Performance Monitoring for Live Systems with Soft FEC and Multilevel Modulation

Abstract

Performance monitoring is an essential function for margin measurements in live systems. Historically, system budgets have been described by the Q-factor converted from the bit error rate (BER) under binary modulation and direct detection. The introduction of hard forward error correction (FEC) did not change this. In recent years, technologies have changed significantly to comprise coherent detection, multilevel modulation and soft FEC. In such advanced systems, different metrics such as (nomalized) generalized mutual information (GMI/NGMI) and asymmetric information (ASI) are regarded as being more reliable. On the other hand, Q budgets are still useful because pre-FEC BER monitoring is established in industry for live system monitoring.The pre-FEC BER is easily estimated from available information of the number of flipped bits in the FEC decoding, which does not require knowledge of the transmitted bits that are unknown in live systems. Therefore, the use of metrics like GMI/NGMI/ASI for performance monitoring has not been possible in live systems. However, in this work we propose a blind soft-performance estimation method. Based on a histogram of log-likelihood-values without the knowledge of the transmitted bits, we show how the ASI can be estimated.We examine the proposed method experimentally for 16-and 64-ary quadrature amplitude modulation (QAM) and probabilistically shaped 16-, 64-, and 256-QAM in recirculating loop experiments. We see a relative error of 2.4%, which corresponds to around 0.5 dB signal-to-noise ratio difference for binary modulation, in the regime where the ASI is larger than the assumed FEC threshold. For this proposed method, the digital signal processing circuitry requires only the minimal additional function of storing the L-value histograms before the soft FEC decoder.

Required and Received SNRs in Coded Modulation

“…We investigate the SNR performance by offline experiments with a setup, which is almost the same as the one in Ref. [14] . In this work, we performed noise loading experiments only.…”

Soft Q-factor in optical communication systems

“…These metrics work better than pre-FEC BER, but are not so suitable for describing the performance for live systems because of following two reasons: (1) the soft information cannot be derived without the knowledge of transmitted bits, which are usually unknown in live systems, and (2) there is no standardized Q-factor for quantifying the soft information. The issue (1) was addressed by introducing the blind method based on SD-FEC decoder input statistics in [6], and (2) was discussed in, e.g., [7,8]. So far, Q-factor based on soft information, called soft Q-factor, have been quantified by the approximation method shown in [9], which is rather inaccurate than the requirement (e.g., aproximation error is 0.1 dB) for applications of fiber-optic communication sytems.…”

On the Performance under Hard and Soft Bitwise Mismatched-Decoding

“…The pre-FEC BER can be estimated from the fraction of flipped bits in the FEC decoding, converted to Q-factor, and used in the system margin description. Alternatively, the ASI can be estimated blindly [12].…”

“…Note that this work focuses on the true performance behavior under various mismatched decoding. We will not use any blind performance monitoring methods like [12] to avoid the uncertainty due to the estimation error. Fig.…”

“…We need to set an SNR for the demapping, SNR aux , and the performance is maximized when SNR aux = SNR tr . The ASI under matched decoding is equivalent to the normalized GMI as shown in [5,12]. We will investigate how the Q-factors depend on SNR tr and SNR aux .…”