Monitoring burst errors in forward error correction frames

Abstract

In an example of the disclosed technology, a method includes calculating, by a computing system, a symbol error distribution in a forward error correction (FEC) frame transmitted in an optical signal. The method further includes adjusting, in the optical transceiver module, one or more of a front cursor setting, a rear cursor setting, or a signal amplitude of the optical signal in response to determining that the symbol error distribution indicates a burst error that meets the threshold.

Description

[0001] This application claims the benefit of U.S. Patent Application No. 17 / 820,456, filed on August 17, 2022, the entire contents of which are incorporated herein by reference. Technical Field

[0002] The present disclosure relates generally to optical networks and, more particularly, to optical transceiver modules. Background Art

[0003] A network service exchange provider or hosting provider ("provider") may employ communication facilities such as a data center, in which multiple customers of the provider locate networks, servers, and storage devices, and interconnect with various telecommunications and (multiple) other network service providers with minimal cost and complexity. Such customers are referred to as "hosting" customers of the data center. A data center may be shared by multiple tenants of network equipment located within the data center. With secure and reliable IT and communication facilities, telecommunications, Internet, application service providers, cloud service providers, content providers, and other providers and enterprises can enjoy less latency and be free to focus on their core business.

[0004] In some cases, a communications facility provides interconnection services by which the provider's customers can interconnect with each other over the communications facility infrastructure, or by which the provider's customers can interconnect their spatially and / or geographically distributed customer network equipment over the communications facility infrastructure. In this case, the communications facility has been referred to as an "interconnection facility" or "hosting facility." The facility provider can provide services accessible to customers at a shared location via interconnection, such services including, for example, cloud exchange, Internet access, Internet exchange, cross-connect from one customer to another, and other interconnection services.

[0005] One example of interconnection between different networks within an interconnection facility is a physical cross-connect, where a provider installs physical cabling (such as CAT5 / 6, coaxial cable or fiber optic cable) and then provides connections between physical (e.g., "layer 0") customer networks for inter-customer interconnection (between different customers) or intra-customer interconnection (between (multiple) interconnection center areas operated / leased by a single customer). For example, fiber optic cross-connects may be installed between floors across an interconnection facility, between different floors, across campuses, between distributed interconnection facilities within a metropolitan area (or "metropolis"), and between distributed interconnection facilities within a connection facility across different metropolitan areas. Summary of the invention

[0006] In general, a technique for monitoring burst errors in forward error correction (FEC) frames transmitted in optical signals is described. For example, Reed-Solomon (RS) FEC is widely used in 100-Gigabit Ethernet (GE) and 400GE interfaces to improve the signal-to-noise ratio of optical signals. Burst errors can reduce the performance of RS FEC and can be presented in the electrical channel connecting the host and the module. In the absence of burst errors, the symbol error distribution within the FEC frame will follow a Poisson distribution. In some examples, the technique includes comparing the symbol error distribution within the measured FEC frame with a Poisson distribution to detect and quantify burst errors. In some examples, the pre-FEC bit error rate (BER) can be monitored by measuring the error distribution in the received optical signal. In addition, by using the monitored and measured error distribution as feedback, signal integrity and power consumption can be effectively rebalanced. For example, the pre-cursor, post-cursor, and signal amplitude can be adjusted to try to reduce the incidence of burst errors and thereby improve FEC performance. The techniques described in this disclosure include burst error detection, pre-FEC BER estimation, signal integrity optimization, and power consumption optimization.

[0007] The technology can provide one or more technical advantages for realizing one or more practical applications. For example, in an optical channel, random errors such as additive white Gaussian noise (AWGN) dominate. In an electrical channel, burst errors severely reduce net code gain (NCG) and reduce system performance, also reduce the performance of FEC and reduce system margin. Therefore, it is crucial to detect the presence of burst errors and understand their impact. The bit error rate is an important parameter in any communication channel, which affects the forward correction error, and the forward correction error can only correct errors within a certain threshold. Exceeding this threshold, uncorrected codewords appear, resulting in optical signal frame loss. Therefore, it is crucial to monitor BER in real time and with high precision. Although the received optical signal can be compared with the transmitted optical signal to obtain an accurate error count, it is not feasible in practice. The technology described in the present disclosure can be used to monitor BER in real time and with high precision. A variety of methods have been introduced to improve signal integrity in electrical channels. At present, the total BER can be optimized in a laboratory environment by adjusting the corresponding parameters. Then, these parameters can be applied in a set and forget manner. However, the signal transmission performance through the optical channel will degrade due to differences between devices, environmental drift, aging of components, etc. The technology described in this disclosure can dynamically improve signal integrity based on a feedback mechanism.

[0008] The technology described in this disclosure can be used to detect and monitor burst errors in real time to be used as feedback to adjust signal amplitude / integrity while also balancing this goal with power consumption. Power consumption is a key issue for optical transceiver modules on both the transmitting and receiving ends. High power consumption limits port density and increases total cost of ownership (TCO). The technology described in this disclosure can dynamically optimize power consumption based on a feedback mechanism that balances low power consumption and reliable signal transmission goals. The technology described in this disclosure can be used to improve future 800GE interfaces.

[0009] In one example, the present disclosure describes a method that includes calculating, by a computing system, a symbol error distribution in a forward error correction (FEC) frame transmitted in an optical signal. The method also includes adjusting, in an optical transceiver module, one or more of a pre-cursor setting, a post-cursor setting, or a signal amplitude for the optical signal in response to determining that the symbol error distribution indicates a burst error that satisfies a threshold.

[0010] In another example, the present disclosure describes determining, by a computing system, a pre-forward error correction (FEC) bit error rate (BER) based on a bit interleaved parity (BIP) field within an alignment marker of an electrical signal; and adjusting, in an optical transceiver module, a signal amplitude for the electrical signal through an optical channel in response to determining that the pre-FEC BER based on the BIP field exceeds an estimated BER.

[0011] In another example, the present disclosure describes a computing system including a processing circuit having access to a memory device, the processing circuit being configured to: calculate a symbol error distribution in a forward error correction (FEC) frame transmitted in an optical signal; and in response to determining that the symbol error distribution indicates a burst error that satisfies a threshold, adjust a control setting in the optical transceiver module for modifying the optical signal in the optical transceiver module.

[0012] The details of one or more examples of the technology of the present disclosure are set forth in the accompanying drawings and the description below. Other features, objects, and advantages of the technology will be apparent from the description and drawings, and from the claims. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1A is a block diagram illustrating an example interconnect system including network devices having optical transceiver modules in accordance with one or more techniques of the present disclosure.

[0014] Figure 1B is a block diagram illustrating an example interconnect system including a network device having an optical transceiver module and an optimization module according to one or more techniques of the present disclosure.

[0015] Figure 2is a block diagram illustrating an example system including one or more optical transceiver modules, according to one or more techniques of this disclosure.

[0016] Figure 3A is a block diagram illustrating an example network device including one or more switch ASICs and one or more optical transceiver modules, according to one or more techniques of this disclosure.

[0017] Figure 3B is a block diagram illustrating another example network device including one or more switch ASICs and one or more optical transceiver modules, according to one or more techniques of this disclosure.

[0018] Figure 3C is a block diagram illustrating another example network device including one or more switch ASICs and one or more optical transceiver modules, according to one or more techniques of this disclosure.

[0019] Figure 3D is a block diagram illustrating yet another example network device including one or more switch ASICs and one or more optical transceiver modules, according to one or more techniques of this disclosure.

[0020] Figure 4 is a block diagram illustrating further details of the operation of an example computing device in accordance with one or more techniques of this disclosure.

[0021] Figure 5 is a flow chart of an example operating mode for a computing system in accordance with one or more techniques of this disclosure.

[0022] Figure 6 is a flowchart of another example operating mode for a computing system in accordance with one or more techniques of this disclosure.

[0023] Figure 7 is a flowchart of yet another example operating mode for a computing system, in accordance with one or more techniques of this disclosure.

[0024] Like reference numerals refer to like elements throughout the drawings and embodiments. DETAILED DESCRIPTION

[0025] Optical connections between clients within or between data centers can be provided by optical switches. Optical connection services can typically be used for connections within a cluster of data centers or data centers that are close to each other, for example, within the same campus or region, such as a metropolitan area.

