Kazutaka Hara
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Research Interests
Kazutaka Hara has a total of 29 co-authors. Here are the top co-authors by number of shared publications:
- Jun-Ichi Kani
NTT, Inc
8 shared publications - Tomoaki Yoshida
NTT Corporation
7 shared publications - Ryo Koma
NTT Access Network Service Systems Laboratories, NTT Corporation, Yokosuka-shi, Kanagawa, Japan
5 shared publications - Ryo Igarashi
NTT Access Network Service Systems Laboratories, NTT Corporation, 1-1 Hikari-no-oka, Yokosuka-Shi, Kanagawa, 239-0847, Japan
3 shared publications - Takuya Kanai
NTT Innovative Devices Corporation, Atsugi, Japan
3 shared publications - Yasunari Tanaka
Access Network Service Systems Laboratories, NTT, Inc., Kanagawa, Japan
3 shared publications
Latest Publications
- Open Access
- 16 October 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Photonics Journal
- Vol. 17 (6) , 1-10
Fast QoT Estimation Based on Probability Density Calculation of Electric Field Waveform
For dynamic optical path provisioning in optical networks, real-time path design based on fast quality-of-transmission (QoT) estimation is highly desirable. We propose an electric field feature propagation based QoT estimation technique using a new feature extraction method. The proposed feature extraction method can reduce the calculation load by a factor of 10000 compared with the conventional approach, enabling comprehensive datasets for various networks to be created in realistic time. This study confirms that the proposed QoT estimation method can accurately estimate the bit error rate (BER) for different combinations of input power, span length, and received optical power.
- 16 October 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Photonics Journal
- Vol. 17 (6) , 1-10
Open AccessFast QoT Estimation Based on Probability Density Calculation of Electric Field Waveform
For dynamic optical path provisioning in optical networks, real-time path design based on fast quality-of-transmission (QoT) estimation is highly desirable. We propose an electric field feature propagation based QoT estimation technique using a new feature extraction method. The proposed feature extraction method can reduce the calculation load by a factor of 10000 compared with the conventional approach, enabling comprehensive datasets for various networks to be created in realistic time. This study confirms that the proposed QoT estimation method can accurately estimate the bit error rate (BER) for different combinations of input power, span length, and received optical power.
- 11 August 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in Journal of Lightwave Technology
- Vol. 43 (19) , 9137-9147
Fiber Parameter-Independent Zero-Dispersion Wavelength Estimation Method and Its Feasibility With Reduced Quantization Bits and Sampling Rates
We propose a method for estimating the zero-dispersion wavelength (ZDW) of an optical fiber independently of fiber parameters and for tuning the emission wavelength for the ZDW to suppress waveform distortion in high-speed intensity modulation and direct detection (IM-DD) systems. By using two emission wavelengths and unique notch frequencies caused by chromatic dispersion (CD) calculated from the frequency spectra (FSs) of each received signal, the proposed method can estimate the ZDW independent of fiber parameters such as fiber length and dispersion slope. The ZDW can be estimated by using the emission wavelength on either the shorter-wavelength side or the longer-wavelength side of the ZDW. To evaluate the feasibility of the method, we estimate the ZDW of 60-km fiber using 50-Gbps pseudorandom binary sequence 211−1 PAM4 signals at 1300-nm and 1305-nm emission wavelengths. The obtained estimation accuracy of −0.78 nm is sufficiently good when using an 8-bit analogue-to-digital converter operating at a sampling rate of 200 GS/s. We also demonstrate that results similar to the estimated ZDW are obtained even when the number of quantization bits is reduced from 8 to 4 and the sampling rate is reduced from 200 to 50 GS/s. These results show that the functions of the proposed method can be implemented in optical transceivers. On the basis of this configuration, we also achieve penalty-free and equalizer-free transmission by using an emission wavelength tuned to the estimated ZDW. Furthermore, we discuss the challenges associated with implementing the method in practical applications. We show that the BtB FSs required for estimating the ZDW can be replaced with simulated data instead of experimental data, that the method can be applied even at the minimum receiver sensitivity, and that there is no CD penalty when the notch frequency does not appear below the baud rate. Finally, we modify the procedure of the method to be applicable when the notch frequency does not appear within the value of the baud rate. Since this method can suppress waveform distortion due to CD independently of baud rate, it is applicable to higher speed transmission systems with a larger capacity of 100 Gbps or more.
