Search

In this paper we demonstrate a novel optical receiver for visible light communications that combines a fluorescent antenna and a silicon photomultiplier (SiPM). The fluorescent antenna is configured in a slab geometry to collect and absorb an incoming optical data signal, and to waveguide the resulting fluorescence to an edge where it is detected by the SiPM. The antenna incorporates the fluorophore, pentafluorene, selected for its very high bandwidth of 245 MHz, high photoluminescence quantum yield of 90%, and emission spectrum that matches the wavelengths most efficiently detected by the SiPM. The performance of a receiver, comprising the fluorescent antenna and a J-series 30020 SiPM manufactured by ON Semiconductor, was assessed in a 30 cm data link with a 405 nm GaN laser diode transmitter, both in the dark and under 500 lux of ambient white light. The fluorescent antenna successfully rejects ambient light by a factor of 20, limited by extrinsic scattering in the thin film. Using on-off-keying with decision feedback equalization, a maximum data rate of 1.4 Gbps was demonstrated in 500 lux of ambient light. Using the pentafluorene antenna plus a BG3 filter, provides a 200 fold suppression of ambient light, and can increase the data rate of the receiver at low signal powers by up to a factor of 2. When operating in ambient light the composite receiver requires only 2.5 times more signal power than when it operates in the dark, with this power penalty reducing to a factor of 1.2 at 1 Gbps.

High precise IQ skew calibration of the coherent optical transceiver is crucial for high baud rate and high-order modulation formats optical transmission system. In this paper, a novel high precise optical coherent transceiver calibration method based on the specially designed interleaved in-phase (I) and quadrature-phase (Q) multi-tone signals is proposed. This method can simultaneously characterize the transmitter (Tx) and receiver (Rx) side IQ skew by just one measurement without using any additional device or training process. Because of the non-overlapping characteristic of these interleaved I and Q signals in the frequency domain, the IQ crosstalk in the coherent optical transceiver can be effectively avoided, contributing to high-precision calibration. Since the calibration process only focuses on the phase frequency response characteristic, the dynamic range of IQ skew calibration can easily reach dozens of picoseconds. Theoretical analysis and simulation results indicate that the measurement accuracy of the Tx skew and Rx skew is within 0.2 ps in the proposed scheme, and the dynamic range of IQ skew calibration can reach 40 ps. With the help of this IQ skew calibration method, the transmission performance improvement is experimentally verified in both single-polarization 30 Gbaud 32QAM signal and dual-polarization 30 Gbaud 16QAM signal transmission systems.

Compensation of Kerr nonlinearity-induced distortions has been shown to allow for increasing transmission rate and reach, with optical compensation techniques particularly attractive for broadband wavelength-division multiplexed (WDM) scenarios. However, they normally require additional devices within the link, which is particularly challenging for already deployed systems, and even more so for unrepeatered transmission. In this work, we focus on providing lumped optical nonlinearity suppression, either at the transmitter or at the receiver side, based on optical phase conjugation (OPC). The theory to design scaled-down OPC compensation modules is derived, and it allows to move beyond the standard mid-link OPC approach and to explore compensation modules with fibers types not directly linked to the dispersion properties of the transmission link. This design method is then validated both through numerical investigations and experimental demonstrations for a number of systems with a varying degree of complexity. Ultimately, significant performance improvement is shown by employing short OPC-based compensation structures which are carefully designed to match much longer transmission links.

We propose indoor network with THz-band wireless link for 6G application. In our proposed indoor network, an optical hub unit (OHU) that controls whole system is optically linked to THz remote nodes (RNs) over optical distribution fibers. The THz RN communicates with a user equipment through a THz wireless link. The function of the THz RN provides an interface between optical link and THz wireless link. For downlink transmission, photonics-based THz-band signal generation method is adopted to take advantage of broad-band characteristics of optical components. On the other hand, electronics-based THz mixer is also used for uplink transmission because of its cost-effectiveness and low-energy consumption property. Digital signal processor (DSP) is designed to recover the original transmitted baseband signal. The DSP provides a frequency offset compensation in wide frequency range and reduces a probability of cyclic slip. The performance of the proposed system is investigated by experiment with commercially available components. We demonstrate 100 Gb/s 2.5 m wireless transmission with 16-quadrature amplitude modulation (16-QAM) signal for configuring downlink. The optical transmission distance is set to 10 km, and the measured power penalty by the optical transmission is negligible. We also investigate the scalability and the tunability of the photonics-based THz transmitter to confirm the upgradability of our proposed indoor network for considering future capacity expansion. For establishing uplink, 25 Gb/s 1.5 m wireless transmission with quadrature phase shift keying (QPSK) signal is shown. Directly modulated laser is used for cost-effective optical transmission. Unlike the downstream transmission, the measured bit error rate (BER) penalty caused by the optical transmission is observed. This is due to the interplay between frequency chirp of directly modulated laser and chromatic dispersion in fiber. Despite of this penalty, BERs less than the soft-decision forward error correction (FEC) threshold (2 10-2) is achieved. We discuss about remaining several technical challenges for real-field deployment. They are THz Tx power improvement, photonic integration, reducing form-factor, polarization insensitiveness, and automatic beam steering. Our recent efforts to solve these issues are also introduced and examined.

This paper is the first demonstration of multiplexed wideband data transmission in the millimeter-wave range using rectangular-coordinate orthogonal multiplexing (ROM) antennas. This spatial wireless multiplex communication method can be applied at several hundred GHz for further improvements in the data rate because much wider bandwidth is available and this multiplexing method does not require any signal processing. The multiplexing is achieved through the spatial eigenmodes of a novel antenna based on a rectangular coordinate system and magic-T which eliminates the need for computational signal processing efforts. The aperture distributions of these spatial eigenmodes are designed to have different polarities to avoid crosstalk and operate over a wide bandwidth range. We demonstrate their performance with four eigenmodes, achieving crosstalk between modes below -37.8 dB over a 14.6% relative bandwidth (57-66 GHz). We have introduced these antennas on a photonics-enabled real-time wireless data transmission, transmitting over two channels simultaneously, without any signal processing at the transmitter (multiplex) or the receiver (demultiplex). The two multiplexed channels show a total data rate up to 9.0 Gbps at most (5.875 Gbps and 3.125 Gbps for each channel) limited by the bandwidth of the low noise amplifiers at the receiver. The measured bit error rate (BER) is below the forward error correction (FEC) limit. This work was supported in part by SEI Group CSR Foundation and the Murata Science Foundation.

Visible light positioning (VLP) is a promising technology since it can provide high accuracy indoor localization based on the existing lighting infrastructure. However, existing approaches often require dense LED distributions and persistent line-of-sight (LOS) between transmitter and receiver. What's more, sensors are imperfect, and their measurements are prone to errors. Through multi sensors fusion, we can compensate the deficiencies of stand-alone sensors and provide more reliable pose estimations. In this work, we propose a loosely-coupled multi-sensor fusion method based on VLP and Simultaneous Localization and Mapping (SLAM), using light detection and ranging (LiDAR), odometry, and rolling shutter camera. Our multi-sensor localizer can provide accurate and robust robot localization and navigation in LED shortage/outage situations. The experimental results show that our proposed scheme can provide an average accuracy of 2.5 cm with around 42 ms average positioning latency.