[0026] Direct, dedicated, carrier-grade network links between customers can be spread over a wide geographic area and use dense wavelength division multiplexing (DWDM) optical transponders to transmit data, and can include multiple types or categories of transmission services with different reliability and availability levels. For example, transmission services can include unprotected services, protected services, and dual-diversified services. The provision, grooming, protection, and recovery of services using optical switches can be automated. These types of services can operate on multiple data center locations in data centers across multiple sites in a metropolitan or regional optical network. For example, client signals can come from different client platforms and are received from various sources (e.g., routers, other switches, data center interconnects (DCIs), etc.) at the optical network. The optical network can receive client traffic at 10G, 100G, 400G, or other rates. Incoming client traffic can also have different service categories and quality of service (QoS) requirements, such as unprotected, protected, or dual diversity. The number of clients that an optical network can serve can range from up to one thousand (at 100G speeds) to less than one thousand (e.g., at 10G speeds). An optical network of NxN optical switches can switch incoming client signals to any one of the DCI ports, and the DCI port can adapt / map the client signal to a DWDM signal. The DWDM signal can then be transmitted through a line system to another node in another data center or location. The network / domain controller can provide signaling and control functions that allow the system to perform functions such as service-independent DCI and DWDM channel allocation, service management across multiple optical networks and different networks, optical network decomposition, rerouting client signals from paths with faulty elements to other working paths in the network by using spare capacity (e.g., ports, switches) for fault detection and mitigation to improve network reliability and availability, network optimization and defragmentation, and zero-touch automatic provisioning and reconfiguration of client paths.

[0027] The ever-increasing demand for Internet data has driven the deployment of high-speed Ethernet interfaces in Internet Protocol (IP) router and switch technologies. The most advanced system currently is 400GE (400 Gigabit Ethernet per second), which is standardized in IEEE 802.3b. In addition, 800GE (800 Gigabit Ethernet per second) is under active development. The demand for 400GE and 800GE is expected to grow rapidly in the next few years. The quality of service for clients depends on high-speed Ethernet interfaces for communication within and between data centers.

[0028] The physical layer of 400GE and 800GE interfaces is mainly developed as optical transceiver modules. The most common form factors are quad small form factor pluggable - double density (QSFP-DD) and octal small form factor pluggable (OSFP). Optical transceiver modules can be connected to application specific integrated circuits (ASICs) that switch / route Ethernet packets. The connection is through electrical printed circuit board (PCB) traces and electrical connectors. In addition, optical transceiver modules perform electronic to optical conversion and vice versa. Optical transceiver modules are connected to each other through optical fibers. For end-to-end systems, both electrical connections (through PCB traces) and optical connections (through optical fibers) are required.

[0029] Noise is present on any communication channel, whether electrical or optical. A degradation in the signal-to-noise ratio (SNR) will result in errors within the signal that may be unrecoverable. In addition, retransmitting the damaged signal requires time and resource allocation, resulting in increased costs. To address this issue, forward error correction (FEC) is widely used in optical networks to improve receiver sensitivity and increase the transmission distance of the signal. In addition, when FEC is used, various requirements for optical transmitters, receivers and / or transceivers can be relaxed, resulting in a cost-effective approach. The IEEE has recognized two FEC codes, KR and KP. Both codes belong to Reed-Solomon (RS) codes, which are linear block codes with critical parameters (n, k, m). Here, n is the number of code symbols in a frame, including parity symbols; k is the number of message symbols from the source message; and m is the number of bits per symbol. The message symbols and code symbols belong to elements of a Galois field.

[0030] KR FEC is mainly used in non-return-to-zero (NRZ) on-off keying (OOK) modulation format. Here, n=528, k=514, m=10. The net coding gain (NCG) is 5.3dB, and the bit error rate (BER) threshold for the uncorrected codeword (UCW) is 2.1e -5 . To further increase the total data rate during transmission, pulse amplitude modulation 4 levels (PAM4) is widely used. However, the sensitivity loss for PAM4 signals is 4.77 dB. Therefore, strong KP FEC was developed, where n=544, k=514, and m=10. The net coding gain is 6.4 dB, and the BER threshold for arbitrary UCW is 2.2e -4 .

[0031] KP FEC and KR FEC are used in short-distance optical transceiver modules with optical wavelengths centered around 1310 nm. For long-distance optical transceiver modules with wavelengths centered around 1550 nm, other types of FEC are used. For example, coherent FEC (OFEC) is standardized in 400GE-ZR modules and OFEC is being considered in 400GE-ZR+ modules. In addition, a variety of proprietary FEC schemes are used in advanced coherent optical transponders.

[0032] A burst error is a continuous sequence of bits that is different from the same bit sequence defined in the source signal. When the burst error reaches a threshold number, the source signal is unreadable and must be retransmitted. Currently, there are some theoretical methods to understand the impact of burst errors on digital signals. Theoretical methods include the Gilbert-Elliot channel model, Monte-Carlo simulation, and Markov chain model. However, in practice, there is no public method for monitoring and detecting burst errors.

[0033] Advantageously, the technology of the present disclosure describes a solution for monitoring and mitigating burst errors: monitoring a digital signal for errors, generating an error distribution, comparing the generated error distribution with an expected error distribution for random background noise, and adjusting the pre-cursor, post-cursor, and / or signal amplitude based on the comparison. For example, the technology of the present disclosure includes monitoring symbol errors of an FEC frame, generating an error distribution of the monitored symbol errors, comparing the error distribution with an expected error distribution, and then adjusting one of the pre-cursor, post-cursor, and signal amplitude based on the comparison. As another example, the technology of the present disclosure includes monitoring pre-FEC BER by relying on the bit interleaved parity (BIP) used in the alignment marker. Generate an error distribution of the pre-FEC BER and then compare it with the error distribution of the pre-FEC BER. Adjust one of the pre-cursor, post-cursor, and / or signal amplitude based on the comparison. In one example, we compare two pre-FEC BER values: one is obtained from the BIP field in the alignment marker, and the second is obtained from the error distribution.

[0034] Alignment markers can be separate 66-bit blocks that are periodically inserted into the Ethernet stream. However, due to the low occurrence rate of alignment markers, the accuracy of BER monitoring through BIP is low. In addition, over 1e -6The BER result is saturated. Currently, the total BER is used to optimize the parameters of signal integrity. These parameters are applied in a set-and-forget manner. In addition, multiple signal integrity improvement methods are prone to burst errors, including continuous time domain equalizers (CTLE), decision feedback equalizers (DFE), reflection cancellers (RC), slicers, and level-dependent equalizers (LDEQ). The general total BER reading does not provide insights into the optimization of these parameters.

[0035] For host systems with optical transceiver modules, power consumption is a limiting factor in many scenarios. In addition, any increase in power consumption in the optical transceiver module results in an approximately 5x increase in power consumption for system cooling. The signal amplitude from the optical transceiver module plays a role in determining power consumption. Currently, signal amplitude is applied in a set-and-forget manner, without taking into account the varying factors that cause signal degradation.

[0036] Advantageously, the present disclosure describes a method for detecting and monitoring burst errors in real time for innovation, estimating BER with high accuracy, and accurately and adaptively optimizing signal integrity and power consumption. The technology disclosed in the present disclosure has unique advantages over existing technologies. The technology can be applied in a variety of ways. For example, in the certification stage, the technology can be used to select superior optical transceiver modules and host systems, resulting in longer transmission distances for optical transceiver modules and improved signal-to-noise ratios, thereby achieving a reduction in capital expenditures. In the deployment stage, the technology disclosed in the present disclosure can monitor the performance of optical transceiver modules and host systems. End users can receive key information such as pre-FEC BER, burst errors, etc. In addition, an alarm can be triggered when an error threshold is met. Signal integrity and power consumption can be optimized in real time based on a feedback mechanism, resulting in a reduction in operating expenses and total cost of ownership.

[0037] Figure 1A2 is a block diagram of an example interconnection system including network devices with optical transceiver modules according to one or more techniques of the present disclosure. The interconnection system 2 depicts interconnection facilities 8A-8B (collectively, "interconnection facilities 8") operated by an interconnection system provider, the interconnection facilities 8 being connected by an optical network 10 that can be configured to cross-connect customer networks 6A-6B (collectively, "customer networks 6"). In some instances, each of the customer networks 6 can be associated with a different customer of the interconnection system provider. In some instances, the customer networks 6 are each associated with a single customer of the interconnection system provider. As used herein, the term "customer" of the interconnection system provider can refer to a tenant of the interconnection facility 8 deployed by the interconnection system provider, whereby the customer leases space within the interconnection facility 8 to be co-located with other tenants for purposes of increasing the efficiency of an independent facility, and interconnecting network devices with network devices of other tenants within the interconnection facility or campus to reduce latency / jitter and improve reliability, performance, and security (relative to the transport network), among other things. In some cases, the interconnection system provider can be referred to as the interconnection facility provider, i.e., the two are generally the same entity.