- 11 August 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in Journal of Lightwave Technology
- Vol. 43 (19) , 9137-9147
Fiber Parameter-Independent Zero-Dispersion Wavelength Estimation Method and Its Feasibility With Reduced Quantization Bits and Sampling Rates
We propose a method for estimating the zero-dispersion wavelength (ZDW) of an optical fiber independently of fiber parameters and for tuning the emission wavelength for the ZDW to suppress waveform distortion in high-speed intensity modulation and direct detection (IM-DD) systems. By using two emission wavelengths and unique notch frequencies caused by chromatic dispersion (CD) calculated from the frequency spectra (FSs) of each received signal, the proposed method can estimate the ZDW independent of fiber parameters such as fiber length and dispersion slope. The ZDW can be estimated by using the emission wavelength on either the shorter-wavelength side or the longer-wavelength side of the ZDW. To evaluate the feasibility of the method, we estimate the ZDW of 60-km fiber using 50-Gbps pseudorandom binary sequence 211−1 PAM4 signals at 1300-nm and 1305-nm emission wavelengths. The obtained estimation accuracy of −0.78 nm is sufficiently good when using an 8-bit analogue-to-digital converter operating at a sampling rate of 200 GS/s. We also demonstrate that results similar to the estimated ZDW are obtained even when the number of quantization bits is reduced from 8 to 4 and the sampling rate is reduced from 200 to 50 GS/s. These results show that the functions of the proposed method can be implemented in optical transceivers. On the basis of this configuration, we also achieve penalty-free and equalizer-free transmission by using an emission wavelength tuned to the estimated ZDW. Furthermore, we discuss the challenges associated with implementing the method in practical applications. We show that the BtB FSs required for estimating the ZDW can be replaced with simulated data instead of experimental data, that the method can be applied even at the minimum receiver sensitivity, and that there is no CD penalty when the notch frequency does not appear below the baud rate. Finally, we modify the procedure of the method to be applicable when the notch frequency does not appear within the value of the baud rate. Since this method can suppress waveform distortion due to CD independently of baud rate, it is applicable to higher speed transmission systems with a larger capacity of 100 Gbps or more.
- 21 February 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in Journal of Lightwave Technology
- Vol. 42 (10) , 3736-3746
Burst-Mode Multi-Level-CPFSK Signal Detection Using DD-LMS Based Adaptive Equalizer for High-Speed TDM-PON Upstream
This paper evaluates the link budget for burst-mode single wavelength 50-Gbps 4-level continuous phase frequency shift keying (CPFSK) signal reception in time domain multiplexing (TDM) based digital coherent optical access networks. The key advances of the burst-mode CPFSK transmitter and receiver are as follows; simple EA-DFB-SOA based transmitter that suppresses the intensity modulation component generated by direct modulation of the DFB-LD, a burst-mode DSP that reduces convergence time of the adaptive finite impulse response (FIR) filter-based equalizer by using the tap coefficient handover method with simple decision-directed least mean square (DD-LMS) algorithm. The results of burst-mode experiments show the proposal offers high link budget of 40.7 dB with 32.8 ns short preamble. In addition, we calculate and compare the BER characteristics for higher baud rate 4-level CPFSK signals (37.5 Gbaud, and 50 Gbaud) and higher level CPFSK signals (8- and 16- level CPFSK signals) detection as a feasibility study for beyond 50-Gbps optical access networks.
