This disclosure relates to multiplexing and demultiplexing optical signals from one bit rate to a different bit rate or data rate.
Dense wavelength division multiplexing (DWDM) technology is typically used in core optical networks for transporting voice, data, and multimedia traffic. Local networks may be based on different technology such as SDH (Synchronous Digital Hierarchy), SONET (Synchronous Optical Networking), and PoS (Packet over SONET). Whilst existing OTN (Optical Transport Network) core networks utilising DWDM technology are designed for supporting STM-64/OC-192 (10G bit rate or data rate), newer technologies applied to local networks may support 40G data rates. Access to core networks from a local network is normally based on particular interfaces fitted to routers which provide suitable adaptations. The capacity of router ports have recently increased up to 40G using PoS technology (STM-256C/OC-768c). From a local network operator's view, this represents an advantage because it allows the reduction of the number of ports on the router. However, because existing OTN core networks are designed for supporting STM-64/OC-192 (10G data rate); they are not suitable for directly accepting these 40G signals.
Existing OTN technology allows for demultiplexing higher rate signals by mapping the higher rate signal into transport containers, adding a digital overhead wrapper to each transport container, and transmitting these digitally wrapped transport containers separately. This process is illustrated in
There is provided a method of demultiplexing client data from a constant bit rate (CBR) optical client serial link into lower bit rate CBR signals for transmitting over a plurality of wavelength division multiplexed (WDM) optical channels. The method comprises generating a plurality of CBR transport frames for transmitting a client frame received from the optical client serial link over respective WDM optical channels. Mapping blocks of bytes from the client frame into respective CBR transport frames for transmitting over the respective WDM optical channel, wherein the number of bytes in the client frame is the same as the number of bytes in the plurality of respective CBR transport frames. Modifying a number of non-payload bytes from the client frame to carry CBR transport frame control data in each of the CBR transport frames.
Modifying some of the non-payload bytes from the client frame to carry CBR transport frame control data allows the payload bytes of the client frame together with overhead data for the WDM optical channels to be carried in the same number of bytes as the received client frame. By selecting appropriate byte locations within the client frame for the CBR transport frame control data, client frame control data (non-payload) need not be transported separately. For example, unused byte locations may be used, or byte locations containing client frame control data which is easily replicated at the receiver. An example of this type of client frame control rate data is the frame alignment bytes of a 40G PoS client frame. This arrangement allows the signal from the optical client serial link to be demultiplexed into the WDM optical channels without using extra bytes. This compares with the OTN demultiplexing approach which would require additional bytes for the digital wrappers for each transport container. Thus the bytes from the client frame can be divided into OTN sized frames, but with no additional bytes required to carry overhead information for the WDM optical channels. For example a 40G PoS signal can be demultiplexed into only four 10G WDM optical channels whilst maintaining the same overall bit rate or data rate. This reduces the delay in adapting signals from one network into the other, and additionally reduces the signal processing load required when compared with the digital wrapping functions used in the standard OTN method illustrated in
The method may be implemented as a computer programme, running on an appropriate computer such as a DSP (digital signal processor) core, or a more general purpose processor. The computer programme may be carried on any suitable computer readable media such as a CD-ROM or a download signal for example. Alternatively, the method may be implemented on an ASIC (application specific integrated circuit) or a FPGA (field programmable gate array).
There is also provided a corresponding multiplexing method and hardware for implementing these demultiplexing and/or multiplexing methods.
Embodiments will now be described with reference to the following drawings, by example only and without intending to be limiting, in which;
a-d illustrates CBR transport frames for four respective WDM optical channels according to an embodiment;
Referring to
The core network 230 is implemented using wave division multiplexing (WDM) such as DWDM or coarse WDM (CWDM) and provides a 10G data rate across each wavelength. The core network 230 may comprise a network of fibre optic cable together with various nodes not explicitly shown for simplicity and receives 40G PoS signals 225R over a constant bit rate (CBR) optical client serial link 215R from the IP router 220R coupled to the first local network 210R. The core network also transmits 40G PoS signals 225T over a second CBR optical client serial link 215T to the router 220T coupled to the second local network 210T. The core network 230 also comprises a constant bit rate (CBR) transport frame (TF) demultiplexer 235 coupled to the first optical client serial link 215R and a CBR TF multiplexer 245 coupled to the second optical client serial link 215T. The CBR TF demultiplexer 235 demultiplexes received 40G PoS signals 225R into a plurality of lower bit rate CBR signals 245A-D each at a bit rate of 10G. Each 10G CBR signal 245A-D is transmitted across a separate respective wavelength optical channel of the core network 230. These 10G CBR signals 245A-D are received and multiplexed together by the CBR TF multiplexer 245 and transmitted as 40G PoS signals 225T onto the second optical client serial link 215T. These transmitted 40G PoS signals 225T are received by the second IP router 220T and routed and transmitted over the second local network 210T.
Whilst for simplicity a separate demultiplexer 235 and multiplexer 245 have been described, typically both demultiplexing and multiplexing functions (235 and 245) will be performed on either side of the core network 230 in order to transport higher bit rate signals (40G PoS) in either direction across the core network 230. Similarly, whilst the embodiment has been described with respect to 40G PoS signals, other STM-256/OC-768 signals can also be supported. Furthermore, whilst the embodiment has been described with respect to 40G bit rate local networks 210R and 210T, and a 10G bit rate WDM core network 230, the principles of the embodiment may also be applied to different bit rates on both the local and core networks. For example, local network signals at 120G bit rate may be transported across a 10G WDM core network utilising more wavelengths, or a 40G WDM core network. Similarly, lower bit rate local networks at, for example, 10G may be transported across a 2.5G WDM core network.
