUpTH表示设定的播放缓存上限，简称第二门限。本发明实施例中，将各用户终端的上报的视频播放缓存中的视频帧数与所述BDownTH和BUpTH进行比较，并根据比较结果对用户终端进行分类。具体可以根据比较结果将用户终端分入以下三个集合中：
SD＝{k|B(k)<BDownTH}
SM＝{k|BDownTH≤B(k)≤BUpTH}
SU＝{k|B(k)>BUpTH}
其中，B(k)表示用户终端k的视频播放缓存信息，即用户终端k的视频 播放缓存中的视频帧数。
第一集合SD为B(k)<BDownTH的用户终端的集合，即将B(k)小于BDownTH的用户终端分入第一集合SD中，第一集合SD中的用户终端的B(k)小于BDownTH，表示第一集合SD中的用户终端的视频播放缓存中的视频帧数很少，所述第一集合SD中的用户终端存在播放过程中断的可能性很大；
第二集合SM为BDownTH≤B(k)≤BUpTH的用户终端的集合，即将B(k)大于等于BDownTH、且小于等于BUpTH的用户终端分入第二集合SM中，第二集合SM中的用户终端的B(k)大于等于BDownTH、且小于等于BUpTH，表示第二集合SM中的用户终端的视频播放缓存中的视频帧数较少，所述第二集合SM中的用户终端存在播放过程中断的一定可能性，但可能性比第一集合SD中的用户终端小；
第三集合SU为B(k)>BUpTH的用户终端的集合，即将B(k)大于BUpTH的用户终端分入第三集合SU中，第三集合SU中的用户终端的B(k)大于BUpTH表示第三集合SU中的用户终端的视频播放缓存中的视频帧数较多，第三集合SU中的用户终端在一段时间内不会出现播放过程的中断。
步骤102，根据用户终端的分类结果，确定网络的工作状态。
根据网络内的用户终端在各类(即前述第一集合SD、第二集合SM、第三集合SU)中的分布情况，确定当前网络的工作状态，以便采取不同的调度策略。具体的网络工作状态可以分为：
Case a：
Figure PCTCN2014089822-appb-000013
Case a表示第一状态，当第一集合SD不为空时，确定网络处于第一状态下，第一状态表示SD中的用户终端的视频播放缓存中的视频帧数很少，SD中的用户终端即将或已经出现播放过程的中断，为了提高视频流用户的服务体验，此种情况下的调度优化目标是避免播放中断的出现或减少播放中 断的持续时间；
Case b：
Figure PCTCN2014089822-appb-000014
and
Figure PCTCN2014089822-appb-000015
Case b表示第二状态，当第一集合SD为空、且第二集合SM不为空时，确定网络处于第二状态下，第二状态表示SM中的用户终端的视频播放缓存中的视频帧数较少，SM中的用户终端易于出现播放过程的中断，此种情况下的调度优化目标是一方面要避免播放中断的出现，另一方面要保证系统频谱效率不会因为单一考虑用户体验而出现严重下降；
Case c：
Figure PCTCN2014089822-appb-000016
and
Figure PCTCN2014089822-appb-000017
Case c表示第三状态，当第一集合SD为空、且第二集合SM也为空时，确定网络处于第三状态下，第三状态表示网络中的所有用户终端都属于第三集合SU中，即第三集合SU中的用户终端的视频播放缓存中的视频帧数都较大，一段时间不会出现播放过程的中断，此种情况下的调度优化目标是提高系统频谱效率。
步骤103，根据网络的工作状态，确定相应的调度集和调度指标。
在网络处于不同工作状态下的调度集为：
Figure PCTCN2014089822-appb-000018
也就是说，当网络处在第一状态Case a时，确定的相应调度集为第一集合SD；当网络处在第二状态Case b时，确定的相应调度集为第二集合SM；当网络处在第三状态Case c时，确定的相应调度集为第三集合SU。
与所述调度集相对应的调度指标为：
Figure PCTCN2014089822-appb-000019
也就是说，当网络处在第一状态Case a时的调度指标为rmin(k)，rmin(k)针 对第一集合SD中的用户终端；当网络处在第二状态Case b时的调度指标为
Figure PCTCN2014089822-appb-000020
针对第二集合SM中的用户终端；当网络处在第三状态Case c时的调度指标为r(k,n)，r(k,n)针对第三集合SU中的用户终端。
其中，r(k,n)表示第k个用户终端在第n个RB上的传输能力(即传输码流)；rmin(k)表示用户终端需要保证播放过程连续所需的最低平均传输码率，可以表示为
