ImageVerifierCode 换一换
格式:DOCX , 页数:32 ,大小:35.74KB ,
资源ID:4009429      下载积分:3 金币
快捷下载
登录下载
邮箱/手机:
温馨提示:
快捷下载时,用户名和密码都是您填写的邮箱或者手机号,方便查询和重复下载(系统自动生成)。 如填写123,账号就是123,密码也是123。
特别说明:
请自助下载,系统不会自动发送文件的哦; 如果您已付费,想二次下载,请登录后访问:我的下载记录
支付方式: 支付宝    微信支付   
验证码:   换一换

加入VIP,免费下载
 

温馨提示:由于个人手机设置不同,如果发现不能下载,请复制以下地址【https://www.bdocx.com/down/4009429.html】到电脑端继续下载(重复下载不扣费)。

已注册用户请登录:
账号:
密码:
验证码:   换一换
  忘记密码?
三方登录: 微信登录   QQ登录  

下载须知

1: 本站所有资源如无特殊说明,都需要本地电脑安装OFFICE2007和PDF阅读器。
2: 试题试卷类文档,如果标题没有明确说明有答案则都视为没有答案,请知晓。
3: 文件的所有权益归上传用户所有。
4. 未经权益所有人同意不得将文件中的内容挪作商业或盈利用途。
5. 本站仅提供交流平台,并不能对任何下载内容负责。
6. 下载文件中如有侵权或不适当内容,请与我们联系,我们立即纠正。
7. 本站不保证下载资源的准确性、安全性和完整性, 同时也不承担用户因使用这些下载资源对自己和他人造成任何形式的伤害或损失。

版权提示 | 免责声明

本文(MPCP 多点控制协议.docx)为本站会员(b****3)主动上传,冰豆网仅提供信息存储空间,仅对用户上传内容的表现方式做保护处理,对上载内容本身不做任何修改或编辑。 若此文所含内容侵犯了您的版权或隐私,请立即通知冰豆网(发送邮件至service@bdocx.com或直接QQ联系客服),我们立即给予删除!

MPCP 多点控制协议.docx

1、MPCP 多点控制协议IEEE802.3ah协议学习(3)MPCP多点控制协议 64.3 Multi-Point Control Protocol (MPCP)As depicted in Figure 643, the Multi-point MAC Control functional block comprises the following functions:a) Discovery Processing. This block manages the discovery process, through which an ONU is discovered and register

2、ed with the network while compensating for RTT.b) Report Processing. This block manages the generation and collection of report messages, through which bandwidth requirements are sent upstream from the ONU to the OLT.c) Gate Processing. This block manages the generation and collection of gate messag

3、es, through which multiplexing of multiple transmitters is achieved.多点MAC控制功能模块包括以下功能:a) 发现处理(Discovery Processing):该模块管理发现流程,通过该进程发现ONU 并使其注册到网络中,同时对其RTT 进行补偿。 b) 报告处理(Report Processing):该模块管理报告消息的产生和搜集,ONU 通过该模块在上行方向向OLT 发送带宽请求。 c) 选通处理(Gate Processing):该模块管理GATE 消息的产生和搜集,通过该模块实现多个发送端的复用。As depict

4、ed in Figure 643, the layered system may instantiate multiple MAC entities, using a single physical layer. Each instantiated MAC communicates with an instance of the opcode specific functional blocks through the Multi-point MAC Control. In addition some global variables are shared across the multipl

5、e instances. Common state control is used to synchronize the multiple MACs using MPCP procedures. Operation of the common state control is generally considered outside the scope of this document.如图64-3所示,分层的系统可以实例化多个MAC实体,而只使用一个物理层。每个实例化的MAC通过MPCP和与操作码(opcode)相关的功能模块的实例进行通信。此外,多个实例共享一些全局变量。通过MPCP使用公

6、共状态控制来同步多个MAC。有关公共状态控制的操作不在本标准规范的范围内。64.3.1 Principles of Multi-Point Control ProtocolMulti-point MAC Control enables a MAC Client to participate in a point-to-multi-point optical network by allowing it to transmit and receive frames as if it was connected to a dedicated link. In doing so, it employ

7、s the following principles and concepts:多点MAC控制通过控制MAC客户端的接收和发送使其工作在点对多点光网络中,而对MAC客户端来说就好象连接到一个独享的链路上。为了达到此目的,多点MAC控制协议遵循以下原则: a) A MAC client transmits and receives frames through the Multi-point MAC Control sublayer.b) The Multi-point MAC Control decides when to allow a frame to be transmitted usi

8、ng the client interface Control Multiplexer.c) Given a transmission opportunity, the MAC Control may generate control frames that would be transmitted in advance of the MAC Clients frames, utilizing the inherent ability to provide higher priority transmission of MAC Control frames over MAC Client fr

