CN113076199A - RNN-based EPON dynamic bandwidth allocation algorithm - Google Patents

RNN-based EPON dynamic bandwidth allocation algorithm Download PDF

Info

Publication number
CN113076199A
CN113076199A CN202110467814.6A CN202110467814A CN113076199A CN 113076199 A CN113076199 A CN 113076199A CN 202110467814 A CN202110467814 A CN 202110467814A CN 113076199 A CN113076199 A CN 113076199A
Authority
CN
China
Prior art keywords
bandwidth
queue
onu
olt
service
Prior art date
Legal status (The legal status 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 status listed.)
Pending
Application number
CN202110467814.6A
Other languages
Chinese (zh)
Inventor
秦攀科
付岩岩
尤俊茹
刘飞扬
韩尚雅
王家伟
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Henan University of Technology
Original Assignee
Henan University of Technology
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 Henan University of Technology filed Critical Henan University of Technology
Priority to CN202110467814.6A priority Critical patent/CN113076199A/en
Publication of CN113076199A publication Critical patent/CN113076199A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00—Arrangements for program control, e.g. control units
    • G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/46—Multiprogramming arrangements
    • G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
    • G06F9/5005—Allocation of resources, e.g. of the central processing unit [CPU] to service a request
    • G06F9/5027—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resource being a machine, e.g. CPUs, Servers, Terminals
    • G06F9/5038—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resource being a machine, e.g. CPUs, Servers, Terminals considering the execution order of a plurality of tasks, e.g. taking priority or time dependency constraints into consideration
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Software Systems (AREA)
  • General Physics & Mathematics (AREA)
  • Business, Economics & Management (AREA)
  • General Engineering & Computer Science (AREA)
  • Computational Linguistics (AREA)
  • Health & Medical Sciences (AREA)
  • Evolutionary Computation (AREA)
  • General Health & Medical Sciences (AREA)
  • Molecular Biology (AREA)
  • Computing Systems (AREA)
  • Biophysics (AREA)
  • Biomedical Technology (AREA)
  • Mathematical Physics (AREA)
  • Artificial Intelligence (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Data Mining & Analysis (AREA)
  • Strategic Management (AREA)
  • Economics (AREA)
  • Human Resources & Organizations (AREA)
  • Game Theory and Decision Science (AREA)
  • Development Economics (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Marketing (AREA)
  • Operations Research (AREA)
  • Quality & Reliability (AREA)
  • Tourism & Hospitality (AREA)
  • General Business, Economics & Management (AREA)
  • Small-Scale Networks (AREA)

Abstract

The invention relates to an EPON dynamic bandwidth allocation algorithm based on RNN, comprising the following steps: the method comprises the steps that firstly, an ONU reports a bandwidth request to an OLT, then the OLT judges whether the bandwidth request is an EF service, the bandwidth prediction and bandwidth allocation are carried out on the EF service, actually required bandwidth is generated for the bandwidth requests of AF and BE services, the bandwidth allocation is carried out, then the remaining bandwidth requests of the AF and BE services are collected, and the bandwidth allocation is carried out on the bandwidth requests in a queuing and scheduling mode. The invention carries out bandwidth prediction and bandwidth allocation on the high-priority business EF on the basis of the RNN structural model, improves the prediction accuracy, carries out bandwidth allocation on AF and BE businesses on the premise of ensuring the high-priority business, and finally carries out bandwidth allocation on the residual bandwidth requests of the three businesses in a queuing scheduling mode, thereby ensuring the service quality of other businesses to a certain extent, improving the prediction accuracy, reducing the queue delay and improving the uplink bandwidth utilization rate.

