The Internet of Things (IoT) system gathers data through diverse smart devices and sensors to make thorough decisions tailored to specific needs. Yet, in intricate IoT setups, privacy infringement occurs through various means like data collection, initial data handling, and data sharing. Therefore, the concealment of data during transmission should receive sufficient attention. The communication approach that merges blockchain technology with covert communication has shown progress in addressing the aforementioned issues. However, this integration has also led to challenges, such as low-data embedding rates and distinctive features in blockchain transactions containing covert data. To seek a solution with high-embedding rates that do not make generated transactions stand out distinctly, this article analyzes the Ethereum transaction field formats, identifies the input data field with high concealment and large capacity as the embedding target, then proposes a data covert transmission scheme based on hybrid embedding in contract fields. This scheme utilizes LSB steganography to embed high-capacity covert data in images, and embeds the URL of the image into the input data field of the Ethereum smart contract transaction, thereby increasing the embedding rates. Subsequently, to further enhance the concealment of this scheme, a data embedding method based on contract relationships is proposed. Through this technique, for the first time, covert data transmission is achieved solely through the invocation relationships of smart contracts within the blockchain covert communication environment, instead of directly embedding covert data into transactions. This method results in transactions that are theoretically indistinguishable from regular transactions, greatly enhancing the security of the scheme. Finally, an evaluation of undetectability, embedding rate, and scalability was conducted for the proposed schemes, concluding that the schemes presented in this article have significant advantages in all three areas.

The rapid development of the Internet of Things has increased the importance of IoT data privacy. Traditional encryption and access control mechanisms are insufficient for ensuring privacy. Blockchain-based covert communication offers enhanced concealment, anonymity, and immutability for secure information exchange over open networks. However, existing blockchain-based schemes face limitations in point-to-point communication and low screening efficiency and concealment, as well as challenges when extended to group scenarios, such as the existence of leakers. To address these issues, we propose a Blockchain-based Group Covert Communication (BGCC) scheme. BGCC leverages broadcast encryption to revoke leakers and introduces an efficient covert filtering mechanism based on the decisional ℓ-BDHE assumption. We prove its concealment through security reduction, statistical tests, and machine learning test. Experimental results demonstrate that BGCC outperforms existing schemes.

Blockchain-based censorship-resistant systems enable the user to access the blocked content through a covert channel while avoiding a suspicious network connection between the user and the proxy. However, state-of-the-art blockchain-based censorship-resistant schemes cannot satisfy both low communication fees and unobservability, and their method of identifying transactions with covert data may inadvertently expose the covert channel. In this paper, we present Hades, a blockchain-based covert channel framework that aims to circumvent censorship. Hades allows users to encode covert data as a transaction field, and identify transactions with covert data by using another transaction field as a label. We also present the security model for Hades, which defines the unobservability of Hades as the indistinguishability of transactions with covert data from normal transactions. We further propose two cost-friendly and unobservable instantiations of Hades: the basic RDSAC and the improved DDSAC. RDSAC uses private keys to encode covert data and utilizes random factors in the signing process as labels, while incurring a communication delay. DDSAC avoids the delay by encoding covert data into random factors and sampling a transaction amount from normal transactions as the label. We implement a prototype system of Hades and evaluate its performance. Experiment results show that our Hades prototype is unobservable, robust, and efficient. RDSAC and DDSAC can identify 1,654 transactions in 6.054 seconds and 0.071 seconds, respectively. Hades supports 1KB data transfer at $0.44 on the Bitcoin mainnet and cost-free data transfer on the Bitcoin testnet.

