HK1169769A - Physical layer chip and method for distributing rates of epon mac traffic - Google Patents

Description

Method for carrying out rate division on EPON MAC flow and physical layer chip

Technical Field

The present invention relates to Ethernet Passive Optical Network (EPON) communications.

Background

Passive Optical Networks (PONs) are single shared fibers that are split into separate strands using inexpensive optical splitters to feed individual subscribers individually. Ethernet PON (epon) is a PON based on the ethernet standard. EPONs provide a simple, manageable link to ethernet-based IP equipment, either at the client or at a central office (central office). Like other gigabit ethernet media, EPONS is well suited to carry packet traffic.

Existing EPON Optical Link Terminals (OLTs) implement the EPON MAC layer. An EPON MAC may provide various packet processing capabilities, quality of service (QoS) functions, and management functions.

To ensure the most efficient use of EPON MAC functionality, it is desirable that the EPON MAC be end-to-end from the OLT to the Optical Network Unit (ONU). This means that the standard defined EPON MAC data rate (1Gbps or 10Gbps) must be used between the OLT and the ONUs. In practice, however, the physical medium between the OLT and the ONUs (which includes, for example, hybrid fiber-coax networks) does not always maintain the full data rate of EPONMAC. Thus, EPON MAC communication links need to be sub-rated in either upstream or downstream traffic.

Disclosure of Invention

According to one aspect of the present invention, there is provided a method for de-rating Ethernet Passive Optical Network (EPON) Medium Access Control (MAC) traffic on a communication link, the method comprising:

receiving a first control frame having a packet queue length value from a first EPON MAC layer;

increasing the packet queue length value in the first control frame;

transmitting the first control frame with the increased packet queue length value to a second EPON MAC layer;

receiving a second control frame having a transmission slot value from the second EPON MAC layer;

reducing the transmission slot value; and

forwarding the second control frame with the reduced timeslot transmission value to the first EPON MAC layer.

Preferably, the first control frame is a reporting multipoint control protocol data unit (MPCPDU) defined by ieee802.3ah or ieee802.3 av.

Preferably, the second control frame is a gate multipoint control protocol data unit (MPCPDU) defined by ieee802.3ah or ieee802.3 av.

Preferably, the first EPON MAC layer is located in an Optical Network Unit (ONU).

Preferably, the second EPON MAC layer is located in an Optical Link Terminal (OLT).

Preferably, the step of adding includes up-converting the packet queue length value based on at least one of an actual supported data rate of the communication link and a specified operating EPON MAC data rate.

Preferably, the specified operational EPON MAC data rate is one of 10Gbps for 10G-EPON and 1Gbps for 1G-EPON.

Preferably, the transmission slot value is determined according to the increased packet queue length value.

Preferably, the reducing step comprises downconverting the transmission slot value in dependence upon at least one of the data rate actually supported by the communications link and a specified operating EPON MAC data rate.

Preferably, said reducing step comprises down-converting said transmission slot value in inverse proportion to said up-converting of said packet queue length value.

Preferably, the method further comprises:

receiving data packets from the first EPON MAC layer at the specified operating EPON MAC data rate; and

transmitting the data packet over the communication link at the actual supported data rate of the communication link.

Preferably, the first and second EPON MAC layers are 10G-EPON MAC layers defined by IEEE802.3av or 1G-EPON MAC layers defined by IEEE802.3 ah.

Preferably, the communication link comprises a coaxial cable.

Preferably, the communication link comprises an optical fibre link.

Preferably, the communication link comprises a Hybrid Fiber Coaxial (HFC) network.

Preferably, the method is performed at the PHY layer.

Preferably, the PHY layer connects an Optical Network Unit (ONU) to the communication link.

Preferably, the PHY layer is located in a Coax Media Converter (CMC) located between an Optical Network Unit (ONU) and an Optical Link Terminal (OLT).

Preferably, the PHY layer is an EPON MAC layer or a coax PHY layer.

Preferably, the method is performed in a PHY device.

According to an aspect of the present invention, there is provided a physical layer (PHY) chip including:

a Media Access Control (MAC) interface to communicate with a first Ethernet Passive Optical Network (EPON) MAC layer;

the intercepting module is used for intercepting a first control frame from the first EPON MAC layer and increasing the length value of a packet queue in the first control frame;

and the transceiver module is used for transmitting the first control frame to a second EPON MAC layer.

Preferably, the transceiver module is further configured to receive a second control frame from the second EPON MAC layer, and the interception module is further configured to intercept the second control frame and reduce a transmission slot value in the second control frame.

According to one aspect of the invention, there is provided a computer program product comprising a computer usable hardware medium having computer program logic recorded therein, the computer logic when executed by a processor being operable to rate Ethernet Passive Optical Network (EPON) Medium Access Control (MAC) traffic on a communication link according to a method comprising:

receiving a first control frame having a packet queue length value from a first EPON MAC layer;

increasing the packet queue length value in the first control frame;

transmitting the first control frame with the increased packet queue length value to a second EPON MAC layer;

receiving a second control frame having a transmission slot value from a second EPON MAC layer;

reducing the transmission slot value; and

forwarding the second control frame with the reduced timeslot transmission value to the first EPON MAC layer.

According to one aspect of the present invention, there is provided a method of de-rating downstream Ethernet Passive Optical Network (EPON) Medium Access Control (MAC) traffic, the method comprising:

receiving a frame;

determining whether the received frame is a data frame or a control frame;

discarding the received frame if the received frame is a control frame and a control frame threshold has been exceeded; and

discarding the received frame if the received frame is a data frame and a data frame threshold has been exceeded.