Figure PCTCN2014089822-appb-000021
假设F(k)是用户终端k还未解码的所有视频帧中最紧急的帧，RS(k)表示帧F(k)正确解码所需要的视频信息的大小(单位是比特数或字节数)，为经验值，
Figure PCTCN2014089822-appb-000022
表示用户终端k的缓存中视频数据能够支撑的连续播放时间，FR表示帧率，是系统初始化时的已知变量。
步骤104，分别计算调度集中的所有用户终端的调度指标取值，计算所得各用户终端的调度指标取值形成调度矩阵。
以Sava表示当前可以使用的RB集合，对k∈S,n∈Sava计算调度指标，调度指标的计算方法为：
M(k,n)＝FI(k)*m(k,n),k∈S,n∈Sava
其中，S表示调度集，M(k,n)表示调度集中的第k个用户终端在第n个RB上的调度指标的取值；FI(k)表示公平性参数，
Figure PCTCN2014089822-appb-000023
用以表征用户终端之间的公平性，IT(k)表示第k个用户终端的过去总中断时长；
Figure PCTCN2014089822-appb-000024
*表示相乘。
步骤105，在调度矩阵中选择最大调度指标取值所对应的用户终端，并将最大调度指标取值所对应的RB分配给所选的用户终端。
具体的，将调度矩阵中满足以下条件的第n*个RB分配给第k*个用户：
Figure PCTCN2014089822-appb-000025
即将FI(k)*m(k,n)最大的第n*个RB分配给第k*个用户，其中，
Figure PCTCN2014089822-appb-000026
用以表征用户终端之间的公平性，IT(k)表示第k个用户终端的过去总中断时长。
步骤106，根据资源分配结果，更新被调度用户终端(即步骤105中所选的用户终端)的视频播放缓存信息，并根据更新后的视频播放缓存信息，对所述用户终端进行重新分类。
当一个RB被分配给用户终端后，需要根据RS(k)和R(k)更新相应用户终端的视频播放缓存信息，其中，R(k)为当前所有分配给用户终端k的RB所能提供的传输能力；如果R(k)>RS(k)，则B(k):＝B(k)+1，B(k):表示更新后的视频播放缓存信息(即更新后的视频播放缓存中的视频帧数)，B(k)表示更新前的视频播放缓存信息(即更新前的视频播放缓存中的视频帧数)。假设F(k)是用户终端k还未解码的所有视频帧中最紧急的帧，RS(k)表示帧F(k)正确解码所需要的视频信息的大小。
步骤107，判断网络中是否还有未分配的RB，如果有，返回步骤102；否则，结束流程。
在执行完网络中一RB的分配后，如果判断网络中还有其他未分配的RB，则按照步骤102～106同样的操作进行下一RB的分配，直到网络中所有的RB都分配完毕。
需要说明的是，本发明实施例的资源调度方法可应用于基站，如可以应用于家庭基站中。
对应本发明实施例的资源调度方法，本发明实施例还提供了一种资源调度装置，如图2所示，所述装置包括：
分类单元10，配置为根据用户终端的视频播放缓存信息，对用户终端进行分类；
状态确定单元20，配置为根据用户终端的分类结果，确定网络的工作状态；
调度集和调度指标确定单元30，配置为根据网络的工作状态，确定相应的调度集和调度指标；
调度指标取值计算单元40，配置为分别计算调度集中的所有用户终端的调度指标取值，计算所得各用户终端的调度指标取值形成调度矩阵；
分配单元50，配置为在调度矩阵中选择最大调度指标取值所对应的用户终端，并将最大调度指标取值所对应的RB分配给所选的用户终端。
在一实施方案中，所述装置还包括更新单元60，配置为在分配单元50将最大调度指标取值所对应的RB分配给所选的用户终端后，所述更新单元60更新所选的用户终端的视频播放缓存信息，并通知分类单元10根据更新后的视频播放缓存信息，对所选的用户终端进行重新分类。
在一实施方案中，所述装置还包括判断单元70，配置为在分类单元10对所选的用户终端进行重新分类后，判断单元70判断网络中是否还有未分配的RB，并在判断结果为有时，通知状态确定单元20、调度集和调度指标确定单元30、调度指标取值计算单元40、分配单元50、更新单元60；状态确定单元20会重新根据用户终端的分类结果，确定网络的工作状态；调度集和调度指标确定单元30会重新根据状态确定单元20确定的网络的工作状态，确定相应的调度集和调度指标；调度指标取值计算单元40会重新根据调度集和调度指标确定单元30确定的调度集和调度指标，计算调度集中的所有用户终端的调度指标取值；分配单元50会重新在调度指标取值形成的调度矩阵中选择最大调度指标取值所对应的用户终端，并将最大调度指标取值所对应的RB分配给所选的用户终端；更新单元60会重新更新所选的用户终端的视频播放缓存信息，并通知分类单元10根据更新后的视频 播放缓存信息，对所选的用户终端进行重新分类；