5G defines below 100 GHz as the millimeter-wave bands, whereas 100 GHz - 3 THz is categorized as THz band in 6G. Deep leraning (DL) is expected to enable a significant paradigm shift in 6G wireless networks. In this paper, D-band 90-Gbps single channel PAM-4 signal generation and transmission over 10-km SMF and 3-m wireless link at 140-GHz can be achieved. A novel complex-valued neural network (CVNN) equalizer using ‘CReLU’ activation function to directly recover PAM-4 signals from received noised signals is demonstrated. There are three DSP options in the experiment. In Opt .1, firstly conducted cascaded multi-modulus algorithm (CMMA) pre-equalization (pre-EQ) at transmitter, then processed via frequency offset estimation (FOE), carrier phase recovery (CPR) and finally real-valued neural network (RVNN) equalization at receiver. Here, the RVNN equalizers include DNN with a softmax output layer, two-step joint-DNN equalizer and LSTM. The experimental results show that LSTM-based equalizer outperforms the other real NN-based equalizers by average 0.5 to 1.5 dB at BER of 10−3 magnitude. Differently from Opt . 1, real-valued CMMA pre-EQ at transmitter is unexpected for CVNN in the other two options. In Opt . 2, we only combine down-conversion and CVNN regarded as ‘pure data-driven’ training at receiver. This pure data-driven CVNN equalizer improves BER a lot and also has a larger computation burden, especially BER is as low as 1 × 10−4 with n 0 = 571 and n 1 = 200 training cells and the time complexity reaches 350000 in one iteration. Thanks to the aid of traditional mathematical-oriented models including FOE and CPR, the computation burden of CVNN in Opt . 3 is released significantly. Furthermore, we compare the performance of CVNN and RVNN in terms of BER decision accuracy, time complexity and receiver sensitivity. Followed by the same DSP with the same complexity, the comparison result between DNN and CVNN in the same structure of [371-260-1] with 11000 samples and 300 epochs shows that CVNN performs better due to its reservation of phase information. Therefore, we believe that the joint use of model-based, e.g., FOE, CPR steps and complex DL-based techniques has a potential for the future 6G wireless physical layer algorithms.

An optical heterodyned 28-GHz carrier generation is demonstrated for synthesizer-free fifth-generation (5G) millimeter wave over fiber (MMWoF) link by using orthogonally polarized dual-mode vertical-cavity surface-emitting laser (VCSEL) as the frontend transmitter, and the power envelope detection is employed for self-heterodyne down-conversion of the generalized frequency division multiplexing (GFDM) data stream. The demonstrated MMWoF link exhibits superior immunity to the residual frequency and phase noises as well as the inter-carrier interference induced by the free-running dual-mode carrier. For back-to-back optical downstream wireline transmission, the receiving GFDM data optimizes its bit error rate (BER) to 2.2×10−4 by adjusting the bias current of the dual-mode VCSEL to 8 mA (3 Ith), and by grouping the GFDM data matrix with total symbols of N = K×M with K = 2 subcarriers and M = 24 timeslots. After a long-reach delivering the optical data stream over 50-km dispersion-managed single-mode fiber (DM-SMF), the qualified receiving power sensitivity is −12 dBm, whereas the decoding GFDM data reveals improved receiving sensitivity with an extremely low power penalty of only 0.3 dB. For the wireless transmission after optical downstream receiving at the remote node, the self-heterodyned MMW carrier exhibits peak power of −62.3 dBm and carrier-to-noise (CNR) of 26.8 dB over long-reach 50-km DM-SMF downstream optical link. In comparison with the mutual heterodyned from two incoherent optical carriers, the self-heterodyned frequency fluctuation can be suppressed from >500 to

This paper explores the effects of sunlight on using a low-cost off-the-shelf silicon solar panel as an optical wireless communication (OWC) receiver. A receiver circuit structure has been proposed to maximize simultaneous energy harvesting and data communication performance. An equivalent circuit model of the solar panel for simultaneous energy harvesting and wireless optical communication is discussed. By using the solar panel model, the effects of varying sunlight conditions on the performance of the model as an OWC receiver are estimated. Furthermore, an experimental setup is developed to study the effects and verify the simulated estimations. The experimental setup consists of a 3.5 m wireless optical link with a full Transmission Control Protocol and Internet Protocol (TCP/IP) network stack. The system uses DC-biased optical orthogonal frequency division multiplexing (DCO-OFDM) to use the available communication bandwidth efficiently. A 940 nm low-cost off-the-shelf laser device along with an off-the-shelf off-axis-parabolic mirror is used as the transmitter. The maximum user throughput achieved over the air is 28.3 Mb/s while simultaneously harvesting energy using the maximum power point tracking (MPPT) technique. The peak power harvested with simultaneous communication is 4.5 W. The harvested energy is stored in a 38 Wh lithium-ion (Li-ion) battery.

In-phase and quadrature (IQ) imbalances in optical transceiver, such as amplitude imbalance, phase imbalance, and IQ skew, can cause serious system performance degradation, especially when high order modulation formats are used. In this work, we present a method to isolate and separately monitor these various impairments from received signals by coherent digital signal processing (DSP). Receiver amplitude and phase imbalances could be compensated and monitored by Gram-Schmidt orthogonalization procedure (GSOP) and receiver skew could be compensated and monitored by a 4 x 4 or 4 x 2 multiple-input and multiple-output (MIMO) equalizer. Transmitter imperfections compensation and monitoring take place behind the residual frequency offset compensation and carrier phase recovery by a 4 x 4 MIMO equalizer based on minimal mean square error (MMSE) criterion. By these methods, we analyze and investigate the interaction of transceiver's impairments impact on each monitor numerically. After that, we conduct an experiment to verify the feasibility of the proposed method. All of these results show that each monitor could operate very robust even when multiple co-exist.

In this paper, we experimentally demonstrate an optical frequency comb (OFC) based transmitter, employing directly modulated active demultiplexers, for data center interconnects. The results validate that the proposed transmitter has the potential to achieve aggregate data rates of 100 Gb/s (8 $ \times $ 12.5 Gb/s) and 200 Gb/s (8 $ \times $ 25 Gb/s) for systems employing 4- and 16- quadrature amplitude modulated (QAM) discrete multi-tone (DMT) modulation. An OFC based on an externally injected gain switched laser (EI-GSL) is used, providing excellent stability and flexibility in free spectral range (FSR). The OFC is followed by an injection locked active demultiplexer, which not only filters, but also amplifies individual comb tones, thus alleviating the need for an external optical amplifier to boost the low powered comb tones. Using the proposed configuration, we experimentally demonstrate a successful transmission of 12.5 Gb/s/ ${\boldsymbol{\lambda }}$ 4-QAM DMT and 25 Gb/s/ ${\boldsymbol{\lambda }}$ 16-QAM DMT signals over 40 km and 25 km of SSMF, respectively. In addition, we show that it is possible to filter and modulate comb lines that are 20 dB below the spectral peak, whilst achieving a BER below the hard decision (HD-) FEC limit of 3.8e-3. This gain flattening or comb expansion feature leads to a significant increase in the channel count, which in turn provides a reduction in the energy consumption and the footprint of the transmitter.

