We report on the first experimental demonstration of 200-Gbps (net rate 166.7-Gbps) 1.55-μm DMT IMDD transmission over 1.6 km fiber using a single monolithically-integrated-EML, DAC and photodiode, achieving an effective electrical spectrum efficiency of 4.93 bit/s/Hz. © 2017 IEEE.

We experimentally demonstrate a DMT transmission system with 1.55-μm EML using nonlinearity-aware time domain super-Nyquist image induced aliasing. Compared with linear equalization, the capacity is improved by ∼16.8%(33.1%) with proposed method for 4(40) km transmission.

We experimentally demonstrate a DMT transmission system with 1.55-μm EML using nonlinearity-aware time domain super-Nyquist image induced aliasing. Compared with linear equalization, the capacity is improved by ~16.8%(33.1%) with proposed method for 4(40) km transmission. © 2017 OSA.

We propose a low complexity timing algorithm for high order PAM. Experimental results demonstrate higher performance and lower complexity than conventional algorithms in a 32 Gbaud PAM-8 transmission over 4 km SMF links.

A thorough analysis of equalization-enhanced phase noise (EEPN) and its impact on the coherent optical system is presented. We show with a time-domain analysis that EEPN is caused due to the interference of multiple delayed versions of the dispersed signal, generated by intermixing of the received dispersed signal, and the noise side bands of the local oscillator (LO) in the photodetectors. We derive statistical properties such as the mean, variance, and error vector magnitude of the received signal influenced with EEPN. We show that in coherent optical systems utilizing electronic dispersion compensation, this noise corresponds to multipath fading in wireless communication systems. Closed-form expressions of necessary LO linewidth and/or mitigation bandwidth for a general system configuration and specified OSNR penalty are given. The expressions for system design parameters, validated with system simulations, show that higher order modulation formats, such as 16-quadrature amplitude modulation and beyond, put stringent demands on the LO linewidth unless a mitigation technique is used.

We present an extensive study of equalization enhanced phase noise (EEPN) in coherent optical system for all practical electronic dispersion compensation configurations. It is shown that there are only eight practicable all-electronic impairment mitigation configurations. The non-linear and time variant analysis reveals that the existence and the cause of EEPN depend on the digital signal processing (DSP) schemes. There are three schemes that in principle do not cause EEPN. Analysis further reveals the statistical equivalence of the remaining five system configurations resulting in EEPN. In three of them, EEPN is due to phase noise of the transmitting laser, while in the remaining two, EEPN is caused by the local oscillator. We provide a simple look-up table for the system designer to make an informative decision regarding practicable configuration choice and design.

A method for mitigating local oscillator (LO) phase noise-induced impairment, also known as equalization-enhanced phase noise, in coherent optical systems is discussed. The method is suitable for real-time implementation and requires hardware with a bandwidth much lower than the signal baud rate, even for a system utilizing conventional semiconductor laser as LO. We evaluate the required parameters like interpolation technique, electrical signal-to-noise ratio at digital coherence enhancement (DCE) front end, for long haul transmission links having quadrature phase shift keying and 16-quadrature amplitude modulation formats. We show that the method can be implemented using a low-speed DCE front end and a simple digital linear interpolator with small (<1 dB) implementation penalty even in cases that would otherwise result inerror floor.

We experimentally demonstrate for a 28 Gbaud 64-QAM metro link that the LO frequency noise causes timing impairment. Results show the existence of LO frequency noise spectrum regimes where different design criteria apply.

We experimentally demonstrate for a 28 Gbaud 64-QAM metro link that the LO frequency noise causes timing impairment. Results show the existence of LO frequency noise spectrum regimes where different design criteria apply.

We for the first time experimentally demonstrate a simple technique to overcome EEPN. Performance recovery from above FEC to <1 dB penalty for 28 Gbd 16-QAM over 520 km with high LO linewidth is achieved.

We present experimental validation of novel analytical expressions essential for the design of coherent optical systems impaired by EEPN. These expressions enable a simple and accurate EEPN analysis for any system specification.

We experimentally investigate the possibility to mitigate local oscillator induced Equalization Enhanced Phase Noise penalty. The results pave the way for the use of even 10 MHz linewidth local oscillator lasers in 28 Gbd 16-QAM metro links.

A theoretical investigation of the equalization-enhanced phase noise (EEPN) and its mitigation is presented. We show with a frequency domain analysis that the EEPN results from the non-linear inter-mixing between the sidebands of the dispersed signal and the noise sidebands of the local oscillator. It is further shown and validated with system simulations that the transmission penalty is mainly due to the slow optical frequency fluctuations of the local oscillator. Hence, elimination of the frequency noise below a certain cut-off frequency significantly reduces the transmission penalty, even when frequency noise would otherwise cause an error floor. The required cut-off frequency increases linearly with the white frequency noise level and hence the linewidth of the local oscillator laser, but is virtually independent of the symbol rate and the accumulated dispersion.

