This PDF file contains the front matter associated with SPIE Proceedings Volume 9008, including the Title Page, Copyright information, Table of Contents, and Conference Committee listing.

The emergence of digital coherent optical transmission technologies is being eagerly awaited by the world. This enables us to develop spectrally-efficient transmission systems by means of polarization-division multiplexing and multilevelmodulation formats such as quadrature-phase-shift keying (QPSK) and higher-order quadrature-amplitude modulation (QAM). Thanks to recent rapid advances in the research and development of electronics, demodulation of such signals can be realized effectively by utilizing sophisticated digital signal processors (DSPs). Such digital coherent technologies have successfully been implemented in commercial systems. However, the transmission performance of photonic networks is limited by system impairments that include crosstalk and spectrum narrowing caused at reconfigurable optical add/drop multiplexers (ROADMs) and the nonlinearity of optical fibers. Current digital coherent technologies do not resolve these problems comprehensively necessitating further research. In this paper, we investigate the impacts of the system impairments through intensive computer simulations and show the maximum transmission distances of multilevel-modulation signals. Various transmission schemes for gridless networks including Nyquist wavelengthdivision- multiplexing (WDM) networks, which need digital coherent technologies, are evaluated. We also discuss DSP algorithms that suit photonic networks and permit digital coherent technologies to become more effective in realizing future networks.

The integration of ultra-wideband (UWB)-over- ber into passive optical network (PON) is of great interest as it bene ts the high bandwidth capability from optical network technologies and the high exibility from wireless network technologies. The later can only be done with a reasonable cost when a universal optical transmitter, which is capable of generation both UWB and PON signals, is available. Direct modulation of semiconductor laser was demonstrated to be suitable for high bit-rate PON systems, however the generation of UWB signals by this technique is still challenging. Using the chirp properties of directly modulated semiconductor lasers, UWB signals are generated. Di erent UWB signal waveforms and polarities are obtained. The received electrical spectra conform to the requirements of indoor UWB systems.

Advanced multi-level modulation is an attractive modulation technique for beyond 100 Gbps short reach optical transmission system. Above all, discrete multi-tone (DMT) technique and pulse amplitude modulation (PAM) technique are the strong candidates.

We compared the 100 Gbps transmission characteristics of DMT and PAM by simulation and experiment. The comparison was done by using same devices and only the digital signal processing was changed. We studied the transmission distance dependence for 0.5 to 40 km and the impact of the frequency responses of the optical devices. Finally we discuss the features of the both modulation techniques.

This paper describes recent progress in relation to the key optical devices for high-speed digital coherent transmission, namely coherent receivers and advanced-format modulators. Miniaturization and higher performance have been achieved on a silica-based planar lightwave circuit (PLC) platform and it has been integrated with other key materials.

Results and analysis of semiconductor optical amplifiers (SOA) are presented as applied to Photonic switching nodes in OPS/OBS future optical networks. Detailed characterization is provided to investigate physical constraints of optical power, gain and noise figure of SOAs. Two different lasers, one external cavity tunable laser and one DFB laser, verify that although the SOA gain is not significantly sensitive to input source a clear difference on the noise figure (NF) is observed. Another important result is that by limiting the average number of hops in the network accumulated ASE power from the amplifiers should not impair signal quality.

We demonstrate wavelength-tunable SFF transceivers operating at 1550 nm using a tunable VCSEL with a high contrast grating (HCG) as the output mirror. Tunable HCG VCSELs with a ~25 nm mechanical tuning range and over 2 mW output power were realized. Error-free operation of an optical link using directly-modulated tunable HCG VCSELs transmitting at 1.25 Gbps over 18 channels spaced by 100 GHz and transmitted over 20 km of single mode fiber is demonstrated, showing the suitability of the HCG tunable VCSEL as a low cost source for next generation DWDM communications systems in access networks and data centers.

The Internet is entering an era of cloud computing to provide more cost effective, eco-friendly and reliable services to consumer and business users. As a consequence, the nature of the Internet traffic has been fundamentally transformed from a pure packet-based pattern to today’s predominantly flow-based pattern. Cloud computing has also brought about an unprecedented growth in the Internet traffic. In this paper, a hybrid optical switch architecture is presented to deal with the flow-based Internet traffic, aiming to offer flexible and intelligent bandwidth on demand to improve fiber capacity utilization. The hybrid optical switch is capable of integrating IP into optical networks for cloud-based traffic with predictable performance, for which the delay performance of the electronic module in the hybrid optical switch architecture is evaluated through simulation.