In this paper, we investigate two categories of linear frequency-modulated chirp signals suitable for single-photodiode based coherent light detection and ranging (LiDAR) systems, namely, the frequency-modulated continuous-wave (FMCW) single-sideband (SSB) signal and the amplitude-modulated double-sideband (DSB) signal, and compare their achievable receiver sensitivity performance. The DSB signal requires a simpler transmitter design, as it is real-valued and can be generated using a single-drive Mach-Zehnder modulator (MZM), while the SSB signal, which is frequency/phase modulated, requires an in-phase and quadrature modulator (IQM)-based transmitter. A theoretical analysis of direct-detection (DD) beating interference (BI) especially the local oscillator (LO) beating with itself, known as LO-LO BI, is presented. Both Monte Carlo simulations and experimental demonstrations are carried out. Good agreement between simulations and experiments is achieved. In comparison with the SSB system, the DSB signal-based system is affected by laser phase noise-induced power fluctuation, and also suffers a significant sensitivity penalty due to nonlinear LO-LO BI. A spectral guard band for mitigating LO-LO BI is necessary for the DSB signal, achieved at the expense of requiring a larger electrical bandwidth. In system tests with a delay line of 385 m, the SSB signal outperforms the DSB signal with a 10 dB better receiver sensitivity in the case with a guard band, and 25 dB better sensitivity without a guard band.

Multi-subcarrier (MSC) modulation is recently standing out as a notable new feature in the latest generations of high-speed coherent optical transceivers, providing distinctive advantages over legacy single-carrier systems. However, the transition from single-carrier to MSC modulation still requires significant optimization of digital signal processing subsystems, in order to maximize the inherent potential of subcarrier multiplexing. In this article, the penalty associated with the usage of carrier phase estimation (CPE) methods inherited from single-carrier systems is quantitatively studied and the need for enhanced MSC-tailored solutions is evidenced. A novel pilot-based joint-subcarrier CPE method based on a dual-reference subcarrier (DRS) approach is described and numerically assessed. The proposed DRS-CPE exploits the frequency-dependent walk-off effect imposed by chromatic dispersion (CD) to enable the separation of phase noise processes incurred by the transmitter and local oscillator lasers, thereby allowing to effectively apply joint-subcarrier CPE independently of the symbol-rate per subcarrier. Our numerical results demonstrate that the proposed DRS-CPE enables to operate with near-optimum performance at symbol-rates per subcarrier as low as 1 Gbaud over ultra-long-haul distances.

We propose and experimentally demonstrate a dual-polarization (DP) transceiver imbalance monitoring scheme, which is applicable for the coherent system using square QAM signals. The scheme can separate/monitor various kinds of transceiver imbalances and has technical advantages of multi-dimension, wide monitoring range, and modulation-format-transparency. In the proposed scheme, a polarization scrambler is configured between transmitter (Tx) and receiver (Rx). After coherent detection, the proposed digital signal processing (DSP) modules are implemented to monitor transceiver imbalance based on the received signals. To simplify the DSP modules, the implementation sequence of DSP modules is in reverse order compared to the order in which the sequence of impairments is introduced. Firstly, Godard timing error detection (TED) and Gram–Schmidt orthogonalization procedure (GSOP) is used to monitor and compensate Rx imbalances. After compensating the Rx imbalances, Tx imbalance monitoring is implemented. Complex/real maximum likelihood independent component analysis (ML-ICA) and Godard-TED are used to monitor Tx imbalance. Finally, the amplitude ratios of the final output signal and Tx in-phase/quadrature (IQ) amplitude imbalance are fed back to mitigate the interaction of Tx imbalances on the signal power and improve monitoring accuracy. The transceiver IQ and x-polarization/y-polarization (XY) imbalances are separated for the first time by using the frequency offset (FO) naturally existing in the system and the continuous polarization rotation induced by a polarization scrambler. The separation method combined with the reverse algorithm design and amplitude ratio feedback mechanism makes the proposed scheme achieve the most multi-dimensional imbalance monitoring by far. The simulation and experimental results verify that the proposed scheme is modulation-format-transparent and can separate/monitor multi-dimensional transceiver imbalances within a wide range.

We propose a comb-locked multi-channel transmitter with high-speed VCSELs injection-locked by an optical frequency comb. VCSEL dynamics are analyzed with various injection parameters including number of comb tones, frequency detuning, and injection ratio. We evaluate the performance of filtering and amplification effect when multiple comb tones injection into single VCSEL and estimate the transmission performance considering the reflection from VCSEL. Compared to intrinsic parameters of VCSEL, a 2-fold increase in frequency response and a $2.6\times 10^4$ -fold reduction in linewidth have been verified experimentally when the bandwidth-limited VCSEL locked to an incident comb line. 14-Gb/s single-lane transmission is demonstrated through 20 km single-mode fiber based on a 3.4-GHz VCSEL without any dispersion compensation. The similar performance can be achieved for all generated 25 comb tones, verifying a potential high-capacity transmitter with distinct feature of two-dimensional VCSELs.

We demonstrate a technology to extend the bandwidth of optical transmitters based on optical time interleaving. An optical time-interleaving in-phase-and-quadrature (IQ) modulator (TI-IQM), which consists of a 1-to-2 optical selector followed by two IQMs, generates optical signals with a bandwidth of each side of the optical carrier that is up to twice that of the electronic sub-signals sent to the IQMs. The sub-signals are generated by a digital spectral weaver followed by digital-to-analog converters and tailored so that the targeted extended-bandwidth signal is obtained as the final optical output from the TI-IQM. We verified the concept in a transmission experiment with 150-GBaud 4, 8, and 16-level quadrature amplitude modulation. We successfully generated optical signals with a bandwidth of each side of optical carrier of around 76 GHz, which is almost twice that of the electronic sub-signals.

We report on a 16-nm CMOS DAC based transmitter optimization enabling bitrates up to 1.1 Tb/s. We describe the transmitter optimization and derive the maximization of the net bitrate. Based on back-to-back experiments we optimize the entropy of the probabilistically shaped QAM format, which we further apply for subsequent transmission experiments over data center interconnect distances. Finally, we successfully demonstrate DWDM transmission of 17 channels in a 112.5 GHz grid operated at 105 GBaud over 80 km up to 118 km achieving a net bitrate of 1.1 Tb/s and spectral efficiencies of 9.8 bit/s/Hz.