[0038] Each of the interconnection facilities 8 may represent a different data center in a geographically different metropolitan area to provide a resilient and independent interconnection service exchange through which customers in one metropolitan area and customers in different metropolitan areas may connect to receive and provide services respectively.

[0039] In each of the interconnection facilities 8, space may be divided and rented in flexible increments to customers in the form of cages (areas of common floor space enclosed by fences or other demarcations), cabinets, racks, suites (enclosed rooms that are not part of the common floor space), or other spaces, in which customers may place their network equipment to provide and / or receive network services to (multiple) other customers co-located in the same interconnection facility 8 and / or (multiple) other customers co-located in another interconnection facility 8.

[0040] Optical network 10 (including network devices 12) represents an optical and photonic switching infrastructure that is configured by programmable network platform 3 to create optical channels 7 that connect customer network 6A and customer network 6B in co-located interconnection facilities 8A and 8B, respectively. Optical network 10 may include pre-installed optical fiber between customer network 6 and at least one wiring facility (or "patch panel") in interconnection facility 8, and may also include pre-installed optical fiber between multiple wiring facilities in interconnection facility 8. Optical network 10 may also include programmable photonic switches located within wiring facilities of interconnection facility 8 and / or located at customer space boundaries within interconnection facility 8. Although dynamically configurable as described herein, much of optical network 10 may be statically configured by a network operator associated with an interconnection facility provider, a network service provider, or other entity that manages aspects of the optical network.

[0041] Although described in the context of optical paths connecting networks located in different interconnection facilities, the technology can also be applied to the context of optical paths connecting networks located within the same interconnection facility (e.g., a single data center). In addition, the technology can be applied to optical paths connecting any two networks, and not just customer networks for customers of the interconnection facility provider.

[0042] The optical network 10 may include one or more optical network devices, such as DCI repeaters, muxponders, optical transceivers, channel multiplexers / demultiplexers, amplifiers, and / or optical switches, such as wavelength selective switches (WSS), photonic cross-connects (PXC), optical cross-connects (OXC), optoelectronic-based switches (OEO), or other types of optical signal switching devices. The optical switch may route optical signals (light beams or wavelengths) between optical fibers coupled to the optical switch.

[0043] In this example, the system 2 also includes a programmable network platform (PNP) 3, which may also be referred to herein as an "interconnection platform" alternatively. The programmable network platform 3 may expose a software interface that defines methods, fields, and / or other software primitives through which the (multiple) applications 5 may call the PNP 3 to dynamically provision optical channels 7 between the customer networks 6A, 6B. In this way, the PNP 3 allows customers, operators, network service providers (NSPs), and / or interconnection system providers themselves to have the programmable ability to configure the optical network 10, thereby interconnecting the customer network 6A to the customer network 6B via the optical channel 7.

[0044] The programmable network platform 3 may represent an application executed within one or more data centers of the system 2, or alternatively, an application executed off-site / remotely, such as at a back office or branch of a provider of the interconnection system 2. The programmable network platform 3 may be distributed in whole or in part among interconnection facilities. As illustrated in FIG. 1 , the programmable network platform 3 may control service provisioning for multiple different interconnection facilities. Alternatively or additionally, multiple separate instances of the programmable network platform 3 may control service provisioning for a corresponding multiple different interconnection facilities.

[0045] Application(s) 5 represent at least one application that communicates with the PNP 3 to request and dynamically provision interconnections between the interconnection facilities 8 of the customer networks 6A, 6B. Application(s) 5 represent client-side software for interacting with the PNP 3 and may include a customer portal, a customer application, an interconnection system 2 provider application accessible only to the operator of the interconnection facility 8, a console such as a command line interface or a graphical user interface. For example, users or clients of the application(s) 5 may include enterprise customers, cloud service and content providers, operators, network service providers (NSPs), and / or the interconnection system providers themselves.

[0046] An application of the application(s) 5 may issue an interconnect request 11 to the PNP 3, the interconnect request 11 specifying parameters for interconnection between the interconnection facilities 8 of the customer networks 6A and 6B. In response to receiving the interconnect request 11, the PNP 3 determines a path for interconnection, such as an optical path through the optical network 10, and dynamically provisions a path for interconnection of the optical channels 7 in the optical network 10 to provide a fiber-based interconnection between the customer networks 6A, 6B, thereby satisfying the interconnect request 11. Example parameters for the interconnect request 11 may include requested bandwidth, type of service (e.g., protected, unprotected, dual redundant, etc.), port identifiers, etc.

[0047] The PNP 3 issues configuration commands directly to elements within the optical network 10 through the configuration interfaces of the elements, or indirectly to these elements via a software defined networking (SDN) controller (e.g., an SDN controller) that manages the elements on behalf of the PNP 3, thereby configuring the elements of the optical network 10. In some examples, the PNP 3 may alternatively be referred to as a coordinator, where the PNP 3 coordinates one or more SDN controllers to configure the optical network devices 4 (e.g., photonic switches) and / or other elements of the optical network 10 to establish the optical channel 7.

[0048] The programmable network platform 3 configures the optical network device so that it exchanges optical signals along an optical path, each optical path starting from an optical transmitter and ending at an optical receiver, and one of the optical transmitter and the optical receiver is associated with a customer network in the customer network 6. In this way, the customer network 6 can exchange frames and packets for L0 / L1 / L2 / L3 services via a dynamically supplied optical fiber cross-connection. The optical path can alternatively be referred to as an optical path, an optical path, a lambda or an optical transmission network wavelength, or a composite DWDM optical signal. The example bandwidth for the optical path for cross-connection can include, for example, 2.5Gbps, 10Gbps, 40Gbps, 100Gbps, 400Gbps, or even a multi-terabit DWDM optical signal. For example, the optical fiber can include multiple optical paths, for example, that is, the optical fiber can be used as different lambda transmissions for optical signals for multiple optical paths. The switch based on optoelectronics can limit the switching application according to the data rate, protocol and format supported by the electronic components of the switch.

[0049] The optical channel 7 may represent a layer 0 ("L0") connection, i.e., a fiber cross-connection that is the basis of the Open Systems Interconnection (OSI) model or the TCP / IP model for packet networking. In particular, the optical channel 7 may be the basis of an optical transport network (OTN) layer, which (usually in conjunction with a SONET / SDH layer) provides layer 1 ("L1") connectivity between optical network devices 4, such as wavelength division multiplexing (WDM) devices. For example, the optical channel 7 may represent one or more wavelengths (or "lambdas") that are photon-switched by elements of an optical switch construction to provide an optical path (or "lightpath") for optical signals exchanged between customer networks 6A, 6B associated with different customers / tenants of the provider of the interconnection system 2. In this manner, the optical network 10 is configured to provide non-blocking switching through high-speed fiber connections between respective customer networks 6A, 6B for customers of the provider of the interconnection system 2, and in this case it may provide reliable and low-latency communications between the customer networks 6A, 6B. For example, optical channels 7 may be provisioned statically or manually by an interconnect system provider, or dynamically using PNP 3 .

[0050] The customer networks 6A, 6B use corresponding access links 9A, 9B to access the optical channel 7 to exchange data. In some examples, the access links 9A, 9B may represent gray links / light, where routers or switches of the customer networks 6A, 6B exchange gray (colorless) optical signals with coherent optical devices that convert between gray optical signals and optical signals of a specific wavelength (color) exchanged with optical network devices of the optical network 10.

[0051] In some cases, the optical channels 7 may be the basis for providing layer 2 ("L2") or layer 3 ("L3") services interconnecting customer networks 6A, 6B according to the L2 or L3 services provided by the interconnection system 2 provider. For example, the interconnection facility 8 may operate a network service exchange, such as an Ethernet exchange, an Internet exchange, and / or a cloud exchange, and each of these network service exchanges may use the dynamically provisioned optical channels 7 to transmit L2 / L3 packet data between the customer networks 6A, 6B.

