Standard Forward Error Correction Research Articles

XPM estimation and compensation in the WDM system based on a low-complexity inner FEC.

In this work, we focus on compensation of inter-channel nonlinearity in optical wavelength division multiplexed (WDM) systems. We consider pre-coding with a high-rate forward error correction (FEC) code and a low-complexity and low-latency decoding and equalization scheme at the receiver. The 2 × 2 multiple-input multiple-output (MIMO) recursive least square (RLS) equalizer is applied to reduce the inter-symbol interference (ISI) caused by the inter-channel nonlinearity, while the decoder improves the quality of the data feedback. Decisions based on feedback reliability have been implemented to prevent error propagation. We demonstrate that our system can readily utilize the inner FEC part of a mainstream concatenated optical FEC designs such as the 400ZR standard FEC. We evaluate the proposed algorithm for the dual-polarization 16QAM and 256QAM optical links. Based on our simulation results, the Q-factor has been increased by 0.36 and 0.67 dB for the 16QAM and 256QAM, respectively. The optimum launch power has been increased by 0.4 dB for both cases.

PAM-4 implementation study for future high-speed links

With the ever-increasing amount of data produced by the high energy physics experiments, the transmission rates from the detectors to the back-end stages must keep up. To mitigate the exponential growth of the total loss due to the increased bandwidth, the 4-Level Pulse-Amplitude Modulation (PAM4) could be envisaged. Where the line rate of Non-Return to Zero (NRZ) modulated signals is capping at 28 Gbps per lane, PAM4 allows the rate of 53.125 Gbps or above in extremely high-end applications. Investigating the implementation of such links in FPGA is one of the activities carried out by the Work Package 6 of the CERN EP Research and Development programme. A proof-of-concept system of high-speed links using PAM4-53.125 Gbps has been built, based on a Xilinx Virtex evaluation platform and various commercial optoelectronics transceivers. PAM4 standard Forward Error Correction (FEC) codes have also been implemented and characterized over electrical and optical layers in terms of coding gain and latency. Finally, the telecom and datacom markets were investigated to identify development perspectives for the research and development for future links. In this paper, the performance of the proof-of-concept implementing these high-speed links is presented and the current and future challenges for an error-free communication are discussed.

Analog Low-Latency Kramers-Kronig optical Single-Sideband Receiver

Kramers-Kronig (KK) receivers have been shown to be able to cancel the signal-signal beat interference that occurs upon direct-photodetection in optical single-sideband systems, where the transmitted optical field is proportional to the modulation signal (Field-Modulated Direct-Detection systems). They can support complex modulation formats, and being single-sideband, electronic dispersion compensation, without the need for a coherent receiver. All current research has used digital-signal processing (DSP) to implement the KK algorithm. Radio-frequency KK receivers using analog processing were first introduced in the 1960's, and offer an alternative to DSP, particularly in their approximate form. Analog processing is inherently real-time and offers the lowest latency (processing delay). In this paper, we show that with minor changes, analog processing can also be used with signals from a photodetector. We have built an experimental receiver, using standard parts that have functionality that approximates to the desired algorithm. In particular, we show that a differential pair of bipolar transistors can approximate a square-root function, that the Hilbert transform can be approximated by an 8-tap transmission-line delay, and that a commercial analog multiplier chip can be used for a squaring and subtraction function. The system is demonstrated with 16-QAM at 500 Mbit/s with the photocurrent emulated in software and the KK algorithm performed in hardware. Finally, software band-limits the signal, down-converts it to form a constellation, and calculates symbol error rates. We show that the analog KK reduces the symbol error rate by at least a factor of ten, enabling low carrier-to-signal power ratios to be used with standard forward error correction. Much higher data rates could be supported using custom-designed microwave ASICs, which could be incorporated in packaged receivers. This would enable miniaturized receivers using standard DSP to work with dispersion-limited links. Interestingly, half the benefit could be gained by simply re-biasing the differential stage of a standard photoreceiver.