上述各模块执行，直至网络中的所有RB都分配完毕。
在一实施方案中，分类单元10配置为，将用户终端的视频播放缓存的视频帧数与预设的第一门限和第二门限进行比较，如果用户终端的视频播放缓存的视频帧数小于第一门限，则将对应的用户终端分入第一集合；如果用户终端的视频播放缓存的视频帧数大于第一门限、且小于第二门限，则将对应的用户终端分入第二集合；如果用户终端的视频播放缓存的视频帧数大于第二门限，则将对应的用户终端分入第三集合。
在一实施方案中，状态确定单元20配置为，在所述第一集合不为空时，确定网络处于第一状态；在所述第一集合为空、且第二集合不为空时，确定网络处于第二状态；在所述第一集合和第二集合都为空时，确定网络处于第三状态。
在一实施方案中，调度集和调度指标确定单元30配置为，在网络处于第一状态时，确定相应的调度集为所述第一集合；在网络处于第二状态时，确定相应的调度集为所述第二集合；在网络处于第三状态时，确定相应的调度集为所述第三集合；
并通过以下方式确定调度集相对应的调度指标：
Figure PCTCN2014089822-appb-000027
其中，r(k,n)表示调度集中第k个用户终端在第n个RB上的传输能力，rmin(k)表示用户终端需要保证播放过程连续所需的最低平均传输码率，
Figure PCTCN2014089822-appb-000028
RS(k)表示帧F(k)正确解码所需要的视频信息的大小，F(k)表示调度集中第k个用户终端还未解码的所有视频帧中最紧急的帧，
Figure PCTCN2014089822-appb-000029
示用户终端k的缓存中视频数据能够支撑的连续播放时间，B(k) 表示调度集中第k个用户终端的视频播放缓存信息。
在一实施方案中，调度指标取值计算单元40配置为，通过以下方式计算调度集中的用户终端的调度指标取值：
M(k,n)＝FI(k)*m(k,n),k∈S,n∈Sava
其中，S表示调度集，表示当前可以使用的RB集合；M(k,n)表示调度集中的第k个用户终端在第n个RB上的调度指标取值；FI(k)表示公平性参数，
Figure PCTCN2014089822-appb-000030
用以表征用户终端之间的公平性，IT(k)表示第k个用户终端的过去总中断时长；*表示相乘。
在一实施方案中，分配单元50配置为，将调度矩阵中满足以下条件的第n*个RB分配给第k*个用户：
Figure PCTCN2014089822-appb-000031
其中，*表示相乘。
需要说明的是，上述分类单元10、状态确定单元20、调度集和调度指标确定单元30、调度指标取值计算单元40、分配单元50、更新单元60和判断单元70可以由资源调度装置的中央处理器(CPU，Central Processing Unit)、微处理器(MPU，Micro Processing Unit)、数字信号处理器(DSP，Digital Signal Processor)或可编程逻辑阵列(FPGA，Field－Programmable Gate Array)实现。
需要说明的是，本发明实施例的资源调度装置可应用于基站，如可以应用于家庭基站中。从而，本发明实施例还提供一种包括所述资源调度装置的基站，所述基站中资源调度装置的内部结构和功能参见前述实施例，此处不再赘述。
本发明实施例提供的针对LTE网络下视频流的资源调度方法和装置，首先给出了最小化网络中所有用户终端播放中断时间和的优化目标式，并 根据此目标确定用户的调度顺序和无线资源的分配策略。根据用户终端周期性反馈的播放缓存信息(具体为播放缓存中待播放的视频帧数)，可以将用户终端分成三类：即将或已经出现播放中断的用户终端；有较大概率出现播放中断的用户终端和短期内不会出现播放中断的用户终端。网络中三类用户终端的分布情况可以用来说明当前网络的工作状态。为了有效提高网络中视频流用户的QoE，所采用的策略会根据不同的网络状态动态调整。
具体来说，网络状态可以分为三类：a.网络中存在视频播放过程即将或已经出现中断的用户终端，b.网络有部分用户终端的缓存中视频帧数较少，有较高概率出现播放过程中断，c.网络中所有用户终端的缓存视频帧数都较大，能够有效抵抗网络传输能力的波动，一段时间内不会出现播放过程的中断。在每一种情况下，都会确定具体的调度集合和调度指标。