[0052] According to the techniques described in the present disclosure, the optical network 10 includes at least one network device 12. The network device 12 may be an optical patch panel switch for a customer cage area patch panel located within the interconnection facility 8A and storing the customer network 6A. The network device 12 may represent a packet switch / router, an optical transmission device such as a data center interconnect (DCI) transponder, and the like. The network device 12 may be communicatively coupled to an optical module 15, which may be an optical transceiver module inserted into the network device 12. The optical module 15 may convert electrical signals transmitted within a printed circuit board via a switch into optical signals for transmission through the optical network 10 along a fiber cross-connect 7. In addition, another network device 12 may be configured to receive a signal and may be communicatively coupled to another optical module 15 (which may be an optical receiver module, an optical transmitter module, or an optical transceiver module), which is configured to receive the transmitted optical signal and convert the optical signal into an electrical signal for transmission through another printed circuit board, as described with respect to FIG. Figure 2 Further detailed discussion of FIG1 is not shown in FIG1 for ease of illustration. Network device 12 may be programmable and, in some cases, may be located within the wiring facility of interconnection facility 8A and / or 8B.

[0053] In some examples, the optical module 15 may be communicatively coupled to the network controller 18. The network controller 18 may execute the optimization module 13 to monitor burst errors within a forward error correction (FEC) frame transmitted in the optical signal. The optical module 15 may transmit error data 17 to the optimization module 13, and the optimization module 13 may transmit optical channel configuration data 19 to the optical module 15. The error data 17 may include information related to symbol errors within the forward error correction (FEC) frame, pre-FEC bit error rate, FEC frame, etc. The optimization module 13 may process the error data 17 to calculate symbol error distributions, pre-FEC BER distributions, or other error markers in the forward error correction (FEC) frame for the optical signal transmitted through the optical network 10 via the optical channel 7. These error distributions may be compared with expected error distributions (e.g., random distributions or Poisson distributions) to detect burst errors within the optical signal. The optical channel configuration data 19 may be configuration data for the optical module 15 calculated by the optimization module 13 to mitigate burst errors on the optical channel 7. For example, the optical channel configuration data 19 may include pre-cursor settings, post-cursor settings and / or signal amplitude settings, which, when configured in the optical module 15 , causes the optical module 15 to adjust the optical signal transmitted through the optical network 10 via the optical channel 7 according to the optical channel configuration 19 .

[0054] Figure 1B is a block diagram of an example interconnect system including a network device having an optical transceiver module and an optimization module according to one or more techniques of the present disclosure. Figure 1A In contrast to the example of FIG. 1 , the optimization module 13 may be integrated within the network device 12. For example, the optimization module 13 may be executed by a processing circuit of the network device 12. For example, the optimization module 13 may be electrically coupled to a printed circuit board (e.g., as described with respect to FIG. 1 ). Figure 2 The optimization module 13 may receive error data 17 from the optical module 15, calculate a symbol error distribution within a forward error correction (FEC) frame transmitted in the optical signal, compare the calculated symbol error distribution with an expected symbol error distribution (as described below), and generate optical channel configuration data 19 based on the comparison to detect burst errors. The optimization module 13 configures the optical module 15 using the optical channel configuration data 19 to adjust the optical signal transmitted through the optical network 10 via the optical channel 7 according to the optical channel configuration data 19. The error data 17 and the optical channel configuration data 19 are related to Figure 1A Similar to what is described in .

[0055] As reference Figures 1A to 1BAs discussed, the optimization module 13 can determine the expected symbol error distribution as follows. Forward error correction (FEC) is a digital signal technology used to enhance the reliability of digital signals on networks that may be unreliable or noisy. Connecting FEC frames to digital signals allows receivers to correct bit errors within digital signals that traverse the network without the need to retransmit data. Reed-Soloman (RS) FEC, such as KP and KR FEC, is widely used in high-speed Ethernet interface technology. After detecting and correcting errors, the optimization module 13 (e.g., via the burst error estimator 450) can provide a statistical analysis of symbol errors within the FEC frame. If the symbol errors are caused by random noise, the number of symbol errors x in the FEC frame will follow a Poisson distribution. The theoretical (expected distribution) symbol error distribution ("theoretical symbol error distribution" is used interchangeably with "expected symbol error distribution" hereinafter) Pr_theo(x) is determined by:

[0056] Pr_theo(x) = λ x e -λ / x! (1)

[0057] Where λ is the average value of expected symbol errors within the FEC frame. For RS FEC, a symbol is formed by 10 bits. The symbol error rate (SER) can be determined as:

[0058] SER=1-(1-BER) m ≈m*BER (2)

[0059] Where m is the number of bits (e.g., 10, etc.). Therefore, within a FEC frame containing n symbols, λ is determined as:

[0060] λ=n*SER≈n*m*BER (3)

[0061] The computing system may determine the measured symbol error distribution Pr_meas(x) as follows:

[0062] Pr_meas(x)=CW(x) / CW total (4)

[0063]

[0064] Where CW(x) is the acquired FEC codeword (eg, acquired by the optical module 13 (which may be a receiver optical module)), x represents the number of errors, and CW total is the total number of symbols including both correctable and uncorrectable codewords, UCW count is the count of uncorrectable codewords UCW, and (nk) / 2 is the maximum number of symbol errors that can be corrected within the FEC frame.

[0065] Next, the computing system may compare Pr_theo(x) and Pr_meas(x), as described herein. If Pr_theo(x) deviates from Pr_meas(x) by a predetermined amount, then a threshold may be met, indicating the impact of a burst error within the FEC frame.

[0066] The techniques of the present disclosure allow for the selection of superior optical modules (e.g., fixed or pluggable optical transceiver modules, etc.) and host systems. For example, by minimizing burst errors, a computing system can extend the transmission distance of an optical module beyond its nominal specifications. In practice, without prior knowledge, a system (e.g., system 2) cannot accurately measure the BER of a digital signal. In contrast, the techniques of the present disclosure allow for the estimation of BER using real-time data. One approach is to rely on a bit-interleaved parity (BIP) field within an alignment marker. However, alignment markers are typically not inserted into a data stream because of implementation cost.

[0067] It can be shown that BER can be estimated from the error distribution. Based on the Poisson distribution, equation (1) is simplified by inputting the number of symbol errors for 0 and 1 to obtain:

[0068] Pr_theo(0)=e -λ , Pr_theo(1)=λe -λ (6)

[0069] Next, the computing system can estimate the BER (defined as BER est ):

[0070]

[0071] BER est =λ est / n / m (7)

[0072] In this way, the computing system only needs to monitor CW(1) and CW(0) to get an accurate estimate of the BER. Without the corrected codeword CW(0), there are no errors. In CW(1), any effect of a burst error is limited to one symbol (10 bits). Therefore, the effect of a burst error on the above estimate is negligible.

[0073] In systems dominated by random noise, the above method is highly accurate. The optimization module (e.g., optimization module 13, 424) can use λ est and BER estTo plot a theoretical (expected) curve and compare it with the measured results obtained. For example, the optimization module 13 may receive the error data 17 from the optical module 15 and use a burst error estimator (e.g., reference Figure 4 The burst error estimator 424 discussed above is used to calculate the estimated symbol error distribution of the FEC frame, and then the calculated results are plotted against the expected symbol error distribution. A tight curve can indicate that the expected symbol error distribution is close to the calculated symbol error distribution, suggesting that there are no burst errors.

[0074] Alternatively, the computing system may perform a curve fit on the first few data points in the symbol error distribution. The computing system may perform a curve fit on the first few data points in the symbol error distribution by minimizing λ est and Err fit To estimate λ est , as shown in equation (8). Here, the upper limit l is the number of data points used for curve fitting. Next, the computing system determines the BER accordingly est This approach will remove any uncertainty created by measurement errors on a single data point and provide even greater accuracy. For example, the computing system can calculate Pr meas(0) , Pr meas(1) , and Pr meas(2) Perform curve fitting.

[0075]

[0076] On the other hand, for systems with burst errors, the estimated BER from CW(1) and CW(2) will be inaccurate. This observation allows us to develop a simpler method to monitor the performance of errors within the system, where the system is calculated based on the BIP field (defined as BER BIP ) to estimate the BER. For example, the alignment marker has a fixed data pattern defined by IEEE standard 802.3b. The BIP field is part of the alignment marker. The receiver optical module can compare the received signal with the known pattern and determine the BER. The accuracy can be improved by using a longer measurement period. At the same time, the computing system determines the BER based on equation (7) est If BER est Close to BER BIP , then the computing system can conclude that the system is dominated by random noise. BIP Greater than BER est , the computing system can determine that a burst error is present and take certain actions to improve performance. BIP and BER est If the difference is above a certain threshold, the computing system can warn the user. For example, the computing system can be communicatively coupled to the handheld device, operated by a technician, and send an alarm to the handheld device.

