Abstract
Forested regions such as Nigeria's Sambisa Forest conceal victims of displacement, kidnapping and escape from violence, making search and rescue (SAR) extremely difficult[27]Reporting on Boko Haram's use of the Sambisa Forest and the difficulty of aerial searches for abducted civilians. Source →. Conventional UAV sensors (EO/IR, thermal, LiDAR) generally cannot see under dense canopy[38]Studies showing optical and thermal sensors degrade severely under dense forest canopy. Source →. Military-grade foliage-penetrating radar (FPR) systems exist[33]DARPA/U.S. military foliage-penetrating radar (FOPEN/FORESTER) programs designed for large aircraft. Source →, but are large, expensive and restricted to defense use[30]Example of large UAV platforms capable of carrying traditional military radar payloads. Source →. We propose a first-principles design of a new, lightweight FPR sensor suite built from the ground up for small drones, with the goal of affordable, short-range victim detection and cueing in humanitarian SAR. Key features include small size and weight, low power, low cost, modular components, and detection-focused performance rather than full imaging. This approach rejects direct miniaturization of military FPR pods; instead it leverages modern low-cost radar chips[50]Compact radar work and automotive FMCW radar chips relevant to UAV-scale sensing. Source → and synthetic processing to achieve limited foliage penetration at UAV scale. Constraints (size, weight, power, affordability, simplicity) and ethics (civilian SAR only, privacy) are explicitly accounted for.
Introduction and Motivation
In northeast Nigeria, thousands of civilians have been displaced, abducted or forced into hiding by Boko Haram. Forest enclaves, including the Sambisa Forest, provide cover both for insurgents and for kidnapped or fleeing civilians[27]Reporting on Boko Haram's use of the Sambisa Forest and the difficulty of aerial searches for abducted civilians. Source →. Aerial drones and manned aircraft have been deployed in rescue efforts, but dense jungle canopy severely limits their effectiveness. Optical and thermal cameras fail under foliage occlusion[38]Studies showing optical and thermal sensors degrade severely under dense forest canopy. Source →, and even advanced vision-based systems struggle in random canopy environments. As a result, many victims remain unlocated for extended periods, prolonging humanitarian crises.
This creates an urgent need for new sensing approaches capable of providing information beneath forest canopy. While foliage-penetrating radar has demonstrated such capability[33]DARPA/U.S. military foliage-penetrating radar (FOPEN/FORESTER) programs designed for large aircraft. Source →, existing systems are inaccessible to civilian responders. This work focuses on addressing this gap through a fundamentally different design philosophy.
Limitations of Existing Systems
Existing foliage-penetrating radar systems are almost exclusively developed for military or intelligence applications. Programs such as airborne FOPEN and UHF SAR platforms[33]DARPA/U.S. military foliage-penetrating radar (FOPEN/FORESTER) programs designed for large aircraft. Source → are mounted on large aircraft or heavy unmanned aerial vehicles[30]Example of large UAV platforms capable of carrying traditional military radar payloads. Source →, often requiring antennas several meters in length, significant power budgets, and specialized processing hardware. These systems cost millions of dollars and are unavailable for humanitarian or civilian use.
Small commercial UAVs used by non-governmental organizations typically carry payloads of only a few kilograms and operate under strict power and cost constraints. Attempting to directly miniaturize military FPR systems for such platforms is impractical. Meanwhile, thermal and LiDAR-based systems, while useful in open terrain, provide limited effectiveness under dense foliage[38]Studies showing optical and thermal sensors degrade severely under dense forest canopy. Source →. As a result, there is currently no affordable, lightweight foliage-aware sensing solution available for civilian SAR operations.
First-Principles Design Vision
Rather than shrinking existing military hardware, this project adopts a first-principles approach. The core question is not how to recreate full radar imaging under canopy, but what minimal sensing capability is required to assist human rescuers.
The design goal is a cueing system rather than an imaging system. Such a system aims to detect anomalies or signatures consistent with human presence—such as motion, periodic micro-movements, or reflection patterns—without producing detailed images. This allows significant reductions in antenna size, transmit power, and processing complexity.
The proposed approach emphasizes modularity. Small sensor modules can be attached as add-ons to commercial drones, allowing flexible configuration and easy replacement. Multiple drones or sensor passes may be combined in software to synthesize useful information. This distributed sensing philosophy trades spatial resolution for accessibility, cost reduction, and deployment flexibility.
Technical Considerations and Constraints
The system must satisfy strict constraints to be viable for civilian SAR.
Payload size and weight must be compatible with drones under five kilograms total mass. Sensor packages should ideally weigh no more than a few hundred grams. Power consumption must be low enough to preserve useful flight times, requiring efficient radar architectures and duty-cycled operation.
Foliage penetration requires operation at relatively low radio frequencies compared to consumer radar systems[41]Radar engineering references explaining how lower frequencies penetrate foliage better but require larger antennas. Source →. Lower frequencies penetrate vegetation more effectively but demand larger antennas. The design therefore balances frequency selection with antenna miniaturization techniques and short-range operational assumptions.
Affordability is critical. The system must rely on commodity electronics, such as automotive radar chips or software-defined radio platforms[50]Compact radar work and automotive FMCW radar chips relevant to UAV-scale sensing. Source →, rather than custom military hardware. Modular construction allows incremental upgrades and repair without full system replacement.
Signal processing will prioritize anomaly detection and temporal change analysis rather than high-resolution imaging. Outputs are designed to be interpretable by human operators, such as flagged regions of interest, rather than autonomous decisions.
Intended Impact
If successful, this system would provide humanitarian responders in Nigeria with a new capability previously limited to state-level military actors. By enabling detection cues beneath dense forest canopy, the technology could significantly reduce search times and improve the likelihood of locating missing civilians.
The approach is explicitly humanitarian. It is intended to assist search and rescue teams in locating abducted or displaced civilians, not to support combat operations or surveillance. By making foliage-aware sensing accessible and affordable, this work aims to democratize life-saving technology and reduce reliance on foreign military assets.
Responsible Use and Ethics
This research is guided by a strict civilian-use mandate. The technology is intended solely for humanitarian search and rescue operations. It will not include features designed for targeting, tracking, or surveillance of populations.
All deployments would involve human operators and appropriate authorization. Data collection would be limited to mission-specific needs, and privacy-preserving design principles will be applied throughout development. Transparency and accountability are core design requirements to ensure ethical use.
Citations Reference
| Citation | Description | Source | |----------|-------------|--------| | [27] | News & investigative reporting on Boko Haram's use of Sambisa Forest | BBC News | | [30] | UAV platform documentation for military radar payloads | Lockheed Martin | | [33] | Defense radar program documentation (DARPA FORESTER/FOPEN) | DARPA | | [38] | Academic research on EO/IR/thermal sensor limitations under canopy | IEEE | | [41] | Academic radar engineering on frequency selection and penetration | IEEE | | [50] | UAV radar miniaturization and automotive FMCW radar ICs | Texas Instruments |