Phase-Based DAS: Advancing Distributed Acoustic Sensing for Real-Time Infrastructure Intelligence

Distributed Acoustic Sensing (DAS) has become a critical tool for monitoring and protecting infrastructure at scale. As deployments have grown, so too have the expectations to move beyond simple detection toward higher accuracy, better event identification, and fewer false alarms. This evolution has driven the adoption of phase‑based DAS, such as VIAVI’s patented true‑phase DAS, as the next step in DAS technology.

Rather than replacing existing DAS approaches, phase‑based DAS builds on them and enables more precise, quantitative insight that drives new applications and more confident operational decisions.

Phase‑Based DAS: An Evolution of DAS Technology

Phase‑based DAS, also referred to as phase‑sensitive DAS, phase DAS, or quantitative DAS, is not a new category of sensing, but an advancement of traditional DAS techniques. While conventional intensity‑based DAS systems detect changes in backscattered light intensity, phase‑based DAS measures the optical phase of the Rayleigh backscatter signal.

VIAVI true-Phase DAS enables high-fidelity sensing through optical phase recovery, using patented techniques like phase-stepping interferometry.

Recovering optical phase allows the system to move beyond simple disturbance detection and into quantitative measurement. With true‑phase DAS, physical parameters such as strain, vibration amplitude, frequency, and true acoustic power can be measured directly. The result is higher‑fidelity data that is more repeatable, objective, and comparable across deployments.

Why Phase‑Based Measurement Matters

The ability to recover optical phase delivers several important advantages:

• Higher sensitivity, enabling detection of smaller acoustic signals
• Wider usable frequency range, supporting more diverse event types
• Improved performance in noisy environments, with greater resilience to background noise
• More accurate localization and tracking, especially for moving or closely-spaced events

These characteristics translate directly into higher detection rates with fewer false alarms—a critical requirement for large‑scale, always‑on monitoring systems.

That said, intensity‑based DAS still plays an important role. In higher‑amplitude environments, or in use cases driven purely by detection rather than identification or classification, intensity‑based systems can be an effective and economical choice. Phase‑based DAS extends the DAS toolbox, enabling applications where accuracy, discrimination, and confidence are paramount.

Expanding DAS Applications Across Industries

The number of DAS applications has increased significantly in recent years. What began primarily as a pipeline and perimeter monitoring tool now spans a broad set of infrastructure use cases, including:

• Threat identification for data centers and telecom network cables
• Security surveillance of borders and sensitive perimeters
• Monitoring of critical infrastructure, from pipeline leak detection to undersea power cable wear
• Cable health, dynamic strain to assess fatigue

New business models are also emerging. Traditional telecommunications operators are increasingly exploring ways to leverage existing fiber assets for additional revenue streams. In some deployments, DAS data from telecom fibers running near municipal infrastructure—such as water pipelines—is used to detect and locate leaks, with the sensing data sold to nearby utilities. This approach allows operators to extract new value from fiber already in the ground.

At the same time, classic OTDR, temperature, strain, and acoustic monitoring can now be performed within a single fiber core, supporting both dark and in‑service fiber links. This convergence simplifies deployment while expanding insight.

True‑Phase DAS Meets AI and ML at the Network Edge

As DAS systems become more capable, the volume and richness of data they generate increases dramatically. Extracting full value from that data requires advanced analytics—particularly AI and machine learning.

Modern phase‑based DAS systems increasingly utilize AI and ML at some level, usually through a centralized AI/ML model. While allowing for extra insight to be gained it does require the backhaul of the raw DAS data, placing additional demands on network links (capacity, latency, up-time) and essentially working in a post processing model which can introduce delays in event identification and alarm generation, not to mention extending system tuning and commissioning times.

Implementing AI/ML directly at the network edge can therefore bring advantages, using on‑board GPU processing and embedded models that operate in real time to increase autonomy and reduce delay. Trained on decades of historical data, VIAVI models enable automatic event detection, classification, and tracking without reliance on centralized processing or manual intervention.

Utilizing AI/ML at the network edge also reduces data transport requirements, eases connectivity constraints, and minimizes commissioning time by up to 50%. Allowing models can be updated and adapted automatically, with continuous improvement supported through private or VIAVI‑managed cloud ecosystems.

The Combined Value of True‑Phase DAS and AI/ML

The combination of true‑phase DAS and edge‑based AI/ML delivers tangible operational benefits:

• Better detection of events against background noise
• Improved discrimination of multiple events in close proximity
• More accurate classification and labeling of events
• Faster alarm generation and response times
• Higher locational accuracy and reliable tracking of moving targets
• Automatic adaptation to seasonal and environmental changes
• Fewer false alarms, lower operational overhead, and more successful call‑outs

Together, these capabilities move DAS beyond simple awareness toward actionable, real‑time intelligence.

Bringing It All Together

By combining on‑device AI and Machine Learning alongside a patented true‑phase DAS technique, the VIAVI DAS solution with FTH‑DAS delivers the enhanced measurement accuracy and interpretation needed for real‑time infrastructure intelligence. The result is higher confidence, faster response, and a scalable foundation for next‑generation sensing applications across critical infrastructure.

Douglas Clague is currently solutions marketing manager for fiber optic field solutions at VIAVI. Doug has over 20 years of experience in test and measurement with a primary focus on fiber optics and cable technologies, supporting the telecommunications industry. Prior to VIAVI, Doug held positions as manufacturing engineer, solutions engineer and business development manager. Doug has participated on numerous industry panels around fiber and cable technology trends. He attended Brunel University in London and graduated with an honors degree in electrical and electronic engineering.