Safe and Inclusive E-Society • Lithuanian Science Development Programme • Sub-Project 6

Solar-Powered
Border Probe Mesh

Early Warning Alert System for National Border Security

A distributed network of autonomous, solar-powered detection probes with AI-driven classification, LoRa networking, and precision triangulation for the Lithuanian national border.

Explore the System
Scroll

Securing Europe's Eastern Border

Autonomous detection where traditional surveillance fails

The Challenge

Lithuania's extensive forested border presents unique surveillance challenges: dense vegetation blocks line-of-sight, wildlife triggers cause high false alarm rates, extreme seasonal temperatures strain equipment, and traditional infrastructure is costly to deploy and maintain in remote areas.

The Solution

A mesh of small, autonomous "border probes" deployed at strategic intervals along the border. Each probe combines solar power for year-round energy autonomy, thermal AI detection for intelligent threat classification, compass-based directional awareness for precision triangulation, and LoRa mesh networking for reliable long-range communication without cellular infrastructure.

The AI-powered thermal pipeline filters 70-90% of wildlife false alarms before alerting operators, dramatically reducing operator fatigue while maintaining detection reliability.

Dense forest border environment
$110–150 Per Unit at Scale
30–60m Probe Spacing
365 Days Energy Autonomy
<25m Triangulation Accuracy

Always-On AI Detection Pipeline

Hardware CNN accelerator classifies every thermal frame in real time

Thermal Monitoring

MLX90640 always on
0.5–1 Hz / ~120 µA

Continuous

CNN Classification

MAX78000 64-core CNN
~2 ms / 0.01 mJ per frame

Every frame
70–90% Wildlife → Continue

Detection Ray

Compass heading +
thermal centroid bearing

Bearing computed

Alert Broadcast

LoRa alert +
detection ray bearing

ALERT
0.01 mJ CNN classification per frame
~128 mJ Alert broadcast per event
70–90% Wildlife filtered by CNN

Custom Single-Board Design

All sensing, compute, and communication integrated on one purpose-built PCB

Custom Border Probe PCB — single-board design integrating MAX78000 AI MCU, LoRa, GPS, thermal sensor, IMU, and power management
Single integrated PCB — replaces 7 separate modules

Integrated On-Board

MAX78000 AI MCU

Cortex-M4F + RISC-V with 64-core CNN accelerator. 442KB weight storage for always-on AI classification.

64-core CNN

MLX90640 Thermal Array

32×24 far-infrared pixels, 55° FOV. Day/night detection at 15–30 m with 0.86° angular accuracy.

32×24 px

SX1262 LoRa Radio

868 MHz EU band, +22 dBm TX. AES-256 encrypted communication via custom protocol, 2–15 km range.

868 MHz

L76K GNSS Module

GPS + GLONASS with SBAS/EGNOS correction. Position averaged over 100+ readings for ~2.5 m accuracy.

2.5 m CEP

LSM303AGR IMU

Magnetometer + accelerometer for compass heading (±2°) and tamper detection.

±2°

MPPT Solar Input

On-board charge controller for 10W monocrystalline panel. Optimized for LiFePO4 chemistry.

10 W

LiFePO4 Battery Interface

LiFePO4 battery pack with cold-weather tolerance to −20°C and voltage monitoring.

LiFePO4

Industrial MicroSD Slot

Crash-resistant local storage for event data and detection logs. Circular buffer with wear leveling.

32 GB

IP67 Enclosure Ready

Board designed for sealed ABS housing with cable glands. Rated −40°C to +60°C operating range.

IP67
Estimated cost per probe unit (board + enclosure + peripherals)
$237 – $440
At scale (100+ units, custom PCB): $110 – $150

Technical Specifications

Power budget, solar production, and sensor capabilities

Power Budget

ModeCurrentEnergy
Always-on monitoring~184 µA0.6 mW
Thermal CNN (per frame)~5 mW, 2 ms0.01 mJ
Alert broadcast~400 mW, 300 ms~128 mJ

Solar vs Consumption

Summer
20–28 Wh/day
~0.02
1000x+
Spring / Autumn
8–14 Wh/day
~0.02
400–700x
Winter
1.5–4 Wh/day
~0.02
75–200x

10W panel, Lithuania latitude, with forest canopy factor

Sensor Capabilities

Thermal resolution32 × 24 pixels
Thermal FOV55°
Angular accuracy~0.86° (sub-pixel)
Detection range15–30 m (human)
AI accelerator64-core CNN (MAX78000)
CNN inference~2 ms / 0.01 mJ
LoRa frequency868 MHz (EU)
LoRa spreadingSF10
GPS accuracy~2.5 m (averaged)
Compass heading±2° (calibrated)

Base Station / Hub (per 10–20 probes)

LoRa receiver + high-gain antenna$50–70
Raspberry Pi 4 or mini PC$50–80
4G/LTE modem$30–50
Solar + battery$80–150
Enclosure$30–50
Total$250–400

LoRa Mesh Network Architecture

Resilient communication without cellular infrastructure

BORDER LINE P1 P2 P3 P4 P5 mesh relay BASE STATION DASHBOARD Direct-to-base Mesh relay

Direct-to-Base

Preferred mode. Probes send alerts straight to the nearest border post receiver for lowest latency.

Mesh Relay

When direct range is limited, alerts hop through neighboring probes using a custom relay protocol.

Health Monitoring

Heartbeat every 15–30 min reports battery, solar rate, GPS quality, heading, temperature, and event count.

Detection Ray Triangulation

Pinpointing intrusions with sub-degree angular precision

P1 54.6831°N 25.2799°E P2 54.6831°N 25.2803°E Detection Ray Detection Ray 55° FOV TARGET ZONE ~80-100 m²
1

GPS + Compass Heading

Each probe knows its exact position and the direction it faces, establishing a geographic reference frame.

2

Thermal Centroid Bearing

The MLX90640 determines the angular position of a heat source within its 55° FOV to ~0.86° accuracy via sub-pixel centroid fitting.

3

Detection Ray

Combining GPS position + compass heading + thermal bearing produces a "detection ray" — a precise direction line in absolute geographic coordinates.

4

Multi-Probe Intersection

Two or more probes observing the same target produce intersecting rays, pinpointing the target's location to an ~80–100 m² zone. More probes = higher confidence.

Regulatory Compliance

Designed for full EU regulatory alignment

EU Fundamental Rights Agency

Non-intrusive surveillance approach. No facial recognition. Thermal-only classification preserves individual privacy while detecting threats.

AI Act

Transparent classification pipeline. All AI decisions logged with confidence scores. Human-in-the-loop for critical alerts.

GDPR

Thermal data only — no personal images captured. Event logs stored locally on encrypted MicroSD. Privacy by design.

NIS2

AES-256 encrypted LoRa communications. Secure boot on MAX78000. Tamper detection with accelerometer alerts.

MVP Targets & Deliverables

36-month project milestones and measurable outcomes

TRL 6–9 Technology Readiness Level
70–90% Wildlife False Alarm Filtering
<25m Triangulation Position Error
365 Days Energy Autonomy
1 Patent Filed
24+ Q1/Q2 Journal Articles

MVP Deliverables

✓ 10–20 probes deployed at 30–60 m spacing
✓ Always-on thermal AI detection (MAX78000 CNN accelerator)
✓ LoRa alert messages with location, heading, and detection ray bearing
✓ Event correlation engine with automatic spatio-temporal clustering
✓ Map-based control dashboard
✓ Multi-probe triangulation with weighted least-squares estimation
✓ Local data storage on industrial MicroSD
✓ Health monitoring dashboard with probe telemetry