Patent Grant
7643753
The present invention relates generally to communication using broadband passive optical networks (PONs), and more particularly to implementing PON processing on a single integrated circuit.
As the demand from users for bandwidth is rapidly increasing, optical transmission systems, where subscriber traffic is transmitted using optical networks, is installed to serve this demand. These networks are typically referred to as fiber-to-the-curb (FTTC), fiber-to-the-building (FTTB), fiber-to-the-premise (FTTP), or fiber-to-the-home (FTTH). Each such network provides an access from a central office (CO) to a building, or a home, via optical fibers installed near or up to the subscribers' locations. As the transmission quantity of such an optical cable is much greater than the bandwidth actually required by each subscriber, a passive optical network (PON), shared between a plurality of subscribers through a splitter, was developed.
An exemplary diagram of a typical PON 100 is schematically shown in
In the related art, PONs are classified into one of the following: an ATM PON (APON), a broadband PON (BPON), an Ethernet PON (EPON or GE-PON), and a Gigabit PON (GPON). The APON uses the ATM protocol; the BPON is designed to provide broadband services over an ATM protocol; the EPON accommodates an Ethernet protocol; and the GPON is utilized when accommodating both the ATM protocol and the Ethernet protocol. Each type of PON is a standardized technology as detailed in Table 1.
OLT 130 and ONUs 120 provide the interface between the optical network and homes or businesses, and thus define the type of the PON. For example, in order to establish a BPON the OLT 130 and ONUs 120 should be compatible with the BPON standards, i.e., ITU-T-G.983.X series. Specifically, an ONU includes a PON processor utilized to receive downstream traffic from an OLT, and provide the contents of the downstream traffic to one or more subscriber devices. Similarly, the processor of the ONU is deigned to receive and transmit upstream data from the one or more subscriber devices to the OLT via the passive optical network. An example of such PON processor may be found in U.S. patent application Ser. No. 10/340,635 entitled “Integrated PON processor” which is incorporated herein by reference for its useful background description of the state of the art heretofore.
A limitation of PON processors known in the art is the inability to operate in multiple PON modes. For example, the PON processor described in U.S. Ser. No. 10/340,635 is compliant only with the ITU G.983.X recommendations. This requires from service providers additional investments in infrastructure when upgrading their PONs, for example, from EPON to GPON.
Therefore, in the view of the limitations introduced in the prior art it would be advantageous to provide an enhanced PON processor adapted to operate in a mixed-mode. It would be further advantageous if the PON processor were compatible with the GPON standard.
It is therefore an object of the present invention to provide an enhanced PON processor adapted to operate in a mixed-mode.
It is a further object to provide such a PON processor that is compatible with the GPON standard.
These objects are realized in accordance with an aspect of the invention by an enhanced passive optical network (PON) processor adapted to operate in a mixed-mode, said PON processor comprising:
a microprocessor;
a packet processor for processing PON traffic processing PON traffic and being coupled to said microprocessor for providing instructions thereto;
an Ethernet media access control (MAC) adapter coupled to the packet processor for interfacing with a plurality of subscriber devices; and
a PON MAC adapter coupled to the packet processor for handling a plurality of PON operating modes.
According to a second aspect of the invention there is provided a method for processing data flows of a plurality of passive optical network (PON) operating modes, the method comprising:
processing upstream data flows of said plurality of PON operating modes; and
processing downstream data flows of said plurality of PON operating modes.
In order to understand the invention and to see how it may be carried out in practice, an embodiment will now be described, by way of non-limiting example only, with reference to the accompanying drawings, in which:
In the following description, reference will be made to devices of which more than one are referenced in the drawings by a generic numeric descriptor N followed by a numeric suffix of the form N−1, N−2 . . . N−m. To the extent that reference is made to the devices without regard to a specific one thereof, such device(s) will be referenced by the generic numeric descriptor N only.
The present invention discloses an enhanced optical network (PON) processor adapted to serve a plurality of PON applications. The processor is a highly integrated communications processor that can operate in different PON modes including, but not limited to, a GPON, a BPON, an EPON, or any combination thereof. In an exemplary embodiment of the present invention the provided PON processor is fabricated on a single integrated circuit (IC).
