Wataru Kohno is a Collaborator from NEC Japan and works with our Optical Sensing & Solutions team. Dr. Kohno earned his Ph.D. in Physics from Hokkaido University in Japan, where he built a strong foundation in the physics and engineering principles underlying optical communications.
He has also authored a number of research papers in condensed matter physics, where he investigated fundamental problems in electronic structures, correlated electron systems, and quantum materials, contributing to a deeper theoretical understanding of how microscopic physical principles give rise to novel material properties; a selection of these works can be found on his Google Scholar profile. With his background in fundamental theoretical condensed matter physics, his work focuses on distributed acoustic sensing (DAS) and fiber-optic communication technologies, where he develops methods to transform optical fibers into highly sensitive, passive sensors that detect vibration, movement, and pressure across vast geographic areas.
These innovations enable real-time situational awareness without requiring active electronics at the sensing points, making large-scale monitoring systems both scalable and unobtrusive. His research supports critical applications, including perimeter security, seismic monitoring, and infrastructure protection. His academic training continues to inform his applied research at NEC, where he bridges theoretical advances in photonics with real-world sensing challenges.
At NEC Laboratories America, he has contributed to multiple cutting-edge projects that expand the capabilities of distributed acoustic sensing. His recent work includes developing advanced vibrometry techniques, recognition systems that adapt fiber sensing for downstream applications, and AI-enhanced methods for real-time detection in critical infrastructure such as power grids. By combining deep expertise in optics with practical engineering approaches, his research is helping create intelligent sensing platforms that can deliver reliable, real-time monitoring solutions for global security and infrastructure needs.
Posts
Knowledge distillation establishes a learning paradigm that leverages both data supervision and teacher guidance. However, determining the optimal balance between learning from data and learning from the teacher is challenging, as some samples may be noisy while others are subject to teacher uncertainty. This motivates the need for adaptively balancing data and teacher supervision. We propose Beta-weighted Knowledge Distillation (Beta-KD), an uncertainty-aware distillation framework that adaptively modulates how much the student relies on teacher guidance. Specifically, we formulate teacher–student learning from a unified Bayesian perspective and interpret teacher supervision as a Gibbs prior over student activations. This yields a closed-form, uncertainty-aware weighting mechanism and supports arbitrary distillation objectives and their combinations. Extensive experiments on multimodal VQA benchmarks demonstrate that distilling student Vision-Language Models from a large teacher VLM consistently improves performance. The results show that Beta-KD outperforms existing knowledge distillation methods.
Distributed Fiber-Optic Sensing (DFOS) is a promising technique for large-scale acoustic monitoring. However, its wide variation in installation environments and sensor characteristics causes spatial heterogeneity. This heterogeneity makes it difficult to collect representative training data. It also degrades the generalization ability of learning-based models, such as fine-tuning methods, under a limited amount of training data. To address this, we formulate Sound Event Classification (SEC) as data assimilation in an embedding space. Instead of training models, we infer sound event classes by combining pretrained audio embeddings with simulated DFOS signals. Simulated DFOS signals are generated by applying various frequency responses and noise patterns to microphone data, which allows for diverse prior modeling of DFOS conditions. Our method achieves out-of-domain (OOD) robust classification without requiring model training. The proposed method achieved accuracy improvements of 6.42, 14.11, and 3.47 percentage points compared with conventional zero-shot and two types of fine-tune methods, respectively. By employing the simulator in the framework of data assimilation, the proposed method also enables precise estimation of physical parameters from observed DFOS signals.
While distributed acoustic sensing based on Rayleigh backscattering has been widely studied and deployed across various use cases, its adaptation in existing telecom infrastructure is fundamentally limited by inline optical amplifiers which block the backscattering light. Recently, vibration sensing based on forward transmission has emerged as a promising alternative, offering wider frequency response, higher signal-to-noise ratio, longer sensing distance, and compatibility with inline amplifiers. However, forward-transmission approach faces the challenge in signal identification and reconstruction when multiple vibration events and interference occur along the fiber simultaneously. To address this issue, we propose an equivalent dual-sensor phased array model to the bidirectional forward-transmission fiber sensing, which reformulates the problems in fiber sensing into the tasks in array processing. We introduce two beamforming techniques, including generalized sidelobe canceller (GSC) and null steering beamformer (NSBF) to reconstruct the signal of interest while suppressing noise and interference. We demonstrate the effectiveness of dual-sensor beamforming over 100km fiber experiment without amplifiers, achieving over 20 dB suppression of stationary interference using GSC and 18 dB jamming rejection using NSBF. To our knowledge, this is the first demonstration of high-fidelity signal recovery in forward-transmission sensing under strong noise and interference. This work bridges the gap between fiber sensing and adaptive array processing, opening new directions for blind source separation and integrated communication-sensing systems.
