skip to main content

Title: Performance studies of evolutionary transfer learning for end-to-end QoT estimation in multi-domain optical networks [Invited]

This paper proposes an evolutionary transfer learning approach (Evol-TL) for scalable quality-of-transmission (QoT) estimation in multi-domain elastic optical networks (MD-EONs). Evol-TL exploits a broker-based MD-EON architecture that enables cooperative learning between the broker plane (end-to-end) and domain-level (local) machine learning functions while securing the autonomy of each domain. We designed a genetic algorithm to optimize the neural network architectures and the sets of weights to be transferred between the source and destination tasks. We evaluated the performance of Evol-TL with three case studies considering the QoT estimation task for lightpaths with (i) different path lengths (in terms of the numbers of fiber links traversed), (ii) different modulation formats, and (iii) different device conditions (emulated by introducing different levels of wavelength-specific attenuation to the amplifiers). The results show that the proposed approach can reduce the average amount of required training data by up to13×<#comment/>while achieving an estimation accuracy above 95%.

; ; ;
Award ID(s):
Publication Date:
Journal Name:
Journal of Optical Communications and Networking
Page Range or eLocation-ID:
Article No. B1
1943-0620; JOCNBB
Optical Society of America
Sponsoring Org:
National Science Foundation
More Like this
  1. Abstract

    Objective.High-density electromyography (HD-EMG) decomposition algorithms are used to identify individual motor unit (MU) spike trains, which collectively constitute the neural code of movements, to predict motor intent. This approach has advanced from offline to online decomposition, from isometric to dynamic contractions, leading to a wide range of neural-machine interface applications. However, current online methods need offline retraining when applied to the same muscle on a different day or to a different person, which limits their applications in a real-time neural-machine interface. We proposed a deep convolutional neural network (CNN) framework for neural drive estimation, which takes in frames of HD-EMG signals as input, extracts general spatiotemporal properties of MU action potentials, and outputs the number of spikes in each frame. The deep CNN can generalize its application without retraining to HD-EMG data recorded in separate sessions, muscles, or participants.Approach.We recorded HD-EMG signals from the vastus medialis and vastus lateralis muscles from five participants while they performed isometric contractions during two sessions separated by ∼20 months. We identified MU spike trains from HD-EMG signals using a convolutive blind source separation (BSS) method, and then used the cumulative spike train (CST) of these MUs and the HD-EMG signals to train andmore »validate the deep CNN.Main results.On average, the correlation coefficients between CST from the BSS and that from deep CNN were0.983±0.006for leave-one-out across-sessions-and-muscles validation and0.989±0.002for leave-one-out across-participants validation. When trained with more than four datasets, the performance of deep CNN saturated at0.984±0.001for cross validations across muscles, sessions, and participants.Significance.We can conclude that the deep CNN is generalizable across the aforementioned conditions without retraining. We could potentially generate a robust deep CNN to estimate neural drive to muscles for neural-machine interfaces.

    « less
  2. Efficient resource allocation and management can enhance the capacity of an optical backbone network. In this context, spectrum retuning via hitless defragmentation has been presented for elastic optical networks to enhance efficient spectrum accommodation while reducing the unused fragmented spaces in the spectrum. However, the quality of service committed in a service level agreement may be affected due to spectrum retuning. In particular, for transmission beyond the conventional C band, the presence of inter-channel stimulated Raman scattering can severely degrade the quality of the signal during defragmentation. To conquer this problem, this paper proposes, for the first time to our knowledge, a signal-quality-aware proactive defragmentation scheme for theC+Lband system. The proposed scheme prioritizes the minimization of the fragmentation index and quality of transmission (QoT) maintenance for two different defragmentation algorithms, namely, nonlinear-impairment (NLI)-aware defragmentation (NAD) and NLI-unaware defragmentation (NUD). We leverage machine learning techniques for QoT estimation of ongoing lightpaths during spectrum retuning. The optical signal-to-noise ratio of a lightpath is predicted for each choice of spectrum retuning, which helps to monitor the effect of defragmentation on the quality of ongoing lightpaths (in terms of assigned modulation format). Numerical results show that, compared to amore »baseline algorithm (NUD), the proposed NAD algorithm provides up to 15% capacity increment for smaller networks such as BT-UK, while for larger networks such as the 24-node USA network, a capacity benefit of 23% is achieved in terms of the number of served demands at 1% blocking.