然后，对于调度集中的每一个用户终端计算其在每一个可用RB上的调度指标值，形成调度矩阵。接着，在调度矩阵中搜索具有最大调度指标值的用户终端和对应的RB，并将该RB分配给该用户终端。之后，需要根据分配情况更新该用户终端的播放缓存信息。
重复上述过程，直至网络中所有的RB都被分配，进而结束该过程。
最后，根据分配结果，本发明实施例中可以采用先入先出(FIFO，First In First Out)的方式进行队列包调度。
本发明实施例的资源调度方法和装置，综合考虑了每个用户终端在RB上的信道质量、用户终端播放缓存中的视频帧数和用户之间的公平性等因素，既可以有效地避免视频播放中断的发生或缩短中断的持续时间，还能保证系统的频谱效率不发生明显下降。
下面给出本发明实施例的仿真设置和实验结果分析。
1、小区仿真参数设置
网络中随机生成用户位置、移动速度(0，3km/h)和用户使用的业务 类型，其中，宏基站用户上随机生成语音或数据业务，家庭基站用户随机生成数据或视频业务。对于数据业务用户，在(0，10M)区间内随机生成其所需下载数据量的大小，精确到1k。对于视频用户，其动态因子αi在集合{2，5，10，15}内随机选取，编码速率REQi在集合{0.8Mbps，0.9Mbps，1Mbps，1.1Mbps，1.2Mbps}中随机选取。对于宏小区的仿真测试环境参照TR 25.814协议实现，家庭基站的仿真测试环境采用TS 36.814中对于家庭基站的仿真环境设置，详细的仿真参数设置如表2、表3、表4所示。实验结果中，现存的算法为MOSMax算法，本发明实施例对应的算法为proposed算法。
表1 LTE网络仿真参数对照表
参数 设定值
RB数目 10
物理层传播模型 Macro-Cell Urban Mode
可选调制方式 QPSK，16QAM，64QAM
可选Turbo编码码率 1/2，3/4
基站总发射功率 43dbm，平均分配在每个RB上
CQI反馈模式 全频带CQI
CQI反馈周期 1 TTI
表2视频流仿真参数对照表
参数 设定值
视频编码软件及版本 JM16.2
帧率 30帧/秒
GOP长度 12帧
视频序列长度 3000帧
播放缓存上限BUpTH 6帧
播放缓存下限BDownTH 2帧
表3视频流请求配置表1
  视频序列 分辨率 量化参数 比特率(kbps)
用户1 Highway CIF 28 266.936
用户2 Coastguard QCIF 28 220.0776
用户3 Highway CIF 29 223.1672
用户4 Foreman CIF 32 235.1504
用户5 Highway QCIF 28 90.552
用户6 Coastguard QCIF 32 113.6112
表4视频流请求配置表2
  视频序列 分辨率 量化参数 比特率(kbps)
用户1 Highway CIF 28 266.936
用户2 Foreman CIF 32 235.1504
用户3 Highway CIF 29 223.1672
用户4 Foreman CIF 32 235.1504
用户5 Highway CIF 32 137.1416
用户6 Highway CIF 31 162.3952
2、实验结果和分析
图3比较了在视频流请求配置表1所示环境下采用六种算法时各自的总中断时长，这六种算法分别是：RR(轮询)算法、MAXCI(最大载干比) 算法、PF(比例公平)算法、RG(均衡)算法、PB算法、以及本发明实施例提出的PBA算法。从图3中可以看出，RR可以有效保证用户之间资源分配的公平性，但是因为所请求的视频序列不同，同时每个用户的信道质量也不同，造成了用户得到的服务质量出现了较大差异。其中，请求大码率视频或者信道质量差的用户在观看视频时会出现更长的播放中断。相较于RR，MAXCI会放大信道质量差异对于用户服务体验的影响：信道条件好的用户几乎没有发生中断，而信道条件较差的用户几乎没有机会获得资源的使用权，带来了很大的等待时长。PF算法在前四个用户(请求相近码率的视频序列)中表现出较好的公平性，但对于最后两个请求码率较小视频的用户，表现出了明显的偏袒。RG算法通过考虑每个视频流具体的传输速率需求，可以有效避免PF算法的缺陷，能够获得不错的性能；然而，RG保证的是长期的传输速率，并没有考虑视频流畅播放的实时码率要求，仍然会造成视频播放出现中断。PB考虑了缓存的实时状态，可以更好的满足视频流畅播放的实时码率需求。但PB考虑缓存内数据是以bit为单位，而并不是以视频帧为单位，这样会使PB与PF有相同的缺陷，即会对小码率用户造成偏袒。
从图3中可以看出，本发明实施例提出的PBA算法，能够得到很好的性能，不仅能够有效降低中断时长，还能够保证用户之间的公平性。