Preferably, the method further comprises:

sending the received frame to a frame buffer if the received frame is a control frame and the control frame threshold is not exceeded;

sending the received frame to the frame buffer if the received frame is a data frame and the data frame threshold is not exceeded.

Preferably, the control frame threshold is greater than the data frame threshold.

Drawings

The accompanying drawings, which are incorporated herein and form a part of the specification, illustrate the present invention and, together with the description, further serve to explain the principles of the invention and to enable a person skilled in the pertinent art to make and use the invention.

Fig. 1 illustrates an exemplary Ethernet Passive Optical Network (EPON) -ethernet over coax passive optical network (EPOC) hybrid network architecture according to embodiments of the present invention;

fig. 2 illustrates another exemplary EPON-EPOC hybrid network architecture in accordance with embodiments of the present invention;

fig. 3 illustrates an exemplary end-to-end layered communication architecture between an Optical Link Termination (OLT) and a Coaxial Network Unit (CNU) in accordance with an embodiment of the present invention;

fig. 4 illustrates another exemplary end-to-end layered communication architecture between an Optical Link Termination (OLT) and a Coaxial Network Unit (CNU) in accordance with an embodiment of the present invention;

fig. 5 illustrates an exemplary process for de-rating upstream EPON MAC traffic in accordance with an embodiment of the present invention;

fig. 6 illustrates an exemplary flow of upstream EPON MAC traffic in accordance with an embodiment of the present invention;

fig. 7 is a process flow diagram of a method for de-rating upstream EPON MAC traffic in accordance with an embodiment of the present invention;

fig. 8 illustrates an exemplary process for de-rating downstream EPON MAC traffic in accordance with an embodiment of the present invention;

fig. 9 is a process flow diagram of a method for de-rating downstream EPON MAC traffic in accordance with an embodiment of the present invention.

The present invention will be described with reference to the accompanying drawings. In general, the drawing in which an element first appears is representatively illustrated by the leftmost digit(s) in the corresponding reference number.

Detailed Description

EPON-EPOC hybrid network embodiments

Fig. 1 illustrates an exemplary Ethernet Passive Optical Network (EPON) -ethernet over coax passive optical network (EPOC) hybrid network architecture according to embodiments of the present invention. As shown in fig. 1, exemplary network architecture 100 includes an Optical Link Termination (OLT)102, an optional passive optical splitter 106, a communication node 110 including a Coax Media Converter (CMC) 122, an optional amplifier 116, an optional coax splitter 118, a Coax Network Unit (CNU)122, and a plurality of user Media devices 124.

The OLT102 is located at a Central Office (CO) of the network and is connected to a fiber link 104. The OLT102 may implement a DOCSIS (data over cable service interface specification) mediation layer (DML) that may enable the OLT102 to provide DOCSIS provisioning and management of network components (e.g., CMCs, CMUs, Optical Network Units (ONUs)). In addition, OLT102 may implement an EPON Medium Access Control (MAC) layer (e.g., ieee802.3ah or ieee802.3 av).

Optionally, a passive splitter 106 may be used to split the fiber link 104 into multiple fiber links 108. This may enable multiple users in different geographical areas to be served by the same OLT102 in a point-to-multipoint topology.

The communication node 110 may act as a bridge between the EPON side and the EPOC side of the network. Thus, node 110 is connected to fiber link 108a from the EPON side of the network and to coaxial cable 114 from the EPOC side of the network. In an embodiment, communications node 110 includes a Coaxial Media Converter (CMC)112, CMC112 enabling EPON to EPOC bridging and conversion (and vice versa).

CMC112 performs physical layer (PHY) conversion from EPON to EPOC, and vice versa. In an embodiment, the CMC112 comprises a first interface (not shown in fig. 1) connected to the optical fiber link 108 for receiving the first optical signal from the OLT102 and generating a first bit stream having a first physical layer (PHY) encoding. In an embodiment, the first PHY encoding is EPON PHY encoding. The CMC112 also includes a PHY conversion module (not shown in fig. 1) coupled to the first interface to perform PHY layer conversion of the first bitstream and generate a second bitstream having a second PHY encoding. In an embodiment, the second PHY encoding is EPOC PHY encoding. Further, the CMC112 includes a second interface (not shown in fig. 1) coupled to the PHY conversion module and the coaxial cable 114 to generate a first Radio Frequency (RF) signal from the second bitstream and to transmit the first RF signal over the coaxial cable 114.

In EPOC to EPON conversion (i.e., during upstream communications), a second interface of the CMC112 is to receive a second RF signal from the CNU122 and generate a third bit stream having a second PHY encoding (e.g., EPOC PHY encoding) from the second RF signal. The PHY conversion module of the CMC112 is to perform PHY layer conversion of the third bitstream to generate a fourth bitstream having the first PHY encoding (e.g., EPON PHY encoding). The first interface of the CMC112 is then configured to generate a second optical signal from the fourth bitstream and send the second optical signal to the OLT102 via the optical fiber link 108.

Optionally, the amplifier 116 and the second splitter 118 may be placed in the path between the communication node 110 and the CNU 122. The amplifier 116 amplifies the RF signal on the coaxial cable 114 before the coaxial cable 114 is split by the second optical splitter 118. The second optical splitter 118 splits the coaxial cable 114 into a plurality of coaxial cables 120 to serve several subscribers in the same or different geographic areas over the coaxial cables.