- 21 February 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in Journal of Lightwave Technology
- Vol. 42 (10) , 3736-3746
Burst-Mode Multi-Level-CPFSK Signal Detection Using DD-LMS Based Adaptive Equalizer for High-Speed TDM-PON Upstream
This paper evaluates the link budget for burst-mode single wavelength 50-Gbps 4-level continuous phase frequency shift keying (CPFSK) signal reception in time domain multiplexing (TDM) based digital coherent optical access networks. The key advances of the burst-mode CPFSK transmitter and receiver are as follows; simple EA-DFB-SOA based transmitter that suppresses the intensity modulation component generated by direct modulation of the DFB-LD, a burst-mode DSP that reduces convergence time of the adaptive finite impulse response (FIR) filter-based equalizer by using the tap coefficient handover method with simple decision-directed least mean square (DD-LMS) algorithm. The results of burst-mode experiments show the proposal offers high link budget of 40.7 dB with 32.8 ns short preamble. In addition, we calculate and compare the BER characteristics for higher baud rate 4-level CPFSK signals (37.5 Gbaud, and 50 Gbaud) and higher level CPFSK signals (8- and 16- level CPFSK signals) detection as a feasibility study for beyond 50-Gbps optical access networks.
- 30 October 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in Journal of Lightwave Technology
- Vol. 42 (6) , 1844-1852
Zero-Dispersion Wavelength Estimation and Optimization Method for Penalty-Free and Equalizer-Free 53-Gbps/λ IM-DD Transmission Over 100 Km
We propose the method to estimate the zero-dispersion wavelength (ZDW) and to optimize the emission wavelength for high-speed intensity modulation and direct detection (IM-DD) transmission. The feature of this method can estimate the ZDW by using the unique notch frequency caused by chromatic dispersion from the frequency spectrum of the received signal, and reduce chromatic dispersion by tuning the emission wavelength to the estimation value of the ZDW in a configuration that using only general transmitters/receivers without additional dedicated equipment. With the proposed method, the limitation of transmission distance can be relaxed in high-speed IM-DD transmission without using a dispersion compensation device or an equalizer. We experimentally demonstrated that the notch frequency was observed at a condition of the pseudorandom binary sequence (PRBS) 2 15 -1 and 2 31 -1 non-return-to-zero (NRZ) signals at 53.125 Gbps for a total 80 and 100 km, and the ZDW was estimated with fast Fourier transformation sizes of 256, 512, and 1024. The results indicated that the proposed method can be estimated with an accuracy of about ±1 nm from the true value of the ZDW. In addition, to confirm the feasibility of the proposed method, we experimentally demonstrated that PRBS 2 15 -1 and 2 31 -1 NRZ signals at 53.125 Gbps were transmitted over 100 km by tuning the emission wavelength to the estimated ZDW value. For the ZDW with a true value of 1312.20 nm, the chromatic dispersion penalty was up to 1.5 dB under the conditions of a 53.125 Gbps NRZ signal with an emission wavelength of 1303.73 nm and a transmission distance of 100 km. Penalty-free and equalizer-free transmission was demonstrated by applying this method.
- 30 October 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in Journal of Lightwave Technology
- Vol. 42 (6) , 1844-1852
Zero-Dispersion Wavelength Estimation and Optimization Method for Penalty-Free and Equalizer-Free 53-Gbps/λ IM-DD Transmission Over 100 Km
We propose the method to estimate the zero-dispersion wavelength (ZDW) and to optimize the emission wavelength for high-speed intensity modulation and direct detection (IM-DD) transmission. The feature of this method can estimate the ZDW by using the unique notch frequency caused by chromatic dispersion from the frequency spectrum of the received signal, and reduce chromatic dispersion by tuning the emission wavelength to the estimation value of the ZDW in a configuration that using only general transmitters/receivers without additional dedicated equipment. With the proposed method, the limitation of transmission distance can be relaxed in high-speed IM-DD transmission without using a dispersion compensation device or an equalizer. We experimentally demonstrated that the notch frequency was observed at a condition of the pseudorandom binary sequence (PRBS) 2 15 -1 and 2 31 -1 non-return-to-zero (NRZ) signals at 53.125 Gbps for a total 80 and 100 km, and the ZDW was estimated with fast Fourier transformation sizes of 256, 512, and 1024. The results indicated that the proposed method can be estimated with an accuracy of about ±1 nm from the true value of the ZDW. In addition, to confirm the feasibility of the proposed method, we experimentally demonstrated that PRBS 2 15 -1 and 2 31 -1 NRZ signals at 53.125 Gbps were transmitted over 100 km by tuning the emission wavelength to the estimated ZDW value. For the ZDW with a true value of 1312.20 nm, the chromatic dispersion penalty was up to 1.5 dB under the conditions of a 53.125 Gbps NRZ signal with an emission wavelength of 1303.73 nm and a transmission distance of 100 km. Penalty-free and equalizer-free transmission was demonstrated by applying this method.