图4比较了在视频流请求配置表2所示环境下采用六种算法时各自的总中断时长。由于此种配置条件下，用户请求的码率差异较小，所以性能提升不如配置1环境下明显。但从图4中仍可以看出，PBA的性能仍然优于其他对比算法的。
图5说明了在视频流请求配置表1所示环境下，采用六种算法时各个用户的随播放时间的累积中断时长。RR算法中，各用户的中断时间连续增 长，说明了RR资源分配的公平性。在MAXCI算法，可以看到明显的资源分配偏向，即信道条件较差的用户等待了很长时间之后才会得到服务。同时，可以看到PF和PB对于小码率用户的偏向。本发明实施例提出的PBA算法可以保证各用户得到服务的公平性，并将中断时间控制在一个较低的水平上。
图6说明了在视频流请求配置表1所示环境下，六种算法的吞吐量情况。可见，虽然PBA的吞吐量性能不如MAXCI，PF和RR，但比RG和PB要好。这个结果可以说明，PBA在保证用户QoE的同时，并没有使系统的吞吐量严重下降。
图7说明了采用不同的缓存状态反馈周期时，PBA算法的性能变化。因为视频播放的帧率通常选为30帧/秒或更低，那么，一帧播放的时长就是33ms或更长。相对于CQI反馈周期1ms来说，用户终端播放缓存状态改变是很慢的。这样，为了减少缓存状态信息反馈带来的开销，可以适当放大反馈周期。具体的实验结果可见图7，反馈周期为1个TTI，11个TTI和22个TTI时，性能差异不大。但当超过33个TTI后，就随着反馈周期变长而恶化。所以，缓存反馈周期确实可以选择一个较大的值，以此来减少信令反馈的开销。
综上所述，本发明实施例综合考虑了每个用户终端在RB上的信道质量、用户终端播放缓存中的视频帧数和用户之间的公平性等因素，既可以有效地避免视频播放中断的发生或缩短中断的持续时间，还能保证系统的频谱效率不发生明显下降。
与传统的RR算法、MAXCI算法、PF算法和考虑用户缓存信息的PB算法相比，本发明实施例可以有效降低视频流用户观看视频过程中的中断 时长，同时还能解决对比算法中对于特定用户的偏袒问题，例如PF和PB会偏袒请求小码率视频的用户，MAXCI会偏袒具有较好信道质量的用户。
本发明实施例还提供一种计算机可读存储介质，所述存储介质包括一组计算机可执行指令，所述指令用于执行本发明实施例所述的资源调度方法。
本领域内的技术人员应明白，本发明的实施例可提供为方法、系统、或计算机程序产品。因此，本发明可采用硬件实施例、软件实施例、或结合软件和硬件方面的实施例的形式。而且，本发明可采用在一个或多个其中包含有计算机可用程序代码的计算机可用存储介质(包括但不限于磁盘存储器和光学存储器等)上实施的计算机程序产品的形式。
本发明是参照根据本发明实施例的方法、设备(系统)、和计算机程序产品的流程图和/或方框图来描述的。应理解可由计算机程序指令实现流程图和/或方框图中的每一流程和/或方框、以及流程图和/或方框图中的流程和/或方框的结合。可提供这些计算机程序指令到通用计算机、专用计算机、嵌入式处理机或其他可编程数据处理设备的处理器以产生一个机器，使得通过计算机或其他可编程数据处理设备的处理器执行的指令产生用于实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能的装置。
这些计算机程序指令也可存储在能引导计算机或其他可编程数据处理设备以特定方式工作的计算机可读存储器中，使得存储在该计算机可读存储器中的指令产生包括指令装置的制造品，该指令装置实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能。
这些计算机程序指令也可装载到计算机或其他可编程数据处理设备上，使得在计算机或其他可编程设备上执行一系列操作步骤以产生计算机实现的处理，从而在计算机或其他可编程设备上执行的指令提供用于实现在流 程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能的步骤。
以上所述，仅为本发明的较佳实施例而已，并非用于限定本发明的保护范围。
工业实用性
本发明实施例综合考虑了每个用户终端在RB上的信道质量、用户终端播放缓存中的视频帧数和用户之间的公平性等因素，既可以有效地避免视频播放中断的发生或缩短中断的持续时间，还能保证系统的频谱效率不发生明显下降。
与传统的RR算法、MAXCI算法、PF算法和考虑用户缓存信息的PB算法相比，本发明实施例可以有效降低视频流用户观看视频过程中的中断时长，同时还能解决对比算法中对于特定用户的偏袒问题，例如PF和PB会偏袒请求小码率视频的用户，MAXCI会偏袒具有较好信道质量的用户。