Description

RNN-based EPON dynamic bandwidth allocation algorithm
Technical Field
The invention belongs to the technical field of bandwidth utilization, and particularly relates to an EPON dynamic bandwidth allocation algorithm based on RNN.
Background
RNN is a cyclic neural network, has memory ability, can obtain prediction ability through priori knowledge learning, and improves the prediction accuracy of nonlinear data by modifying parameters in real time through online learning. As shown in fig. 1, in the process of sending data to the OLT by the ONU, during the waiting time T1-T5, the ONU still receives the upstream data, and the data amount is not included in the request frame, which causes the phenomenon that the request information does not conform to the actual demand, so that the data packet can only be accumulated to be sent in the next polling period, which not only increases the average queue delay, but also affects the real-time performance of the service.
Compared with other access network technologies, the high capacity characteristic of the multi-access optical fiber network is a main driving force for the progress of optical technologies, more and more low-latency intensive tasks such as real-time communication, high-definition video playing, games and the like appear in daily life, the increasing bandwidth requirement brings a serious challenge to the construction of network infrastructure, resource allocation is a basic task of the PON, an effective scheme is needed to reduce latency, the bandwidth utilization rate is maximized, the resource waste is minimized, and along with the increasing increase of users, the selection of a bandwidth allocation mode is very critical to avoid conflicts and ensure the service requirement of high-priority services. Therefore, the invention provides an EPON dynamic bandwidth allocation algorithm which is based on an RNN structural model, can predict high-priority services, ensures the service quality of other services to a certain extent on the premise of ensuring the high-priority services, and simultaneously reduces queue delay and improves the utilization rate of uplink bandwidth.
Disclosure of Invention
The invention aims to overcome the defects of the prior art and provides an EPON dynamic bandwidth allocation algorithm based on an RNN.
The technical scheme adopted by the invention is as follows: an EPON dynamic bandwidth allocation algorithm based on RNN, comprising the following steps:
the method comprises the following steps: the bandwidth request is reported and the bandwidth request is reported,
respectively recording N ONUs uploading data to an OLT as ONUsi(i＝1,2，……N)，ONUi(i-1, 2, … … N) respectively buffering three different services of EF, AF, BE into three queues according to service priority, and reporting bandwidth requests of the three queues to the OLT;
step two: the OLT determines whether it is EF traffic,
if yes, executing bandwidth prediction, and going to the third step, and if not, not executing bandwidth prediction, and going to the fourth step;
step three: the bandwidth actually required for the EF traffic is calculated,
OLT aggregates actual bandwidth requests of EF queues
Figure BDA0003043973220000021
Using RNN structure model to wait time TwaitPredicting the flow of the internal arrival, carrying out differential calculation on historical bandwidth requests, and setting the data packets arriving in the j period as
Figure BDA0003043973220000022
Get the average rate of packet arrival in j period as
Figure BDA0003043973220000023
Wherein, TcycleIn order to be a polling period,
the newly added flow data is obtained by calculation
Figure BDA0003043973220000024
The predicted bandwidth result of the EF service is
Figure BDA0003043973220000025
ONUiThe actually required bandwidth of the medium EF service is:
Figure BDA0003043973220000026
allocating bandwidth with the size equal to the actually required bandwidth for the EF service, and turning to the seventh step;
step four: the OLT generates the bandwidth actually needed by the AF and BE services,
OLT summarizes actual bandwidth requests of AF queue and BE queue respectively
Figure BDA0003043973220000027
And
Figure BDA0003043973220000028
the actual required bandwidth is equal to the bandwidth requested by the queue, and the formula is:
ri AF＝Ri AF
ri BE＝Ri BE
wherein r isi AFFor ONUiBandwidth actually required by the medium AF queue, ri BEFor ONUiBandwidth actually required by the BE queue;
step five: the allocation of the bandwidth is guaranteed and,
the three queues of EF, AF and BE are respectively recorded as a queue k (k is 1,2 and 3), and for EF traffic, the ONUiBandwidth G actually allocated to the corresponding queue ki k＝ri kFor AF, BE service, ONUiThe bandwidth allocated to the queue k is the bandwidth r actually required by the corresponding queue ki kAnd minimum guaranteed bandwidth Bi kGet ONUiBandwidth G to which queue k is actually allocatedi k＝min{ri k，Bi k}；