CNUs 122 are typically located at the subscriber end of the network. In an embodiment, CNU122 implements the EPON MAC layer, whereby the end-to-end EPON MAC link is terminated by OLT 102. Thus, the CMC112 enables end-to-end assurance, management, and quality of service (QoS) functionality between the OLT102 and the CNUs 122. CNU122 may also provide GigE (gigabit ethernet) and 100M ethernet ports to connect user media devices 124 to the network. Thus, the CNU122 enables gateway integration of various services including VOIP (voice over IP), MoCA (multimedia over coax alliance), HPNA (home phoneline networking alliance), Wi-Fi (Wi-Fi alliance), etc. At the physical layer, CNU122 may perform physical layer conversion from coax to another medium while preserving the EPON MAC layer.

Depending on the desired services or infrastructure for the network, EPON-EPOC conversion may be performed anywhere in the path between OLT102 and CNU122 to provide various service configurations, according to an embodiment. For example, the CMC112 may not be integrated in the node 110, but rather integrated in the OLT102, the amplifier 116, or integrated in an Optical Network Unit (ONU) (not shown in fig. 1) located between the OLT102 and the CNUs 122.

Fig. 2 illustrates another exemplary EPON-EPOC hybrid network architecture 200, according to an embodiment of the present invention. In particular, the exemplary network architecture 200 enables synchronizing FTTH (fiber to the home) and multi-tenant building EPOC service configurations.

The exemplary network architecture 200 includes similar components in the exemplary network architecture 100 described above, including the OLT102, the passive splitter 106, the CMC112, and one or more CNUs 122 located in a CO hub. The OLT102, splitter 106, CMC112, and CNU122 operate in the same manner as described above with respect to fig. 1.

The CMC112 is located, for example, in a basement of the multi-tenant building 204. Thus, the EPON side of the network extends as far as possible to the subscribers, while the EPOC side of the network provides only a short coaxial connection between the CMC112 and the CNU units 122 located in the individual apartments of the multi-tenant building 204.

Furthermore, the exemplary network architecture 200 includes an Optical Network Unit (ONU) 206. The ONUs 206 are connected to the OLT102 via an all-fiber link, which includes fiber links 104 and 108 c. The ONU206 can provide FTTH service to the home 202, with the fiber link 108c reaching the boundary of the living space of the home 202 (e.g., a box on the outside wall of the home 202).

Accordingly, the exemplary network architecture 200 may enable an operator to service ONUs and CNUs using the same OLT. This includes end-to-end provisioning, management, and QoS for a single interface for both fiber and coaxial cable users. In addition, the exemplary network architecture 200 eliminates the conventional two-tier management architecture that uses media cells (media cells) on the end-user side to manage users and the OLT to manage media cells.

2. Hybrid Fiber Coax (HFC) network based on end-to-end EPON MAC

Fig. 3 illustrates an exemplary end-to-end layered communication architecture 300 between an Optical Link Termination (OLT) and a coaxial Cable Network Unit (CNU) in accordance with an embodiment of the present invention. The exemplary architecture 300 may enable two-way EPON-EPOC communication between the OLT102 and CNUs 122 via the CMC 112.

As shown in fig. 3, OLT102 and CNU122 are connected via a Hybrid Fiber Coaxial (HFC) network. The HFC network includes fiber optic link 302, CMC112, and coaxial cable 304. Those skilled in the art will appreciate that the HFC network of fig. 3 is used as an example only. In practice, the HFC network may include other components including amplifiers, splitters, and the like. The fiber optic link 302 and the coaxial cable 304 each include one or more links/cables that connect between network components.

The OLT102 implements an EPON PHY layer 306 and an EPON MAC layer 310. The CNU122 implements a coax PHY layer 308 and an EPON MAC layer 310. The CMC112 is located between the OLT102 and the CNUs 122 and only performs conversion at the PHY level between the OLT102 and the CNUs 122 and vice versa. In particular, CMC112 converts between EPOM PHY and coax PHY, and vice versa.

By implementing the same EPON MAC layer 310 at both the OLT102 and CNU122, the example network architecture 300 may enable an EPON MAC to be used in an end-to-end manner (i.e., from the OLT102 to the CNU122) to fully utilize the packet processing capabilities, QoS functions, and management functions of an EPON MAC over a Hybrid Fiber Coax (HFC) network.

Fig. 4 illustrates another exemplary end-to-end layered communication architecture 400 between an Optical Link Termination (OLT) and a Coaxial Network Unit (CNU) in accordance with an embodiment of the present invention.

Similar to exemplary architecture 300, exemplary architecture 400 may enable two-way EPON-EPOC communication between OLT102 and CNUs 122 via CMC 112. Still further, the exemplary architecture 400 may enable epon mac to be used in an end-to-end manner (i.e., from OLT102 to CNU 122).

As shown in fig. 4, OLT102 and CNU122 implement the same layer 2(L2) functionality 402, layer 2 including the EPON MAC layer. However, since the OLT102 and CNUs 122 are connected to different physical media (i.e., fiber and coax), the OLT102 and CNUs 122 implement different physical layers (PHYs) (layer 1).

The CMC112 is located between the OLT102 and the CNUs 122 and only performs conversion at the PHY level between the OLT102 and the CNUs 122 and vice versa. In particular, the CMC112 converts a first bit stream having PHY encoding of the OLT102 (e.g., EPON PHY encoding) into a second bit stream having PHY encoding of the CNU122 (e.g., EPOC PHY encoding), and vice versa. Accordingly, the conversion by the CMC112 does not affect or alter any frames in the received bitstream produced by layer 2 and above, including any frames produced by the EPON MAC layer implemented at the OLT102 or CNU 122. In other words, the data packets contained in the first bit stream and the second bit stream have the same MAC layer. In an embodiment, the MAC layer is an EPON MAC layer (e.g., an ieee802.3ahmac layer).