[0077] The techniques of the present disclosure can be applied to both the certification phase and the deployment phase. During the certification phase, the techniques can allow the selection of superior components to achieve optimal performance, resulting in a reduction in capital expenditures. During the deployment phase, the computing system can continue to monitor key parameters, such as pre-FEC BER and burst errors within FEC frame symbols, and provide valuable information to the end user.

[0078] In addition to comparing the calculated symbol error distribution with the expected symbol error distribution, the optimization module 13 can derive an estimated BER using the equation described above to improve the signal integrity in the electrical channel. Currently, the parameter optimization of those methods is based on eye diagrams, which require expensive external instruments, such as digital communication analyzers (DCA). Alternatively, the total BER that does not consider burst errors and random errors can be used. After the initial calibration, those parameters are stored in the optical transceiver module or the host system and will not be dynamically changed afterwards, for example, based on feedback. For example, as discussed above, a feedback mechanism can be used to adjust at least one of CTLE, DFE, RC, Slicer, and LDEQ. However, due to differences between devices, environmental offsets, and aging of components, system performance will be reduced. The technology disclosed in the present invention solves these problems by dynamically adjusting the front cursor setting, the rear cursor setting, and / or the signal amplitude based on the indication of the burst error. For example, a microcontroller with an electrically erasable programmable read-only memory (EEPROM) in the optical transceiver module. Some registers in the EEPROM correspond to the settings of the front cursor, the rear cursor, and the signal amplitude. There may be an integrated circuit (IIC) bus between the host and the optical transceiver module. The host may communicate with the optical transceiver module through the IIC bus and adjust each of the registers to change any one of the pre-cursor setting, post-cursor setting, and / or signal amplitude.

[0079] The electrical signal amplitude plays a role in determining the amount of power allocated for consumption by the optical transceiver module. When the electrical signal amplitude is large, the power consumption of the module will increase. In addition, high signal amplitude will increase the crosstalk between adjacent electrical channels. On the other hand, low signal amplitude reduces the signal-to-noise ratio (SNR) in the electrical channel, resulting in high BER. Therefore, depending on the environmental conditions, there is an optimal setting for the electrical signal amplitude for any particular system. In existing computing systems, the electrical signal amplitude can be optimized in a laboratory environment and applied in a set-and-forget manner. However, setting and forgetting does not allow real-time adjustment in response to changing environmental conditions. For the purpose of this disclosure, the term "optimization" should not be confused with the concept of making it perfect, and includes reasonable interpretations of optimization (especially from the perspective of the inventor), such as increasing the signal amplitude (or pre-cursor or post-cursor setting) to reduce power consumption (including power consumption of external systems, such as server cooling units), increasing SNR, and other attributes that positively affect the system and system resource allocation.

[0080] Figure 2 1 is a block diagram of an example system including one or more optical transceiver modules according to one or more techniques of the present disclosure. Optical system 20 includes network device 12A, which can be used to send and convert digital signals from electrical signals to optical signals and send them to network device 12B. Network device 12B receives the optical signal transmitted over an optical network (e.g., Figure 1A to Figure 1B The optical network 10 in the optical network 10 can receive optical signals transmitted on the optical network 10 and can convert received optical signals into electrical signals. The network devices 12A to 12B can represent packet switches / routers, optical transmission devices such as data center interconnect (DCI) repeaters, etc.

[0081] For example, the network device 12A may include a switch application specific integrated circuit (ASIC) 26A and an optical transceiver module 22A. The network device 12B may include a switch ASIC 26B and an optical transceiver module 22B. The optical transceiver modules 12A to 12B are each connected to a corresponding switch ASIC 26A, and the ASIC 26A switches / routes Ethernet packets on the optical network 10 via an optical channel 7. The optical transceiver modules 22A to 22B are connected by electrical PCB (printed circuit board) 21A to 21B traces and electrical connectors. In addition, the optical transceiver modules 22A to 22B perform electronic to optical conversion, and vice versa. For example, the optical transceiver module 22A receives an electrical signal from the switch ASIC 26A, and then, converts the electrical signal into an optical signal for transmission to the optical transceiver module 22B, wherein the optical signal is converted into an electrical signal, and then the electrical signal is transmitted to the switch ASIC 26B. The optical transceiver modules 22A to 22B are communicatively coupled via the optical channel 7. As Figure 2 The pin-to-end optical system 20 shown requires both electrical connections (through PCB 21A-21B traces) and optical connections (through optical channel 7).

[0082] The optical system 20 also includes switch control planes 25A to 25B and an optimization module 13 (see reference Figures 1A to 1B 2A to 25B). The switch control planes 25A to 25B control the forwarding of digital signals throughout the optical system 20. For example, the switch control plane 25A controls the forwarding of electrical signals transmitted from the switch ASIC 25A to the optical transceiver module 22A. The switch control plane 25A also controls the forwarding of optical signals from the optical transceiver module 22A to the optical transceiver module 22B through the optical channel 7. The switch control plane 25B is communicatively coupled to the switch control plane 25A via the optimization module 13. The switch control plane 25B may receive information about the forwarding of digital signals from the switch control plane 25A, such as routing tables including protocols that identify network paths, digital signal delivery, etc. The switch control plane 25B may obtain errors within the received digital signals. For example, the switch control plane 25B may obtain symbol errors, bit errors, etc. within the FEC frame, and generate error data (e.g., error data 17) that is sent to the optimization module 13. Switch control planes 25A- 25B may include processing circuitry, such as a general-purpose processor, and in some cases, one or both of switch control planes 25A- 25B may execute an instance of optimization module 13 .

[0083] In some examples, the optimization module 13 can act as an intermediary between the switch control planes 25A to 25B. In addition, the optimization module can receive the acquired errors in the form of error data 17 from the switch control plane 25B, calculate a symbol error distribution of the received FEC frame, compare the calculated symbol error distribution in the FEC frame with an expected symbol error distribution (e.g., as discussed above), determine that the symbol error distribution indicates a burst error based on the comparison satisfying a threshold, and then determine configuration data (e.g., optical channel configuration data 19) for the switch control plane 25A. The switch control plane 25A can receive the optical channel configuration data 19, configure one or more of the ASIC 26A or the optical transceiver module 22A to adjust one or more of a pre-cursor setting, a post-cursor setting, or a signal amplitude for an optical signal transmitted via the optical transceiver module 22A. For example, adjusting the pre-cursor setting, post-cursor setting, or signal amplitude can modify the signal-to-noise ratio, modify the power consumption of network device 12A, and / or have the effect of modifying the power consumption of an external cooling system (e.g., within a data center) required to maintain the temperature of optical system 20 or other systems.

[0084] FIG. 3A to FIG. 3D Discussed together. FIG. 3A to FIG. 3D Multiple configurations of different types of forward error correction (FEC) encoders and decoders integrated into various types of optical transceiver modules and / or switch ASICs are illustrated. Certain types of optical transceiver modules may be more suitable than other types, depending on parameters (distance between optical switches, bandwidth, etc.). The techniques described herein apply to various configurations of network devices and networks. For example, an optical transceiver module may perform FEC, a switch ASIC may perform FEC, and so on.

[0085] Figure 3A 30A is a block diagram illustrating an example of a computing system having a forward error correction (FEC) encoder (FECE) and decoder (FECD) within a corresponding switch ASIC. The computing system 30A includes a network device 34A, the network device 34A includes a switch ASIC 31A (e.g., switch ASIC 26A), and the switch ASIC 31A includes an optical transceiver module 32A (e.g., optical transceiver module 22A). The network devices 34A to 34B may represent packet switches / routers, optical transmission devices such as data center interconnect (DCI) transponders, etc. The optical transceiver module 32A may be an optical transceiver that is capable of receiving an electrical signal transmitted from a KR FEC encoder (FECE) 33, converting the electrical signal into an optical signal, and then transmitting the optical signal to an optical receiver 37 of the optical transceiver module 32B via an optical transmitter 35 on an optical channel 36 (e.g., optical channel 7). The optical transceiver module receiver 32B converts the received optical signal into an electrical signal for transmission to the KR FEC decoder (FECD) 39 integrated in the switch ASIC 31B. In an example, the optical transceiver modules 32A to 32B may be 100GE coarse wavelength division multiplexer (CWDM) 4, 100GE-SR4, etc.