- 18 October 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in Journal of Lightwave Technology
- Vol. 41 (1) , 159-168
Novel Data-Aided Carrier Frequency Offset Compensation Methods Using Asymmetric-Shape Constellations for Burst-Mode Coherent Reception
This paper introduces a novel data-aided carrier frequency offset (CFO) estimation and compensation method for m-array phase shift keying (m-PSK) signals in burst-mode digital coherent reception. Its key advances are as follows; the use of newly designed training symbols (TS) that exhibit asymmetric distributed constellations after differential decoding and a vector summation based feed-forward CFO estimation algorithm. These offer short response time and strong robustness to training symbols (TS) timing detection error with wide CFO estimation range of ±B/2, where B is signal baudrate. We evaluate the feasibility of the proposal by simulations of and experiments on 12.5-Gbaud SP-QPSK signals that target optical budget enhancement of 25-Gbps/λ-class PONs to realize long-reach and high splitting ratio optical access. The 100-km burst-mode transmission experiments confirm the proposal offers high sensitivity detection of better than −43 dBm with much wider CFO range of above ±10 GHz.
- 18 October 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in Journal of Lightwave Technology
- Vol. 41 (1) , 159-168
Novel Data-Aided Carrier Frequency Offset Compensation Methods Using Asymmetric-Shape Constellations for Burst-Mode Coherent Reception
This paper introduces a novel data-aided carrier frequency offset (CFO) estimation and compensation method for m-array phase shift keying (m-PSK) signals in burst-mode digital coherent reception. Its key advances are as follows; the use of newly designed training symbols (TS) that exhibit asymmetric distributed constellations after differential decoding and a vector summation based feed-forward CFO estimation algorithm. These offer short response time and strong robustness to training symbols (TS) timing detection error with wide CFO estimation range of ±B/2, where B is signal baudrate. We evaluate the feasibility of the proposal by simulations of and experiments on 12.5-Gbaud SP-QPSK signals that target optical budget enhancement of 25-Gbps/λ-class PONs to realize long-reach and high splitting ratio optical access. The 100-km burst-mode transmission experiments confirm the proposal offers high sensitivity detection of better than −43 dBm with much wider CFO range of above ±10 GHz.
- 30 June 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in Journal of Lightwave Technology
- Vol. 40 (17) , 5783-5792
Extraction of AMCC Signal Superposed by SOA-Integrated EA-DFB Laser for In-Service Monitoring in All-Photonics Network
We propose an in-line monitoring method that enables fully passive extraction of the control signal for end-to-end optical direct connection on the All-Photonics Network (APN), which is one of the elements of the Innovative Optical and Wireless Network (IOWN). In our approach, the auxiliary management and control channel (AMCC) signal superposed on the client signal as the control signal is extracted by a fully passive configuration without any optical to electrical conversion. By installing the proposed configuration in the Photonic Gateway (Ph-GW), which is an edge optical node of the APN located at the boundary between the access network and the local full-mesh network, the AMCC signal can be extracted in the middle of a route even when the users are communicating via direct optical connection. The proposed configuration enables in-service user terminal monitoring for end-to-end optical direct connection, which is the core concept of the APN. We conducted transmission experiments on the area management of cyber physical system (CPS) in which we assumed that an intensity modulation-direct detection (IM-DD) system with low power consumption and economy is used. We then determined the target transmission distance and bandwidth and conducted 60-km optical transmission in the local full-mesh network. The results demonstrated that a loss budget of 36 dB for the client signal was achieved. We also investigated extending the transmission distance of the client signal by adjusting the branch ratio of the passive extractor so that the optical power of the AMCC signal reached the minimum receiver sensitivity. As a result, we were able to extend the transmission distance to 80 km.