In an embodiment, the CMC112 includes two physical layers (PHYs) implementing first and second PHY stacks, respectively, the first PHY stack for transmitting raw bits over a fiber link and the second PHY stack for transmitting raw bits over a coax cable. Typically, the first PHY stack matches the PHY stack used by OLT102, while the second PHY stack matches the PHY stack used by CNU 122. In an embodiment, the first PHY stack is configured as an EPON PHY stack and the second PHY stack is configured as a coax PHY stack. In addition, the CMC112 includes a two-way conversion module to adjust an incoming bitstream received by a first PHY stack for transmission over a second PHY stack, and vice versa.

In an embodiment, as shown in fig. 4, the first PHY stack includes two sublayers 404 and 406. Sub-layer 404 performs power-related transmission functions over the optical fiber link, including determining and setting transmission power levels. The sub-layer 406 performs link coding functions including determining a link coding rate of an input bitstream received by the first PHY from the optical fiber link 302, removing (striping) the link coding of the input bitstream, and adding the link coding to the bitstream output from the first PHY. In an embodiment, the first PHY uses 8b/10b link coding.

The second PHY stack includes sublayers 408, 410, 412, and 414. The sublayer 408 performs link coding and packet framing functions including determining the link coding rate of the input bitstream received by the second PHY from the coax cable 304, removing the link coding of the input bitstream, and adding the link coding to the bitstream output from the second PHY. In an embodiment, the second PHY uses 64b/66b link coding. In addition, sublayer 408 may perform framing functions including adding framing bits to the output bitstream from the second PHY and removing framing bits of the input bitstream received by the second PHY. The framing bits may determine the start and end of a data packet in the bitstream.

Sublayer 410 may perform Forward Error Correction (FEC) functions including adding inner and/or outer FEC bits to the output bitstream from the second PHY, FEC correction (correct), and de-FEC bits from the input bitstream received by the second PHY.

Sublayer 412 performs Sub-Band Division Multiplexing (Sub-Band Division Multiplexing) functions including determining that a Sub-Band transmits an output bitstream from the second PHY, dividing the output bitstream into a plurality of Sub-bands (described further below with reference to fig. 5), determining widths of the Sub-bands, and assembling (assembling) the bitstream received by the second PHY over the plurality of Sub-bands to generate an input bitstream. According to an embodiment, the sub-layer 412 may implement any one of Subband Division Multiplexing (SDM), wavelet Orthogonal Frequency Division Multiplexing (OFDM), and Discrete Wavelet Multitone (DWMT), for example.

Sublayer 414 performs power-related transmission functions on the coaxial cable. Sublayer 414 may be a proprietary sublayer or other sublayer that may be employed by standard agencies.

The first PHY and the second PHY of the CMC112, along with optional other modules of the CMC112 (e.g., a link or interface module between the first and second PHYs), form a two-way conversion module that can condition an input bitstream received by the first PHY for transmission by the second PHY, and vice versa. In an embodiment, an input bitstream received by a first PHY over an optical fiber link is processed by sublayers 404 and 406 of a first PHY stack to generate an intermediate bitstream. The intermediate bit stream is then successively processed by sublayers 408, 410, 412, and 414 of the second PHY stack to generate an output bit stream for transmission by the second PHY over the coax. The input bitstream received by the second PHY over the coax cable is similarly conditioned for transmission by the first PHY over the fiber link.

It should be appreciated by those skilled in the art that the above descriptions of the example architectures 300 and 400 are provided by way of example only and are not limiting upon the embodiments of the present invention. For example, in other embodiments, layer 1(PHY) and layer 2(MAC) stacks and sublayers may be used to perform the media conversion functions described above.

3. sub-Rate (sub-rating) embodiments of EPON MAC traffic

Embodiments are discussed above that enable EPON MAC to be used in an end-to-end fashion on HFC networks. These embodiments enable the packet processing capabilities, QoS functions, and management functions of the EPON MAC layer to be leveraged on HFC networks.

The IEEE standard defines two data rates (upstream or downstream) for EPON MAC. These two data rates are 10Gbps (10G-EPON, IEEE802.3av) and 1Gbps (1G-EPON, IEEE802.3 ah). Thus, to run an EPON MAC end-to-end, the EPON MAC layers at the OLT and CNUs must use one of these two data rates.

In practice, an HFC network may not be able to support EPON MAC full data rates. For example, in some HFC networks, the EPOC portion of the network (e.g., the coaxial cable between the CMC and CNU) may not have sufficient bandwidth (due to limited spectrum or noise impairments) to support an EPON MAC full data rate. In other cases, cost factors force low cost transceivers to be unable to support the use of EPON MAC full data rates, for example, at the OLT, CMC, or CNU. Accordingly, to continue to run EPON MAC end-to-end on the network (per standard specification, EPON MAC itself is unmodified), EPON MAC traffic must be sub-rated according to the available bandwidth and/or transport capabilities of the physical medium.

As will be described further below in embodiments, EPON MAC traffic is rate-split according to available bandwidth and/or transmission capacity of the physical medium. The de-rating of EPON MAC traffic is performed in the PHY layer. The PHY layer may perform the rate splitting anywhere in the communication path that supports the end-to-end EPON MAC link, e.g., in the CNU or CMC. Because of the performance of the split-rate in the PHY layer, the EPON MAC layer at either end of the EPON MAC link does not find the split-rate being performed, and thus continues to operate in accordance with the IEEE standard to which it typically adheres.

