3.1.1. IoVT Application

The upper-level IoVT application utilizes an EconLedger fabric to enable decentralized video analytic services and information visualization at the edge. All devices and users must be registered to join the IoVT system as required by the permissioned network, which can provide basic security primitives such as public key infrastructure (PKI), identity authentication [30], and access control [31], etc. Real-time video streams generated by cameras are transferred to on-site/near-site edge devices for lower level analytic tasks, such as object detection and situational contextual features extraction. Thus, cameras associated with edge devices act as IoVT service units at the network of edge. Then, IoVT service units send raw video data and extracted contextual information to the information visualization unit, which provides video recordings and smart applications for authorized users.

**Figure 1.** The EconLedger system architecture.

To prevent visual layer attacks, IoVT service extracts ENF signals from video streams as an environmental fingerprint, which is stored into DDB and secured by *EconLedger fabric*. At any given time instant, variation trends of ENF-containing multimedia signals from all synchronous cameras on the same power grid are almost identical. Therefore, using ENF fluctuations recorded on EconLedger laid solid ground truth for video authenticity verification. By calculating correlation coefficients among ENF signals extracted from video recordings with an agreed ENF estimate recorded on distributed ledger, the information visualization unit verifies whether or not live/offline video streams are generated by cameras within the same power grid [32,33].