Claims (18)

  1. 一种资源调度方法，所述方法包括：
    根据用户终端的视频播放缓存信息，对用户终端进行分类；
    根据用户终端的分类结果，确定网络的工作状态；
    根据网络的工作状态，确定相应的调度集和调度指标；
    分别计算调度集中的所有用户终端的调度指标取值，计算所得各用户终端的调度指标取值形成调度矩阵；
    在调度矩阵中选择最大调度指标取值所对应的用户终端，并将最大调度指标取值所对应的资源块RB分配给所选的用户终端。
  2. 根据权利要求1所述资源调度方法，其中，在将最大调度指标取值所对应的RB分配给所选的用户终端后，所述方法还包括：
    更新所选的用户终端的视频播放缓存信息，并根据更新后的视频播放缓存信息，对所选的用户终端进行重新分类。
  3. 根据权利要求2所述资源调度方法，其中，在对所选的用户终端进行重新分类后，所述方法还包括：
    判断网络中是否还有未分配的RB，如果有，则重新执行所述根据用户终端的分类结果确定网络的工作状态、以及后续流程；如果无，则结束流程。
  4. 根据权利要求1、2或3所述资源调度方法，其中，所述根据用户终端的视频播放缓存信息，对用户终端进行分类，包括：
    将用户终端的视频播放缓存的视频帧数与预设的第一门限和第二门限进行比较，如果用户终端的视频播放缓存的视频帧数小于第一门限，则将对应的用户终端分入第一集合；如果用户终端的视频播放缓存的视频帧数大于第一门限、且小于第二门限，则将对应的用户终端分入第二集合；如果用户终端的视频播放缓存的视频帧数大于第二门限，则将对应的用户终 端分入第三集合。
  5. 根据权利要求4所述资源调度方法，其中，所述根据用户终端的分类结果，确定网络的工作状态，包括：
    当所述第一集合不为空时，确定网络处于第一状态；
    当所述第一集合为空、且第二集合不为空时，确定网络处于第二状态；
    当所述第一集合和第二集合都为空时，确定网络处于第三状态。
  6. 根据权利要求5所述资源调度方法，其中，所述根据网络的工作状态，确定相应的调度集包括：
    网络处于第一状态时，确定相应的调度集为所述第一集合；网络处于第二状态时，确定相应的调度集为所述第二集合；网络处于第三状态时，确定相应的调度集为所述第三集合。
  7. 根据权利要求6所述资源调度方法，其中，通过以下方式确定调度集相对应的调度指标：
    Figure PCTCN2014089822-appb-100001
    其中，r(k,n)表示调度集中第k个用户终端在第n个RB上的传输能力，rmin(k)表示用户终端需要保证播放过程连续所需的最低平均传输码率，
    Figure PCTCN2014089822-appb-100002
    RS(k)表示帧F(k)正确解码所需要的视频信息的大小，F(k)表示调度集中第k个用户终端还未解码的所有视频帧中最紧急的帧，
    Figure PCTCN2014089822-appb-100003
    示用户终端k的缓存中视频数据能够支撑的连续播放时间，B(k)表示调度集中第k个用户终端的视频播放缓存信息。
  8. 根据权利要求7所述资源调度方法，其中，通过以下方式计算调度集中的用户终端的调度指标取值：
    M(k,n)＝FI(k)*m(k,n),k∈S,n∈Sava
    其中，S表示调度集，表示当前可以使用的RB集合；M(k,n)表示调度集中的第k个用户终端在第n个RB上的调度指标取值；FI(k)表示公平性参数，
    Figure PCTCN2014089822-appb-100004
    用以表征用户终端之间的公平性，IT(k)表示第k个用户终端的过去总中断时长；*表示相乘。
  9. 根据权利要求8所述资源调度方法，其中，所述在调度矩阵中选择最大调度指标取值所对应的用户终端，并将最大调度指标取值所对应的RB分配给所选的用户终端，包括：
    将调度矩阵中满足以下条件的第n*个RB分配给第k*个用户：
    Figure PCTCN2014089822-appb-100005
    其中，*表示相乘。
  10. 一种资源调度装置，所述装置包括：
    分类单元，配置为根据用户终端的视频播放缓存信息，对用户终端进行分类；
    状态确定单元，配置为根据用户终端的分类结果，确定网络的工作状态；
    调度集和调度指标确定单元，配置为根据网络的工作状态，确定相应的调度集和调度指标；
    调度指标取值计算单元，配置为分别计算调度集中的所有用户终端的调度指标取值，计算所得各用户终端的调度指标取值形成调度矩阵；
