An Examination of LPWAN Security in Maritime Applications

LoRaWAN’s role in global maritime logistics has allowed for efficient monitoring of ships and cargo, but it also comes with critical cybersecurity vulnerabilities. Experimental validation of three attack vectors—replay attacks, narrowband jamming and metadata inference—is conducted using a reproducible digital-twin LoRaWAN dataset reflecting Rotterdam port-like operational patterns (N = 20,000 baseline transmissions). Using controlled simulations and Kolmogorov–Smirnov statistical analysis, we show that: (1) replay attacks are feasible under Activation by Personalization (ABP) configurations lacking enforced frame-counter validation and exhibit no univariate separation from legitimate traffic under Kolmogorov–Smirnov analysis (p > 0.46 for all evaluated radio features); (2) narrowband jamming leads to significant SNR degradation (p = 2.36 × 10−5) on targeted channels without inducing broad distributional anomalies across other radio features; and (3) metadata-only analysis supports elevated metadata-based re-identification susceptibility (median Rd=0.834), indicating high predictability under passive observation which can reveal operationally relevant signals even when AES-128 is employed. Our proposed layered mitigation framework consists of mandatory Over-the-Air Activation (OTAA), cryptographic key rotation, channel diversity incorporating Adaptive Data Rate (ADR), gateway hardening, and protocol-level enforcement considerations, customized for maritime LPWAN scenarios. We provide experiment-backed evidence and actionable recommendations to connect academic LPWAN security research to that of industrial maritime practice.