    « less
  3. This paper presents a cognitive flexible-bandwidth optical interconnect architecture for datacom networks. The proposed architecture leverages silicon photonic reconfigurable all-to-all switch fabrics interconnecting top-of-rack switches arranged in a Hyper-X-like topology with a cognitive control plane for optical reconfiguration by self-supervised learning. The proposed approach makes use of a clustering algorithm to learn the traffic patterns from historical traces. We developed a heuristic algorithm for optimizing the intra-pod connectivity graph for each identified traffic pattern. Further, to mitigate the scalability issue induced by frequent clustering operations, we parameterized the learned traffic patterns by a support vector machine classifier. The classifier is trained offline by self-labeled data to enable the classification of traffic matrices during online operations, thereby facilitating cognitive reconfiguration decision making. The simulation results show that compared with a static all-to-all interconnection, the proposed approach can improve the throughput by up to1.62×<#comment/>while reducing the end-to-end packet latency and flow completion time by up to3.84×<#comment/>and20×<#comment/>, respectively.

  4. Abstract

    We perform path-integral molecular dynamics (PIMD), ring-polymer MD (RPMD), and classical MD simulations of H$$_2$$2O and D$$_2$$2O using the q-TIP4P/F water model over a wide range of temperatures and pressures. The density$$\rho (T)$$ρ(T), isothermal compressibility$$\kappa _T(T)$$κT(T), and self-diffusion coefficientsD(T) of H$$_2$$2O and D$$_2$$2O are in excellent agreement with available experimental data; the isobaric heat capacity$$C_P(T)$$CP(T)obtained from PIMD and MD simulations agree qualitatively well with the experiments. Some of these thermodynamic properties exhibit anomalous maxima upon isobaric cooling, consistent with recent experiments and with the possibility that H$$_2$$2O and D$$_2$$2O exhibit a liquid-liquid critical point (LLCP) at low temperatures and positive pressures. The data from PIMD/MD for H$$_2$$2O and D$$_2$$2O can be fitted remarkably well using the Two-State-Equation-of-State (TSEOS). Using the TSEOS, we estimate that the LLCP for q-TIP4P/F H$$_2$$2O, from PIMD simulations, is located at$$P_c = 167 \pm 9$$Pc=167±9 MPa,$$T_c = 159 \pm 6$$Tc=159±6 K, and$$\rho _c = 1.02 \pm 0.01$$ρc=1.02±0.01 g/cm$$^3$$3. Isotope substitution effects are important; the LLCP location in q-TIP4P/F D$$_2$$2O is estimated to be$$P_c = 176 \pm 4$$Pc=176±4 MPa,$$T_c = 177 \pm 2$$Tc=177±2 K, and$$\rho _c = 1.13 \pm 0.01$$ρc=1.13±0.01 g/cm$$^3$$3. Interestingly, for the water model studied, differences in the LLCP location from PIMD and MD simulations suggest that nuclear quantum effectsmore »(i.e., atoms delocalization) play an important role in the thermodynamics of water around the LLCP (from the MD simulations of q-TIP4P/F water,$$P_c = 203 \pm 4$$Pc=203±4 MPa,$$T_c = 175 \pm 2$$Tc=175±2 K, and$$\rho _c = 1.03 \pm 0.01$$ρc=1.03±0.01 g/cm$$^3$$3). Overall, our results strongly support the LLPT scenario to explain water anomalous behavior, independently of the fundamental differences between classical MD and PIMD techniques. The reported values of$$T_c$$Tcfor D$$_2$$2O and, particularly, H$$_2$$2O suggest that improved water models are needed for the study of supercooled water.

    « less
  5. Abstract

    Ultra-pure NaI(Tl) crystals are the key element for a model-independent verification of the long standing DAMA result and a powerful means to search for the annual modulation signature of dark matter interactions. The SABRE collaboration has been developing cutting-edge techniques for the reduction of intrinsic backgrounds over several years. In this paper we report the first characterization of a 3.4 kg crystal, named NaI-33, performed in an underground passive shielding setup at LNGS. NaI-33 has a record low$$^{39}$$39K contamination of 4.3 ± 0.2 ppb as determined by mass spectrometry. We measured a light yield of 11.1 ± 0.2 photoelectrons/keV and an energy resolution of 13.2% (FWHM/E) at 59.5 keV. We evaluated the activities of$$^{226}$$226Ra and$$^{228}$$228Th inside the crystal to be$$5.9\pm 0.6~\upmu $$5.9±0.6μBq/kg and$$1.6\pm 0.3~\upmu $$1.6±0.3μBq/kg, respectively, which would indicate a contamination from$$^{238}$$238U and$$^{232}$$232Th at part-per-trillion level. We measured an activity of 0.51 ± 0.02 mBq/kg due to$$^{210}$$210Pb out of equilibrium and a$$\alpha $$αquenching factor of 0.63 ± 0.01 at 5304 keV. We illustrate the analyses techniques developed to reject electronic noise in the lower part of the energy spectrum. A cut-based strategy and a multivariate approach indicated a rate, attributed to the intrinsic radioactivity of the crystal, of$$\sim $$1 count/day/kg/keV in the [5–20] keV region.