Referring to
The microprocessor 210 executes commands received from the packet processor 230. The microprocessor 210 performs fast processing, where the execution of each command is preferably completed in one clock cycle. In an exemplary embodiment of the present invention, the microprocessor 210 may be a high-performance MIPS 4KEC RISC microprocessor, with 16 Kbytes of two-way instruction cache and eight (8) Kbytes of two-way data cache. The SIU 220 provides both microprocessor 210 and packet processor 230 with interface and control signals.
The broad bus 240 interfaces between the Ethernet MAC adapter 250 as well as PON MAC adapter 260 and the packet processor 230. The broad bus 240 transfers data at high rates and its architecture is based on a push-ahead mechanism, using a binary tree topology. The broad bus 240 supports parallelism in read and write transactions and allows simultaneous transfer of data from various units at the same time. A more detailed description of the broad bus 240 can be found in U.S. patent application Ser. No. 11/459,553 commonly assigned to the same assignee as the present application, and whose contents are hereby incorporated by reference.
The Ethernet MAC adapter 250 includes a plurality of Ethernet interfaces for interfacing with a plurality of subscriber devices. These interfaces may be, but are not limited to, 100 Mbit Ethernet or 1 Gigabit Ethernet. The Ethernet MAC adapter 250 is capable of receiving upstream data flow from subscriber devices and transmitting downstream data to subscribers. Either upstream or downstream data flows are respectively forwarded to or received from the packet processor 230 via the broad bus 240.
The PON MAC adapter 260 is capable of processing traffic in accordance with the various PON modes that include, but are not limited to, a GPON, a BPON, an EPON, or any combination therefore. Specifically, the PON adapter 260 is adequate to serve the needs of a multi-service ONU operating in a point to multi point optical network. The GPON MAC adapter 260-1, the BPON MAC adapter 260-2, and the EPON MAC adapter 260-3 respectively support the GPON, BPON and EPON standards. It should be noted that the PON MAC adapter 260 ought to include at least one of the GPON, BPON, or EPON adapters. If two or more adapters are installed, then the PON processor 200 can be adapted to operate in a mixed-mode. An enhanced PON processor 200, configured to operate in a mixed-mode, allows service providers to easily upgrade their networks without any additional investments on infrastructure. One technique for upgrading the PONs is disclosed in U.S. provisional application No. 60/687,442 commonly assigned to the same assignee as the present application, and which is hereby incorporated by reference. Each of the MAC adapters 260-1, 260-2, and 260-3 includes a receiver to handle downstream data flow and a transmitter for handling upstream data flows. The operation of the PON MAC adapter 260 will be described in greater detail below. The SERDES unit 270 deserializes the data and converts it into a format compliant with the respective standard. The SERDES unit 270 also serializes data before transmitting the data to the optical network.
The packet processor 230 is adapted to perform PON processing tasks, such as filtering, forwarding-and-learning, flow classification, packets classification, ATM queuing and shaping, reassembling of packets, and so on. Data processed by the packet processor 230 may be either an upstream flow, i.e., data sent from a subscriber device to an OLT or a downstream flow, i.e., data sent from an OLT to a subscriber device.
The packet processor 230 includes a core processor 310, a plurality of hardware (HW) accelerators 320-1 through 320-6, and a memory 330. The core processor 310 may be, for example, a RISC machine that is designed to execute processing tasks with minimal latency. For this purpose, all arithmetic and logic operations as well as source and destinations variables are registers based. The only operations that require access to the memory 330 are load and store operations. Furthermore, the core processor 310 is designed with separate channels utilized for program, data, and context accesses. Specifically, the memory units included in the memory 330 are high speed synchronous memories that are used for program, data and context. The program memory 330-1 is a read only memory that holds tasks' instructions. The program memory 330-1 is accessible by the microprocessor 210. The data memory 330-2 is a read/write memory that keeps data of the various tasks. The context memory 330-3 is a special memory that holds instances of registers used by core processor 310. When switching contexts, the previous context is saved in memory 330-3 and a new context is fetched. The context memory 330-3 is also accessible by the microprocessor 210.