Embodiments will be described below with reference to an HFC network having an architecture similar to that described above in fig. 3, for example, based on the teachings herein, one skilled in the art will appreciate that embodiments are not limited to this example architecture. Furthermore, embodiments are not limited to use in HFC networks. For example, embodiments may be applied to any EPON MAC based network that requires EPON MAC traffic split rates upstream or downstream. For example, embodiments may be applied to non-hybrid EPON networks where limitations in the network components of the OLT or ONU necessitate an EPON MAC traffic split rate (e.g., a low cost 2/3 x 10Gbps laser may be used instead of a 10Gbps laser to operate a 10G-EPON).

Fig. 5 illustrates an exemplary process 500 for split-rate upstream EPON MAC traffic in accordance with an embodiment of the present invention. The exemplary process 500 is described with reference to an exemplary HFC network having an architecture similar to that described above in fig. 3. Specifically, the exemplary HFC network includes OLT102, CMC112, and CNU 122. The OLT102 uses an EPON MAC layer 310 and an EPON PHY layer 306. CMC112 performs conversion between EPON PHY and coax PHY, and vice versa. The CNU122 uses a coax PHY layer 308 and an EPON MAC layer 310. Since CNUs 122 implement the EPON MAC layer, the CNUs appear to the OLT as ONUs. The example process 500 may be performed by a processor included in a PHY implementing subrate.

Generally, in an EPON, an OLT manages a plurality of ONUs. The EPON standard thus specifies that the OLT specify time slots during which the ONUs transmit. In particular, the EPON standard specifies a mechanism for allocating transmission timeslots to ONUs. The mechanisms include "Report" operations performed by the ONUs and "Gate" operations performed by the OLT. In the reporting operation, the ONU (having a packet to be transmitted) generates and transmits a report MPCPDU (multipoint control protocol data unit) (report frame) to the OLT. The report frame includes an indication of a packet queue length at the ONU. In response to the report frame, the OLT generates and transmits a gate MPCPDU (gate frame) to the ONU. The gate frame includes a time stamp, a transmission start time, and a transmission slot. The OLT determines the transmission time slot based on a specified operating EPON MAC data rate (e.g., 1Gbps or 10Gbps) and the packet queue length indicated in the report frame. Typically, transmission slots are specified in increments of 16ns time quanta (16ns is the time to transmit 2 bytes at 1Gbps in an EPON).

Embodiments utilize the EPON standard allocation mechanism described above. Specifically, as shown in fig. 5, the process 500 begins with the EPON MAC layer 310 of the CNU122, which CNU122 generates and sends report frames to the coax PHY layer 308 for transmission. The report frame includes a packet queue length value, which is the value that the packet queue length exhibits at the EPON MAC layer 310, and the destination of the report frame is the OLT 102. The report frame is forwarded over the MAC-PHY interface.

The coax PHY layer 308 intercepts the report frame from the EPON MAC310 prior to transmission on the coax 304 and upconverts (i.e., scales up) the packet queue length value in the report frame based on the data rate actually supported on the coax 304. In an embodiment, the packet queue length value is upconverted according to a ratio of a specified operating EPON MAC data rate to the data rate actually supported on the cable 304 (an upconversion ratio). For example, if the specified data rate for running an EPON MAC is 1Gbps and the actual supported data rate is 333Mbps, the packet queue length value is multiplied by a factor of 3. Based on the teachings herein, one of ordinary skill in the art will appreciate that other ways of up-converting the packet queue length value may be used depending on the embodiment. In an embodiment, the coax PHY layer 308 includes a MAC interface (not shown in fig. 5) for communicating with the epon MAC310, and an intercept module (not shown in fig. 5) that performs interception of the report frame and up-conversion of the packet queue length value prior to transmission of the report frame over the coax 304.

In an embodiment, PHY layer 308 determines the data rate actually supported on coaxial cable 304 based on one or more of the transmit/receive data rate of the transceiver at either end of coaxial cable 304, the available bandwidth of coaxial cable 304, the modulation efficiency on cable 304, and the measured transmission quality (e.g., SNR) of coaxial cable 304. It will be appreciated by those skilled in the art, in light of the teachings herein, that other parameters or measurements may also be utilized to determine the data rate actually supported on the coaxial cable 304. In an embodiment, coax PHY layer 308 includes a data rate determination module (not shown in fig. 5) for determining the data rate actually supported on coax 304 as described above.

After upconverting the packet queue length in the report frame, the coax PHY308 transmits the report frame to the OLT 102. The report frame is forwarded by the CMC112 to the OLT 102. At the OLT102, the report frame is received by the EPON PHY306 and forwarded to the EPON MAC 310.

The EPON MAC310 receives the report frames and processes the report frames in the usual manner in accordance with the EPON standard. Specifically, the EPON MAC310 determines a transmission start time and a transmission slot from the report frame, and generates a time-stamped Gateframe (time-stamped Gateframe) having the determined start time and transmission slot. The transmission time slot is determined based on a specified data rate (e.g., 1Gbps or 10Gbps) at which the EPON MAC is operating and an up-converted packet queue length value in the report frame. When the EPON MAC310 determines the start time, the RTT (round trip time: round trip delay) between the OLT102 and the CNU122 may also be calculated. RTT can be measured using gate and reporting operations according to the EPON standard.

The EPON MAC310 then sends a gate frame to the CNU122 (i.e., ONU-initiated report frame). Gate frames are transmitted by EPON PHY306 to coax PHY308 through CMC 112.