9、ames.d) Multiple MACs operate on a shared medium by allowing only a single MAC to transmit upstream at any given time across the network using a time-division multiple access (TDMA) method.e) Such gating of transmission is orchestrated through the Gate Processing function.f) New devices are discover

10、ed in the network and allowed transmission through the Discovery Processing function.g) Fine control of the network bandwidth distribution can be achieved using feedback mechanisms supported in the Report Processing function.h) The operation of P2MP network is asymmetrical, with the OLT assuming the

11、 role of master, and the ONU assuming the role of slave.a) MAC客户端通过多点MAC 控制子层发送和接收帧。 b) 多点MAC 控制决定何时允许客户端接口控制复用器发送一个帧。 c) 在某些情况下,当MAC 客户端发送帧时MAC 控制可能产生控制帧,并利用MAC 控制帧传送优先级高于MAC 客户帧的特性,优先发送MAC 控制帧。 d) 通过使用时分复用(TDMA)的方法,在任意给定时刻仅允许一个MAC 在网络中发送上行帧,使得多个MAC 可以在共享介质上操作。 e) 通过GATE 处理功能协调此类发送的GATE。 f) 通过发现处理功

12、能在网络中发现新设备并允许其发送数据。 g) 利用报告处理功能提供的反馈机制可以合理地控制网络带宽的分配。 h) P2MP 网络的操作是非对称的,其中OLT 为主设备,ONU 为从设备。64.3.2 Compatibility considerations64.3.2.1 PAUSE operationEven though MPCP is compatible with flow control, optional use of flow control may not be efficient in the case of large propagation delay. If flow

13、control is implemented, then the timing constraints in Clause 31B supplement the constraints found at 64.3.2.4.虽然MPCP和流量控制兼容,但是在传输时延很大的情况下可选的流量控制使用效率不高。 NOTEMAC at an ONU can receive frames from unicast channel and single-copy-broadcast (SCB) channel. If the SCB channel is used to broadcast data fra

14、mes to multiple ONUs, the ONUs MAC may continue receiving data frames from SCB channel even after the ONU has issued a PAUSE request to its unicast remote-end.注意:ONU端的MAC可以接收来自单播信道和单拷贝广播(SCB)信道的帧。如果SCB信道用于向多个ONU广播数据帧,那么即使ONU已向它的单播远端发出了PAUSE请求,ONU的MAC仍然可以继续从SCB信道接收数据帧。64.3.2.2 Optional Shared LAN Emu

15、lationBy combining P2PE, suitable filtering rules at the ONU, and suitable filtering and forwarding rules at the OLT, it is possible to emulate an efficient shared LAN. Support for shared LAN emulation is optional, and requires an additional layer above the MAC, which is out of scope for this docume

16、nt. Thus, shared LAN emulation is introduced here for informational purposes only.通过结合P2PE,ONU端的适当过滤规则以及OLT端的适当过滤转发规则,可以仿真一个有效的共享LAN。共享LAN仿真的支持是可选的,它需要MAC层上额外的一层,这超出了本标准规定的范围。因此,这里的共享LAN仿真仅仅做为一般性说明。Specific behaviour of the filtering layer at the RS is specified in 65.1.3.3.2.64.3.2.3 Multicast and

17、single copy broadcast supportIn the downstream direction, the PON is a broadcast medium. In order to make use of this capability for forwarding broadcast frames from the OLT to multiple recipients without multiple duplication for each ONU, the single-copy broadcast (SCB) support is introduced.在下行方向,

18、PON是一种广播介质。为利用PON的这一能力,使OLT向ONU发送广播帧时不对每个ONU都拷贝该帧,因此引入单拷贝广播(SCB)支持。The OLT has at least one MAC associated with every ONU. In addition one more MAC at the OLT is marked as the SCB MAC. The SCB MAC handles all downstream broadcast traffic, but is never used in the upstream direction for client traffi

19、c, except for client registration. Optional higher layers may be implemented to perform selective broadcast of frames. Such layers may require additional MACs (multicast MACs) to be instantiated in the OLT for some or all ONUs increasing the total number of MACs beyond the number of ONUs + 1.OLT针对每个

20、ONU至少有一个MAC与之对应。此外,OLT还有一个标记为SCB的MAC。SCBMAC处理所有下行广播业务,但是不能用于上行方向的业务流(客户注册除外)。组播功能可通过可选的上层功能实现。这些层可能需要在OLT中进行为某些或者所有的ONU实例化额外的MAC(组播MAC),从而使MAC的总数超过(ONU数+1)个。When connecting the SCB MAC to an 802.1D bridge port it is possible that loops may be formed due to the broadcast nature. Thus it is recommende