- 30 June 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in Journal of Lightwave Technology
- Vol. 40 (17) , 5783-5792
Extraction of AMCC Signal Superposed by SOA-Integrated EA-DFB Laser for In-Service Monitoring in All-Photonics Network
We propose an in-line monitoring method that enables fully passive extraction of the control signal for end-to-end optical direct connection on the All-Photonics Network (APN), which is one of the elements of the Innovative Optical and Wireless Network (IOWN). In our approach, the auxiliary management and control channel (AMCC) signal superposed on the client signal as the control signal is extracted by a fully passive configuration without any optical to electrical conversion. By installing the proposed configuration in the Photonic Gateway (Ph-GW), which is an edge optical node of the APN located at the boundary between the access network and the local full-mesh network, the AMCC signal can be extracted in the middle of a route even when the users are communicating via direct optical connection. The proposed configuration enables in-service user terminal monitoring for end-to-end optical direct connection, which is the core concept of the APN. We conducted transmission experiments on the area management of cyber physical system (CPS) in which we assumed that an intensity modulation-direct detection (IM-DD) system with low power consumption and economy is used. We then determined the target transmission distance and bandwidth and conducted 60-km optical transmission in the local full-mesh network. The results demonstrated that a loss budget of 36 dB for the client signal was achieved. We also investigated extending the transmission distance of the client signal by adjusting the branch ratio of the passive extractor so that the optical power of the AMCC signal reached the minimum receiver sensitivity. As a result, we were able to extend the transmission distance to 80 km.
- Open Access
- 19 August 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in Journal of Lightwave Technology
- Vol. 39 (21) , 6814-6823
Network Design for Bus-Type Optical Access Using Distributed Raman Amplification With Asymmetric Power Splitter
This paper proposes loss budget improvement of the bus-type optical access network (NW) with cost-effective non-repeater configuration. To achieve drastic improvements, we propose two approaches; (i) application of a low noise distributed Raman amplification (DRA) technique to upstream signal transmission (ii) utilization of asymmetric power splitters at each drop point to enhance Raman gain and to suppress transmission loss of far branches. We optimize the asymmetric power splitter drop ratio to maximize the number of accommodated users. Then, we clarify the number of accommodated users and improvements based on DRA through calculations that examine the inter-drop distance and total branch number. Experiments evaluate the bit error rate performance of the proposal. We confirm that measured values correspond to the simulation results. Based on the simulation and the experiment, we clarify the optimal network configuration for the user distribution. It is found that bus-type NWs support more accommodated users with shorter total fiber length than star-type NWs for user range examined. The potential advantage of bus-type NWs is revealed.
- 19 August 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in Journal of Lightwave Technology
- Vol. 39 (21) , 6814-6823
Open AccessNetwork Design for Bus-Type Optical Access Using Distributed Raman Amplification With Asymmetric Power Splitter
This paper proposes loss budget improvement of the bus-type optical access network (NW) with cost-effective non-repeater configuration. To achieve drastic improvements, we propose two approaches; (i) application of a low noise distributed Raman amplification (DRA) technique to upstream signal transmission (ii) utilization of asymmetric power splitters at each drop point to enhance Raman gain and to suppress transmission loss of far branches. We optimize the asymmetric power splitter drop ratio to maximize the number of accommodated users. Then, we clarify the number of accommodated users and improvements based on DRA through calculations that examine the inter-drop distance and total branch number. Experiments evaluate the bit error rate performance of the proposal. We confirm that measured values correspond to the simulation results. Based on the simulation and the experiment, we clarify the optimal network configuration for the user distribution. It is found that bus-type NWs support more accommodated users with shorter total fiber length than star-type NWs for user range examined. The potential advantage of bus-type NWs is revealed.