Cycle-slip-less low-complexity phase recovery algorithm for coherent optical receivers.

We propose and experimentally validate a blind phase recovery algorithm based on tracking low-frequency components of the phase noise, which we call "filtered carrier-phase estimation (F-CPE)." Tracking only the low-frequency components allows F-CPE to reduce the computational complexity by using a frequency-domain equalizer and to simplify the partitioning of a 16 quadrature amplitude modulation (16QAM) constellation. Further, this approach eliminates cycle slips by suppressing the impact of amplified spontaneous emission on phase noise estimation. The experimental results demonstrate cycle-slip-free operation for 15 and 32GBd 16QAM signals. Additionally, the proposed method showed similar or better sensitivity compared with the blind-phase-search algorithm, near standard forward error correction thresholds of modern wavelength division multiplexing systems.

Low noise frequency comb carriers for 64-QAM via a Brillouin comb amplifier.

Optical frequency comb lines with poor carrier to noise ratio (CNR) are significantly improved by Brillouin amplification using its extreme narrow bandwidth gain to suppress out of band noise, enabling higher quality signal modulation. Its application to spectral lines of narrow 10 GHz pitch and poor CNR is shown to suppress the otherwise strong phase distortion caused by poor CNR after encoding with 96 Gb/s DP-64-QAM signals and restore the bit error rate (BER) to below the limit for standard forward error correction (FEC). This is also achieved with the required frequency shifted optical pump for amplification obtained by seeding it from the comb itself, sparing the need for lasers and frequency locking. Simultaneous CNR improvement for 38 comb lines is also achieved with BER restored to below the FEC limit, enabled by a multi-line pump that is pre-dispersed to suppress its spectral distortion from the Kerr effect in the gain medium. Carrier performance at minimum BER shows minimal noise impact from the Brillouin amplifier itself. The results highlight the unique advantage of Brillouin gain for phase sensitive communications in transforming otherwise noisy spectral lines into useful high quality signal carriers.

Experimental demonstration of a 10 Gb/s RSOA-based 16-QAM subcarrier multiplexed WDM PON

We experimentally demonstrate the performance of a 10 Gb/s subcarrier multiplexed (SCM) wavelength-division multiplexed (WDM) passive optical network (PON) using 2.5 Gbaud 16-QAM transmission signals. Digital signal processing (DSP) and square-root raised cosine (SRRC) pulse shaping enable both the uplink and downlink channels to achieve net spectral efficiencies up to 2.8 bit/s/Hz per channel using reflective semiconductor optical amplifier (RSOA)-based optical network units (ONUs) and economical 10 GHz intensity modulation and direct detection transceivers. We characterize the system's bit error rate (BER) performance over a 20 km single feeder PON with both remote continuous-wave (CW) seeding and full-duplex transmission scenarios, assuming standard forward error correction (FEC) codes.

Adaptive Unicast Video Streaming With Rateless Codes and Feedback

Video streaming over the Internet and packet-based wireless networks is sensitive to packet loss, which can severely damage the quality of the received video. To protect the transmitted video data against packet loss, application-layer forward error correction (FEC) is commonly used. Typically, for a given source block, the channel code rate is fixed in advance according to an estimation of the packet loss rate. However, since network conditions are difficult to predict, determining the right amount of redundancy introduced by the channel encoder is not obvious. To address this problem, we consider a general framework where the sender applies rateless erasure coding to every source block and keeps on transmitting the encoded symbols until it receives an acknowledgment from the receiver indicating that the block was decoded successfully. Within this framework, we design transmission strategies that aim at minimizing the expected bandwidth usage while ensuring successful decoding subject to an upper bound on the packet loss rate. In real simulations over the Internet, our solution outperformed standard FEC and hybrid automatic repeat request approaches. For the quarter common intermediate format Foreman sequence compressed with the H.264 video coder, the gain in average peak signal to noise ratio over the best previous scheme exceeded 3.5 dB at 90 kb/s.

