⚡ electrical engineering

Multi-Sensor Integrated Sensing and Communication for Critical Infrastructure Protection

This paper proposes a distributed multi-sensor Integrated Sensing and Communication (MS-ISAC) architecture using cooperative passive sniffers to enhance critical infrastructure protection against drone threats, demonstrating its superior coverage and geometric precision compared to traditional quasi-monostatic approaches.

Original authors: Reiner Thomä, Gerd Sommerkorn, Christian Schneider, Thomas Dallmann

Published 2026-06-30

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Reiner Thomä, Gerd Sommerkorn, Christian Schneider, Thomas Dallmann

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

The Big Picture: Protecting Sensitive Sites from Sneaky Drones

Imagine you have a very valuable garden (like a solar farm or a power plant) that you need to protect from unauthorized drones. These drones are tricky: they don't turn on their own radios, so you can't "hear" them coming with a standard radio scanner. You need a way to see them without them knowing they are being watched.

The paper proposes a new way to do this using the existing cell phone network (4G/5G) that is already everywhere. Instead of building expensive, dedicated radar towers, the authors suggest turning the cell network into a giant, invisible radar system.

The Problem with the "Old" Way (The Single Flashlight)

Currently, many people think about doing this by placing a special listening device right next to the cell phone tower.

The Analogy: Imagine the cell tower is a giant flashlight shining a beam of light. The listening device is a camera sitting right next to the flashlight.
The Flaw: If you stand right next to a blinding flashlight, it's hard to see anything else in the beam because the light is too bright and overwhelms your eyes (this is called "self-interference"). Also, a camera right next to the light can only tell you how far away an object is, but it struggles to tell you exactly where it is or how fast it's moving sideways. It's like trying to judge the speed of a car driving straight at you versus one driving past you; it's much harder when you are standing right in front of the headlights.

The New Idea: The "Distributed Sniffers" (The Team of Spies)

The authors propose a different setup. Instead of one listener next to the tower, they suggest placing three or more small listening devices (called "sniffers") right around the perimeter of the protected garden.

The Analogy: Imagine the cell tower is still the flashlight, but instead of one camera next to it, you have three spies standing at different corners of the garden.
How it works:
1. The cell tower shines its signal (the light) over the garden.
2. The drones reflect this signal (like a mirror reflecting light).
3. The three spies catch the reflected signal from their different spots.
4. They send this data back to the tower to be combined.

Why This Team Approach is Better

The paper argues that this "Multi-Sensor" approach has three main superpowers compared to the single listener:

1. Better Hearing (Signal Strength)

The Analogy: If you are standing far from a speaker, the music is quiet. But if you have three friends standing right next to the stage, they hear the music loud and clear.
The Science: Because the sniffers are close to the garden (the target) and the tower is far away, the signal doesn't have to travel as far to get to the listener. This means the "echo" from the drone is much stronger and clearer than if the listener were stuck next to the tower.

2. No Blinding Light (Less Interference)

The Analogy: The spies are far enough away from the flashlight that the direct beam doesn't blind their cameras. They can clearly see the reflection off the drone without the glare of the source.
The Science: This removes the "self-interference" problem. The system can easily distinguish between the signal coming straight from the tower and the signal bouncing off the drone.

3. Perfect 3D Tracking (Triangulation)

The Analogy: If you have one person looking at a bird, they can guess how far away it is, but they might be wrong about which way it's going. If you have three people at different angles, they can all point at the bird, and where their fingers meet is exactly where the bird is.
The Science: By combining the data from three different angles, the system can calculate the drone's exact 3D position and speed without needing complex, expensive equipment to guess the direction. It uses math (called "multilateration") to pinpoint the target perfectly.

The "Passive" Secret

A key part of this system is that the sniffers are passive.

The Analogy: The sniffers don't shout out to the drone to see if it's there (which would reveal their presence). Instead, they just listen to the cell tower's normal conversation with other phones. They use the existing "noise" of the network to catch the reflection of the drone.
The Benefit: This means the system doesn't need to change the cell towers or the phones. It just adds these small listening devices near the protected site. It's like adding a motion sensor to a door without having to rebuild the house.

The Conclusion

The paper concludes that for protecting specific, critical areas (like a solar field or a power plant), this "Distributed Multi-Sensor" approach is much better than the traditional "Single Sensor" approach.

It sees further and clearer.
It tracks movement more accurately.
It uses the existing cell network without needing expensive hardware upgrades to the towers.

The authors suggest that while this might not cover an entire country at once, it is the perfect solution for creating a secure "fence" around important infrastructure, and it could be expanded later by using more cell sites as listening posts.