- 1 November 2020
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEICE Transactions on Communications
- Vol. E103.B (11) , 1249-1256
Design for Long-Reach Coexisting PON Considering Subscriber Distribution with Wavelength Selective Asymmetrical Splitters
A long-reach coexisting PON system (1G/10G-EPON, video, and TWDM-PON) that uses the Wavelength Selective Asymmetrical optical SPlitter (WS-ASP) without any active devices like optical amplifiers is proposed. The proposal can take into account the subscriber distribution in an access network and provide specific services in specific areas by varying the splitting ratios and the branch structure in the optical splitter. Simulations confirm the key features of WS-ASP, its novel process for deriving the splitting ratios and greater transmission distance than possible with symmetrical splitters. Experiments on a prototype system demonstrate how wavelengths can be assigned to specific areas and optical link budget enhancement. For 1G-EPON systems, the prototype system with splitting ratio of 60 % attains the optical link budget enhancement of 4.2 dB compared with conventional symmetrical optical splitters. The same prototype offers the optical link budget enhancement of 4.0 dB at the bit rate of 10G-EPON systems. The values measured in the experiment agree well with the simulation results with respect to the transmission distance.
- 1 November 2020
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEICE Transactions on Communications
- Vol. E103.B (11) , 1249-1256
Design for Long-Reach Coexisting PON Considering Subscriber Distribution with Wavelength Selective Asymmetrical Splitters
A long-reach coexisting PON system (1G/10G-EPON, video, and TWDM-PON) that uses the Wavelength Selective Asymmetrical optical SPlitter (WS-ASP) without any active devices like optical amplifiers is proposed. The proposal can take into account the subscriber distribution in an access network and provide specific services in specific areas by varying the splitting ratios and the branch structure in the optical splitter. Simulations confirm the key features of WS-ASP, its novel process for deriving the splitting ratios and greater transmission distance than possible with symmetrical splitters. Experiments on a prototype system demonstrate how wavelengths can be assigned to specific areas and optical link budget enhancement. For 1G-EPON systems, the prototype system with splitting ratio of 60 % attains the optical link budget enhancement of 4.2 dB compared with conventional symmetrical optical splitters. The same prototype offers the optical link budget enhancement of 4.0 dB at the bit rate of 10G-EPON systems. The values measured in the experiment agree well with the simulation results with respect to the transmission distance.
- 1 January 2011
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEICE Transactions on Communications
- Vol. E94-B (7) , 1845-1852
Ultra Fast Response AC-Coupled Burst-Mode Receiver with High Sensitivity and Wide Dynamic Range for 10G-EPON System
- 1 January 2011
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEICE Transactions on Communications
- Vol. E94-B (7) , 1845-1852
Ultra Fast Response AC-Coupled Burst-Mode Receiver with High Sensitivity and Wide Dynamic Range for 10G-EPON System
- 1 January 2011
- Published by Optica Publishing Group
- p. OTuB5
Simple and Scalable WDM/TDMA-PON using Spectral Slicing and Forward Error Correction
We describe a simple and scalable WDM/TDMA PON system that uses spectral slicing and FEC in burst-mode upstream transmissions. The sensitivity was improved to -31.5 dB at a sliced bandwidth of 400 GHz with FEC.
- 1 January 2011
- Published by Optica Publishing Group
- p. OTuB5
Simple and Scalable WDM/TDMA-PON using Spectral Slicing and Forward Error Correction
We describe a simple and scalable WDM/TDMA PON system that uses spectral slicing and FEC in burst-mode upstream transmissions. The sensitivity was improved to -31.5 dB at a sliced bandwidth of 400 GHz with FEC.