At the CNU122, the coax PHY308 intercepts the gate frames before forwarding them to the EPON MAC310, downconverts the transmission slot values indicated in the gate frames, and forwards the gate frames to the EPON MAC310 again. In an embodiment, the coax PHY308 downconverts the transmission slots by a ratio (downconversion ratio) inversely proportional to the upconversion ratio used to upconvert the packet queue length value in the report frame. In other words, the product of the up-conversion ratio and the down-conversion ratio is equal to one (1). In an embodiment, the same (or a different) intercept module that performs the upconversion of the packet queue length performs these steps.

Through the above process, the transmission time slots allocated for CNUs 122 to transmit their data packets are long enough so that all packets can reach OLT102 before any other predetermined ONU in the network begins transmitting with PLT102 on the same physical medium. The EPON MAC layer 310 at either end of the link operates under the same conditions if the physical medium supports an EPON MAC full data rate. In other words, the EPON MAC layer 310 transmits only at the full EPON MAC data rate and does not reduce its traffic in any way. Indeed, at the CNU122, the EPON MAC310 does not know that the transmission slot of the upconverted queue packet length value has been granted (grant) by the OLT102, so that transmission occurs at the EPON MAC full data rate, assuming a smaller transmission slot value is available to it.

When the EPON MAC layer 310 of the CNU122 receives gate frames from the coax PHY layer 308, the EPON MAC layer 310 processes the gate frames in accordance with the EPON standard and in accordance with its usual methods to conform normally. Specifically, the EPON MAC layer 310 updates a time stamp register according to a time stamp included in the gate frame, a slot start register (slot start register) according to a start time included in the gate frame, and a slot length register (slot length register) according to a down-conversion transmission slot value included in the gate frame. The EPON MAC layer 310 then waits the assigned start time to begin packet transmission.

When the allocated start time is reached, the EPON MAC layer 310 begins transmission of data packets at the specified operational EPON MAC data rate (e.g., 1Gbps or 10 Gbps).

Fig. 6 illustrates an exemplary flow 600 for upstream EPON MAC traffic in accordance with an embodiment of the present invention. Exemplary flow 600 occurs after the process shown in fig. 5 has been performed, exemplary flow 600 is described with reference to an HFC network that is the same HFC network as in fig. 3 and 5 above.

The example flow 600 illustrates the same segment of EPON MAC traffic sent from the EPON MAC310 layer of the CNU122 to the EPON MAC layer 310 of the OLT 102. An EPON MAC traffic segment may be a single EPON MAC packet or a packet stream having multiple EPON MAC packets.

Between the EPON MAC layer 310 and the coax PHY layer 308 of the CNU122, the EPON

The MAC traffic segment is located in the time threshold of segment 602 in fig. 6 for transmission at a specified operational EPON MAC data rate (e.g., 1Gbps or 10 Gbps). As above, the EPON MAC layer 310 in operation assumes that the time slot allocated by the OLT102 is equal to the down-converted transmission time slot contained in the gate frame.

Between coax PHY layer 308 of CNU122 and coax PHY layer 308 of CMC112, the EPON MAC traffic segment, which is transmitted at the data rate actually supported by coax 304, is within the time threshold of segment 602 in fig. 6. As above, the actual supported data rate of the coax cable 304 may be lower than the specified operating EPON MAC data rate. However, since the coax PHY layer 308 has previously upconverted the packet queue length value in the report frame sent to the OLT102, the coax PHY308 has gained more time to send the EPON MAC traffic fragment over the coax cable 304 to the CMC 112.

CMC112 receives EPON MAC traffic segments over coax 304 at data rates actually supported by coax 304. However, the CMC112 must retransmit EPON MAC traffic segments over the fiber link 302 at the specified operational EPON MAC data rate. Because the data rate actually supported by the coax cable 304 is lower than the specified operating EPON MAC data rate, the CMC112 cannot immediately retransmit the received EPON MAC traffic segment. Accordingly, in an embodiment, the CMC112 buffers the received EPON MAC traffic segment and, after a predetermined delay, begins transmitting the EPON MAC traffic segment over the fiber link 302 at the specified EPON MAC data rate. According to an embodiment, buffering may be performed in coax PHY layer 308 or in EPON PHY306 of CMC 112.

The transmission of the EPON MAC traffic segment over fiber link 302 occurs in the time domain of segment 606 in fig. 6. As above, transmissions are done at a specified operational EPON MAC data rate. The exemplary segment 608 in fig. 6 shows the overall traffic format on the fiber link 302. For example, segment 608 in fig. 6 depicts the entire traffic pattern, with idle periods followed by usage periods due to buffering at the CMC 112. Typically, the percentage of usage of the fiber link 302 is approximately equal to the down-conversion ratio used to down-convert the assigned transmission slot value at the coax PHY308 of the CNU 122.

Based on the teachings herein, one skilled in the art should appreciate that one or more of the steps of the above embodiments may be performed at different entities along the end-to-end EPON MCA link between the EPON MAC layer 310 of the CNU122 and the EPON MAV layer 310 of the OLT 102. For example, in the above-described embodiments, the subrate process (i.e., the up-conversion of queue packet length values, the down-conversion of transmission slot values, etc.) is described as being performed in the coax PHY layer 308 of the CNU 122. However, the present invention is not limited thereto. For example, one or more steps of the split-rate procedure may be performed in coax PHY308 of CMC112, EPON PHY306 of CMC112, or both (in addition to or instead of being performed in coax PHY layer 308 of CNU 122).