Heating of poly(methyl methacrylate) ridge optical waveguides slightly above glass transition temperature minimizes surface roughness and provides cylindrical shape. We experimentally demonstrate propagation loss decrease and polarization insensitivity as a result of waveguide thermal treatment.

Heating of poly(methyl methacrylate) ridge optical waveguides slightly above glass transition temperature minimizes surface roughness and provides cylindrical shape. We experimentally demonstrate propagation loss decrease and polarization insensitivity as a result of waveguide thermal treatment.

Integrated polymer photonics brings low cost and high fabrication flexibility to optoelectronic industry. However, this platform needs to overcome several issues to be effective enough for practical applications. In this work, we experimentally demonstrate a decrease of propagation losses and polarization sensitivity of polymer waveguide-based devices as a result of thermal treatment. Heating of poly(methyl methacrylate) strip optical waveguides above the glass transition temperature initiates a waveguide surface reflow due to a decrease of the polymer viscosity and surface tension energy. This results in a decrease of surface roughness and shape change from rectangular to cylindrical. Thus, scattering losses and polarization sensitivity are minimized. IEEE

A novel two-stage n-PSK partitioning carrier phase recovery (CPR) scheme for circular multilevel quadrature amplitude modulation (C-mQAM) constellations is presented. The first stage of the algorithm provides an initial rough estimation of the received constellation, which is utilized in the second stage for CPR. The performance of the proposed algorithm is studied through extensive simulations at the forward error correction bit error rate targets of 3.8 10-3 and 1 × 10-2 and is compared with different CPR algorithms. A significant improvement in the combined linewidth symbol duration product (ΔnTs) tolerance is achieved compared to the single-stage n-PSK partitioning scheme. Superior performance in the ΔnTs tolerance compared to the blind phase search algorithm is also reported. The relative improvements with respect to other CPR schemes are also validated experimentally for a 28-Gbaud C-16QAM back-to-back transmission system. The computational complexity of the proposed CPR scheme is studied, and reduction factors of 24.5 | 30.1 and 59.1 | 63.3 are achieved for C-16QAM and C-64QAM, respectively, compared to single-stage BPS in the form of multipliers | adders.

The phase noise tolerance of circular multilevel quadrature amplitude modulation (C-mQAM) constellations employing different carrier phase recovery (CPR) algorithms is studied. A differential decoding scheme and a bit mapping for this type of constellations are proposed. A novel CPR scheme for C-mQAM constellations is also presented. The particular distribution of the constellation points in a C-mQAM signal is exploited to reduce the required Nth power for the removal of the modulation component by a factor of two. Hence, the computational complexity of the proposed algorithm is drastically reduced. The combined linewidth symbol duration product (??Ts) tolerance of different CPR algorithms for C-mQAM constellations is studied and compared with the proposed CPR scheme. The results are analyzed at 3.8e-3 and 1e-2 bit error rate forward error correction limits. The proposed CPR scheme achieves similar ??Ts tolerance compared to single stage BPS algorithm while its computational complexity is reduced by group factors of 27.2 | 32.3, and 30.5 | 32.6 (in the form of multipliers | adders) for C-16QAM and C-64QAM, respectively.

We experimentally validate two novel CPR schemes outperforming existing CPRs in complexity and performance. A complexity reduction of at least a factor of 4 is reported compared to the BPS algorithm for a 64QAM system.

The inherent discrete phase search nature of the conventional blind phase search (C-BPS) algorithm is found to introduce angular quantization noise in its phase noise estimator. The angular quantization noise found in the C-BPS is shown to limit its achievable performance and its potential low complexity implementation. A novel filtered BPS algorithm (F-BPS) is proposed and demonstrated to mitigate this quantization noise by performing a low pass filter operation on the C-BPS phase noise estimator. The improved performance of the proposed F-BPS algorithm makes it possible to significantly reduce the number of necessary test phases to achieve the C-BPS performance, thereby allowing for a drastic reduction of its practical implementation complexity. The proposed F-BPS scheme performance is evaluated on a 28-Gbaud 16QAM and 64QAM both in simulations and experimentally. Results confirm a substantial improvement of the performance along with a significant reduction of its potential implementation complexity compared to that of the C-BPS.

We propose a novel carrier phase recovery scheme for circular 16QAM coherent transmission systems. Experimental results show low penalties (<;1.4 dB) even for linewidths where square 16QAM exhibits irretrievable (above the FEC limit) error floor.