We propose a hybrid remote and distributed vibration sensing system based on phase-sensitive optical time-domain reflectometry with collimator-based sensor heads. We demonstrate dual-laser vibrometers that detects nm-scale displacements of remote targets.
Deep neural network (DNN)-based models for environmental sound classification are not robust against a domain to which training data do not belong, that is, out-of-distribution or unseen data. To utilize pretrained models for the unseen domain, adaptation methods, such as finetuning and transfer learning, are used with rich computing resources, e.g., the graphical processing unit (GPU). However, it is becoming more difficult to keep up with research trends for those who have poor computing resources because state-of-the-art models are becoming computationally resource-intensive. In this paper, we propose a trainingless adaptation method for pretrained models for environmental sound classification. To introduce the trainingless adaptation method, we first propose an operation of recovering timefrequency-ish (TF-ish) structures in intermediate layers of DNN models. We then propose the trainingless frequency filtering method for domain adaptation, which is not a gradient-based optimization widely used. The experiments conducted using the ESC-50 dataset show that the proposed adaptation method improves the classification accuracy by 20.40 percentage points compared with the conventional method.
Recent advancements in unique acoustic sensing devices and large-scale audio recognition models have unlocked new possibilities for environmental sound monitoring and detection. However, applying pretrained models to non-conventional acoustic sensors results in performance degradation due to domain shifts, caused by differences in frequency response and noise characteristics from the original training data. In this study, we introduce a text-guided framework for generating new datasets to retrain models specifically for these non-conventional sensors efficiently. Our approach integrates text-conditional audio generative models with two additional steps: (1) selecting audio samples based on text input to match the desired sounds, and (2) applying domain transfer techniques using recorded impulse responses and background noise to simulate the characteristics of the sensors. We demonstrate this process by generating emulated signals for fiber-optic Distributed Acoustic Sensors (DAS), creating datasets similar to the recorded ESC-50 dataset. The generated signals are then used to train a classifier, which outperforms few-shot learning approaches in environmental sound classification.
Contrastive Language-Audio Pretraining (CLAP) models have demonstrated unprecedented performance in various acoustic signal recognition tasks. Fiber-optic-based acoustic recognition is one of the most important downstream tasks and plays a significant role in environmental sensing. Adapting CLAP for fiber-optic acoustic recognition has become an active research area. As a non-conventional acoustic sensor, fiberoptic acoustic recognition presents a challenging, domain-specific, low-shot deployment environment with significant domain shifts due to unique frequency response and noise characteristics. To address these challenges, we propose a support-based adaptation method, CLAP-S, which linearly interpolates a CLAP Adapter with the Support Set, leveraging both implicit knowledge through fine-tuning and explicit knowledge retrieved from memory for cross-domain generalization. Experimental results show that our method delivers competitive performance on both laboratory recorded fiber-optic ESC-50 datasets and a real-world fiber optic gunshot-firework dataset. Our research also provides valuable insights for other downstream acoustic recognition tasks.
We demonstrate fiber-optic acoustic data transmission using distributed acoustic sensing technology in an underwater environment. An acoustic orthogonal frequencydivisionmultiplexing (OFDM) signal transmitted through a fiber-optic cable deployed in a standard 40-meter-scale underwater testbed.
We present adaptive beamforming techniques to forward-transmission multi-event vibration sensing in environments with interference and jamming. Experimental validation over 100km fiber demonstrates significant improvements on signal reconstruction, noise reduction, and interference rejection from other locations.
Contrastive Language-Audio Pretraining (CLAP) models have demonstrated unprecedented performance in various acoustic signal recognition tasks. Fiber optic-based acoustic recognition is one of the most important downstream tasks and plays a significant role in environmental sensing. Adapting CLAP for fiber-optic acoustic recognition has become an active research area. As a non-conventional acoustic sensor, fiber-optic acoustic recognition presents a challenging, domain-specific, low-shot deployment environment with significant domain shifts due to unique frequency response and noise characteristics. To address these challenges, we propose a support-based adaptation method, CLAP-S, which linearly interpolates a CLAP Adapter with the Support Set, leveraging both implicit knowledge through fine-tuning and explicit knowledge retrieved from memory for cross-domain generalization. Experimental results show that our method delivers competitive performance on both laboratory-recorded fiber-optic ESC-50 datasets and a real-world fiber-optic gunshot-firework dataset. Our research also provides valuable insights for other downstream acoustic recognition tasks.