[0086] exist Figure 3A In the example of , KR FEC is used to encode the symbols within the forward error correction frame. KR FEC is a weak Reed-Soloman FEC. In addition, compared with other types of 100GE modules, this example does not require high-end optics and is more cost-effective, for example, because Figure 3A The detector output power and sensitivity used in the example computing system are lower. The optical transceiver modules 32A to 32B include second generation technology for 100GE interfaces. Because weak FEC is used instead of strong FEC, the computing parameters need to be changed. For example, KR FEC can correct up to 7 bit errors in a frame, and KP FEC can correct up to 15 bit errors in a frame.

[0087] Figure 3B 30A is a block diagram illustrating an example of a computing system having a forward error correction (FEC) encoder and decoder within a corresponding optical transceiver module. Here, KP FEC is used instead of the KR FEC used in computing system 30A. KPFEC is a strong Reed-Soloman FEC. Network device 34A includes a switch ASIC 41A and an optical transceiver module 42A, which includes a KP FECE 43 and an optical signal 45. Network device 34B includes a switch ASIC 41B and an optical transceiver module 42B, which includes an optical signal receiver 47 and a KP FECD 49.

[0088] Here, with Figure 3A Instead, FEC encoding and decoding occurs within the optical transceiver modules 42A-42B. Furthermore, the switch ASICs 41A-41B may be substantially similar to the switch ASICs 31A-31B for backward compatibility. Figure 3B The digital signal path in includes a switch ASIC 41A, which transmits an electrical signal to an electrical signal receiver (not shown) of an optical transceiver module 42A. KP FECE 43 encodes the FEC frame and connects it with the electrical signal. The optical transceiver module 42A converts the electrical signal including the encoded FEC frame into an optical signal, and the optical transmitter 45 transmits the optical signal to the optical receiver 46 of the optical transceiver module receiver 42B through the optical channel 46. The optical transceiver module 42B converts the optical signal into an electrical signal. KP FECD decodes the encoded FEC frame and transmits the decoded electrical signal to the switch ASIC 41B. In the example, the optical transceiver modules 42A to 42B can be 100GE single lambdas, etc. Figure 3B In one example, the switch ASICs 31A to 31B may include KR FECE and KR FECD, respectively.

[0089] Figure 3C is a block diagram illustrating one example of a computing system having an FEC encoder and decoder within a corresponding switch ASIC. Figure 3A However, the switch ASICs 51A to 51B include KP FECE 53 and KP FECD 59, respectively, instead of KR FECE 33 and KR FECD 39. KP FEC is a strong Reed-Solomon FEC. The optical transceiver modules 52A to 52B may be 400GE non-coherent interfaces.

[0090] The electrical signal is transmitted from the switch ASIC 51A to the optical transceiver module 52A, where the KP FECE 53 encodes the electrical signal with the FEC frame. Then, the optical transmitter 55 transmits the optical signal (converted from the electrical signal by the optical transceiver module 52A) to the optical receiver 57 of the optical transceiver module 52B through the optical channel 56. The optical transceiver module 52B converts the optical signal into an electrical signal sent to the switch ASIC 51B, where the KP FECD 59 decodes the FEC frame.

[0091] Figure 3D is a block diagram illustrating one example of a computing system having a forward error correction (FEC) encoder and a FEC decoder within a corresponding switch ASIC. Figure 3D The system 30D includes network devices 34A to 34B. FIG. 3A to FIG. 3C Differently, each of the network devices 34A-34B includes one or both of FECE, FECD within the switch ASICs 61A-61B and the optical transceiver modules 62A-62B.

[0092] After KP FECE 63A encodes the FEC frame, the electrical signal is transmitted from switch ASIC 61A to optical transceiver module 62A. Here, KP FECD 64A decodes the encoded FEC frame, and CFECE 68A encodes the decoded FEC frame. Optical transmitter 65 transmits the encoded optical signal to optical receiver 67 of optical transceiver module 62B through fiber optic connection cross 66, where CFECD 68B decodes the optical signal. The digital signal and FEC frame are then encoded by KP FECE 64B. The encoded electrical signal is then transmitted to switch ASIC 61B for decoding by KP FECD 63B.

[0093] Coherent FEC (including CFEC, OFEC, proprietary SDFEC) is suitable for coherent signals through long-distance cables, while KPFEC is suitable for shorter distances. KP FEC can be used between the host (e.g., switch ASIC 61A to 61B) and the optical transceiver module 62A to 62B. Within the optical transceiver module 62A to 62B, KP FEC is stripped and replaced with coherent FEC encoding / decoding. KP FEC encoding / decoding is adopted to provide backward compatibility. Short-distance optical devices use KP FEC in ASIC switches 61A to 61B, and there is currently a reluctance in the industry to develop new ASICs for coherent modules. As Figure 3D As shown, with this design, the switch ASICs 61A-61B for both the non-coherent modules and the coherent modules are the same. The optical transceiver modules 62A-62B include second generation technology for the 400GE interface. Figure 3D In one example, the techniques described herein may be applied to a KP FECE / D section.

[0094] Figure 4 is a block diagram illustrating further details of the operation of an example computing device in accordance with one or more techniques of this disclosure. Figure 4 A specific example of a server or other computing device 400 may be illustrated that includes one or more processors 402 for executing one or more PNPs in PNP 3 or any other computing device described herein. Other examples of computing device 400 may also be used in other situations. Although for illustrative purposes, Figure 4 400, which may be any component or system including one or more processors or other suitable computing environment for executing software instructions, and for example, need not necessarily include Figure 4 One or more of the elements shown in 404 (eg, communication unit 406; and in some examples, components such as storage device(s) 408 may not be located in the same location or in the same chassis as other components).

[0095] like Figure 4 As shown in the specific example of, computing device 400 includes one or more processors 402, one or more input devices 404, one or more communication units 406, one or more output devices 412, one or more storage devices 408, user interface (UI) device 410 and communication unit 406. In one example, computing device 400 also includes one or more applications 422, optimization module 424 and operating system 416 executable by computing device 400. Optimization module 424 can be implemented using one or more computer processes. Each component in components 402, 404, 406, 408, 410 and 412 is coupled (physically, communicatively and / or operationally) for communication between components. In some examples, communication channel 414 can include system bus, network connection, inter-process communication data structure or any other method for data communication. As an example, components 402, 404, 406, 408, 410 and 412 can be coupled through one or more communication channels 414.

[0096] In one example, processor 402 is configured to implement functions within computing device 400 and / or process instructions for execution. For example, processor 402 may be capable of processing instructions stored in storage device 408. Examples of processor 402 may include any one or more of a microprocessor, a controller, a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA), or equivalent discrete or integrated logic circuits.

[0097] One or more storage devices 408 may be configured to store information during operation of computing device 400. In some examples, storage device 408 is described as a computer-readable storage medium. In some examples, storage device 408 is a temporary memory, meaning that its primary purpose is not long-term storage. In some examples, storage device 408 is described as a volatile memory, and when the computer is turned off, storage device 408 does not retain the stored content. Examples of volatile memory include random access memory (RAM), dynamic random access memory (DRAM), static random access memory (SRAM), and other forms of volatile memory known in the art. In some examples, storage device 408 is used to store program instructions for processor 402 to execute. In one example, software or applications running on computing device 400 are used in storage device 408 to temporarily store information during program execution.

[0098] In some examples, storage device 408 also includes one or more computer-readable storage media. Storage device 408 can be configured to store a greater amount of information than volatile memory. Storage device 408 can also be configured for long-term storage of information. In some examples, storage device 408 includes a non-volatile storage element. Examples of such non-volatile storage elements include a magnetic hard disk, an optical disk, a floppy disk, a flash memory, or an electrically programmable read-only memory (EPROM) or an electrically erasable programmable read-only memory (EEPROM).

[0099] In some examples, computing device 400 also includes one or more communication units 406. In one example, computing device 400 uses communication unit 406 to communicate with external devices via one or more networks (such as one or more wired / wireless / mobile networks). Communication unit 406 may include a network interface card, such as an Ethernet card, an optical transceiver, a radio frequency transceiver, or any other type of device that can send and receive information. Other examples of such network interfaces may include 3G and WiFi radio modules. In some examples, computing device 400 uses communication unit 406 to communicate with external devices.

[0100] In one example, computing device 400 also includes one or more user interface devices 410. In some examples, user interface device 410 is configured to receive input from a user through tactile, audio, or video feedback. Examples of (multiple) user interface devices 410 include a touch-sensitive display, a mouse, a keyboard, a voice response system, a camera, a microphone, or any other type of device for detecting user commands. In some examples, the touch-sensitive display includes a touch screen.