This article presents a 2-channel 4-level pulse amplitude modulation (PAM4) transmitter front-end consisting of a 2-channel PAM4 shunt laser diode (LD) driver and flip-chip-bonded 1.3-μm LD array-on-Si that has the highest power efficiency yet reported for PAM4 transmitter front-ends. The driver was designed and fabricated using 65-nm CMOS technology, and the LD array-on-Si was fabricated using 1.3-μm membrane LD array-on-Si technology. To decrease the power consumption, the front-end does not use a high-speed digital-to-analog converter (DAC) or linear driver for generating the PAM4 signals. Instead, to generate clear PAM4 signals, the shunt LD driver incorporates DAC, and equalizing functions. The driver was designed with a photonic-electronic conversion system in SPICE, and the transmitter front-end was implemented with flip-chip-bonding interconnection between the driver and LD array-on-Si. The resulting 2-channel PAM4 transmitter front-end has high power efficiency. Our experiments show that in simultaneous operation, each channel could transmit 53-Gbps PAM4 signals, resulting in a power efficiency of 0.57 mW/Gbps. They also show that feed-forward equalization (FFE) with only 10 taps is enough for 2-km-long standard single-mode fiber (SSMF) transmission of 53-Gbps PAM4 signals with a bit error rate (BER) below 2.3 × 10−4 when the two channels are operated simultaneously.

We investigate high-throughput, multi-band transmission in a 4-core multi-core fiber (MCF) with the same 125 μm cladding diameter of standard single-mode fiber (SMF). A single wideband comb source is used to transmit up to 561 wavelength channels with 25 GHz spacing over a 120 nm bandwidth in S, C, and L bands. We demonstrate a maximum decoded throughput of 610 Tb/s in PDM-256QAM and PDM-64QAM signals over a 54 km fiber, transmitting more than 155 Tb/s in a single core and measuring a per-core average throughput exceeding record transmission demonstrations in SMF. In addition, we use noise loading measurements to characterize the achievable signal quality across the wideband transmitter. These results show that a single comb source can enable high-spectral efficiency modulation over wide bandwidths and further that low-core count homogeneous MCF technology can offer the same transmission performance as single-mode fibers without sacrificing mechanical reliability, and still offering the benefits of shared resources and greater efficiency that drives SDM technologies.

In this article, we firstly analyze the specific effect of time and phase synchronization errors on the coherent gain of microwave photonic wideband distributed coherent aperture radar (DCAR), and further propose a robust design based on bandwidth segmentation. In our design, the intermediate-frequency narrowband linear frequency modulated waveform (LFMW) sub-pulses are mixed with the stepped-frequency local oscillator signals at the transmitter, generating the stepped-frequency LFMW. To ensure the waveform purity, a photonics-based single-sideband up-converter is proposed as the mixer. At the receiver side, the echo wave is processed by a photonics-based down-converter combined with the matched-filtering processing. The relatively low-resolution range profiles with low coherent gain loss can be achieved by the narrowband sub-pulses, and the high-resolution range profile with low coherent gain loss can be synthesized from these low-resolution range profiles. The coherent gain of the proposed system is less sensitive to time synchronization error comparing to the system where the wideband LFMW is transmitted. Higher level of robustness also enables easier implementation of such system. To show the effectiveness of our design, we conduct simulations and further establish an X-band (8–12 GHz) microwave photonic system with two transmitters and one receiver. The coherent gain under different time synchronization errors are measured. Both the simulation and experimental results confirm that the proposed system achieves a higher robustness to time synchronization error. For instance, to keep the coherent gain loss less than 0.3 dB, the required time synchronization accuracy can be relaxed from 35 ps to 127 ps.

We investigate the feasibility of pulse amplitude modulation-N (PAM-N) signal transmission over a photonics-based cost-effective THz wireless link. A direct detection-based THz wireless link was established by using commercialized optical and electrical components designed for the WR 3.4 frequency band (220–330 GHz). At the THz signal transmitter, a uni-traveling carrier photodiode (UTC-PD) was used as a photomixer. The maximum output power of the UTC-PD was ∼47 μW when it was supplied with +17 dBm of optical power. At the THz signal receiver, the Schottky barrier diode was used as a diode detector; an offline digital signal processor (DSP) with a decision feedback equalizer was also used to mitigate the inter-symbol interference penalty. In the back-to-back configuration (i.e., antenna of Tx and Rx are placed at a distance of ∼1 cm), the driving condition of the optical modulator was optimized to obtain the best transmission performances of the PAM-N signal. A carrier frequency of 300 GHz is carefully selected considering the bit error rate (BER) dependency on the carrier frequency. After the optimization procedure, the BER performances were analyzed by varying the PAM order, baud rate, and wireless transmission distance. Finally, a 90 Gb/s PAM-8 signal is successfully transmitted over a wireless distance of 1.4 m, in which the measured BER is below the 20% soft-decision forward error correction threshold (2.0 × 10–2).

Self-coherent detection is a promising solution for extending the transmission reach for direct detection system. For these schemes, high carrier-to-signal power ratio (CSPR) is usually required to mitigate the inherent signal-signal beat interference (SSBI) induced from direct detection. Recently, carrier-assisted differential detection (CADD) receiver has been proposed to recover the optical field of complex-valued double sideband (DSB) signals via direct detection. Aiming at relaxing the high CSPR requirement for self-coherent detection systems, we propose an interleaved subcarrier loading scheme for double-sideband signals. At transmitter, only odd subcarriers are loaded with data while even subcarriers are left unused. By utilizing the CADD receiver, signal can be retrieved without the degradation of SSBI, which enables optical field recovery at low CSPRs. In this article, we first show by theory and simulation that the proposed scheme can work under lower CSPRs than the previous SSB self-coherent systems. Meanwhile, the electrical spectral efficiency of the proposed interleaved CADD is almost the same as SSB self-coherent systems. We also illustrate the process of parameter optimization for interleaved CADD, including optical delay, CSPR, and frequency gap. We successfully transmit 100-Gbps 16-QAM OFDM signals over 160-km uncompensated standard single mode fiber (SSMF) at 3.5-dB CSPR.

Transmitter and dispersion eye closure quaternary (TDECQ) penalty has replaced eye mask and transmitter dispersion penalty (TDP) methodologies for qualifying PAM-4 transmitters. TDECQ correlates better than the eye mask test with BER, and its implementation of receiver equalizers abstracts the metric from receiver features, enabling robust and systematic component assessments. However, assessing TDECQ is computationally intensive and time-consuming. Here, we present the use of machine learning (ML) to dramatically accelerate TDECQ assessments of PAM-4 transmitter signals. Explored techniques include convolutional neural networks and long short-term memory. We demonstrate that these methods provide comparable assessment accuracies compared to the conventional method, while tremendously reducing computational time. Some ML methods were ∼4500 times faster than the conventional method. Described architectures are generic and can be modified to accelerate any class of optical component assessments.

Visible light positioning has the potential to be a cost-effective technology for accurate indoor positioning. However, existing approaches often require large amounts of incoming data, usually in the form of high resolution images or dense lighting distributions. Additionally, a line of sight between transmitter and receiver is generally required at all times. In this work, we present a positioning approach that combines measurements from a camera, encoders and a gyroscope. We compare multiple algorithms for fusing these data, namely an extended Kalman filter, a particle filter and a hybrid approach. The end result is a system that provides location estimates even with sparse lighting distributions and temporary outages, yet achieves an average accuracy of 2 to 4 cm. Even in the 95th percentile of the cumulative error distribution, accuracy can be as low as 2 cm and is often lower than 10 cm. Moreover, due to the use of a low-resolution camera (640x480 pixels) and efficient fusion algorithms, the latency is relatively low on a standard laptop (between 5.6 and 21 milliseconds). Even on a low-cost embedded board, latency generally does not exceed 100 milliseconds. We validate our approach experimentally and show that it is robust under a wide range of illumination conditions.