21、d that this MAC not be connected to an 802.1D bridge port.当把SCB MAC连接到802.1D桥接端口时,由于广播自身特点从而有可能形成回路。因此建议不把SCB MAC连接到802.1D桥接端口。SCB channel configuration as well as filtering and marking of frames for support of SCB is defined in 65.1.3.3.2.65.1.3.3.2节定义了SCB通道配置以及用于SCB支持的帧过滤和标记。64.3.2.4 Delay require

22、mentsThe MPCP protocol relies on strict timing based on distribution of timestamps. A compliant implementation needs to guarantee a constant delay through the MAC and PHY in order to maintain the correctness of the timestamping mechanism. The actual delay is implementation dependent, however, a comp

23、lying implementation shall maintain a delay variation of no more than 16 bit times through the implemented MAC stack.MPCP协议依赖于严格定时,这是通过分发时间戳来实现。为了维护时间戳机制的正确性,相应的实现必须保证从MAC到PHY的时延不变。实际的时延大小取决于实现方式。但是,遵循规范的实现应该使所实现的MAC栈时延维持在一定范围内,而该范围变化不超过16比特时间。The OLT shall not grant less than 1024 time_quanta into

24、the future, in order to allow the ONU processing time when it receives a gate message. The ONU shall process all messages in less than this period. The OLT shall not issue more than one message every 1024 time_quanta to a single ONU. The unit of time_quantum is defined as 16 ns.为了使得ONU在接收到GATE消息后有足够

25、的处理时间,OLT不能发布起点时刻早于当前时刻加上1024个时间量子(time_quantum)的授权。ONU应该用小于这个时间段的时间来处理所有的消息。OLT不应该以大于每1024个时间量子(time_quantum)一个消息的速度向一个ONU发布消息。时间量子(time_quantum)的单位是16ns。64.3.3 Discovery ProcessingDiscovery is the process whereby newly connected or off-line ONUs are provided access to the PON. The process is drive

26、n by the OLT, which periodically makes available Discovery Time Windows during which off-line ONUs are given the opportunity to make themselves known to the OLT. The periodicity of these windows is unspecified and left up to the implementer. The OLT signifies that a discovery period is occurring by

27、broadcasting a discovery gate message, which includes the starting time and length of the discovery window. Off-line ONUs, upon receiving this message, wait for the period to begin and then transmit a REGISTER_REQ message to the OLT. Discovery windows are unique in that they are the only times where

28、 multiple ONUs can access the PON simultaneously, and transmission overlap can occur. In order to reduce transmission overlaps, a contention algorithm is used by all ONUs. Measures are taken to reduce the probability for overlaps by artificially simulating a random distribution of distances from the

29、 OLT. Each ONU shall wait a random amount of time before transmitting the REGISTER_REQ message that is shorter than the length of the discovery time window. It should be noted that multiple valid REGISTER_REQ messages can be received by the OLT during a single discovery time period. Included in the

30、REGISTER_REQ message is the ONUs MAC address and number of maximum pending grants. Upon receipt of a valid REGISTER_REQ message, the OLT registers the ONU, allocating and assigning new port identities (LLIDs), and bonding corresponding MACs to the LLIDs.发现是指新连接或者非在线的ONU接入PON的进程。该进程由OLT发起,它周期性地产生合法的发

31、现时间窗口(Discovery Time Windows),使OLT有机会检测到非在线的ONU。发现时间窗口的周期没有定义,由厂商决定。OLT通过广播一个发现GATE消息来通知ONU发现窗口的周期。发现GATE消息包含发现窗口的开始时间和长度,非在线ONU接收到该消息后将等待该周期的开始,然后向OLT发送REGISTER_REQ消息。发现窗口是唯一有多个ONU同时访问PON的窗口,因此这些发送可能发生冲突。为了减少发送冲突,所有的ONU应使用同一种竞争算法。通过模拟ONU到OLT距离的随机分布等措施可以减少冲突发生的概率。每个ONU在发送REGISTER_REQ消息前应等待一段随机大小的时间,

32、该时间段小于发现时间窗口的长度。值得注意的是在一个发现时间周期内OLT可能会接收到多个有效的REGISTER_REQ消息。REGISTER_REQ消息中包括ONU的MAC地址以及最大等待授权(Pending Grant)的数目。OLT接收到有效的REGISTER_REQ消息后,将注册该ONU,分配和指定新端口的标识(LLID),并将相应的MAC和LLID绑定。The next step in the process is for the OLT to transmit a Register message to the newly discovered ONU, which contains the ONUs LLID, and the OLTs required synchronization time. Also, the OLT echoes the maximum number of pending grants. The OLT now has enough information to schedule the ONU for access to the PON and

copyright@ 2008-2022 冰豆网网站版权所有

经营许可证编号:鄂ICP备2022015515号-1