    分配单元，配置为在调度矩阵中选择最大调度指标取值所对应的用户终端，并将最大调度指标取值所对应的资源块RB分配给所选的用户终端。
  11. 根据权利要求10所述资源调度装置，其中，所述装置还包括更新单元，配置为在所述分配单元将最大调度指标取值所对应的RB分配给所选的用户终端后，所述更新单元更新所选的用户终端的视频播放缓存信息， 并通知所述分类单元根据更新后的视频播放缓存信息，对所选的用户终端进行重新分类。
  12. 根据权利要求10所述资源调度装置，其中，所述装置还包括判断单元，配置为在所述分类单元对所选的用户终端进行重新分类后，所述判断单元判断网络中是否还有未分配的RB，并在判断结果为有时，通知所述状态确定单元、调度集和调度指标确定单元、调度指标取值计算单元、分配单元；
    所述状态确定单元重新根据用户终端的分类结果，确定网络的工作状态；所述调度集和调度指标确定单元重新根据网络的工作状态，确定相应的调度集和调度指标；所述调度指标取值计算单元重新计算调度集中的所有用户终端的调度指标取值；所述分配单元重新在调度指标取值形成的调度矩阵中选择最大调度指标取值所对应的用户终端，并将最大调度指标取值所对应的RB分配给所选的用户终端；
    直至网络中的所有RB都分配完毕。
  13. 根据权利要求10、11或12所述资源调度装置，其中，所述分类单元配置为，将用户终端的视频播放缓存的视频帧数与预设的第一门限和第二门限进行比较，如果用户终端的视频播放缓存的视频帧数小于第一门限，则将对应的用户终端分入第一集合；如果用户终端的视频播放缓存的视频帧数大于第一门限、且小于第二门限，则将对应的用户终端分入第二集合；如果用户终端的视频播放缓存的视频帧数大于第二门限，则将对应的用户终端分入第三集合。
  14. 根据权利要求13所述资源调度装置，其中，所述状态确定单元配置为，在所述第一集合不为空时，确定网络处于第一状态；在所述第一集合为空、且第二集合不为空时，确定网络处于第二状态；在所述第一集合和第二集合都为空时，确定网络处于第三状态。
  15. 根据权利要求14所述资源调度装置，其中，所述调度集和调度指 标确定单元配置为，在网络处于第一状态时，确定相应的调度集为所述第一集合；在网络处于第二状态时，确定相应的调度集为所述第二集合；在网络处于第三状态时，确定相应的调度集为所述第三集合；
    并通过以下方式确定调度集相对应的调度指标：
    Figure PCTCN2014089822-appb-100006
    其中，r(k,n)表示调度集中第k个用户终端在第n个RB上的传输能力，rmin(k)表示用户终端需要保证播放过程连续所需的最低平均传输码率，
    Figure PCTCN2014089822-appb-100007
    RS(k)表示帧F(k)正确解码所需要的视频信息的大小，F(k)表示调度集中第k个用户终端还未解码的所有视频帧中最紧急的帧，
    Figure PCTCN2014089822-appb-100008
    示用户终端k的缓存中视频数据能够支撑的连续播放时间，B(k)表示调度集中第k个用户终端的视频播放缓存信息。
  16. 根据权利要求15所述资源调度装置，其中，调度指标取值计算单元配置为，通过以下方式计算调度集中的用户终端的调度指标取值：
    M(k,n)＝FI(k)*m(k,n),k∈S,n∈Sava
    其中，S表示调度集，表示当前可以使用的RB集合；M(k,n)表示调度集中的第k个用户终端在第n个RB上的调度指标取值；FI(k)表示公平性参数，
    Figure PCTCN2014089822-appb-100009
    用以表征用户终端之间的公平性，IT(k)表示第k个用户终端的过去总中断时长；*表示相乘。
  17. 根据权利要求16所述资源调度装置，其中，所述分配单元配置为，将调度矩阵中满足以下条件的第n*个RB分配给第k*个用户：
    Figure PCTCN2014089822-appb-100010
    其中，*表示相乘。
  18. 一种计算机可读存储介质，所述存储介质包括一组计算机可执行指令，所述指令用于执行权利要求1-9任一项所述的资源调度方法。
PCT/CN2014/089822 2014-06-30 2014-10-29 一种资源调度方法、装置和计算机可读存储介质 WO2016000366A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201410309485.2A CN105282848A (zh) 2014-06-30 2014-06-30 一种资源调度方法和装置
CN201410309485.2 2014-06-30