In this article, we experimentally demonstrate the amplifier-less transmission of beyond 100-Gbit/s signal with advanced modulation formats including discrete multi-tone (DMT) and pulse amplitude modulation (PAM) in an intensity modulation direct detection (IM-DD) system for 40-km datacenter interconnects edge (DCI-Edge) utilizing a low cost commercial 10G-class O-band directly modulated laser (DML). During the transmission of DMT signal, Chow Cioffi Bingham algorithm-based bit-loading (CCB-BL) and discrete-Fourier-transform spread (DFT-spread) techniques are both discussed in the bandwidth-limited communication system. For the PAM single carrier communication system, look-up table (LUT) pre-distortion and time-domain pre-equalization schemes are jointly employed at the transmitter side to mitigate nonlinear distortions and insufficient bandwidth induced impairments, respectively. The results show that 105-Gbit/s DFT-spread DMT, 115-Gbit/s CCB-BL DMT, and 120-Gbit/s PAM-8 signals are successfully transmitted over 40-km single mode fiber (SMF) with the bit-error-rate (BER) below hard decision forward error correction (HD-FEC) threshold of 3.8 × 10–3. 120-Gbit/s DFT-spread DMT and CCB-BL DMT signals allocated within 20-GHz bandwidth can be successfully transmitted over 20-km SMF, and the receiver sensitivity of DFT-spread DMT outperforms CCB-BL DMT by 1.5-dB at the BER of 3.8 × 10–3. Due to the flexibility of bit-loading technique, when the allocated bandwidth of 120-Gbit/s CCB-BL DMT signal extends from 20-GHz to 25-GHz, the required received optical power (ROP) of CCB-BL DMT can be lower than DFT-spread DMT by 1-dB at the BER of 3.8 × 10–3 over 20-km SMF transmission. In 120-Gbit/s PAM-8 system, LUT pre-distortion technique improves the system receiver sensitivity by more than 1-dB over 40-km SMF transmission. This is the first time to discuss and implement different advanced modulation formats to achieve 100-Gbit/s/λ transmission in a 40-km amplifier-less IM-DD system with a low-cost O-band 10G class DML. Both CCB-BL DMT and LUT pre-distortion assisted PAM-8 can achieve a BER lower than HD-FEC threshold of 3.8 × 10–3 when the net rate per optical carrier is higher than 100-Gbit/s, which are promising solutions for 40-km four-lane 400-GbE DCI-Edge applications.

Underwater optical wireless communication (UOWC) is of great interest to the academic and the industry community. In this article, we propose a low-complexity and effective joint transmitter-side digital signal processing (DSP) including geometric shaping, time-domain tone reservation (TR), and clipping. The peak-to-average power ratio reduction performance and implementation complexity of the proposed time-domain TR are extensively analyzed. We then develop a real-time 2.2-Gb/s system, which is optimized for its shaping ratio and clipping ratio. With the help of the efficient DSPs, an 8-dB received optical power enhancement is realized under the bit-error-rate threshold of 3.8 × 10−3. We successfully demonstrate a time-multiplexed four 4K video transmission in real-time using the proposed scheme over a 3.6-m underwater and 8-m free-space channel. The implementation details, as well as the analyses of system stability, resource utilization, and latency, are presented. The results validate the feasibility and effectiveness of the proposed scheme and a 2.2-Gbit/s real-time OWC system is demonstrated for a water-air communication link.

In this article, we consider indoor visible light positioning (VLP) based on received signal strength (RSS). Under the assumption of unknown receiver coefficient, we propose a robust VLP scheme which jointly estimates the receiver position and receiver coefficient. Through mathematical derivations, the original estimation problem is equivalently transformed into the position estimation problem of maximizing the angle between the received signal vector and RSS feature vector, where the direction of RSS feature vector is shown to be a one-to-one feature characterizing the receiver position. Compared with existing schemes, the proposed VLP scheme is more robust to the variations of transmission environments, as it only requires the ratios of transmitter coefficients rather than specific values of transceiver coefficients. Besides, for implementation of the proposed scheme, we also present a method for estimating the required parameters. Experimental and simulation results demonstrate the robustness of the proposed VLP scheme. Specifically, experimental results show that the proposed VLP scheme retains a stable positioning accuracy about 4 cm in a 150 × 60 × 60 cm $^3$ space and about 8 cm in a 200 × 60 × 60 cm $^3$ space.

Insufficient system bandwidth and chromatic dispersion (CD) of fiber are major limitations for high-speed four-level pulse amplitude modulation (PAM4) direct detection systems. In this article, we investigate C-band 112-Gb/s PAM4 optical transport systems over 10-km and 20-km standard single-mode fiber (SSMF) transmission, which suffers from both bandwidth limitation and CD-induced power fading. Advanced digital signal processing techniques are proposed to handle those channel distortions. At the transmitter, a low-complexity $1+\alpha Z^{-1}$ finite impulse response (FIR) filter coarsely pre-compensates system bandwidth without a preliminary channel estimation. At the receiver, a modified feed-forward equalizer and decision feedback equalizer (FFE-DFE) is employed to compensate residual insufficient bandwidth and CD-induced power fading. Moreover, a novel multi-symbols joint decision is proposed for receiver-side FFE-DFE to mitigate symbol error propagation and correct possible previous error symbols. Experimental results show that the proposed pre-equalization technique exhibits a better transmission performance, which reduces bit error ratio (BER) from $3.3\times 10^{-3}$ to $5.5\times 10^{-4}$ for optical back to back transmission. As for 10-km transmission case, FFE-DFE with 5-symbols joint decision achieves a BER of $7.3\times 10^{-5}$ , which is better than the BER of $1.6\times 10^{-4}$ with ideal error-propagation-free FFE-DFE (EPF-FFE-DFE). Furthermore, for 20-km transmission scenario, FFE-DFE with 8-symbols joint decision achieves a BER of $6.0\times 10^{-4}$ , which is also a superior result than the BER of $9.1\times 10^{-4}$ with EPF-FFE-DFE.

High throughput coherent optical transmitters are key components in future optical communication infrastructure. However, these transmitters are often distorted with the nonlinearity of their components. A potential approach for compensating nonlinearity is by applying digital pre-distortion methods based on the Volterra series or one of its derivatives. However, the Volterra series-based solution is complex to implement, difficult to scale, and its simplified versions may not yield the desired performance. Recently digital pre distortion solutions based on neural networks were proposed, which may benefit from the generality of neural networks and can be more easily scaled. These solutions are often based on non-standard neural network architectures which require complex neurons-based architectures or being based on indirect training approach which suffer from noise enhancement. In this article, a novel method for neural network-based pre-distortion with direct learning is proposed. The direct learning with neural network does not assume a specific transmitter model and does not suffer from noise enhancement. The method assumes standard neural network inference architecture and is applied to a coherent nonlinear optical transmitted with long-short-term memory neural network. The overall performance and complexity of the direct learning method is compared with the indirect approach and with the Volterra series-based solution, showing significant advantage in performance, especially in cases of severe nonlinearity and noise conditions.