This paper presents a complete design for an optical Internet router based on decomposing the steps required for IP packet forwarding. Implementations of hopcount decrement and header matching are integrated with a simulation-based approach to variable-length packet merging that avoids recirculation, resulting in an all-optical data plane. A method for IPv4 checksum computation is introduced, and this and previously designed components are extended from binary to higher-density (multiple bits per symbol) encodings. The implications of this design are considered, including the potential for chip-level and system integration, as well as the requirements of basic optical processing components.

Analog photonic link (APL) has been considered to be a promising technique due to the low insertion loss, broad bandwidth and immunity to electromagnetic interference. It is essential for many microwave systems, such as avionics, modern electronic warfare, and wireless communication systems. However, polarization effect, chromatic dispersion (CD), fiber Kerr effect and RF nonlinearity are four main problems in APL. All of them degrade the performance of the link. Therefore, APL needs to be optimized according to the different requirements in various applications. In this paper, we firstly establish a propagation model and provide the general expressions for the analog signal in photonic link based on coupled-mode theory and the small-signal analysis. Such model can describe the interaction of polarization effect, CD and nonlinearity. We also investigate the noise figure (NF) and spurious-free dynamic range (SFDR) in dispersive nonlinear link based on the proposed model. Subsequently, we review and introduce different compensation schemes for these impairments, such as CD compensation based on double sideband (DSB) modulation, and simultaneous compensation for CD and nonlinearity by employing a phase modulator (PM). After compensations, the SFDR of the link can be improved greatly. In addition, recent experimental results show that APL might be a supporting technique for the 4G or higher speed optical-wireless communication systems in near future.

We have experimentally demonstrated a new smart grid model which can control DC electric power flow autonomously among individual homes, by using an optical self-organized node with optical non-linear characteristics, and these homes are assumed to be installed by distributed power supplies, and electric power storage devices, and also supposed to be supplied partly by the commercial electric power grid utilities. An electric power network is composed of nodes and devises called Power Gate Unit (PGU). The nodes have optical nonlinearity for self-organizing informations about surplus or shortage of electric power as to individual homes. The PGU is a distributing unit of actual electric power based on above informations of power surplus or shortage at each home. The PGU at each home is electrically connected to both the onsite power supplies and household load such as a solar panel, a DC motor, and a storage battery as well as the commercial electric power grid utilities. In this work, we composed our experimental self-organized DC power grid with above components and supposed the supplied maximum power from the commercial electric power grid utilities to be limited to 5V-0.5A. In this network, information about surplus or shortage of electric power will propagate through the nodes. In the experiments, surplus electric current 0.4A at a particular node was distributed toward a PGU of another node suffering from shortage of electric current. We also confirmed in the experiments and simulations that even when signal propagation path was disconnected accidentally the network could recover an optimized path. The present smart grid system we have attained may be applied by optical fiber link in the near future because our essential components controlling PGU, i.e. the nodes are electro-optical hybrid which are easily applicable to fiber optical link so as to control electric power transmission line.

The dramatic growth of transmitted information in fiber optical networks is leading to a concern about the network latency for high-speed reliable services like financial transactions, telemedicine, virtual and augmented reality, surveillance, and other applications. In order to ensure effective latency engineering, the delay variability needs to be accurately monitored and measured, in order to control it. This paper in brief describes causes of latency in fiber optical metro networks. Several available latency reduction techniques and solutions are also discussed, namely concerning usage of different chromatic dispersion compensation methods, low-latency amplifiers, optical fibers as well as other network elements.

Current communication in optical networks presents a wide range of granularities, making it hard to use the optical spectrum efficiently under the WDM framework. In Spectrum Sliced Optical Networks, the WDM rigid frequency grid is replaced by a more flexible structure, in which the spectrum is organized in frequency slots, and each traffic flow is assigned to an appropriate set of contiguous slots. The classical Routing and Wavelength Assignment (RWA) problem is then replaced by a Routing and Spectrum assignment (RSA) problem. This paper addresses an iterativa approach to balance the network load during the routing decision in Spectrum-Elastic Optical Path Networks. We have built numerical examples to illustrate the performance of our routing approach. Comparisons to other routing techniques show that our approach mitigates the lightpath requests blocking probability.