The hardware accelerators 320 are dedicated hardware processing components designed to increase the packet processor 310 performance by speeding up time consuming tasks. These dedicated processing components including at least a lookup table 320-1, a cyclical redundancy checking (CRC) accelerator 320-2, a scheduler 320-3, a register file 320-4, a direct memory access (DMA) 320-5, and a board bus interface 320-6. The lookup table 320-1 includes MAC addresses used for access both PON and Ethernet MAC adapters 260 and 250. The CRC accelerator 320-2 enables the fast CRC calculation for data received through the PON. The CRC accelerator 320-2 operates off line on data stored in data memory 330-2. The scheduler 320-3 receives requests from the different request generators and determines the next thread number when a context switch is triggered. The next thread is selected according to a predefined priority policy. The DMA 320-5 is responsible for data transfer from and to the memory 330 and an external memory. The register file 320-4 includes all configuration and input/output (I/O) space register. Configuration registers can be read and written by the microprocessor 210, while the input/output (I/O) registers are for the core processor 310 internal uses.
Referring to
The process described above is mainly used for handling Ethernet payloads. However, the present invention is further capable of handling, in a GPON mode, TDM payloads. A TDM downstream payload is sent directly from an ingress of the PON RX 464 to a TDM unit 480, namely without passing through the packet processor 230. Similarly, a TDM upstream payload is directly transferred from Ethernet TX 456 to the TDM unit 480.
Referring to
Referring back to
In accordance with one embodiment, the PON processor 200 is capable of handling a plurality of downstream flows in parallel. The way multiple flows are handled in GPON or BPON is different. For example, in GPON the reassembly of multiple flows is entirely performed by the packet processor 230. The PON RX 464 only generates start flow and end flow per each Port-ID on which a flow is received. In BPON multiple downstream flows are supported by interleaving of ATM cells from different virtual channels (VCCs), i.e., cells having different flow-IDs. The flows reassembly (i.e., performing ATM Adaptation layer 5) is entirely executed by packet processor 230 and the PON RX 464 merely marks, per flow, the start of packet and end of packet cell.
Referring to
The enhanced PON processor has been described with reference to a specific embodiment where the SIU 220, the broad bus 240, the SERDES unit 270, the internal bus 280, and the memory controller 290 are all independent components that are integrated in the processor 200. However, other embodiments will be apparent to those of ordinary skill in the art. For example, a PON processor may be capable of operating in a mixed mode without including all of these components, whose functionality components may be provided externally or internally by other means. Likewise, the functionality of these components may be integrated in the packed processor 230 or adapters 250 and 260. Moreover, the broad bus 240 may be replaced by any internal bus known in the art. Also, the SIU 220, the broad bus 240, and the SERDES unit 270 can be replaced by external devices, and the memory controller 290 can be removed by replacing the external memory with internal memory.
| Number | Name | Date | Kind |
|---|---|---|---|
| 6229788 | Graves et al. | May 2001 | B1 |
| 6330584 | Joffe et al. | Dec 2001 | B1 |
| 6385366 | Lin | May 2002 | B1 |
| 6519255 | Graves | Feb 2003 | B1 |
| 7385995 | Stiscia et al. | Jun 2008 | B2 |
| 20030058505 | Arol et al. | Mar 2003 | A1 |
| 20030137975 | Song et al. | Jul 2003 | A1 |
| 20040136712 | Stiscia et al. | Jul 2004 | A1 |
| 20040202470 | Lim et al. | Oct 2004 | A1 |
| 20040208631 | Song et al. | Oct 2004 | A1 |
| 20040218534 | Song et al. | Nov 2004 | A1 |
| 20040264961 | Nam et al. | Dec 2004 | A1 |
| 20070019956 | Sorin et al. | Jan 2007 | A1 |
| Number | Date | Country | |
|---|---|---|---|
| 20070070997 A1 | Mar 2007 | US |