Still further, based on the teachings herein, one of ordinary skill in the art will appreciate that the above-described embodiments can be implemented in an adaptive manner to accommodate changes in conditions (e.g., transmission quality, noise, etc.). For example, in an embodiment, the data rate actually supported by the coaxial cable 304 is determined periodically and the rate split process is adjusted accordingly.

In other aspects, the HFC network architecture is such that the CMC112 is connected to the plurality of CNUs 122 via coaxial cables 304 and to the OLT102 via fiber optic links 304, respectively. Accordingly, the embodiments described above are implemented in the coax PHY layer 308 of each individual CNU 122. The embodiments implemented in the different CNUs operate independently, each based on a respective coaxial cable 304 connecting the CNU to the CMC 112. Thus, different CNUs can transmit to the CMC112 at different upstream data rates.

Fig. 7 is a process flow diagram 700 of a method for split-rate EPON MAC traffic in accordance with an exemplary embodiment of the present invention. Process 700 may be implemented in a PHY layer. The PHY layer may be a coax PHY layer as described in the exemplary embodiments above, or a PHY layer of another media type (e.g., an optical PHY layer such as an EPON PHY). The PHY layer may be a PHY that connects an ONU to an EPON or to an EPON/EPOC hybrid (i.e., HFC) network. Alternatively, the PHY layer may be located in an AMC between an OLT and an ONU in an EPON/EPOC network. Process 700 may also be performed by multiple PHY layers located at different locations in the network.

Process 700 begins at step 702, where step 702 includes receiving a first control frame having a packet queue length value from a first EPON MAC layer. In an embodiment, the first control frame is a reporting MPCPDU (reporting frame). The first EPON MAC layer may be a MAC layer of an ONU.

Step 704 includes increasing a packet queue length value in the first control frame. In an embodiment, step 704 includes upconverting the packet queue length value based on the actual supported data rate on the physical medium connected to the PHY layer. In an embodiment, the packet queue length value is upconverted according to a specified ratio of the operating EPON MAC data rate to the data rate actually supported on the medium (upconversion ratio).

Step 706 then includes sending the first control frame with the increased packet queue length value to a second EPON MAC layer. In an embodiment, the second EPON MAC layer is located in an OLT that manages the ONUs.

Step 708 includes receiving a second control frame having a transmission slot value from the second EPON MAC layer. In an embodiment, the second control frame is a gated MPCPDU (gate frame).

Step 710 includes reducing a transmission slot value contained in the second control frame. In an embodiment, step 710 includes down-converting the transmission slot value with a ratio (down-conversion ratio) inversely proportional to an up-conversion ratio for up-converting the packet queue length value of the first control frame in step 704.

Finally, step 712 includes forwarding the second control frame with the reduced transmission value of the time slot to the first EPON MAC layer. Subsequently, the first EPON MAC layer starts data transmission based on the reduced slot transmission value but at a normal EPON data rate (1Gbps or 10 Gbps).

As above, the EPON MAC data rates defined by the IEEE standard for upstream and downstream communications are both 1Gbps and 10 Gbps. Thus, in the event that the EPOC portion of the HFC network (e.g., coax 304) is unable to maintain the specified operating EPON MAC data rate, sub-rate downstream EPON MAC traffic (as with upstream EPON MAC traffic) must be split. According to an embodiment, downstream EPON MAC traffic is subject to de-rating according to available bandwidth and/or transmission capacity of the physical medium. In an embodiment, downstream EPON MAC traffic is rate-split in the PHY layer. The PHY layer performing the de-rating may be located anywhere in the communication path supporting the end-to-end EPON MAC link. In an embodiment, the sub-rate is performed on downstream EPON MAC traffic in the OLT (either in the EPON MAC or EPON PHY layer of the OLT), in the CMC, or both. However, in some embodiments, de-rating of EPON MAC traffic cannot be performed in the OLT. Therefore, the rate division can only be performed at other downlinks, e.g., at the CMC. Because the split rate is performed in the PHY layer, the EPON MAC layer at either end of the EPON MAC link is unaware that the split rate is being performed and will continue to operate in accordance with the IEEE standard as per conventional implementations.

Fig. 8 illustrates an exemplary process for sub-rate downstream EPON MAC traffic in accordance with an embodiment of the present invention. The example process 800 illustrates an arrangement in which the CMC performs de-rating of downstream EPON MAC traffic only. The example process 800 may be used when a split rate cannot be performed at the OLT. As shown in fig. 8, the process 800 begins with OLT102 sending downstream traffic over fiber link 302 at a specified operating EPON MAC data rate (e.g., 1 Gbps).

The CMC112 receives downstream traffic transmitted from the OLT102 at a specified operational EPON MAC data rate (e.g., 1 Gbps). However, CMC112 will not transmit on coaxial cable 304 at a data rate greater than that actually supported by coaxial cable 304 (e.g., 500Mbps in example process 800). Thus, when the data rate actually supported by the coax cable 304 is lower than the specified operating EPON MAC data rate, a portion of the downstream traffic EPON MAC must be dropped at the CMC 112.

In an embodiment, as shown in FIG. 8, the CMC112 implements a buffer media module (buffer media module)802 and a frame buffer 804. Frames received by the CMC112 are processed by the buffer medium 802 to determine whether the frames are to be dropped or sent to the buffer 804 for further transmission on the coaxial cable 304. Those skilled in the art will appreciate that this process may also be performed on a packet-by-packet (packet-by-packet) level.

Under certain conditions, the sub-rate of EPON MAC traffic at the CMC112 may result in out-of-order packets at the CNU 122. To address this issue, timestamps are inserted in packets sent by the CMC112 and extracted by the CNU122 to rearrange the received packets as needed.