Effective utilization of fiber capacity in optical communication networks is required to keep up with the increasing traffic demand. Precise optical frequency allocation among carriers is essential for improving the spectral efficiency to utilize the limited spectral resource. In this paper, we show a distributed optical multiplexing scheme, in which data signals are sequentially multiplexed by frequency-division multiplexing on a single-wavelength optical carrier using fiber frequency conversion with locally provided optical subcarrier signals. The scheme achieves dense packing of distributed multi-channel signals with precise frequency allocation using free-running lasers. Using the scheme we demonstrate a precise multiplexing of coherent-optical orthogonal frequency-division multiplexing and Nyquist wavelength-division multiplexing.

With the exponential growth of global internet traffic, it becomes important to increase spectral efficiency for optical transmission systems in order to meet the challenge of continuing capacity growth. Polarization-division multiplexing (PDM) and multi-level quadrature-amplitude modulation (QAM) such as 16QAM allow spectral efficiencies beyond 4 bits/s/Hz. However, for long-haul optical transmission systems, performance is penalized by amplified spontaneous emission (ASE) noise and nonlinearity interference. Moreover, the decreased Euclidean distance between the signal constellation points in high-level QAM modulation formats reduces tolerance to phase noise and distortions. Such reduction in tolerance causes extra performance degradation and limits the transmission distance. In order to counteract this penalty, coded modulation can be used. In coded modulation, part of the redundancy that is usually assigned to forward-error correction (FEC) is moved into symbol modulation to increase the minimum Euclidean distance of a received sequence of symbols. Demodulation is done by a multi-symbol soft-decoder using either maximum likelihood or maximum a posteriori algorithm. In recent publications, LDPC-based bit-interleaved coded-modulation (BICM) with iterative soft-demapping and decoding is used to approach the capacity of AWGN channel, hence increase achievable transmission distance. We review different multidimensional single-parity check (SPC)-based BICM schemes suitable for high spectral efficiency transoceanic transmission systems, and their experimental verification.

Digital signal processing (DSP) for high spectrum efficiency transmission system are investigated in both long-haul and short haul optical networks. For long-haul transmission, two different super-Nyquist WDM systems based on advanced post (receiver side) and pre (transmitter side) DSP are demonstrated and studied. A novel DSP scheme for this optical super-Nyquist filtering 9-QAM like signals based on multi-modulus equalization (MMEQ) without post filter are proposed and experimentally demonstrated, which directly recovers the Nyquist filtered QPSK to a 9-QAM like signal. This improved filtering tolerance and transmission performance are demonstrated in an 8-channel 112-Gb/s wavelengthdivision- multiplexing (WDM) experiment with a 25GHz-grid over 2640-km single-mode fiber (SMF). Alternatively, a novel digital super-Nyquist signal generation scheme is proposed to further suppress the Nyquist signal bandwidth and reduce the channel crosstalk without using optical pre-filtering and using. Only optical couplers are needed for super- Nyquist WDM multiplexing. Using this scheme, we successfully generate and transmit 10 channel 32-GBaud (128-Gb/s) PDM-9-QAM signals within 25-GHz grid over 2975-km at a net SE of 4 bit/s/Hz (after excluding the 20% soft-decision FEC overhead). We extend the DSP for short haul optical transmission networks by using high order QAMs. We propose and experimentally demonstrate a high speed CAP-64QAM system using direct modulation laser (DML) based on direct detection and digital equalizations. Decision-directed least mean squares (DD-LMS) are used to equalize the CAP- 64QAM. Using this scheme, we successfully generate and transmit up to a record 60-Gb/s CAP-64QAM over 20-km stand single-mode fiber (SSMF) based on the DML and direct detection.

Recently, exible grid transport has gained a lot of attentions due to its higher spectral and energy efficiencies in handling dynamic diverse granularity traffic. In this paper, we introduce exible grid transport architecture and address the dynamic routing, wavelength assignment, and spectrum allocation problem. We develop a novel homogeneous irreducible Markovian model to estimate the blocking of point-to-point connections and an analytical model using the reduced load approximation technique to estimate the blocking of multi-hop connections. The numerical results show that the proposed models provide reasonable estimation of blocking for connections with lower state-space complexity and quicker convergence time.

Next generation ultra-broadband elastic optical networks maximize system-bandwidth utilization based on flexible diverse traffic demands accommodation. Research on novel network architectures and system components such as datarate variable transponders and hybrid amplification schemes are needed to allow geographical context adjustability as well as capacity system upgradeability from Gbps to Tbps. In this paper we report experimental validations on deployed Brazilian optical fiber of a data-rate variable transmitter feasible for both metro and long-haul applications. The developed transmission system exhibits spectral efficiencies of 4.58, 6.25 and 3.41 b/s/Hz for 1400, 700 and 4900 km of optical transmission respectively.