Step six: the allocation of the remaining bandwidth is carried out,
after the bandwidth allocation is ensured to BE completed, the OLT collects the remaining bandwidth requests of the AF and BE services, and performs bandwidth allocation on the requests in a queuing and scheduling mode, wherein the remaining total bandwidth which can BE scheduled by the OLT is as follows:
Figure BDA0003043973220000031
s is the remaining total bandwidth schedulable by the OLT, BtotalFor the total bandwidth of the upstream channels,
distributing S to each queue according to the weight of the service, wherein the total residual bandwidth obtained by the queue k in the N ONUs is SkThen, S iskDistributing the data to corresponding queues k in N ONUs;
step seven: the allocation is finished.
Specifically, in the sixth step, S is addedkThe specific allocation steps allocated to the corresponding queues k in the N ONUs are as follows:
a. if SkIf all the remaining bandwidth requests of the corresponding queues in the N ONUs are met, SkAll the data are distributed to the queue of each ONU;
b. if SkIf 0, no residual bandwidth is separable;
c. if SkDistributing the residual bandwidth according to the proportion of the minimum guaranteed bandwidth of the service in each ONU for all residual bandwidth requests smaller than all the residual bandwidth requests of the corresponding queues in the N ONUs, wherein the residual bandwidth distributed to the service corresponding to the queue k in each ONU is
Figure BDA0003043973220000041
Wherein, Bi kFor ONUiMinimum guaranteed bandwidth of medium queue k.
The invention has the beneficial effects that: the invention carries out bandwidth prediction and bandwidth allocation on the high-priority business EF on the basis of the RNN structural model, improves the prediction accuracy, carries out bandwidth allocation on AF and BE businesses on the premise of ensuring the high-priority business, and finally carries out bandwidth allocation on the residual bandwidth requests of the three businesses in a queuing scheduling mode, thereby ensuring the service quality of other businesses to a certain extent, improving the prediction accuracy, reducing the queue delay and improving the uplink bandwidth utilization rate.
Drawings
Fig. 1 is a schematic diagram of data transmission between an ONU and an OLT;
fig. 2 is a system block diagram of an EPON;
FIG. 3 is a flow chart of the present invention.
Detailed Description
The technical solution in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. All other embodiments, which can be obtained by a person skilled in the art without making any creative effort based on the embodiments of the present invention, belong to the protection scope of the present invention, and are specifically described below with reference to the embodiments.
As shown in fig. 2, the system structure diagram of an EPON is disclosed, and on the basis of fig. 2, the present invention discloses an EPON dynamic bandwidth allocation algorithm based on RNN, the flow of which is shown in fig. 3, and specifically includes the following steps:
firstly, a bandwidth request is reported,
respectively recording N ONUs uploading data to an OLT as ONUsi(i＝1,2，……N)，ONUi(i-1, 2, … … N) respectively buffering three different services of EF, AF, BE into three queues according to service priority, and reporting bandwidth requests of the three queues to the OLT;
the OLT determines whether it is EF traffic,
if the data packet is the EF service, performing bandwidth prediction, and calculating the bandwidth actually required by the EF service, as shown in fig. 1, the request frame reports the buffer information of the ONU queue at time T1, the OLT processes the request of the ONU at time T2-T3, and returns an authorization frame in response to the ONU at time T3, the ONU receives the authorization information of the OLT at time T4, and the T5 sends data to the OLT, and during the waiting time T1-T5, the ONU will still receive the upstream data, and the data amount is not included in the request frame, so the data packet can only be accumulated to the next polling cycle for sending, and therefore, bandwidth prediction needs to be performed on the data arriving during the waiting time.
OLT aggregates actual bandwidth requests of EF queues
Figure BDA0003043973220000051
Using RNN structure model to wait time TwaitPredicting the flow of the internal arrival, carrying out differential calculation on historical bandwidth requests, and setting the data packets arriving in the j period as
Figure BDA0003043973220000052
Then the average rate of packet arrival in period j is
Figure BDA0003043973220000053
Wherein, TcycleIn order to be a polling period,
the newly added flow data is obtained by calculation
Figure BDA0003043973220000054
The predicted bandwidth result of the EF service is
Figure BDA0003043973220000055