Fig. 9 is a process flow diagram 900 of a method for sub-rated downstream EPON MAC traffic in accordance with an embodiment of the present invention. For example, the process 900 may be performed by the buffered media module 802.

Process 900 begins at step 902, where step 902 includes determining whether a received frame is a data frame (user data) or a control frame (MPCP frame).

If the received frame is a control frame, process 900 proceeds to step 904, which includes determining if a control frame threshold has been exceeded. In an embodiment, the control frame threshold is exceeded when a first predefined level of the buffer is reached (i.e. the buffer is filled up to or exceeds the first predefined level). If the control frame threshold has been exceeded, the control frame is discarded in step 908. Otherwise, the control frame is sent to the frame buffer in step 910.

In a similar manner, if the received frame is a data frame, process 900 proceeds to step 906, which includes determining if a data frame threshold has been exceeded. In an embodiment, the data frame threshold is exceeded when a second predefined level of the buffer is reached (i.e. the buffer is filled or exceeds the second predefined level). If the data frame threshold has been exceeded, the data frame is discarded in step 912. Otherwise the data frame is sent to the frame buffer in step 910.

Typically, control frames are received at a lower data rate than data frames, and the importance of control frames is generally higher than the importance of data frames. Accordingly, in an embodiment, to ensure that the frequency with which control frames are dropped is less than the frequency with which data frames are dropped, a first predefined level of the buffer (which determines whether the control frame threshold is exceeded) is configured to be higher than a second predefined level of the buffer (which determines whether the data frame threshold is exceeded).

Based on the teachings herein, one of ordinary skill in the art will appreciate that the process 900 may be modified to accommodate OAM (operation, administration, and maintenance) frames in addition to control and data frames. Alternatively, the OAM frame may be considered a control frame in process 900.

The embodiments have been described above with the aid of functional means illustrating the implementation of specific functions and relationships thereof. Here, the boundaries of these functional components have been determined for the convenience of description. Alternate boundaries may be determined so long as the specified functions and relationships thereof are appropriately performed.

The foregoing description of the specific embodiments will so fully reveal the general nature of the invention that others can, by applying knowledge within the skill of the art, readily modify and/or adapt for various applications such specific embodiments, without undue experimentation, without departing from the general concept of the present invention. Therefore, such adaptations and modifications are intended to be within the meaning and range of equivalents of the disclosed embodiments, based on the teaching and guidance presented herein. It is to be understood that the phraseology or terminology herein is for the purpose of description and not of limitation, such that the terminology or phraseology of the present specification is to be interpreted by the skilled artisan in light of the teachings and guidance.

The breadth and scope of an embodiment of the present invention should not be limited by any of the above-described exemplary embodiments, but should be defined only in accordance with the following claims and their equivalents.

Cross Reference to Related Applications

This patent application is entitled to priority of U.S. provisional patent application No.61/472,017, filed 5/2011 (attorney docket No.12/2875.5540000), and is related to U.S. provisional patent application No.12/878,643, filed 9/2010, which is entitled to priority of U.S. provisional patent application No. 61/240,935, filed 9/2009, and U.S. provisional patent application No.61/306,745, filed 2/22/2010, all of which are incorporated by reference herein in their entirety.

Claims (10)

1. A method for de-rating Ethernet Passive Optical Network (EPON) Medium Access Control (MAC) traffic on a communication link, the method comprising:

receiving a first control frame having a packet queue length value from a first EPON MAC layer;

increasing the packet queue length value in the first control frame;

transmitting the first control frame with the increased packet queue length value to a second EPON MAC layer;

receiving a second control frame having a transmission slot value from the second EPON MAC layer;

reducing the transmission slot value; and

forwarding the second control frame with the reduced timeslot transmission value to the first EPON MAC layer.

2. The method of claim 1, wherein the first EPON MAC layer is located in an Optical Network Unit (ONU).

3. The method of claim 1, wherein the second EPON MAC layer is located in an Optical Link Termination (OLT).

4. The method of claim 1 wherein the step of adding comprises up-converting the packet queue length value based on at least one of an actual supported data rate of the communication link and a specified operating EPON MAC data rate.

5. The method of claim 4, wherein the transmission slot value is determined based on the increased packet queue length value.

6. The method of claim 5, wherein the reducing step comprises downconverting the transmission slot value based on at least one of an actual supported data rate of the communication link and a specified operating EPON MAC data rate.

7. The method of claim 5, wherein the reducing step comprises down-converting the transmission slot value in inverse proportion to the up-converting of the packet queue length value.

8. The method of claim 4, further comprising:

receiving data packets from the first EPON MAC layer at the specified operating EPON MAC data rate; and

transmitting the data packet over the communication link at the actual supported data rate of the communication link.

9. A physical layer (PHY) chip, comprising:

a Media Access Control (MAC) interface to communicate with a first Ethernet Passive Optical Network (EPON) MAC layer;

the intercepting module is used for intercepting a first control frame from the first EPON MAC layer and increasing the length value of a packet queue in the first control frame;

and the transceiver module is used for transmitting the first control frame to a second EPON MAC layer.

10. A method for de-rating downstream Ethernet Passive Optical Network (EPON) Medium Access Control (MAC) traffic, the method comprising:

receiving a frame;

determining whether the received frame is a data frame or a control frame;

discarding the received frame if the received frame is a control frame and a control frame threshold has been exceeded; and

discarding the received frame if the received frame is a data frame and a data frame threshold has been exceeded.