Nonlinear interactions between neighboring pulses has always been a fundamental bottleneck in soliton transmission systems. Recently, coherent transceivers, digital signal processing (DSP) and the new nonlinear Fourier transform (NFT) theoretical framework has revived and generalized the field of soliton transmissions into nonlinear frequency division multiplexing (NFDM). We hereby demonstrate analytically and experimentally that one can considerably improve soliton transmission performance by intentionally allowing neighboring solitons to interact and collide during propagation and exchange positions at the receiver followed by standard NFT processing. This can be achieved by designing neighboring solitons’ eigenvalues $\boldsymbol {\lambda }$ to have opposite signs in the real part while the magnitude $| {{\bf \Re}(\boldsymbol {\lambda })} |$ is optimized for a given transmission distance so that neighboring transmitted pulses would have swapped their timing positions at the receiver. Experimental results for 6.13 Gbaud 1-soliton systems demonstrate a transmission reach improvement of 100% for 16APSK and 60% for 8PSK modulated on the b -coefficients. The proposed scheme eliminated a long-standing fundamental limitation in soliton transmissions, opened up new dimensions in transmitter signal design and receiver signal processing for nonlinear optical communication systems.

Arrayed Waveguide Grating Router (AWGR)-based interconnections for Multi-Socket Server Boards (MSBs) have been identified as a promising solution to replace the electrical interconnects in glueless MSBs towards boosting processing performance. In this article, we present an 8-socket glueless optical flat-topology Wavelength Division Multiplexing (WDM)-based point-to-point (P2P) interconnect pursued within the H2020 ICT project ICT-STREAMS and we report on our latest achievements in the deployment of the constituent silicon (Si)-photonic transmitter and routing building blocks, exploiting experimentally obtained performance metrics for analyzing the 8-socket chip-to-chip (C2C) connectivity in terms of throughput and energy efficiency. We demonstrate an 8-channel WDM Si-photonic microring-based transmitter (Tx) capable of providing 400 (8 x 50) Gb/s non-return-to-zero (NRZ) Tx capacity and an 8 x 8 Coarse-WDM (CWDM) Si-AWGR with verified cyclic data routing capability in O-band. Following an overview of our recently demonstrated crosstalk (XT)-aware wavelength allocation scheme, that enables fully-loaded AWGR-based interconnects even for typical sub-optimal XT values of silicon integrated CWDM AWGRs, we validate the performance of a full-scale 8-socket interconnect architecture through physical layer simulations exploiting experimentally-verified simulation models for the underlying Si-photonic Tx and routing circuits. This analysis reveals a total aggregate capacity of 1.4 Tb/s for an 8-socket interconnect when operating with 25 Gb/s line-rates, which can scale to 2.8 Tb/s at an energy efficiency of just 5.02 pJ/bit by exploiting the experimentally verified building block performance at 50 Gb/s line. This highlights the perspectives for up to 69% energy savings compared to the standard QuickPath Interconnect (QPI) typically employed in electronic glueless MSB interconnects, while scaling the single-hop flat connectivity from 4- to 8-socket interconnection systems.

Theoretically, Kramer–Kronig (KK) receivers are able to fully cancel signal-signal beat interference (SSBI), which results from a field-modulated single-sideband (with carrier) transmitter being used with a direct-detection receiver. However, even for strong carriers, which should meet the minimum-phase (MP) condition, bits are frequently erroneous. In this article, we simulate two methods of reducing error rates for weak carriers. The first is to limit the minimum value of the received signal, so that strong negative-going peaks are not produced by the logarithm function of the KK's ‘correction path’. The second is to replace the square-root function in the main signal path with a logarithmic function. This is to allow analog processing using semiconductor diodes. Our numerical simulations show that grossly approximating the square-root function is actually beneficial to the system, as it reduces the power in the signal close to the clipping events. When both the first and second techniques are combined, low error rates can be obtained at 2-dB less carrier power than for conventional processing, and 0.25-dB better receiver sensitivity. When 100-km of standard fiber is included, the carrier can be reduced by 3 dB, and the system has a 0.18 dB better sensitivity, and potentially supports analog processing. This work suggests that there may be other improvements possible, in signal quality and computational load, by moving away from a theoretically perfect implementation of the KK algorithm.

The exhausted radio frequency spectrum in mobile networks calls for light-based wireless solutions. However, visible light communication builds on dedicated receiver sub-systems. As an alternative to traditional photodetectors, we exploit low-cost commercial light emitting diodes (LED) that enjoy abundance in consumer electronics. In particular, surface-mounted LEDs with emission in the yellow spectrum are employed as photodetectors in a line-of-sight optical wireless communication link that is sourced by a green laser diode as a pencil beam emitter. Together with their corresponding transimpedance amplifiers, the LED-based receivers show an opto-electrical 3-dB bandwidth of up to 260 MHz, which enables data rates of 3.1 Gb/s using orthogonally frequency division multiplexed signal transmission over in-door distances. We further show that costly digital signal processing functions can be omitted through a fully analogue design of the optical link. Passive frequency response equalization in combination with simple duobinary signaling allows for 1.25 Gb/s transmission over the same link distance. Moreover, we prove that there is neither a rate nor a quality-of-experience penalty when collapsing an Ethernet connection over the optical wireless link in order to transfer large amounts of data or to stream high-definition video. The high-speed properties of commercial LEDs, together with their established role as a transmitter, promise to open new application possibilities in low-cost, short-reach segments.

In our article to extend a 100 Gbit/s/λ fiber-amplifier-less optical link with a simple and cost-effective transmission system, we have fabricated an O-band wavelength range (1260–1360 nm) SOA assisted extended reach EADFB laser (AXEL) with a 300-μm-long DFB laser and 85-μm-long SOA. The length of the integrated electroabsorption modulator was set to 125 μm to obtain a section with low parasitic capacitance for high speed operation. The fabricated AXEL transmitter simultaneously achieved both the sufficient modulated light output of 8.9 dBm and an excellent quality eye diagram for a 53-Gbaud PAM4 signal with the TDECQ value of 1.52 dB. We investigated the effect of the integrated SOA on the AXEL's performance and found no significant degradation of the quality of the eye diagram or the bit error rate (BER) characteristics of the 53-Gbaud PAM4 back-to-back configuration, even when the AXEL achieved high modulated light output power. We were also able to extend the 53-Gbaud PAM4 fiber-amplifier-less optical link to 60 km by using the AXEL transmitter and APD receiver, and achieved error-free transmission assuming KP4-forward error collection (FEC). These results demonstrate that the enhanced modulated light output using the AXEL can successfully extend the 53-Gbaud PAM4 reach with a simple transmission setup.

