arXiv subjects
Publications and source records attributed to Helmut Griesser.
We demonstrate QoT estimation in a live network utilizing neural networks trained on synthetic data spanning a large parameter space. The ML-model predicts the measured lightpath performance with <0.5dB SNR error over a wide configuration range.
A spectral sweep process is introduced to discover performance issues in optical spectrum services. We detect filtering penalty, spectral ripple/tilt and channel crosstalk in field measurements, potentially leading to increased service robustness in low-margin networks.
We test the concept of extended channel probing in an Optical Spectrum as a Service scenario in coherent optimized flex-grid long-haul and 10Gbit/s OOK optimized 100-GHz fixed-grid dispersion-managed legacy DWDM networks. An estimation accuracy better than +/- 0.1dB in GSNR implementation margin is obtained for both networks by using flexible coherent transceivers on lightpaths up to 822km on regional haul legacy network and up to 5738km on a long-haul network. We also explain, how to detect the operation regime of the channel by analyzing the received GSNR results from constant power spectral density (PSD) probing mode and constant signal power probing mode
We use channel probing to determine the best transponder configurations for spectral services in a long-haul production network. An estimation accuracy better than +/- 0,7dB in GSNR margin is obtained for lightpaths up to 5738km.
The potential to operate third-party terminals over multi-domain transparent optical networks attracts operators and customers to implement Optical Spectrum as a Service (OSaaS). As infrastructure information cannot always be shared with OSaaS end customers, alternatives to off-line Quality of Transmission (QoT) estimation tools are required to assess the performance of the spectrum slot in order to estimate achievable throughput. In this paper, commercially available sliceable coherent transceivers are used to assess the Generalized Signal To Noise Ratio (GSNR) based QoT of the OSaaS in a live production network for both, narrow-band and wide-band OSaaS configurations. Extended channel probing based on symbol rate variability is combined with spectral sweeping and operation regime detection to characterize OSaaS implementations on 17 links with different underlying infrastructure configurations in order to maximize capacity and increase service margins in a low-margin operation regime. We achieve 0.05 dB estimation accuracy in GSNR for a wide-band spectrum services and 0.32 dB accuracy for narrow-band spectrum services. Based on the GSNR profile, spectral misalignment, spectral ripple, and operation regime are detected and service margin improvements are demonstrated. Finally, we discuss the network optimization perspective based on acquired data from channel probing and propose use-cases for continuous channel probing in transparent optical networks.
Probabilistic constellation shaping (PCS) offers a significant performance improvement over uniform signaling. It was recently discovered that long blocks are not required to achieve maximum shaping gain when transmitting over the nonlinear fiber channel because short-length PCS effectively mitigates fiber nonlinear interference (NLI). The reason for this behavior is that short-length PCS implicitly induces some temporal properties in the shaped transmit sequence that are beneficial for the fiber-optic channel. To achieve robust data-aided digital signal processing of high-order QAM, periodic quaternary phase shift keying pilots are typically inserted into the high-order QAM transmit sequence. In this work, we investigate in simulations the effect of such pilot-aided signaling on NLI mitigation. Albeit modifying the temporal properties of the shaped transmit sequence, a pilot rate of 1/32 is found to not alter the beneficial effects of short-length PCS. The operation meaning of this finding is that even with pilot-aided signaling, long PCS block lengths are not required for maximum shaping gain.
We demonstrate up to 12 km, 56 Gb/s DMT transmission using high-speed VCSELs in the 1.5 um wavelength range for future 400Gb/s intra-data center connects, enabled by vestigial sideband filtering of the transmit signal.
We review currently discussed solutions for 80 km DWDM transmission targeting inter data-center connections at 100G and 400G line rates. PDM-64QAM, PAM4 and DMT are investigated, while the focus lies on directly detected solutions.
Discrete multi-tone transmission (DMT) is a promising candidate for future 400G data center interconnects. Eight channels, each carrying 56 Gb/s of data can be combined in a 50-GHz channel grid to form a 400 Gb/s superchannel. For a fully loaded 96 channel DWDM system this leads to a total capacity of 4.8 Tb/s. To meet the requirements of the targeted application in terms of cost efficiency and low power consumption, it is important to keep the complexity for the digital signal processing as low as possible. For DMT, the complexity is mainly determined by the length of the fast Fourier transformation (FFT). Since data center interconnects only have to bridge a distance of typically 80 km, we here investigate among other parameters the influence of the FFT length among other parameters on the achievable performance for 56 Gb/s DMT only for this distance. Transmission is performed in C-band to enable DWDM and no optical dispersion compensation is applied. We consider double sideband (DSB) as well as vestigial sideband (VSB) DMT. It can be seen that an FFT length of 128 is sufficient to reach the required performance in terms of bit error ratio, however, a higher length can significantly improve the performance.
We review three solutions for low-cost data center interconnects with a target reach of up to 80 km. Directly detected DMT, PAM-4 and multi-band CAP are promising modulation schemes, enabling 400 Gbit/s by combining eight channels of 56 Gbit/s.
Enabled by channel adaptive bit and power loading, we experimentally demonstrate discrete multitone transmission at 56Gb/s with simple intensity modulation and direct detection and achieve 50 km reach in the 1.55um window.