Ultra-low-loss optical fiber enabling purely passive 10 Gb/s PON systems with 100 km length

We demonstrate time division multiplexing (TDM) and wavelength division multiplexing/TDM (WDM/TDM) long reach 10 Gb/s passive optical network (PON) architectures of 100 km reach with no infield amplification or dispersion compensation. The purely passive nature of the 100 km systems is enabled by the use of ultra-low-loss optical fiber with average attenuation of 0.17 dB/km and downstream transmission with a 10 Gb/s signal modulated with the duobinary format. The high tolerance of duobinary to dispersion, stimulated Brillouin scattering (SBS), and self-phase modulation (SPM) are all key factors to achieving good system performance at this distance, as is the significantly reduced loss from the ultra-low-loss fiber. We show that this combination of fiber and downstream signal format allow split ratios up to 1:128 for both system architectures. The achievable split ratio is reduced for standard single-mode fiber and/or use of an NRZ modulated downstream signal. Standard strength forward error correction (FEC) is used for the WDM/TDM system but is not required for the TDM system.

A Bandwidth Efficient Turbo Coding Scheme for VDSL Systems

The current forward error correction (FEC) scheme for very high bit-rate digital subscriber line (VDSL) systems in the ANSI standard employs a 16-state four-dimensional (4D) Wei code as the inner code and the Reed-Solomon (RS) code as the outer code. The major drawback of this scheme is that further improvement cannot be achieved without a substantial increase in the complexity and power penalty. Also, a VDSL system employing the 4D Wei-RS scheme operates far below the channel capacity. In 1993, powerful turbo codes were introduced whose performance closely approaches the Shannon limit. In this paper, we propose a bandwidth and power efficient turbo coding scheme for VDSL modems in order to obtain high data rates, extended loop reach and increased transmission robustness. We also propose a pipelined decoding scheme to reduce the latency at the receiver end. The objective of the proposed scheme is to provide a higher coding gain than that given by the 4D Wei-RS scheme, resulting in an improved performance of the VDSL modems in terms of bit rate, loop length and transmitting power. The scheme is investigated for various values of transmitting power, signaling frequencies and numbers of crosstalkers for a targeted bit error rate of 10−5 and is implemented in a system with a quadrature amplitude modulation in which a mixed set partitioning mapping is employed to reduce the decoding complexity. The effects of code complexity, interleaver length, the number of decoding iterations and the level of modulation on the performance of VDSL modems are explored. Simulation results are presented and compared to those of the 4D Wei-RS scheme. The results show that the choice of turbo codes not only provides a significant coding gain over the standard FEC scheme but also efficiently maximizes the loop length and bit rate at a very low transmitting power in the presence of dominant far-end crosstalk and intersymbol interference. In order to compare the hardware complexity, we synthesize the proposed and 4D Wei-RS schemes using SYNOPSYS with the target technology of Xilinx 4020e-3. The Xilinx field programmable gate array statistics of the proposed scheme is compared with that of the 4D Wei-RS scheme.

Joint Source/Channel Coding and MAP Decoding of Arithmetic Codes

In this paper, a novel maximum a posteriori (MAP) estimation approach is employed for error correction of arithmetic codes with a forbidden symbol. The system is founded on the principle of joint source channel coding, which allows one to unify the arithmetic decoding and error correction tasks into a single process, with superior performance compared to traditional separated techniques. The proposed system improves the performance in terms of error correction with respect to a separated source and channel coding approach based on convolutional codes, with the additional great advantage of allowing complete flexibility in adjusting the coding rate. The proposed MAP decoder is tested in the case of image transmission across the additive white Gaussian noise channel and compared against standard forward error correction techniques in terms of performance and complexity. Both hard and soft decoding are taken into account, and excellent results in terms of packet error rate and decoded image quality are obtained.

