TA-BI: A Formal Observation-to-Decision Architecture for Trust-Aware Routing Support in Wireless and Mobile Wireless Sensor Networks
Year: 2026
Type: article
Abstract: TA-BI (trust-aware behavioral intelligence) is a conceptual-formal observation-to-decision contract for trust-aware routing support in reactive multi-hop wireless sensor networks (WSNs) and mobile wireless sensor networks (MWSNs). It specifies evidence provenance and availability, separates current Behavioral Information from historical node Trust, maps these states to route candidates, and controls when RouteScore may affect a native routing decision. In the reference AODV adapter, declared validity checks, destination-sequence freshness, and hop-count precedence are applied before RouteScore on a finite frozen candidate set. A complete common component basis prevents asymmetric missing evidence from favoring a candidate; a closed gate disables TA-BI re-ranking but retains a total routing outcome through baseline ordering or native fallback. Nine propositions establish bounded contract properties, including candidate-admissibility preservation rather than equivalence of the complete AODV transition system. A deterministic Python oracle evaluates ten groups, with explicit T3/T4 subcases for stale, version-inconsistent, missing, low-confidence, and restored evidence. The evidence is separated into formal correctness, executable conformance, historical integration reachability, and empirical network performance. The supplied formal and deterministic artifacts support only the first two levels; the bounded historical OMNeT++/INET record supports integration reachability only. Packet delivery, delay, throughput, energy, overhead, scalability, and mobility performance are not evaluated.
Source: Sensors
DOI: 10.3390/s26175422
Authors: Wiesław Jabłoński, Ksawery Krenc, Adam Kawalec
Language: English
Subfield: Computer Networks and Communications
Field: Computer Science
Domain: Physical Sciences
Open Access status: gold
Estimated APC paid: $3,137
OpenAlex ID: W7204452392