[0101] The computing device 400 may also include one or more output devices 412. In some examples, the output device 412 is configured to provide output to a user using tactile, audio, or video stimulation. In one example, the output device 412 includes a touch-sensitive display screen, a sound card, a video graphics adapter card, or any other type of device for converting signals into a suitable form understandable to a human or machine. Additional examples of output devices 412 include speakers, cathode ray tube (CRT) displays, liquid crystal displays (LCDs), or any other type of device that can generate understandable output to a user.

[0102] The computing device 400 may include an operating system 416. In some examples, the operating system 416 controls the operation of the components of the computing device 400. For example, in one example, the operating system 416 facilitates communication between one or more applications 422 and an optimization module 424 (e.g., the optimization module 13) and the processor 402, the communication unit 406, the storage device 408, the input device 404, the user interface device 410, and the output device 412.

[0103] The application 422 and the optimization module 424 may also include program instructions and / or data executable by the computing device 400. In the illustrated architecture, the example optimization module 424 executable by the computing device 400 includes a burst error estimator 450 and an adjustment module 452.

[0104] The optimization module 424 may include instructions for causing the computing device 400 to perform one or more operations and actions described in the present disclosure with respect to the optical controller 18 and the optical module 15. For example, the error data 17 may be one of a symbol error within an FEC frame, a BER BIP in an alignment marker, and the like. The burst error estimator 450 may include instructions for calculating an error distribution within the signal data 448 (which may be a repository for the acquired error data 17) when analyzing the signal data acquired by the optical module 15, and then comparing the error distribution with an expected error distribution. For example, the burst error estimator 450 may generate a symbol error distribution for the FEC frame and then compare it with an expected symbol error distribution, as discussed with reference to equations (1) to (8). The burst error estimator 450 may generate a bit error distribution for the BIP field and then compare it with an expected error distribution, as discussed with reference to equations (1) to (8).

[0105] In addition, the burst error estimator 450 may include instructions for determining whether at least one burst error exists in the signal data based on the comparison. The adjustment module 452 may include instructions for generating optical channel configuration data (e.g., optical channel configuration data 19) for the optimization module 424 to transmit to the optical module 15. For example, the optical channel configuration data 19 generated by the adjustment data may be based on the determination by the burst error estimator 450. The optical channel configuration data 19 may include adjustment levels for signal amplitude, and adjustments for pre-cursor and post-cursor signals.

[0106] Figure 5 1 is a flow chart of a method 500 for determining whether an optical signal includes a burst error, according to one or more techniques of the present disclosure. The method 500 includes calculating, by the optimization module 13, a symbol error distribution in a forward error correction (FEC) frame transmitted in the optical signal (502). In one example of block 502, the optimization module 13 obtains a FEC frame from the optical signal, and then transmits the obtained FEC frame to the optimization module 13, and the optimization module 13 calculates the symbol error distribution in the obtained FEC frame. In one example of block 502, the receiver optical module (e.g., 22B) transmits error data 17 including symbol errors within the FEC frame to the optimization module 13. In one example of block 502, the burst error estimator 450 within the optimization module 13 calculates the symbol error distribution in the FEC frame.

[0107] The optimization module 13 determines whether the symbol error distribution indicates a burst error that satisfies a threshold (504). In one example of block 504, the optimization module 13 compares the calculated symbol error distribution with an expected symbol error distribution (e.g., based on a Poisson distribution, as discussed above with reference to equations (1) to (8)). The optimization module 13 determines whether the symbol error distribution satisfies a threshold (506). In one example of block 506, the optimization module 13 may perform a curve fit on the calculated symbol error distribution, and then compare the curve fit with the expected symbol error distribution, as discussed above. If the determination result of block 506 does not satisfy the threshold (the "No" branch of 506), the optimization module 13 calculates a symbol error distribution in another FEC frame transmitted in another optical signal (508).

[0108] However, if the determination satisfies the threshold ("yes" branch of 506), the optimization module 13 may adjust one or more of a pre-cursor setting, a post-cursor setting, and a signal amplitude for the optical signal (510). In one example of block 510, the adjustment module 452 generates optical channel configuration data 19 based on an indication of a burst error within the optical signal. The optimization module 13 transmits the optical channel configuration data 19 to the optical module, which includes settings for adjusting control settings in the optical module 15 for modifying the optical signal. For example, the control setting may be one of a pre-cursor setting, a post-cursor setting, or a signal amplitude for the optical signal. For example, the optical module 15 uses the optical channel configuration data 19 to adjust one of a pre-cursor setting, a post-cursor setting, and a signal amplitude for the optical signal. Alternatively, or in addition, the optimization module 13 may use a feedback mechanism to change one of a continuous time domain equalizer (CTLE), a decision feedback equalizer (DFE), a reflection canceller (RC), a limiter, and a level dependent equalizer (LDEQ), etc.

[0109] Figure 6 6 is a flow chart of a method 600 for determining whether an optical signal includes a burst error according to one or more techniques of the present disclosure. The method 600 may include determining a pre-forward error correction (FEC) bit error rate (BER) (602) based on a bit interleaved parity (BIP) field within an alignment marker. For example, the system may insert an alignment marker in at least one of the physical coding sublayers (PCS). In addition, a BIP check may be added for each PCS channel in the PCS channel, and the BIP check is carried in the alignment marker. In one example, the optical module 15 may obtain a pre-FEC BER and generate error data 17 including the pre-FEC BER that is sent to the optimization module 13.

[0110] The method 600 may also include determining whether the pre-FEC BER meets a threshold (604). In one example of step 604, the computing system may estimate the pre-FEC BER based on a random distribution, such as the Poisson distribution discussed above. The computing system may then compare the estimated BER distribution to the determined pre-FEC BER distribution, and if the determined number is equal to or greater than the estimated BER, the computing system may determine that the threshold is met (606). If the computing system determines that the threshold is met (the "yes" branch of 606), the method 600 may continue with the computing system adjusting a control setting in the optical module to modify the optical signal (608). For example, the control setting may be one or more of a pre-cursor setting, a post-cursor setting, or a signal amplitude for the optical signal. In one example of step 608, a feedback mechanism is used to change one of a continuous time domain equalizer (CTLE), a decision feedback equalizer (DFE), a reflection canceller (RC), a limiter, and a level dependent equalizer (LDEQ). However, if the computing system determines that the threshold is not met (“NO” branch of 606 ), method 600 may return to block 602 .

[0111] Figure 7 700 is a flowchart of a method 700 for determining whether an optical signal includes a burst error, according to one or more techniques of the present disclosure. The method 700 includes calculating, by the optimization module 13, a partial error distribution of optical signals from all ports of a network device (e.g., network devices 12A to 12B, 34A to 34B) (702). For example, the network device may include at least an optical transceiver module (e.g., optical transceiver module 15, 22A to 22B, 32A to 32B, 42A to 42B, 52A to 52B, 62A to 62B) and a switch ASIC (e.g., switch ASIC 26A to 26B, 31A to 31B, 41A to 41B, 51A to 51B, 61A to 61B). In one example of block 702, the partial error distribution of the optical signal may be for a set amount of time. In one example of block 702, the distribution of the optical signal may include, for example, Figure 5 The symbol error distribution discussed in Figure 6 , or both. Method 700 may include comparing the calculated partial error distribution of optical signals from all ports of the network device to a first threshold (704). In one example of block 704, the first threshold may be an expected symbol error distribution within the FEC frame, an expected pre-FEC bit error distribution, and the like.

[0112] Method 700 may include determining whether a partial error distribution for at least one particular port among all ports of the network device satisfies a first threshold (706). If the first threshold is satisfied (the "yes" branch of 706), method 700 may proceed to calculate a complete error distribution for the particular port of the network device (708). In one example of block 708, the complete error distribution is for a set amount of time that is longer than the set amount of time for the partial error distribution. However, if the computing system determines that the first threshold is not satisfied (the "no" branch of 706), method 700 may return to block 702.

[0113] The method 700 may also include comparing the calculated complete error distribution to a second threshold (710). The method 700 may include determining whether the calculated complete error distribution satisfies the second threshold (712). If the complete error distribution does not satisfy the second threshold (the "no" branch of 712), the method 700 returns to block 702. However, if the calculated complete error distribution satisfies the second threshold (the "yes" branch of 712), the method 700 may adjust a control setting in the optical module to modify the optical signal (714). For example, the control setting may be one or more of a pre-cursor setting, a post-cursor setting, or a signal amplitude for the optical signal. In one example of step 714, the feedback mechanism is used to change one of a continuous time domain equalizer (CTLE), a decision feedback equalizer (DFE), a reflection canceller (RC), a limiter, and a level dependent equalizer (LDEQ).