Technical Summary: Multi-Sensor Integrated Sensing and Communication for Critical Infrastructure Protection

Problem Statement
The proliferation of Unmanned Aerial Vehicles (UAVs) presents a latent threat to Critical Infrastructure (CI), including energy, transportation, and defense facilities. While remote-controlled drones can be detected via radio surveillance, malicious "passive" drones that transmit no active signals require radar-based detection. However, existing government air-traffic control systems lack the ubiquitous, scalable, and cost-effective coverage necessary to monitor lower airspace (U-Space) comprehensively. Dedicated radar systems face limitations in frequency spectrum availability, regulatory hurdles, and economic feasibility. While Integrated Sensing and Communication (ISAC) offers a solution by reusing mobile network infrastructure, current proposals often rely on a quasi-monostatic architecture. In this setup, a single "sniffer" sensor is located near the Base Station (BS). This approach suffers from significant drawbacks: strong self-interference (direct feed-through), poor Direction of Arrival (DoA) estimation due to limited beamforming resolution, an inability to measure cross-range Doppler shifts, and a lack of geometric degrees of freedom for reliable 3D localization.

Methodology
The authors propose a distributed Multi-Sensor ISAC (MS-ISAC) architecture, specifically termed Multi-Sensor using User Equipment (MSUE)-ISAC. Instead of placing a sensor near the BS, this approach deploys three or more dedicated passive "sniffer" sensors (acting as standard User Equipment) at strategic locations surrounding the Protected Site (PS).

Architecture: The sniffers connect to a distant illumination Base Station (gNB) via standard Downlink (DL) and Uplink (UL) channels. The gNB illuminates the area using existing 4G/5G signaling (PDSCH data streams and reference signals).
Sensing Principle: The system operates on the Cooperative Passive Coherent Location (CPCL) principle. The sniffers receive the DL signal and the target echoes. They perform local processing to estimate the excess Time of Flight (eToF) and excess Doppler (eDoppler) relative to the direct Line-of-Sight (LOS) path from the BS.
Data Fusion: Local estimates are transmitted via the UL to the gNB (or edge cloud), where data from all sniffers is fused. This multilateration process allows for the calculation of a full 3D dynamic target state vector (position and velocity) without requiring DoA estimation.
Signal Processing: To handle the dynamic range requirements of sensing, the paper proposes a "turbo CPCL" retrieval procedure. This involves a two-step equalization process: first using standard Channel State Information (CSI) for initial signal recovery, and second using the estimated communication data to refine the correlation reference, thereby maximizing Signal-to-Noise Ratio (SNR).

Key Contributions

Architectural Shift: The paper advocates for a distributed multi-static architecture over the quasi-monostatic approach, demonstrating that placing sniffers near the target area (rather than the BS) fundamentally improves performance.
Performance Analysis: The authors provide a comparative analysis of coverage and Geometric Dilution of Precision (GDoP). They show that the distributed approach avoids the self-interference issues of monostatic setups and enables full 3D localization through multilateration.
Resource Efficiency: The proposal utilizes existing 4G/5G radio signaling schemes and standard UE hardware, requiring no modifications to the BS hardware or internal Radio Access Network (RAN) protocols.
Scalability Concepts: The paper outlines a hierarchical expansion strategy, suggesting the adaptation of "umbrella cells" and Cooperative Multi-Point (CoMP) networks (MSRU-ISAC) to extend coverage from local hotspots to wide-area monitoring.

Results
Simulations conducted on a realistic scenario (protecting a 500m solar field with a BS located ~800m away) yield the following findings:

Received Power: The multistatic configuration consistently yields higher received power ( $P_r$ ) within the protected area compared to the monostatic case. This is attributed to the shorter target-to-receiver distances ( $R_{s \to r}$ ) in the distributed setup, which reduces the $R^4$ path loss penalty inherent in monostatic radar equations.
Range Ratio: The geometric advantage is quantified by the range ratio $a = R_{t \to s} / R_{s \to r}$ . In the distributed setup, $a > 1$ , providing a gain term of $+20 \log_{10}(a)$ in the received power equation.
Localization Accuracy: The distributed geometry significantly reduces the Horizontal Dilution of Precision (HDoP). By fusing measurements from multiple angles, the system achieves lower estimation variance for position and speed, adhering to the Cramér-Rao Lower Bound (CRLB).
Frequency Bands: The results suggest that lower operating frequencies (FR1 low/mid-band) are viable for ISAC. While they offer lower bandwidth, the higher SNR achieved through optimal sniffer placement compensates for the reduced resolution, and lower antenna directivity relaxes the need for precise target beam search.

Significance and Claims
The paper claims that the distributed MSUE-ISAC approach offers a pragmatic, scalable solution for CI protection that outperforms conventional quasi-monostatic ISAC in coverage and localization accuracy. The significance lies in its ability to:

Provide a "Sensing as a Service" model where Mobile Network Operators (MNOs) can administer sensing quality without altering existing infrastructure.
Overcome the "stealth" problem of passive drones by leveraging the diversity gain of multiple sensors and the high SNR of optimized bistatic geometry.
Enable comprehensive situational awareness by aggregating compressed data from local ISAC clusters, potentially fusing them with external sources.

The authors remain modest regarding wide-area coverage, acknowledging that dedicated sniffers are best suited for defined, finite areas. They propose that for broader coverage, the concept should evolve into a hierarchical network utilizing existing small-cell sites and CoMP architectures, rather than claiming the current distributed sniffer setup alone solves all wide-area monitoring challenges.