DAS: Advancing real-time intelligence for critical infrastructure

Distributed Acoustic Sensing (DAS) is rapidly emerging as a critical tool for real-time monitoring, security, and protection of large-scale infrastructure assets. As DAS deployments expand, expectations are moving beyond basic event detection toward greater precision, better event classification, and fewer false alarms. This progression has accelerated the adoption of phase-based DAS, as the next advancement in technology. Rather than replacing conventional DAS approaches, phase-based techniques complement and enhance existing capabilities by delivering more accurate, quantitative insights that enable new use cases and support more confident, data-driven operational decisions.

Phase-based DAS: Driving greater precision in infrastructure monitoring

Phase-based DAS—also referred to as phase-sensitive DAS, phase DAS, or quantitative DAS—is not a new sensing category, but a significant evolution of traditional DAS techniques. While conventional intensity-based DAS systems rely on detecting variations in backscattered light intensity, phase-based DAS measures the optical phase of the Rayleigh backscatter signal. By recovering the optical phase, the system moves beyond simple disturbance detection and into true quantitative measurement.

• The ability to recover optical phase provides several important performance advantages, including:
• Higher sensitivity, enabling detection of weaker and more subtle acoustic signals
• Broader usable frequency response, supporting a wider range of event types and dynamic behaviors
• Improved performance in noisy environments, with increased resilience to background and operational noise
• More accurate localisation and tracking, particularly for moving targets and closely spaced disturbances

Together, these capabilities deliver enhanced detection performance, lower false alarm rates, and greater operational reliability—key requirements for large-scale, continuously operating monitoring systems.

Enabling wider DAS adoption across industries

The range of DAS applications has expanded significantly in recent years. What began primarily as a solution for pipeline monitoring and perimeter security has evolved into a comprehensive infrastructure intelligence platform supporting a wide variety of use cases, including:

• Threat detection for data centres and telecommunications infrastructure, including fiber routes and critical communication links
• Security monitoring of borders and sensitive perimeters, enabling early identification of intrusion activity
• Monitoring of critical infrastructure, from pipeline leak detection to structural health assessment of subsea power cables
• Cable condition monitoring and dynamic strain analysis, supporting fatigue assessment and long-term asset integrity management

Alongside this expansion, new business models are emerging. Operators are increasingly exploring opportunities to monetise existing fibre assets by enabling additional sensing services. In some deployments, DAS data collected from telecom fibres running adjacent to municipal infrastructure—such as water pipelines—is used to detect and localise leaks, with the resulting insights provided to utility operators. This approach unlocks new value from fibre already deployed in the ground.

At the same time, multiple sensing modalities—including OTDR, temperature, strain, and acoustic monitoring—can now be integrated and performed within a single fiber core, supporting both dark and in-service fiber links. This convergence simplifies deployment architectures while significantly expanding the depth and breadth of actionable insight.

DAS meets AI and ML at the network edge

As DAS systems continue to evolve in capability, the volume, velocity, and richness of data they generate increase significantly. Modern phase-based DAS systems increasingly incorporate AI/ML techniques, typically through centralised processing architectures. While this approach can deliver valuable deeper insight, it relies on backhauling raw DAS data to centralised compute resources. This places additional demands on network infrastructure in terms of capacity, latency, and availability, and often results in a post-processing workflow that can introduce delays in event detection, alarm generation, and overall response time. It can also extend system tuning and commissioning cycles.

Deploying AI/ML directly at the network edge addresses these challenges. By leveraging onboard GPU processing and embedded inference models, edge-based intelligence enables real-time analysis, increasing system autonomy while minimising latency. Edge-based AI/ML also reduces data transport requirements, eases network connectivity constraints, and can reduce commissioning time by up to 50 per cent.

The combination of VIAVI’s True Phase DAS and edge-based AI/ML delivers significant and measurable operational advantages, including:

• Enhanced event detection sensitivity, even in high-background and operational noise environments
• Superior event discrimination, enabling accurate separation of closely spaced or overlapping events
• Improved event classification and labeling, supporting deeper situational awareness and operational insight
• Faster alarm generation and response, reducing overall time to action
• Greater localisation accuracy and reliable target tracking, including dynamic or moving assets
• Adaptive system performance that continuously adjusts to seasonal and environmental changes
• Reduced false alarm rates, lower operator workload, and higher confidence in actionable alerts

Integrated intelligence

By integrating on-device AI and machine learning with patented True-Phase DAS technology, the VIAVI DAS solution with FTH DAS sets a new benchmark for real-time infrastructure intelligence. The solution enhances sensing precision and edge-based interpretation, transforming complex distributed sensing data into accurate, actionable insights in real time as events unfold.

This integrated approach enables operators to detect, analyse, and respond to infrastructure changes with greater speed and accuracy—improving situational awareness and operational confidence. The result is faster decision-making, reduced response times, and a scalable sensing architecture purpose-built for next-generation critical infrastructure monitoring and analytics.