In order to realize probabilistically shaped signaling within the probabilistic amplitude shaping (PAS) framework, a shaping device outputs sequences that follow a certain nonuniform distribution. In case of constant-composition (CC) distribution matching (CCDM), the sequences differ only in the ordering of their constituent symbols, whereas the number of occurrences of each symbol is constant in every output block. Recent results by Amari \textit{et al.} have shown that the CCDM block length can have a considerable impact on the effective signal-to-noise ratio (SNR) after fiber transmission. So far, no explanation for this behavior has been presented. Furthermore, the block-length dependence of the SNR seems not to be fully aligned with previous results in the literature. This paper is devoted to a detailed analysis of the nonlinear fiber interactions for CC sequences. We confirm in fiber simulations the inverse proportionality of SNR with CCDM block length and present two explanations. The first one, which only holds in the short-length regime, is based on how two-dimensional symbols are generated from shaped amplitudes in the PAS framework. The second, more general explanation relates to an induced shuffling within a sequence, or equivalently a limited concentration of identical symbols, that is an inherent property for short CC blocks, yet not necessarily present for long blocks. This temporal property results in weaker nonlinear interactions, and thus higher SNR, for short CC sequences. For a typical multi-span fiber setup, the SNR difference is numerically demonstrated to be up to 0.7dB. Finally, we evaluate a heuristic figure of merit that captures the number of runs of identical symbols in a concatenation of several CC sequences. For moderate block lengths up to approximately 100 symbols, this metric suggests that limiting the number of identical-symbol runs can be beneficial.
In this paper, we study amplitude shaping schemes for the probabilistic amplitude shaping (PAS) framework as well as algorithms for constant-composition distribution matching (CCDM). Huffman-coded sphere shaping (HCSS) is discussed in detail, which internally uses Huffman coding to determine the composition to be used and relies on conventional CCDM algorithms for mapping and demapping. Numerical simulations show that HCSS closes the performance gap between distribution matching schemes and sphere shaping techniques such as enumerative sphere shaping (ESS). HCSS is based on an architecture that is different from the trellis-based setup of ESS. It allows to tailor the used HCSS compositions to the transmission channel and to take into account complexity constraints. We further discuss in detail multiset ranking (MR) and subset ranking (SR) as alternatives to arithmetic-coding (AC) CCDM. The advantage of MR over AC is that it requires less sequential operations for mapping. SR operates on binary alphabets only, which can introduce some additional rate loss when a nonbinary-to-binary transformation is required. However, the binomial coefficients required for SR can be precomputed and stored in a lookup table (LUT). We perform an analysis of rate loss and decoding performance for the proposed techniques and compare them to other prominent amplitude shaping schemes. For medium to long block lengths, MR-HCSS and SR-HCSS are shown to have similar performance to ESS. SR-HCSS and uniform 64QAM are compared in additive white Gaussian noise simulations and shaping gains of 0.5 dB and 1 dB are demonstrated with 1 kbit and 100 kbit LUT size, respectively.
DMT at 25 Gbit/s, 50 Gbit/s and 100 Gbit/s in passive systems with electro-absorption modulators is investigated. Transmission in C- and O-band are compared, yielding up to 22 km reach, 100 Gbit/s in O-band. All results are obtained with a real-time signal processing ASIC.
We report on 112 Gbit/s real-time DMT transmission over up to 60 km, targeted at DCI applications. Chromatic dispersion mitigation by vestigial sideband filtering is compared to the use of dispersion compensating fiber.
We experimentally demonstrate a novel 3-dimensional modulation format with 1.2-dB OSNR tolerance improvement potential compared to DP-BPSK, as verified for the linear case. The non-linearity tolerance is evaluated in single-span transmission over 300km SSMF.
We show that short-length probabilistic shaping reduces nonlinear interference in optical fiber transmission. SNR improvements of up to 0.8 dB are obtained. The shaping gain vanishes when interleaving is employed and not undone before transmission.
For a future 5G Ethernet-based fronthaul architecture, 100G trunk lines of a transmission distance up to 10 km standard single mode fiber (SSMF) in combination with cheap grey optics to daisy chain cell site network interfaces are a promising cost- and power-efficient solution. For such a scenario, different intensity modulation and direct detect (IMDD) Formats at a data rate of 112 Gb/s, namely Nyquist four-level pulse amplitude modulation (PAM4), discrete multi-tone Transmission (DMT) and partial-response (PR) PAM4 are experimentally investigated, using a low-cost electro-absorption modulated laser (EML), a 25G driver and current state-of-the-art high Speed 84 GS/s CMOS digital-to-analog converter (DAC) and analog-to-digital converter (ADC) test chips. Each modulation Format is optimized independently for the desired scenario and their digital signal processing (DSP) requirements are investigated. The performance of Nyquist PAM4 and PR PAM4 depend very much on the efficiency of pre- and post-equalization. We show the necessity for at least 11 FFE-taps for pre-emphasis and up to 41 FFE coefficients at the receiver side. In addition, PR PAM4 requires an MLSE with four states to decode the signal back to a PAM4 signal. On the contrary, bit- and power-loading (BL, PL) is crucial for DMT and an FFT length of at least 512 is necessary. With optimized parameters, all Modulation formats result in a very similar performances, demonstrating a transmission distance of up to 10 km over SSMF with bit error rates (BERs) below a FEC threshold of 4.4E-3, allowing error free transmission.