The prediction process is as follows: reading EF service historical period flow average rate data to a variable EF, taking the first 2126 period flow average rates in the variable EF as training data, taking the last 300 period average rate data in the variable EF as test data, then normalizing the average rates, taking the data of every continuous 60 periods as input characteristics x _ train, taking the data of the 61 st period as corresponding labels y _ train, and generating 2066 groups of training data; and similarly, generating 240 groups of test data, building a neural network layer by layer, configuring a training method to execute a training process and finally obtaining a prediction result.
ONUiThe bandwidth actually required by the medium EF service is:
Figure BDA0003043973220000061
allocating bandwidth with the size equal to the actually required bandwidth for the EF service, and ending the allocation;
if not, not executing bandwidth prediction, OLT generates the bandwidth actually needed by AF and BE service,
OLT summarizes actual bandwidth requests of AF queue and BE queue respectively
Figure BDA0003043973220000062
And
Figure BDA0003043973220000063
the actual required bandwidth is equal to the bandwidth requested by the queue, and the formula is:
ri AF＝Ri AF
ri BE＝Ri BE
wherein r isi AFFor ONUiBandwidth actually required by the medium AF queue, ri BEFor ONUiBandwidth actually required by the BE queue;
the allocation of the bandwidth is guaranteed and,
each queue of each ONU has a certain amount of minimum guaranteed bandwidth, the minimum guaranteed bandwidth is to avoid that the queues of each ONU are starved when the queues of each ONU have bandwidth requests in order to avoid low priority or other problems that the queues of each ONU are not starved, and the minimum guaranteed bandwidth B of each ONU is set according to the specific requirements of each ONUi k。
The three queues of EF, AF and BE are respectively recorded as a queue k (k is 1,2 and 3), and for EF traffic, the ONUiBandwidth G actually allocated to the corresponding queue ki k＝ri kFor AF, BE service, ONUiThe bandwidth allocated to the queue k is the bandwidth r actually required by the corresponding queue ki kAnd minimum guaranteed bandwidth Bi kGet ONUiBandwidth G to which queue k is actually allocatedi k＝min{ri k，Bi k}。
The allocation of the remaining bandwidth is carried out,
after the bandwidth allocation is ensured to BE completed, the OLT collects the remaining bandwidth requests of the AF and BE services, and performs bandwidth allocation on the requests in a queuing and scheduling mode, wherein the remaining total bandwidth which can BE scheduled by the OLT is as follows:
Figure BDA0003043973220000071
s is the remaining total bandwidth schedulable by the OLT, BtotalFor the total bandwidth of the upstream channels,
distributing S to each queue according to the weight of the service, wherein the total residual bandwidth obtained by the queue k in the N ONUs is SkThen, S iskDistributing the data to corresponding queues k in N ONUs, wherein the specific distribution steps are as follows:
a. if SkIf all the remaining bandwidth requests of the corresponding queues in the N ONUs are met, SkAll the data are distributed to the queue of each ONU;
b. if SkIf 0, no residual bandwidth is separable;
c. if SkDistributing the residual bandwidth according to the proportion of the minimum guaranteed bandwidth of the service in each ONU for all residual bandwidth requests smaller than all the residual bandwidth requests of the corresponding queues in the N ONUs, wherein the residual bandwidth distributed to the service corresponding to the queue k in each ONU is
Figure BDA0003043973220000072
Wherein, Bi kFor ONUiThe minimum guaranteed bandwidth of medium queue k, so far the allocation ends.
The algorithm of the invention predicts the high priority service, namely the expedited forwarding service EF and then allocates the bandwidth, the prediction process improves the prediction accuracy of the EF, reduces the queue delay and improves the uplink bandwidth utilization rate by introducing the RNN structural model, and then allocates the bandwidth for ensuring the forwarding service AF and the best effort service BE and all the residual bandwidths of the three services, thereby realizing the fairness and rationality of the bandwidth allocation of the AF and the BE to a certain extent and ensuring the service quality of the services.
It will be evident to those skilled in the art that the invention is not limited to the details of the foregoing illustrative embodiments, and that the present invention may be embodied in other specific forms without departing from the spirit or essential attributes thereof. The present embodiments are therefore to be considered in all respects as illustrative and not restrictive, the scope of the invention being indicated by the appended claims rather than by the foregoing description, and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein. Any reference sign in a claim should not be construed as limiting the claim concerned.

