NKNoriaki Kaneda

Noriaki Kaneda

*This is an algorithmically generated author record based on publicly available information.

Noriaki Kaneda has a total of 34 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Enabling Technologies for Efficient 100-Gb/s Wireless Communication Systems [From the Guest Editor’s Desk]

    by Christian Carlowitz, Noriaki Kaneda

    Recently, major advances in analog front ends for ultrahigh-speed wireless communication systems targeting data rates toward 100 Gb/s have been demonstrated at high carrier frequencies between 100 and 300 GHz. To deliver such performance in a complete system to the end user, various functionalities, such as analog-to-digital converters (ADCs) and high-speed digital signal processors, need to be integrated as well as very high-bandwidth baseband components. Upscaling these components in conjunction with the classic homodyne transceiver front end still comes with major challenges to be addressed, most notably high relative and absolute bandwidth, high frequencies at technological limits as well as low efficiency in terms of power consumption, and system size. Consequently, it is attractive to reconsider central decisions in system architecture to achieve ultrahigh data rates with an efficient implementation. This affects, most notably, the tradeoff between low carrier frequencies with narrow available bandwidth and extremely high spectral efficiency and high frequencies with extremely wide available bandwidth and low spectral efficiency.

  • Artificial Neural Network Assisted Probabilistic and Geometric Shaping for Flexible Rate High-Speed PONs

    by Shuang Yao, Amitkumar Mahadevan, Yannick Lefevre, Noriaki Kaneda, Vincent Houtsma, Doutje van Veen

    In this paper, we employ artificial neural networks (ANNs) to optimize joint probabilistic shaping (PS) and geometric shaping (GS) for a realistic 50G IM/DD passive optical network (PON) link. Apart from being able to find a generalized mutual information (GMI)-maximizing modulation for channel conditions unseen at the training phase, the compatibility of the ANN training with Monte Carlo simulation also enables us to use a more complicated channel model that more closely resembles a real PON system where fiber dispersion, bandwidth limitation and digital signal processing (DSP) are present. The forward error correction (FEC) requirement that must be satisfied in an actual implementation is imposed on the learned modulation by including a normalized GMI (NGMI) penalty term in the loss function. The proposed scheme is demonstrated with simulations. Results show that the ANN can achieve similar performance compared to a case-by-case optimization while also being capable of generalizing to a wide range of received optical power (ROP) from −30 dBm to −18 dBm and/or a broad range of fiber distance from 0 km to 20 km. About 0.1-bits/symbol GMI improvement is attained compared to uniform modulation.


  • Fixed-Point Analysis and FPGA Implementation of Deep Neural Network Based Equalizers for High-Speed PON

    by Noriaki Kaneda, Chun-Yen Chuang, Ziyi Zhu, Amitkumar Mahadevan, Bob Farah, Keren Bergman, Doutje Van Veen, Vincent Houtsma

    Adeep neural network based equalizer is proposed to mitigate the intersymbol interference observed in next generation high speed passive optical network (PON) links. The DNN based equalizer is shown to outperform the best known conventional equalizer, the maximum likelihood sequence estimator (MLSE) both in back-back and through fiber experiments. To reduce the hardware complexity of DNN based equalizer for PON systems, we investigate the use of embedded parallelization within a DNN structure having multiple symbol outputs from one DNN. We further investigate using a classification output stage with cross entropy cost to perform joint decision on multiple symbol outputs and demonstrated that the sensitivity gain of such scheme over regression output. To understand the complexity of hardware implementation, the fixed-point DNN based equalizers are developed and implemented in FPGA. The impact of fixed-point resolution on the receiver sensitivity and hardware resource utilization in FPGA implementation is analyzed and reported in detail. We show that a reduction of over 40% in LUTs (look up table) utilization is possible by reducing the DNN's weight resolution from 8-bit to 4-bit while incurring a small penalty in receiver sensitivity.


  • FLCS-PON – A 100 Gbit/s Flexible Passive Optical Network: Concepts and Field Trial

    by Robert Borkowski, Michael Straub, Yanni Ou, Yannick Lefevre, Željko L. Jelić, Wouter Lanneer, Noriaki Kaneda, Amitkumar Mahadevan, Volker Huckstadt, Dora van Veen, et al.

    We demonstrate concepts and results of a field trial for a flexible-rate passive optical network (FLCS-PON), which delivers bitrates up to 100 Gbit/s and allows for adaptations in the transmission method to match the users’ channel conditions and optimize throughput. FLCS-PON builds on top of the hardware ecosystem that will be developed for ITU-T 50 Gbit/s PON and employs three new ingredients: optical network unit (ONU) grouping, flexible modulation format, and flexible forward error correction (FEC) code rate. Together, these techniques take advantage of the optical distribution network (ODN) statistics to realize a system capable of more than twofold throughput increase compared to the upcoming 50 Gbit/s PON, but still able to support a full array of deployed fiber edge cases, which are problematic for legacy PONs. In this paper we explain the concepts behind enabling techniques of FLCS-PON. We then report on a field trial over a deployed fiber infrastructure, using a system consisting of one FLCS-PON OLT and two ONUs. We report both pre- and post-forward-error-correction (post-FEC) performance of our system, demonstrating achievable net bitrate over an operator's fiber infrastructure. We realize a downlink transmission at double the speed of ITU-T 50 Gbit/s PON for ONUs exhibiting lower optical path loss (OPL), while simultaneously continue to support ONUs at high OPLs. We additionally realize a record-high 31.5 dB loss budget for 100 Gbit/s transmission using a direct-detection ONU with an optical preamplifier.


  • Hybrid-Integrated Wavelength and Reflectivity Tunable III–V/Silicon Transmitter

    by Guilhem de Valicourt, Chia-Ming Chang, Michael S. Eggleston, Argishti Melikyan, Cristian Bolle, Noriaki Kaneda, Mark P. Earnshaw, Young-Kai Chen, Anaelle Maho, Romain Brenot, et al.

    We demonstrate a novel wavelength-tunable hybrid external cavity silicon-based transmitter with integrated reflectivity tunable mirror. First, designs of a novel hybrid laser based on reflective symmetric and asymmetric Mach-Zehnder interferometer based mirrors are studied with single and double ring intra-cavity filters respectively. Asymmetric configuration is used in order to increase the tuning range of a single ring laser by more than 90 %. Furthermore, adjustment of the mirror reflectivity allows for control of laser characteristics such as threshold current, wall-plug efficiency, output power and linewidth. Finally, an ultra-compact hybrid tunable III-V/Si transmitter with more than 28 nm tuning range is then demonstrated based on such hybrid cavity design and one ring modulator. We successfully achieved modulation up to 40 Gbit/s and transmission over 20 (100) km at 25 (10) Gbit/s.


  • Real-time Polarization Division Multiplexed Coherent Optical OFDM Receiver at 9.83-GS/s for 28.6-Gb/s Single-band Intradyne Detection

    by Noriaki Kaneda, Timo Pfau, Stephen Corteselli, Qi Yang, Andreas Leven, Young-Kai Chen

    Real-time polarization division multiplexed coherent optical OFDM receiver is demonstrated to receive 28.6-Gb/s data per optical wavelength in single intradyne detection. QPSK modulated sub-carriers are detected using 9.83-GS/s ADCs and DSP implemented on three FPGAs.