Publications (1)

Publication Number Publication Date
WO2016000366A1 true WO2016000366A1 (zh) 2016-01-07

Family

ID=55018377

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2014/089822 WO2016000366A1 (zh) 2014-06-30 2014-10-29 一种资源调度方法、装置和计算机可读存储介质

Country Status (2)

Country Link
CN (1) CN105282848A (zh)
WO (1) WO2016000366A1 (zh)

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105847033B (zh) * 2016-02-26 2019-02-05 重庆世纪之光科技实业有限公司 一种调度方法
CN107645778A (zh) * 2016-07-22 2018-01-30 中兴通讯股份有限公司 一种无线站点状态信息上报方法和装置
CN110169078A (zh) 2017-01-18 2019-08-23 华为技术有限公司 视频数据处理方法及设备
CN111510479B (zh) * 2020-04-08 2021-09-24 天津大学 一种异构缓存系统的资源分配方法及装置

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102469356A (zh) * 2010-11-05 2012-05-23 乐金电子(中国)研究开发中心有限公司 一种网络电视视频播放方法及网络电视视频接收端
CN102858018A (zh) * 2012-09-24 2013-01-02 西北工业大学 一种HSPDA系统中保证用户QoE性能的调度方法
CN103428862A (zh) * 2012-05-23 2013-12-04 华为技术有限公司 一种资源分配的方法及装置
CN103458520A (zh) * 2013-08-23 2013-12-18 京信通信系统（中国）有限公司 一种上行频域资源的分配方法和装置
CN103476123A (zh) * 2013-08-30 2013-12-25 西安电子科技大学 LTE网络视频业务QoE保障资源分配方法

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR102023582B1 (ko) * 2012-10-11 2019-11-04 삼성전자주식회사 무선 통신 시스템에서 청크 기반 스케줄링 방법 및 장치
AU2013339078B2 (en) * 2012-11-01 2016-09-22 Sharp Kabushiki Kaisha Mobile station apparatus, base station apparatus, communication method and integrated circuit
CN103747274B (zh) * 2013-12-18 2016-08-17 北京邮电大学 一种增设缓存集群的视频数据中心及其缓存资源调度方法

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102469356A (zh) * 2010-11-05 2012-05-23 乐金电子(中国)研究开发中心有限公司 一种网络电视视频播放方法及网络电视视频接收端
CN103428862A (zh) * 2012-05-23 2013-12-04 华为技术有限公司 一种资源分配的方法及装置
CN102858018A (zh) * 2012-09-24 2013-01-02 西北工业大学 一种HSPDA系统中保证用户QoE性能的调度方法
CN103458520A (zh) * 2013-08-23 2013-12-18 京信通信系统（中国）有限公司 一种上行频域资源的分配方法和装置
CN103476123A (zh) * 2013-08-30 2013-12-25 西安电子科技大学 LTE网络视频业务QoE保障资源分配方法

Also Published As

Publication number Publication date
CN105282848A (zh) 2016-01-27

Similar Documents

Publication Publication Date Title
US8737383B2 (en) Techniques for enhanced persistent scheduling with efficient link adaptation capability
Mushtaq et al. QoE/QoS-aware LTE downlink scheduler for VoIP with power saving
Shehada et al. QoE-based cross-layer optimization for video delivery in long term evolution mobile networks
US11297527B2 (en) Method and system for application aware congestion management
JP2016521071A5 (zh)
Nasralla et al. Content-aware downlink scheduling for LTE wireless systems: A survey and performance comparison of key approaches
CN103209494B (zh) 一种基于重要性标记的实时视频业务资源分配方法
Navarro-Ortiz et al. A QoE-aware scheduler for HTTP progressive video in OFDMA systems
WO2013008839A1 (ja) リソース割り当て方法、無線通信システム、基地局及びプログラム
JP5506891B2 (ja) 無線移動ネットワークにおいてスケーラブルビデオ符号化ストリームをスケジューリングする方法
CN103369674B (zh) 空口资源的分配方法及设备
WO2016000366A1 (zh) 一种资源调度方法、装置和计算机可读存储介质
Khan et al. QoE-driven multi-user scheduling and rate adaptation with reduced cross-layer signaling for scalable video streaming over LTE wireless systems
JP2017523650A (ja) 無線通信ネットワークまたはそのネットワーク要素を動作させるための方法
da Mata et al. Resource allocation for the LTE uplink based on Genetic Algorithms in mixed traffic environments
Khorov et al. SEBRA: SAND-enabled bitrate and resource allocation algorithm for network-assisted video streaming
CN110602796A (zh) 移动宽带共享信道上的资源调度方法及电子设备
Yildiz et al. A novel mobility aware downlink scheduling algorithm for LTE-A networks
Mushtaq et al. Qoe-based lte downlink scheduler for voip
WO2016000502A1 (zh) 一种资源分配方法、装置和计算机存储介质
Shojaedin et al. TCP-aware scheduling in LTE networks
Belghith et al. Efficient bandwidth call admission control in 3GPP LTE networks
JP6276206B2 (ja) 帯域割り当て制御装置及び帯域割り当て制御方法
Ramli Performance of maximum-largest weighted delay first algorithm in long term evolution-advanced with carrier aggregation
Mehmeti et al. Resource allocation for improved user experience with live video streaming in 5G

Legal Events

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

Ref document number: 14896640

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 14896640

Country of ref document: EP

Kind code of ref document: A1