Claims (2)

1. An EPON dynamic bandwidth allocation algorithm based on RNN, characterized in that: the method comprises the following steps:
the method comprises the following steps: the bandwidth request is reported and the bandwidth request is reported,
respectively recording N ONUs uploading data to an OLT as ONUsi(i＝1,2，……N)，ONUi(i-1, 2, … … N) respectively buffering three different services of EF, AF, BE into three queues according to service priority, and reporting bandwidth requests of the three queues to the OLT;
step two: the OLT determines whether it is EF traffic,
if yes, executing bandwidth prediction, and going to the third step, and if not, not executing bandwidth prediction, and going to the fourth step;
step three: the bandwidth actually required for the EF traffic is calculated,
OLT aggregates actual bandwidth requests of EF queues
Figure FDA0003043973210000011
Using RNN structure model to wait time TwaitPredicting the flow of the internal arrival, carrying out differential calculation on historical bandwidth requests, and setting the data packets arriving in the j period as
Figure FDA0003043973210000012
Then the average rate of packet arrival in period j is
Figure FDA0003043973210000013
Wherein, TcycleIn order to be a polling period,
the newly added flow data is obtained by calculation
Figure FDA0003043973210000014
The predicted bandwidth result of the EF service is
Figure FDA0003043973210000015
ONUiThe actually required bandwidth of the medium EF service is:
Figure FDA0003043973210000016
allocating bandwidth with the size equal to the actually required bandwidth for the EF service, and turning to the seventh step;
step four: the OLT generates the bandwidth actually needed by the AF and BE services,
OLT summarizes actual bandwidth requests of AF queue and BE queue respectively
Figure FDA0003043973210000017
And
Figure FDA0003043973210000018
the actual required bandwidth is equal to the bandwidth requested by the queue, and the formula is:
ri AF＝Ri AF
ri BE＝Ri BE
wherein r isi AFFor ONUiBandwidth actually required by the medium AF queue, ri BEFor ONUiBandwidth actually required by the BE queue;
step five: the allocation of the bandwidth is guaranteed and,
respectively recording EF, AF and BE queuesFor queue k (k ═ 1,2,3), for EF traffic, the ONUiBandwidth G actually allocated to the corresponding queue ki k＝ri kFor AF, BE service, ONUiThe bandwidth allocated to the queue k is the bandwidth r actually required by the corresponding queue ki kAnd minimum guaranteed bandwidth Bi kGet ONUiBandwidth G to which queue k is actually allocatedi k＝min{ri k，Bi k}；
Step six: the allocation of the remaining bandwidth is carried out,
after the bandwidth allocation is ensured to BE completed, the OLT collects the remaining bandwidth requests of the AF and BE services, and performs bandwidth allocation on the requests in a queuing and scheduling mode, wherein the remaining total bandwidth which can BE scheduled by the OLT is as follows:
Figure FDA0003043973210000021
s is the remaining total bandwidth schedulable by the OLT, BtotalFor the total bandwidth of the upstream channels,
distributing S to each queue according to the weight of the service, wherein the total residual bandwidth obtained by the queue k in the N ONUs is SkThen, S iskDistributing the data to corresponding queues k in N ONUs;
step seven: the allocation is finished.
2. The RNN-based EPON dynamic bandwidth allocation algorithm of claim 1, wherein in the sixth step, S iskThe specific allocation steps allocated to the corresponding queues k in the N ONUs are as follows:
a. if SkIf all the remaining bandwidth requests of the corresponding queues in the N ONUs are met, SkAll the data are distributed to the queue of each ONU;
b. if SkIf 0, no residual bandwidth is separable;
c. if SkAll remaining bandwidth requests less than the corresponding queues in the N ONUs are guaranteed to the minimum according to the service in each ONUThe residual bandwidth is allocated according to the proportion of the bandwidth, and the residual bandwidth allocated to the service corresponding to the queue k in each ONU is
Figure FDA0003043973210000031
Wherein, Bi kFor ONUiMinimum guaranteed bandwidth of medium queue k.
CN202110467814.6A 2021-04-28 2021-04-28 RNN-based EPON dynamic bandwidth allocation algorithm Pending CN113076199A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110467814.6A CN113076199A (en) 2021-04-28 2021-04-28 RNN-based EPON dynamic bandwidth allocation algorithm

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110467814.6A CN113076199A (en) 2021-04-28 2021-04-28 RNN-based EPON dynamic bandwidth allocation algorithm

Publications (1)