We propose a method for cycle-slip mitigation in circular multilevel quadrature amplitude modulation (C-mQAM) coherent optical systems, with constellation rotational asymmetry, based on an adaptive boundaries approach. The impact of cycle-slips in C-mQAM coming from Viterbi-Viterbi algorithm limits the phase noise tolerance. By introducing adaptive boundaries and a differential coding scheme, the ambiguity of asymmetrical rotation of constellation can be effectively removed. Performance of the proposed method is evaluated for a C-16QAM and C-64QAM for various laser linewidths. Results show a noticeable improvement in linewidth symbol duration product (Δν · TS) tolerance compared with the previous studies on C-mQAM and mQAM constellations. The Δν · TS tolerance reaches 4 × 10 -4 and 1.1 × 10 -4 for C-16QAM and C-64QAM, respectively, for 1 dB penalty at a symbol error rate of 10 -3.

The cloud services together with the huge size datasets are driving demand for bandwidth in datacenters. The 400 Gbps client-side links are demanding a solution. The intensity modulation and direct-detection systems together with integrated semiconductor lasers and modulators appear as promising solution in four optical lanes at 100 Gbps net rate in order to reduce complexity, size, power consumption and cost.

We demonstrate 100 Gbaud 4PAM transmission over 400 meters SMF with monolithically integrated 1550 nm DFB-TWEAM having 100 GHz 3-dB bandwidth with 2 dB ripple. We evaluate its capabilities to enable two lanes 400 GbE client-side links for optical interconnects. © 2017 IEEE.

We experimentally evaluate high-speed intensity modulation/direct detection (IM/DD) transmissions with a 1.55-μ text broadband electro-Absorption modulated laser and pulse amplitude modulations (PAM). We demonstrate 80 Gb/s/ λ PAM-4 and 96 Gb/s/ λ PAM-8 transmissions with low-complexity digital equalizers at the receiver. Performance comparison with different types of equalizers are performed, including linear symbol-spaced feed-forward equalizer (FFE), fractional (half-symbol) spaced FFE and decision feedback equalizer (DFE), with different tap number. It is found that for both cases, a 6-Tap symbol-spaced FFE is sufficient to achieve a stable performance with bit-error-rate below the 7% overhead hard decision forward error correction (7%-OH HD-FEC) threshold over a 4 km standard single mode fiber link. Practical considerations including comparison between adaptive and static equalizer implementation and tolerable fiber chromatic dispersion are discussed.

Gigabit free-space transmissions are experimentally demonstrated with a quantum cascaded laser (QCL) emitting at mid-wavelength infrared of 4.65 μm, and a commercial infrared photovoltaic detector. The QCL operating at room temperature is directly modulated using on–off keying and, for the first time, to the best of our knowledge, four- and eight-level pulse amplitude modulations (PAM-4, PAM-8). By applying pre- and post-digital equalizations, we achieve up to 3 Gbit/s line data rate in all three modulation configurations with a bit error rate performance of below the 7% overhead hard decision forward error correction limit of 3.8 × 10−3. The proposed transmission link also shows a stable operational performance in the lab environment.

To tackle the continued increase in capacity demand in optical access, self-homodyne coherent detection (SHCD) is being considered as a candidate solution, utilizing a spatial division multiplexing (SDM) in a low-crosstalk multicore fiber (MCF) [1, 2] and high-order modulation formats such as quadrature phase-shift keying (QPSK), 16-levels quadrature amplitude modulation (16-QAM) or even 64-levels QAM (64-QAM) . The latest imposes strict requirements on both the performance of digital signal processing (DSP) and the spectral purity of the free running lasers [4–6]. Although a SHCD scheme relaxes the laser linewidth requirements, eliminates the need for a carrier frequency tracking, the necessity for a carrier phase estimation remains if high-order constellations are considered. The reasons for that are (1) residual phase noise in a system and (2) a closer location of the constellation points in the complex plane, which leads to a significantly lower phase noise tolerance as compared to QPSK or even 16-QAM signals . Therefore, it is of crucial importance to propose a low-complexity (and low-latency) phase estimation algorithms for these scenarios as well, especially because traditional algorithms cannot be directly employed for 64QAM. During this presentation,

first, we overview our recent activities on designing and experimental validation of a high-performance and a low-complexity DSP block, focusing on carrier phase recovery (CPR) suitable for circular and square m-QAM constellations. For more information about the efficient CPR schemes for m-QAM, we recommend the

OFC2017 and ACP2017 proceedings.

Second, after the efficient CPR scheme is explained, we provide an insight into a solution of SDM-based optical access network where self-homodyne coherent detection (SHCD) scheme with our novel low-complexity CPR is implemented for a 28-Gbaud 16-QAM and 64- QAM transmission in downstream links with a goal to improve their spectral efficiency and to increase a phase-noise tolerance, which paves the way towards higher cost-efficiency. The proposed architecture together with the experimental demonstration is given in

CLEO2018 conference proceedings

We demonstrate the significant impact of crosstalk between add and drop ports at optical nodes on energy-efficiency per Hertz in WDM networks employing 32/64 Gbd DP-16QAM transmission, especially when the isolation is reduced to 30dB.