Identification of Unmanned Aerial Vehicle Intrusion
based on Wi-Fi Analysis

Benchmark vs. Logistic Regression, Random Forests, Boosted Trees & Neural Networks

Design an AI-based decision system that accurately and instantly identifies a type of invading Unmanned Aerial Vehicle (UAV) in a civilian environment, based on Wi-Fi traffic data records, to take appropriate action in a rational way.

Goals & benefits

Identify the Wi-Fi traffic data characterizing an UAV intrusion. Enhance expert knowledge by helping security specialists understand the causal relationships between specific Radio frequency characteristics, their combination, and the type of UAV.

Help security agents qualify the type of UAV to diagnose the intrusion as early as possible.

Avoid a large number of false alarms thanks to transparent diagnosis, in a context of increasing number of consumer UAVs.

  • The top-model is a decision system composed of 4 gradual rules without chaining.
  • Each rule uses some of the 5 variables that XTRACTIS identified as predictors out of the 54 potential predictors from the two-way radio frequency time series recordings preprocessing.
  • Only a few rules are triggered at a time to compute the decision.

It has a perfect Real Performance (on unknown data).

It computes real-time predictions up to 70,000 decisions/second, offline or online (API).

UC10 scores graph
LoR=Logistic Regression
RFo=Random Forests
BT=Boosted Trees
NN=Neural Networks

Detailed results and explanations in full document

Use Case 2024/02 (v3.0)

Results by XTRACTIS® REVEAL 12.2.44533 (2023/01)


  1. Problem Definition
  2. XTRACTIS-induced Decision System
  3. XTRACTIS Process
  4. Top-Model Induction
  5. Explained Predictions for 3 unkown cases
  6. Top-Models Benchmark
  7. Appendices