Publication Number Publication Date
CN113076199A true CN113076199A (en) 2021-07-06

Family

ID=76619062

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110467814.6A Pending CN113076199A (en) 2021-04-28 2021-04-28 RNN-based EPON dynamic bandwidth allocation algorithm

Country Status (1)

Country Link
CN (1) CN113076199A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114039870A (en) * 2021-09-27 2022-02-11 河海大学 Deep learning-based real-time bandwidth prediction method for video stream application in cellular network

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20100254707A1 (en) * 2009-04-07 2010-10-07 Peng Jan-Wen Dynamic Bandwidth Allocation Method of Ethernet Passive Optical Network
CN102629916A (en) * 2012-05-07 2012-08-08 广西大学 Improved EPON (Ethernet passive optical network) dynamic bandwidth allocation method

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20100254707A1 (en) * 2009-04-07 2010-10-07 Peng Jan-Wen Dynamic Bandwidth Allocation Method of Ethernet Passive Optical Network
CN102629916A (en) * 2012-05-07 2012-08-08 广西大学 Improved EPON (Ethernet passive optical network) dynamic bandwidth allocation method

Non-Patent Citations (4)

* Cited by examiner, † Cited by third party
Title
N. A. M. RADZI: "The delay and fairness study of a centralised EPON DBA algorithm", 《2014 IEEE REGION 10 SYMPOSIUM》 *
董传成: "基于突发业务预测且公平的EPON动态带宽分配策略研究", 《中国优秀博硕士学位论文全文数据库(硕士)信息科技辑》 *
邢杨: "支持多业务的EPON动态带宽分配算法", 《通讯世界》 *
高凡等: "一种高带宽利用率10G EPON动态带宽分配算法", 《光通信技术》 *

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114039870A (en) * 2021-09-27 2022-02-11 河海大学 Deep learning-based real-time bandwidth prediction method for video stream application in cellular network

Similar Documents

Publication Publication Date Title
CN113254197B (en) Network resource scheduling method and system based on deep reinforcement learning
CN102594682B (en) Traffic-prediction-based dynamic bandwidth allocation method for gigabit-capable passive optical network (GPON)
CN110381541B (en) Smart grid slice distribution method and device based on reinforcement learning
CN111211830B (en) Satellite uplink bandwidth resource allocation method based on Markov prediction
CN109104373B (en) Method, device and system for processing network congestion
CN101512970A (en) Dynamic bandwidth allocation and service differentiation for broadband passive optical networks
CN111277437A (en) Network slice resource allocation method for smart power grid
CN104168318A (en) Resource service system and resource distribution method thereof
CN107038071A (en) A kind of flexible dispatching algorithm of Storm tasks predicted based on data flow
CN106954232A (en) A kind of resource allocation methods of time delay optimization
CN102158906B (en) Service quality sensory system and task scheduling method thereof
CN100337445C (en) QoS scheduler for voice service in broadband wireless MAN and method thereof
CN115190082B (en) Traffic flow scheduling method and device of TSN (traffic stream network)
CN109743217B (en) Self-adaptive resource adjusting method based on SVRA algorithm
CN114980339B (en) C-V2X multi-service downlink resource allocation method based on variable time slot scheduling
CN110489218A (en) Vehicle-mounted mist computing system task discharging method based on semi-Markovian decision process
CN106095581B (en) Network storage virtualization scheduling method under private cloud condition
CN103209494A (en) Real-time video business resource distribution method based on importance mark
CN113076199A (en) RNN-based EPON dynamic bandwidth allocation algorithm
Shen et al. Goodbye to fixed bandwidth reservation: Job scheduling with elastic bandwidth reservation in clouds
WO2023122954A1 (en) Multi-rendering task scheduling method for quality of service perception of interactive application
CN107665143A (en) Method for managing resource, apparatus and system
CN107769977B (en) Power communication network resource allocation method based on soft switch
CN114390043A (en) Method for achieving multi-path QUIC transmission optimization based on enterprise-level cloud storage system
CN109343940A (en) Multimedia Task method for optimizing scheduling in a kind of cloud platform

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
RJ01 Rejection of invention patent application after publication

Application publication date: 20210706

RJ01 Rejection of invention patent application after publication