The demultiplexer 235 receives a client frame from the optical client serial link (215R) at step 305. In this embodiment the client frame is a 40G PoS frame 425R as illustrated in
Step 315 may be carried out by a mapping function 405 within the demultiplexer 235 and which maps one quarter of the 40G PoS client frame bytes into each CBR transport frame 410A-D. As can be seen, some of the frame alignment bytes 435R are mapped into each CBR transport frame as indicated by 440A-D respectively.
Step 320 will be described in more detail below. The demultiplexer 235 then modifies some non-payload bytes from the client frame to carry CBR transport frame control data in each CBR transport frame at step 325. This step may be carried out by a modify function 415 in the demultiplexer and is indicated by modified bytes 445A-D in respective CBR transport frames 410A-D. The CBR transport frames 410A-D are the same size as optical transport units (OTU) from OTN technology, and may therefore be sent directly to an OTN transponder 420 for conversion into the optical domain as 10G optical signals 245A-245D. These optical domain signals 245A-D are transmitted over respective optical channels 430A-D of the WDM core network 230. This process results in the transmission of the CBR transport frames as parallel CBR optical signals 245A-D over a plurality of respective WDM optical channels at step 330.
Whilst the embodiment has been described with respect to 40G PoS client frames and four respective CBR transport frames 410A-D, different types and data rate client frames could also be accommodated with suitable modifications. Similarly different numbers of CBR transport frames could be used for mapping the client frame. This may even include the possibility of using CBR transport frames which do not match the size of OTU frames. These CBR transport frames may be accommodated using OTN demultiplexing and digital wrapping. As a further alternative, the CBR transport frames provided may be smaller than the client frame resulting in the client frame being transmitted over two or more optical channel time periods and utilising suitable buffering at the receiver and multiplexer side of the core network.
The mapping of blocks of bytes from the client frame 425R into the CBR transport frames 410A-D may be achieved in various ways. In an embodiment, the blocks of bytes are individual columns 510 from the client frame 425R, which are mapped cyclically into the respective CBR transport frames 410A-D. For example, column one is mapped into the first CBR transport frame 410A, column two from the client frame 525R is mapped into the second CBR transport frame 410B, the third column is mapped into the third CBR transport frame 414C, column four is mapped into the fourth CBR transport frame, column five is then mapped into the first CBR transport frame 410A, and so on until the last column in the client frame 425R is mapped into the fourth CBR transport frame. This method of mapping blocks of bytes is partially illustrated in
Bits b7-b5 of the second CBR transport frame control data byte indicate the channel number—001=ch1, 010=ch2, 011=ch3, 100=ch4, 000=ch protection. Bits b4 and b3 convey information for the Link Loss Forwarding (LLF) management by inserting a Forward Incoming Client Fold (FCIF) indication. When an FCIF is recognised at the receiver a consequential action can be performed. 01=FCIF present, 00=FCIF not present, 10 and 11 are unused. The last three bits b2-b0 are unused in this embodiment. Similarly, CBR transport frame control data byte 3 is unused in this embodiment.
CBR transport frame control data bytes 4-6 are used for performance monitoring of each optical channel 430A-D. This embodiment employs bit interleaved parity x 24 code (BIPx24) using even plurality. The BIPx24 is computed over all bits of the previous CBR transport frame after scrambling, and is placed in CBR transport frame control data bytes 4, 5 and 6 of the current frame before scrambling.
Referring to
In this implementation, the interface 1010, apparatus for demultiplexing 1015, serial to parallel converter 1020, and grey transponders 1025A-D, are implemented on a single card. The output from this card is then fed directly into an existing transponder 420 implemented on a separate card. Thus implementing the method of demultiplexing client data according to this embodiment is easily achieved with an additional card 1005 interfacing with a DWDM core network using existing 10G transponders 420. The transponders 420 perform wave division multiplexing and guarantee proper optical performance. Thus the plurality of optical channels is enveloped into a standard G.709 structure for coupling directly into optical add/drop filters. For simplicity of explanation, only an apparatus for demultiplexing client data has been illustrated. However, as the multiplexing method is the reverse of the demultiplexing method, the same hardware may be used for receiving the plurality of wavelength division multiplexed optical channels 430A-D, multiplexing these into a client frame, and transmitting this client frame over a second CBR optical client serial link 215T.
In the multiplexing method 1100, CBR transport frames 410A-D are received from CBR optical signals 245A-D over a plurality of respective WDM optical channels 430A-D at step 105. This step is illustrated in
In order to accommodate differential delay due to chromatic dispersion, buffering may be implemented on the multiplexing card 1305. In order to minimise the buffering requirements and to avoid the need for an aliasing protocol, the optical channels 430A-D (and 930E when used) should be routed along the same optical pathway.
The embodiments provide a mechanism for migrating from 10G data rate signals to 40G data rate signals where local networks and the core network operate at different rates. Compared with the OTN demultiplexing and multiplexing approach using digitally wrapped transport containers, the CBR TF multiplexing and demultiplexing approach of the embodiments reduces overhead processing and memory requirements particularly as 40G signals can be converted into 10G WDM signals and vice versa without the need for digitally wrapping and hence extra overhead bytes. This reduces implementation complexity and cost. The embodiments can also make use of existing 10G OTN networks using any type of suitable 10G transponder in CBR mode.
Various modifications may be made to the described embodiments without departing from the scope of the invention as defined by the appended claims.
Number | Date | Country | Kind |
---|---|---|---|
08100045.7 | Jan 2008 | EP | regional |
Filing Document | Filing Date | Country | Kind | 371c Date |
---|---|---|---|---|
PCT/EP08/53756 | 3/28/2008 | WO | 00 | 10/4/2010 |