A Raman plus linear optical amplifier as an inline amplifier in a long-haul transmission of 16 channels × 10 Gbit/s over single-mode fiber of 1040 km

We have successfully demonstrated a long-haul transmission of 16 channels×10 Gbit/s over single-mode fiber of 1040 km using combined Raman and linear optical amplifiers as inline amplifiers. All the span length used was 80 km (loss of 16 dB), but the span losses varied from 28 to 34 dB according to some additional loss elements, e.g., dispersion-compensating fiber and a variable optical attenuator. The measured Q-factors of the 16 channels after 1040 km (12.7–14.5 dB) were higher than the error-free threshold of the standard forward-error correction, which offers feasibility of the hybrid amplifiers including semiconductor optical amplifiers for the long-haul transmission.

Tolerance to In-Band Coherent Crosstalk of Differential Phase-Shift-Keyed Signal With Balanced Detection and FEC

It is found experimentally that the differential phase-shift-keyed (DPSK) signal with balanced detection has /spl sim/6 dB higher tolerance to in-band coherent crosstalk than on-off-keyed (OOK) signal. We attribute the improved crosstalk tolerance of the DPSK signal to its 6-dB lowered peak signal power as compared to the OOK signal for equal bit-error rate in optical signal-to-noise-ratio-limited systems. The impact of signal coherence on the crosstalk tolerance in a 10-Gb/s DPSK system adopting a standard Reed-Solomon forward-error correction is also investigated.

Performance evaluation of standard FEC in 40 Gbit/s systems with high PMD and prechirped CS-RZ modulation format

The authors present a 40 Gbit/s system model with standard forward error correction (FEC), high polarisation mode dispersion (PMD) and prechirped carrier-suppressed return-to-zero (CS-RZ) format. An approach to evaluation of the system performance is proposed for investigating the coding gain and transmission penalty of standard FEC. The numerical simulation results show that the interaction of the second-order PMD with nonlinearly induced chirp can improve the performance of FEC in some cases.

Performance comparison of standard FEC in 40 Gbit/s optical transmission systems with NRZ, RZ and CS-RZ modulation formats

In this paper, we describe a detailed numerical investigation on performance of standard forward error correction (FEC) in 40 Gbit/s optical transmission systems with nonreturn-to-zero (NRZ), return-to-zero (RZ) and carrier-suppressed RZ (CS-RZ) modulation formats. The results show that FEC performs differently due to the different error distributions for the three formats at different power levels and transmission distances. Furthermore it is found that, with FEC on, NRZ format has a similar tolerance to intersymbol interference (ISI) due to the interactions of dispersion and nonlinearities as RZ and CS-RZ formats without FEC.

Performance comparison of standard FEC and enhanced FEC in coded systems with high polarization mode dispersion

This paper numerically compares the performance of 10- Gbit/s standard forward error correction (SFEC) and enhanced forward error correction (EFEC) in coded systems disturbed by high polarization mode dispersion (PMD). The simulation results show that EFEC, based on the concatenated Reed-Solomon codes RS(247, 239) and RS(255, 247), can increase the system PMD tolerance to 70 ps, com- pared with about 40 ps for SFEC, while keeping the PMD-induced power penalty (at a bit error rate of 10 210 ) below 1 dB. © 2003 Society of Photo-

10- and 40-Gb/s forward error correction devices for optical communications

Two standard forward error correction (FEC) devices for 10- and 40-Gb/s optical systems are presented. The first FEC device includes RS(255, 239) FEC, BCH(4359, 4320) FEC, and standard compliant framing and performance monitoring functions. It can support a single 10-Gb/s channel or four asynchronous 2.5-Gb/s channels. The second FEC device implements RS(255, 239) FEC at a data rate of 40 Gb/s. This paper presents the key ideas applied to the design of Reed-Solomon (RS) decoder blocks in these devices, especially those for achieving high throughput and reducing complexity and power. Implemented in a 1.5-V, 0.16-/spl mu/m CMOS technology, the RS decoder in the 10-Gb/s, quad 2.5-Gb/s device has a core gate count of 424 K and consumes 343 mW; the 40-Gb/s RS decoder has a core gate count of 364 K and an estimated power consumption of 360 mW. The 40-Gb/s RS FEC is the highest throughput implementation reported to date.