Ultrafast optical spectrum monitoring is one of the most challenging tasks in observing ultrafast phenomena, such as the spectroscopy, dynamic observation of the laser cavity, and spectral encoded imaging systems. However, conventional method such as optical spectrum analyzer (OSA) spatially disperses the spectrum, but the space-to-time mapping is realized by mechanical rotation of a grating, so are incapable of operating at high speed. Besides the spatial dispersion, temporal dispersion provided by dispersive fiber can also stretches the spectrum in time domain in an ultrafast manner, but is primarily confined in measuring short pulses. In view of these constraints, here we present a real-time spectrum analyzer called parametric spectro-temporal analyzer (PASTA), which is based on the time-lens focusing mechanism. It achieves a 100-MHz frame rate and can measure arbitrary waveforms. For the first time, we observe the dynamic spectrum of an ultrafast swept-source: Fourier domain mode-locked (FDML) laser, and the spectrum evolution of a laser cavity during its stabilizing process. In addition to the basic single-lens structure, the multi-lens configurations (e.g. telescope or wide-angle scope) will provide a versatile operating condition, which can zoom in to achieve 0.05-nm resolution and zoom out to achieve 10-nm observation range, namely 17 times zoom in/out ratio. In view of the goal of achieving spectrum analysis with fine accuracy, PASTA provides a promising path to study the real-time spectrum of some dynamic phenomena and non-repetitive events, with orders of magnitude enhancement in the frame rate over conventional OSAs.

We review recent progresses of quasi phase-matched (QPM) single-pumped fiber optical parametric amplifiers (FOPA) as a candidate for low noise amplifiers. We show how the QPM operation is realized in the gain fiber where highly nonlinear fibers and pump-phase shifters are alternately concatenated. Then, we show that the QPM operation makes gain of the FOPA flat. Finally, we demonstrate that the QPM-FOPA has noise figure less than 4.5 dB, which is lower than that of conventional EDFAs, and gain more than 20 dB with 1-dB gain bandwidth of 38 nm which covers C-band.

A Novel technique for reducing the OBI noise in optical OFDMA-PON uplink is presented. OFDMA is a multipleaccess/ multiplexing scheme that can provide multiplexing operation of user data streams onto the downlink sub-channels and uplink multiple access by means of dividing OFDM subcarriers as sub-channels. The main issue of high-speed, single-wavelength upstream OFDMA-PON arises from optical beating interference noise. Because the sub-channels are allocated dynamically to multiple access users over same nominal wavelength, it generates the optical beating interference among upstream signals.

In this paper, we proposed a novel scheme using self-homodyne balanced detection in the optical line terminal (OLT) to reduce OBI noise which is generated in the uplink transmission of OFDMA-PON system. When multiple OFDMA sub-channels over the same nominal wavelength are received at the same time in the proposed architecture, OBI noises can be removed using balanced detection. Using discrete multitone modulation (DMT) to generate real valued OFDM signals, the proposed technique is verified through experimental demonstration.

Orthogonal frequency division multiplexing (OFDM) is a modulation technique which is now used in most new and emerging broadband wired and wireless communication systems because it is an effective solution to inter-symbol interference caused by a dispersive channel. Very recently a number of researchers have shown that OFDM is also a promising technology for optical communications. This paper gives a overview of OFDM for long-haul, metro/access and data center highlighting the aspects that are likely to be important in optical applications. To achieve good performance in optical systems OFDM must be adapted in various ways. The constraints imposed by optical channel are discussed and the new forms of optical OFDM which have been developed are outlined. The main drawbacks of OFDM are its high peak to average power ratio and its sensitivity to phase noise and frequency offset. The impairments that these cause are described and their implications for optical systems discussed.

Scientific curiosity to probe the nature of the universe is pushing boundaries of big data transport and computing for radio telescopes. MeerKAT, the South African precursor to Square Kilometre Array (SKA), has 64 antennae separated by up to 12 km. By 2018, each antenna will stream up to 160 Gbps over optical fibre to a central computing engine. The antennae digitizers require highly accurate clock signals distributed with high stability. This paper outlines requirements and key design aspects of the MeerKAT network with timing reference overlay. Fieldwork results are presented into the impact of birefringence and polarization fluctuations on clock stability.