Covert channels serve the construction of cyberspace security. By realizing the secure transmission of data, it is widely used in political and financial fields. Blockchain covert channels have higher reliability and concealment compared to traditional network-based covert channels. However, existing blockchain covert storage channels need to create a large number of transactions to transmit covert information. Creating transactions requires a transation fee, which means that the implementation of blockchain covert storage channels requires a high cost. Besides, created transactions remain on-chain permanently, leading to the threat of covert information being detected. To overcome these limitations, we propose a blockchain covert timing channel framework. Specifically, we utilize inv and getdata messages in the Bitcoin transaction broadcast as carriers and propose three modulation modes to achieve covert channels without cost and leaving no trace. We evaluate the concealment of our modes by K-S, KLD tests, and machine learning approaches. Experimental results show the indistinguishability between traffic carrying covert information and normal traffic. Our channels promise a capacity of 2.4 bit/s.

Abstract With the development of 5G, the wireless Internet of Things (IoT) has become possible; how to provide privacy protections for the communication of IoT devices in a more vulnerable wireless transmission environment is a huge challenge. Thus, steganography is introduced as a safe and effective technology. Blockchain systems have been widely used in the area of steganography. Several works attempted to embed covert data into transactions in public blockchain systems such as Bitcoin, Ethereum and Monero. However, most of them merely focus on putting covert data into certain fields in transactions based on cryptographic algorithms. In this paper, a Covert Transaction Recognition (CTR) model is proposed by the Text Convolutional Neural Networks and Back Propagation Neural Networks. When utilizing the covert data-embedded field for recognizing, our CTR model can attain 0.79 precision and 0.83 recall on average for seven covert transaction construction schemes. The precision and recall can increase by at most 43 and 47%, respectively, if other unembedded fields were additionally exploited for recognition. We further propose a Practical Covert Transaction Construction (PCTC) model. This model fixes the contents in the embedded fields of the constructed transactions, and generates the contents in other fields using Generative Adversarial Networks. Experimental results demonstrated that the precision and recall are greatly decreased when identifying the covert transactions generated by our PCTC model. The data underlying this article are available in ‘covert-transaction-model’, at https://github.com/1997mint/covert-transaction-model.

The quality of covert communications is determined by the choice of communication channels and the design of data embedding schemes. Recently, the Bitcoin system is prevalent as a covert communication channel. The consensus mechanism requires participants to spread their found valid blocks under an adjustable difficulty, which provides a stable periodic broadcast channel. Moreover, senders and receivers are difficult to be traced, because the Bitcoin system is pseudonymous. However, since the historical data in the ledger cannot be removed from the Bitcoin system, the openness and the persistent storage of the ledger in the Bitcoin system post new challenges when designing data embedding schemes. More concreteness, most traditional data embedding schemes either design by heuristic or empirical algorithms or use a fixed field to embed data in the transactions. Therefore, the covert data can be recognized once the algorithm is leaked or the pattern is explored. In this article, we first propose a hash chain-based covert data embedding (HC-CDE) scheme. The embedded transactions are difficult to be discovered. We further propose an elliptic curve Diffie–Hellman chain-based covert data embedding (ECDHC-CDE) scheme to enhance the security of the HC-CDE scheme. Experimental analysis on the Bitcoin Testnet verifies the security and the efficiency of the proposed schemes.

In developing smart city, the growing popularity of machine learning (ML) that appreciates high-quality training data sets generated from diverse Internet-of-Things (IoT) devices raises natural questions about the privacy guarantees that can be provided in such settings. Privacy-preserving ML training in an aggregation scenario enables a model demander to securely train ML models with the sensitive IoT data gathered from IoT devices. The existing solutions are generally server aided, cannot deal with the collusion threat between the servers or between the servers and data owners, and do not match the delicate environments of IoT. We propose a privacy-preserving ML training framework named Heda that consists of a library of building blocks based on partial homomorphic encryption, which enables constructing multiple privacy-preserving ML training protocols for the aggregation scenario without the assistance of untrusted servers, and defending the security under collusion situations. Rigorous security analysis demonstrates the proposed protocols can protect the privacy of each participant in the honest-but-curious model and guarantee the security under most collusion situations. Extensive experiments validate the efficiency of Heda, which achieves privacy-preserving ML training without losing the model accuracy.