In this article we describe an analytical model for estimating performance degradations caused by non-ideal transponder implementation for coherent modulation formats based on parametric fitting of back to back measurements. We discuss in detail several reasons for implementation penalties (IP) such as noise in the transmitter and the receiver, non-ideal filtering, and quantization in the analog-to-digital converter, and how the fitting parameters depend on these penalties. We consider how to combine the IP model and the Gaussian noise (GN) model of nonlinear propagation which includes Amplified Spontaneous Emission (ASE) and Nonlinear Interference (NLI) noise impacts. We modify the Signal to Noise Ratio (SNR) expression of the GN model to include any performance impairments from a non-ideal transponder and additional link components apart from the transponder. In this article we consider transponders based on 16QAM and 2ASK/4PSK-8QAM modulation formats. Then we compare the modeling results obtained with the IP and GN models with experimental data for erbium doped fiber amplifier (EDFA) and Raman-assisted transmission systems. We find that this modeling approach demonstrates good agreement experiment data.

Incorporating radar detection and frequency measurement functions within a single system is highly desirable in modern RF applications. In this article, a microwave photonic system is proposed and demonstrated to simultaneously realize broadband radar detection and frequency measurement. By sharing a light with ± 2nd-order linear frequency-modulated (LFM) sidebands in a polarization-division-multiplexing manner, radar detection and frequency measurement functions can be realized simultaneously. In the transmitter, the shared light is used to generate a radar signal that has a quadrupled bandwidth compared with the input electrical LFM signal. In the receiver, the shared light is applied as the reference to perform photonic de-chirping of the radar echo and to scan the frequency of the signal under test (SUT) along two orthogonal polarization directions. In the experiment, a 4.5–6.5 GHz LFM signal is used to generate the shared light source, based on which radar detection (bandwidth: 18–26 GHz) with a range resolution as high as 2.06 cm, and frequency measurement (measurement range: 28–36 GHz) with a scanning rate of 0.8 GHz/μs and a resolution of 37.6 MHz are achieved. In addition, the reconfigurablilty of the frequency measurement range, and its capability for measuring the multi-tone and wideband signals are also demonstrated.

Beyond 100-Gbit/s/λ high-speed inter-data-center link composed by hybridly integrated electro-absorption modulator (EAM) and distributed feedback laser diodes (DFBLD)-based transmitter at 1306 nm, and received by a PIN-PD + TIA optical receiver is demonstrated to carry 104-Gbit/s pre-emphasized PAM-4 data for 2–10 km single mode fiber (SMF) propagation. The EAM + DFBLD transmitter offers an average power of −1.5 dBm, an excellent side mode suppression ratio of 53 dB, an on/off extinction ratio up to 10 dB, an analog bandwidth of 25 GHz, and a relative intensity noise of -137 dBc/Hz. The PIN-PD + TIA module exhibits 30-GHz receiving bandwidth. For on-off keying (OOK) format, the allowable data rates at 62 and 60/54 Gbit/s for BtB and 2-km/10-km SMF cases obtain receiving BER of 1.48 × 10−13 and 3.4 × 10−14/1.77 × 10−14 qualified for error-free telecommunication. Even at simulated inter-data center link, the receiving power sensitivity and power penalty at 54 Gbit/s can achieve −6.43/−6.26 dBm and 0.07/0.24 dB after 2-/10-km SMF propagation. To efficiently utilize the limited transmitter bandwidth, the directly encoded PAM-4 data transmission reveals qualified BER slightly enlarged from 1.4 × 10−4 to 2.3 × 10−4 when transferred from back-to-back to 2-km SMF transmission. Lengthening the distance to 10-km SMF degrades the maximal PAM-4 data rate to 50 GBaud with a received BER of 2.3 × 10−4 (less than KP4-FEC criterion). The 50 GBaud PAM-4 link provides receiving power penalty of 0.11/0.24 dB after 2-/10-km SMF propagation, which is attributed to the narrow-linewidth DFBLD responsible for extremely low RF fading effect under nearly zero dispersion condition. The transmitting frontend with amplitude/phase pre-emphasis function makes the receiving end compact and cost-effective, which exhibits the higher spectral usage efficiency of 4 bit/s/Hz and the better receiving sensitivity of −5.5 dBm at the lower power budget of 2.8 fJoule/bit than previous works.

In this article, we report on energy efficient (6.2 mW/Gbit/s) transmitter and receiver assemblies capable of NRZ 80 Gbit/s and 72 Gbit/s fiber data transmission through 2 and 50 m of MMF, respectively. The optical link studied consisted of a 130 nm BiCMOS driver and transimpedance amplifier, ∼30 GHz VCSEL, and ∼27 GHz PD.

Spatial multiplexing multiple-input multiple-output (MIMO) visible light communication (VLC) has become a potential candidate for next-generation wireless communication networks. In this paper, a novel superposed odd-order 32QAM constellation scheme is proposed in 2 × 2 MIMO VLC systems. Two independent signals generated from two light emitting diode (LED)-based transmitters with a four-quadrature amplitude modulation (4QAM) and geometrically shaped square 8QAM, respectively, are transmitted over free space and spatially multiplexed in the receiver to obtain a distinctly square-shaped 32QAM signal. The high channel correlation causes failure of MIMO detection in traditional spatial multiplexing (SMP) systems owing to the rank defect of the channel matrix. However, MIMO detection can always be realized by a simple constellation de-mapping based on the concept of superposed constellations, regardless of the extent of channel correlation. Considering the non-linearity of an LED, the square-shaped 8QAM constellation is introduced to reduce the peak to average power ratio (PAPR) at the transmitter, which also benefits from its maximum minimum Euclidean distance (MED) at the receiver compared to those of traditional 8QAM constellations. Moreover, the equal-gain combining (EGC) algorithm is employed to gain the receive diversity. The proposed system is experimentally studied in detail, where the system performance is evaluated under the conditions of different driving peak-to-peak voltages (Vpps) and channel gains. The experimental demonstration also shows that the dynamic range of the driving Vpp can be improved from 0.3 V to 0.6 V when compared with that of the traditional non-geometrically-shaped 8QAM, clearly validating the superiority of the proposed system.

