**1. Introduction**

Thanks to the rapid advancements in artificial intelligence (AI) and Internet of Things (IoT) technologies, the concept of Smart Cites becomes realistic. The information fusion capability provided by these interconnected devices enables situational awareness (SAW), which is essential to ensure a safe and sustainable urban environment. With wide deployment of the exponentially increasing smart Internet of Video Things (IoVT) for safety surveillance purposes, intelligent online video stream processing is becoming one of the most actively researched topics in smart cites [1].

In typical Internet of Video Things (IoVT) systems, a huge amount of raw video data collected by geographically scattered cameras is sent to a remote cloud for aggregation. It provides a broad spectrum of promising applications, including public space monitoring, human behavior recognition [2], and suspicious event identification [3]. However, centralized IoVT solutions suffer from the risk of single points of failure and are not scalable for accommodating the ever growing IoVT networks, which are pervasively deployed with heterogeneous and resource-limited smart devices at the edge of networks. Moreover, online video streams and other offline data, such as situation contextual features, are shared among participants using high-end cloud servers, which are under the control of third-party entities. Such a centralized architecture also raises severe privacy and security concerns that data in storage can be misused or tampered with by dishonest entities.

Evolving from the distributed ledger technology (DLT), blockchain has gained significant attention for its potential to revolutionize multiple areas of the economy and

**Citation:** Xu, R.; Nagothu, D.; Chen, Y. EconLedger: A Proof-of-ENF Consensus Based Lightweight Distributed Ledger for IoVT Networks. *Future Internet* **2021**, *13*, 248. https://doi.org/10.3390/ fi13100248

Academic Editors: Ahad ZareRavasan, Taha Mansouri, MichalKrˇcáland SaeedRouhani

Received: 7 September 2021 Accepted: 22 September 2021 Published: 24 September 2021

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

society. The inherent security guarantees of blockchain lay down the foundations of serverless record keeping, without the need for centralizing trusted third-party authorities [4]. Blockchain runs on a decentralized peer-to-peer (P2P) network in order to securely store and verify data without relying on a centralized trust authority. The decentralization removes the risk of singular point of failures and mitigates bottleneck performances, which were inherent in centralized architectures. In addition, blockchain leverages distributed consensus protocols to enable a verifiable process for fault tolerance and tamper-proof storage on a public distributed ledger. Therefore, transparency, immutability, and auditability guaranteed by blockchain ensure resilience, correctness, and provenance for all data sharing among untrusted participants.

Internet of Video Things (IoVT) provides a broad spectrum of applications, particularly in the area of public safety [5]. Migrating from centralized cloud-based paradigms to decentralized blockchain-based methods renders IoVT systems more efficient, scalable, and secure. However, directly integrating cryptocurrency-oriented blockchains into resource constrained IoVT systems is difficult in terms of handling the blockchain trilemma [6], which points out that decentralization, scalability, and security cannot perfectly co-exist. Most IoVT devices are highly resource constrained. Therefore, computing and storage intensive consensus protocols are not affordable, such as Proof-of-Work (PoW) [7], Proofs-of-Retrievability (PORs) [8], or Practical Byzantine Fault Tolerant (PBFT) [9], which come with high communication complexity and poor scalability. In addition, IoVT systems involve a large volume of real-time transactions. Higher throughput and lower latency become key metrics in blockchain-based systems for IoVT deployed on edge networks. Furthermore, DLTs are not general-purpose databases. The storage overhead is prohibitively high if raw data generated by IoVT transacting networks are stored in the blockchain.

The Electrical Network Frequency (ENF) is the power supply frequency which fluctuates around its nominal frequency (50/60 Hz). The frequency fluctuations vary based on geographical region. The ENF fluctuations estimated from simultaneously recorded audio/video recordings within a power grid have a high correlation similarity [10].

Inspired by spatio-temporal sensitive ENF contained in multimedia signals, this paper proposes *EconLedger*, a novel *Proof-of-ENF* (PoENF) consensus algorithm based lightweight DLT for small scale IoVT networks. Compared to PoW or PoRs, which require high computation or storage resources in mining process, our novel PoENF consensus requires each validator to use extracted ENF variations from simultaneous multimedia recordings as proofs during current consensus round. The validator that presents a valid ENF proof with minimal squared-distance-based score is qualified to generate a new block. Thus, the PoENF consensus mechanism not only achieves efficiency without high demand of mining resource or hardware platform support but it also enhances security by mitigating mining centralization.

In contrast to existing solutions that directly collect ENF fluctuations from power grids and stores audio/video recordings in a centralized location-dependent ENF database [10,11], EconLedger uses *Swarm* [12], which is a decentralized database (DDB) technology, to archive raw ENF-containing multimedia proofs and transactions over IoVT networks. Only hashed references of data are recorded on an immutable and auditable distributed ledger. Thus, it reduces the ever-increasing data storage overhead on the public ledger. The EconLedger ensures correctness, availability, and provenance of data sharing among untrusted devices under a distributed network environment. Moreover, a network with permission ensures that only authorized nodes can access raw data on DDB such that privacy preservation is guaranteed.

In summary, this paper makes the following contributions:


The remainder of this paper is organized as follows: Section 2 briefly discusses background knowledge of ENF, then reviews existing consensus algorithms and state-ofthe-art research on IoT Blockchains. Section 3 introduces the rationale and architecture of EconLedger, as well as core features and security guarantees. A novel PoENF consensus mechanism is explained in Section 4. Section 5 presents prototype implementation and numerical results and discusses performance improvements and security insurances. Finally, a summary is presented in Section 6.
