IEEE Consumer Electronics Magazine - November/December 2021 - 78

Special Section on Advances in Smart Cities
surveillance systems. Visual layer attacks on the
IoVT system include the systemic frame duplication
or replay attacks in temporal regions, frame
modification in spatial regions, and hybrid spatiotemporal
attacks on both.5 Timely detection
of compromised feed to an IoVT system is essential
to the infrastructure security, and having a
reliable authentication system with minimal
delay is a priority.
As a reliable and verifiable environmental
fingerprint associated to local power grid, ENF
signal extracted from video-audio streams provides
a prospective solution to detect visual
layer attacks.5 Thanks to multiple key characteristics
such as decentralization, immutability,
and auditability, blockchain is an ideal trustbased
framework to address security issues in
existing centralized solutions. Through integrating
blockchain and ENF fingerprint into IoVT systems,
the BEVA scheme enables a decentralized
and secure self-executing video authentication
mechanism to prevent against visual layer
attacks at the network of edge.
IoVT-BLOCKCHAIN
CONSIDERATIONS
A) Blockchain Overview: As a form of DLT,
blockchain initially was implemented as an
enabling technology of Bitcoin,6 which aims to
provide a cryptocurrency to record and verify
commercial transactions among trustless entities
in a decentralized manner. In blockchain networks,
a consensus protocol is executed among a
large amount of distributed miners or validators
to ensure integrity, consistence, and total order
of data on the ledger. Leveraging the decentralized
network architecture and cryptographic
mechanisms, all trustless participants maintain
a verifiable, append-only chained blocks structure
of public ledger, instead of relying on a centralized
third-party trust authority. Emerging
from the intelligent property, smart contract (SC)
introduces programmability into blockchain to
support various customized transaction logic
rather than simple cash transactions.15 Through
triggering self-executing procedures of SCs
stored on distributed ledgers, users can achieve
business logic or contract agreements among
trustless parties in a decentralized manner.
78
B) Challenges in Integrating Blockchain With
IoVT: Leveraging blockchain and smart contract is
promising to enable a decentralized security
framework to facilitate sharing, exchanging, and
the integration of information across trustless participants.9
However, directly integrating blockchain
into smart IoVT systems suffers from
various disadvantages, particularly in terms of
computation cost and transactions throughput in
consensus, and the storage requirements of storing
the entire ledger.8 In most cases, an IoVT
device's resources are highly constrained owing
to limited computation and memory capacity. It is
not affordable to execute high complexity and
computing intensive consensus algorithms, such
as Proof of Work (PoW). In addition, many intelligent
tasks in IoVT systems involve a large volume
of transactions that must be performed at high
speed. Therefore, higher throughput and lower
latency become key metrics when introducing
blockchain into IoVT systems. Meanwhile, IoVT
systems for Smart Cities are normally not open to
public. They are deployed on a permissioned network,
which requires identity authentication and
registration to enroll valid participants. It is necessary
to provide basic security primitives, such as
public key infrastructure (PKI) and access control
(AC) strategy, to enhance security properties of
permissioned blockchain network.
BEVA SYSTEM DESIGN
Taking advantages of ENF fingerprint and blockchain
technology, BEVA scheme fights against the
visual layer attacks on cross-devices networking
IoVT systems. Figure 1 presents a conceptual system
architecture of BEVA. Unlike conventional solutions
that rely on a centralized architecture to
verify and save ENF fingerprint,5 BEVA utilizes
blockchain technology to enable a decentralized
security framework for smart IoVT systems.
As the application layer in BEVA, an IoVT
framework provides a distributed video analytic
service and information visualization at the edge
network. All IoVT devices and users must be registered
to join the permissioned network. The realtime
video streams generated by cameras are
transferred to on-site edge devices for lower-level
analytic tasks, such as object-detection and situation
contextual features extraction. Then, raw
IEEE Consumer Electronics Magazine
IEEE Consumer Electronics Magazine - November/December 2021

Table of Contents for the Digital Edition of IEEE Consumer Electronics Magazine - November/December 2021