The upcoming new radio access allows ultra-high data rate using millimeter-wave (mm-Wave) frequencies, while it normally suffers from large path loss. To compensate for path loss, phased arrays for both the transmitter and receiver are used. The 5G new radio (NR) three beam management process proceeds as follows: The transmitted beam is first swept in the downlink direction from the remote radio unit (RRU) to the user equipment (UE), and then the uplink beam is aligned to determine which beam direction has the best reception quality, and vice versa . However, this sequential beam management requires that the RX must be able to perform both beam detection and steering across all the reception angles. Moreover, due to the narrow beamwidth of the phased array operation, a “quantum leap” performance improvement of the receiver operating at mm-Wave is required. In this article, a self-steering array beamformer (SSA-BF) receiving system is proposed, which is composed of a home-designed IC package with zero DC power consumption and a 4-element antenna array. We first conduct the measurement without the antenna, and the SSA-BF receiver shows a significant array factor enhancement with negligible SNR degradation over full field-of-view (FoV) (incidence angle = ±90°)

A novel iterative algorithm for electronic dispersion compensation (EDC) at the transmitter is described and demonstrated for intensity modulation and direct detection (IM/DD) systems. The proposed algorithm is based on the Gerchberg-Saxton (GS) algorithm, where the link dispersion performs time-to-frequency mapping, while the unconstrained phase at the direct detection receiver acts as a degree of freedom. Simulation results for a transmitter employing a single drive Mach-Zehnder modulator (MZM) operating at 28 Gb/s non-return-to-zero on-off keying (NRZ-OOK), show pre-compensation of upto 100 km of single mode fiber (SMF), with 1 dBQ margin above the forward error correction (FEC) bit error rate (BER) limit of $3.8 \times 10^{-3}$ . It is shown that with five iterations of the proposed EDC algorithm in digital-signal processing (DSP) coupled with an 8-bit 56 GSa/s digital-to-analog converter (DAC), the transmission reach is extended by a factor of 5 in unamplified SMF links. Simulations for a 56 Gb/s four-level pulse-amplitude-modulation (PAM-4) show the compensation of upto 10 km with five iterations of the EDC algorithm at 1 dBQ above the FEC limit.

Semiconductor optical amplifier (SOA) is regarded as a promising candidate to amplify the optical signal in next-generation access network, especially for O-band signal transmission. However, the SOA-based system suffers from the pattern-dependent effect due to its gain saturation. Compared to digital signal processing (DSP) at receiver, DSP in the transmitter can reduce implementation complexity, which is attractive for the multi-user system. In this paper, we propose three methods to mitigate the pattern-dependent effect in SOA and compare their performance based on: 1) look-up-table (LUT) considering pattern-dependent effect, 2) pre-distortion without considering pattern-dependent effect, 3) probabilistic shaping (PS). The performance of three mitigation methods is investigated in an intensity modulation with direct detection (IM/DD) 4-ary pulse-amplitude-modulation (PAM-4) transmission system at O-band. Our research results show that all these schemes can improve the receiver dynamic range of PAM-4 signals at 50 Gbit/s and 100 Gbit/s. Moreover, PS and LUT can also improve the receiver dynamic range of 75-Gbit/s PAM-8 signals.

In the paper, bit-interleaved polar-coded modulation discrete multi-tone (DMT) (BIPCM-DMT) with 256-QAM enabled by constant amplitude zero auto-correlation (CAZAC) precoding is proposed and experimentally demonstrated in a low-cost intensity-modulation and direct-detection (IM-DD) system. In order to conquer the fading effect on mapped bit-levels, two types of interleaver including quadratic polynomial permutation (QPP) interleaver and random interleaver are compared to optimize the system performance. However, for 256-QAM BIPCM-DMT system, the signal-to-noise ratio (SNR) fades dramatically as the data subcarrier index increases, it is difficult to match the well-designed operating point of Monte Carlo in the process of polar code construction. Thus, the precoding schemes including CAZAC and orthogonal circulant transform (OCT) are proposed to equalize the uneven SNRs in BIPCM-DMT system. Compared with the bit / power loading and pre-equalization schemes, it will greatly reduce the cost and complexity of the system because channel information obtained by the transmitter via round-trip feedback is not considered. In addition, CAZAC precoding is selected to optimize the BIPCM-DMT system because it is superior to OCT precoding in reducing PAPR. The experimental results show that aided by CAZAC precoding, 20.63 Gb/s BIPCM-DMT signal can be received with an error-free over 50-km standard single-mode fiber (SSMF). Furthermore, compared with BIPCM-free, only CAZAC precoding, and only random interleaver schemes, the proposed scheme can achieve Q-factor enhancement of 4.82 dB, 2.46 and 1.78 dB, respectively, at the received optical power (ROP) of −5 dBm.

We report a promising IQ optical modulator for beyond 100-GBd transmitter. By introducing both a new n-i-p-n heterostructure and an optimized capacitance-loaded traveling-wave electrode (CL-TWE), high-frequency electrical losses of the modulator can be drastically reduced. As a result, we extended an electro-optic (EO) bandwidth without degrading other properties, such as half-wave voltage (Vπ) and optical losses. The 3-dB EO bandwidth of the 1.5-V Vπ modulator reaches 80 GHz. Furthermore, we demonstrated up to 128-GBd IQ modulations by co-assembling with an ultra-broadband InP-based driver IC.

As next-generation data center optical interconnects aim for 0.8 Tb/s or 1.6 Tb/s, serial rates up to 106 GBd are expected. However doubling the bandwidth of current 53 GBd 4-level pulse-amplitude modulation (PAM-4) transmitters is very challenging, and perhaps unfeasible with compact, non-travelling wave, lumped modulators or lasers. In our previous work, we presented an integrated 4:1 optical serializer with electro-absorption modulators (EAMs) in each path. Transmitter (TX) functionality was shown up to 104 GBd non-return-to-zero (NRZ) On-Off Keying (OOK) or PAM-4. However the performance for PAM-4 was limited by the distortion introduced by the EAM non-linearity. We also presented a real-time, DSP-free 128 Gb/s PAM-4 link with a silicon photonic transmitter using binary driven EAMs in a Mach-Zehnder interferometer (MZI) configuration. By combining two of such half-rate (53 GBd) transmitters in an integrated 2:1-serializer, improved 106 GBd PAM-4 performance is expected without needing to compensate the inherent modulator non-linearity and without requiring faster modulators or drivers. In this paper, we present a Silicon integrated 53 GBd PAM-4 TX as a candidate for integration into 106 GBd PAM-4 2:1 serialized TX. The presented TX consists of two EAMs in an MZI configuration, wirebonded to a low-power 55 nm 4-channel SiGe BiCMOS driver, operating at 1.5 pJ/b (excluding laser). With a reference receiver (RX), transmission at or below the KP4-FEC threshold is shown beyond 1km standard single-mode fiber (SSMF) and up to 2 km non-zero dispersion-shifted fiber (NZ-DSF) at 1565 nm. Furthermore, the integrated TX was combined with an Si-integrated RX consisting of the same EAM component, wirebonded to a 55 nm SiGe BiCMOS transimpedance amplifier (TIA). Both TX and RX were wirebonded on an RF-PCB, with electrical connectivity through transmission lines and 6-inch 50 GHz multi-coax cable connectors. With this electrically connectorized all-EAM TX and RX, PAM-4 link operation is shown up to 40 GBd at 3.9 pJ/b (excluding laser), without using DSP or equalization.

Microring modulators (MRMs) with CMOS electronics enable compact low power transmitter solutions for 400G Ethernet and future on-package optical transceivers. In this paper, we present a 112 Gb/s PAM4 transmitter using silicon photonic MRM, on-chip laser and co-packaged 28 nm CMOS driver. We describe the impact of static and dynamic MRM nonlinearity on PAM4 signaling and present a dual path nonlinear pre-distortion technique to compensate both effects. PAM4 measurement results of our transmitter at 112 Gb/s show that TDECQ