[0114] The method 700 may also include determining whether a burst error is present (716). If the computing system determines that a burst error is present (the "yes" branch of 716), the method 700 may return to block 708. However, if a burst error is not present (the "no" branch of 716), the method 700 may circumvent adjusting control settings in the optical module (718).

[0115] The techniques described herein may be implemented in hardware, software, firmware, or any combination thereof. The various features described as modules, units, or components may be implemented together in an integrated logic device, or separately as discrete but interoperable logic devices or other hardware devices. In some cases, the various features of the electronic circuit may be implemented as one or more integrated circuit devices, such as an integrated circuit chip or chipset.

[0116] If implemented using hardware, the present disclosure may be directed to an apparatus such as a processor or an integrated circuit device (such as an integrated circuit chip or chipset). Alternatively or additionally, if implemented using software or firmware, the technology may be implemented at least in part by a computer-readable data storage medium that includes instructions that, when executed, cause a processor to perform one or more of the methods described above. For example, a computer-readable data storage medium may store such instructions for execution by a processor.

[0117] Computer-readable media may form part of a computer program product, which may include packaging materials. Computer-readable media may include computer data storage media, such as random access memory (RAM), read-only memory (ROM), non-volatile random access memory (NVRAM), electrically erasable programmable read-only memory (EEPROM), flash memory, magnetic or optical data storage media, etc. In some examples, an article of manufacture may include one or more computer-readable storage media.

[0118] In some examples, computer-readable storage media may include non-transitory media. The term "non-transitory" may indicate that the storage medium is not embodied in a carrier wave or propagating signal. In some examples, non-transitory storage media may store data that may change over time (e.g., in RAM or cache).

[0119] The code or instructions may be software and / or firmware executed by a processing circuit including one or more processors, such as one or more digital signal processors (DSPs), general-purpose microprocessors, application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other equivalent integrated or discrete logic circuits. Accordingly, the term "processor" as used herein may refer to any of the foregoing structures or any other structure suitable for implementing the techniques described herein. Furthermore, in some aspects, the functionality described in the present disclosure may be provided in software modules or hardware modules.

Claims

1. A method comprising: calculating, by a computing system, a distribution of symbol errors in a forward error correction (FEC) frame transmitted in the optical signal; as well as In response to determining that the symbol error distribution indicates a burst error that satisfies a threshold, a control setting in an optical transceiver module that is used to modify the optical signal is adjusted.

2. The method according to claim 1, wherein the computing system includes a receiver optical transceiver module that receives the optical signal, and Wherein calculating the symbol error distribution in the FEC frame comprises: acquiring, by the receiver optical transceiver module, the FEC frame from the optical signal; The symbol error distribution in the FEC frame is calculated by the receiver optical transceiver module.

3. The method according to claim 1, wherein the threshold comprises a second threshold, Wherein calculating the symbol error distribution in the FEC frame transmitted in the optical signal further comprises: calculating, by the computing system, partial symbol error distribution in the FEC frame from a plurality of ports of a network device, Wherein determining that the symbol error distribution indicates a burst error that satisfies the threshold comprises: determining that the partial symbol error distribution at a specific port of the switch satisfies a first threshold, and the method further comprises: calculating, by the computing system, a complete symbol error distribution of the optical signal for at least the specific port of the network device, Wherein adjusting, in an optical transceiver module, the control setting in the optical transceiver module to modify the optical signal is in response to determining that the calculated complete symbol error distribution indicates that the burst error satisfies the second threshold.

4. The method according to any one of claims 1 to 3, further comprising: acquiring, by a receiver optical transceiver module of a network device, the FEC frame from the optical signal; as well as The FEC frame from the network device is sent to the computing system.

5. The method according to any one of claims 1 to 4, further comprising: comparing the calculated symbol error distribution in the FEC frame with an expected symbol error distribution; as well as The symbol error distribution is determined to indicate the burst error satisfying the threshold based on the comparison. The method of claim 5 , wherein the desired symbol error distribution threshold follows a Poisson distribution.

7. The method according to any one of claims 1 to 6, wherein calculating the symbol error distribution in the FEC frame comprises: FEC codewords with different numbers of errors are obtained.

8. The method according to any one of claims 1 to 7, further comprising: The transmitted optical signal is collected for predetermined time intervals.

9. The method according to any one of claims 1 to 8, further comprising: determining that the indicated amount of burst errors satisfies a burst error threshold by comparing the indicated amount of burst errors within at least the optical signal; as well as In response to determining that the burst error threshold is met, an alert is generated.

10. The method according to any one of claims 1 to 9, wherein the FEC frame comprises a first FEC frame, the method further comprising: calculating, by the computing system, a symbol error distribution in a second FEC frame transmitted in the optical signal after the adjusting; as well as In response to determining that the symbol error distribution in the second FEC frame does not indicate burst errors that meet the threshold, adjusting the control setting in the optical transceiver module to modify the optical signal is bypassed in the optical transceiver module.

11. The method of any one of claims 1 to 10, wherein the control settings in the optical transceiver module include one or more of a pre-cursor setting, a post-cursor setting, or a signal amplitude for the optical signal.

12. A method comprising: determining, by a computing system, a pre-forward error correction (FEC) bit error rate (BER) based on a bit interleaved parity (BIP) field within an alignment marker of the electrical signal; as well as In response to determining that the pre-FEC BER based on the BIP field exceeds an estimated BER, a signal amplitude for the electrical signal is adjusted through the optical channel in an optical transceiver module.

13. The method of claim 12, wherein determining the pre-FEC BER based on the BIP field within an alignment mark of the electrical signal is prior to transmission through an optical channel.

14. The method of any one of claims 12 to 13, wherein determining that the pre-FEC BER exceeds the estimated BER further comprises: Estimate the BER of the FEC frame based on random distribution; as well as The estimated BER is compared to the determined pre-FEC BER.

15. The method according to any one of claims 12 to 14, further comprising: determining that the amount of indicated burst errors satisfies a burst error threshold by comparing the amount of indicated burst errors within at least the optical signal; as well as In response to determining that the burst error threshold is met, an alert is generated.

16. The method of claim 14, wherein estimating the BER of the FEC frame based on a random distribution further comprises: performing a curve fit on a first set of data in the random distribution; as well as The error of the curve fit is minimized according to:

17. A computing system comprising: A processing circuit having access rights to a memory device, the processing circuit being configured to: calculating a symbol error distribution in a forward error correction (FEC) frame transmitted in an optical signal; as well as In response to determining that the symbol error distribution indicates a burst error that satisfies a threshold, a control setting in the optical transceiver module that is used to modify the optical signal is adjusted in the optical transceiver module.

18. The computing system of claim 17, wherein the computing system further comprises a receiver optical transceiver module that receives the optical signal, and Wherein, in order to calculate the symbol error distribution in the FEC frame, the receiver optical transceiver module comprises the processing circuit, and the processing circuit is configured to: acquiring the FEC frame from the optical signal; The symbol error distribution in the FEC frame is calculated.

19. The computing system of any one of claims 17 to 18, wherein the processing circuit is further configured to: comparing the calculated symbol error distribution in the FEC frame to an expected symbol error distribution; and Based on the comparison, it is determined that the symbol error distribution indicates the burst of errors that satisfies the threshold.

20. The computing system of any one of claims 17 to 19, wherein the processing circuit is further configured to: determining that the indicated amount of burst errors satisfies a burst error threshold by comparing the indicated amount of burst errors within at least the optical signal; and In response to determining that the burst error threshold is met, an alert is generated.

21. The computing system of any one of claims 17 to 20, wherein the FEC frame comprises a first FEC frame, wherein the processing circuit is further configured to: calculating a symbol error distribution in a second forward error correction (FEC) frame transmitted in the optical signal after the adjusting; and In response to determining that the symbol error distribution in the second FEC frame does not indicate burst errors that meet the threshold, adjusting the control setting in the optical transceiver module to modify the optical signal is bypassed in the optical transceiver module.

22. The computing system of any one of claims 17 to 21, wherein the control settings in the optical transceiver module include one or more of a pre-cursor setting, a post-cursor setting, or a signal amplitude for the optical signal.

Images