Research On Blockchain-Based Covert Communication Methods

In the traditional network covert communication,the sender and receiver communicate directionally,which leads to the problems including that the covert channel is easy to be detected,the both sides of covert communication may be exposed to others and the robustness of the covert transmission is poor.In this paper,because blockchain has the features of anti-tampering,anonymity and transaction broadcast across the network,we propose a method to construct a covert channel using blockchain,which can protect the covert channel from being listened to and detected,hide the identities of both sides of the covert communication,and ensure the integrity and covert nature of the information transmitted through the covert channel.The main contributions of this paper are as follows.(1)A blockchain covert communication model oriented to plain text information hiding is proposed to improve the amount of secret information embedding and reduce the communication latency.First,we design a space method based on partial order to embed the secret information into plain text carriers;then,the sender constructs a transaction in the blockchain network and uses the blockchain network as a covert channel for covert transmission of secret information based on the broadcast mechanism and data sharing mechanism of the blockchain;finally,the receiver extracts the plain text from the transaction and obtains the secret information through the inverse process of plain text embedding.The experimental results show that the model not only improves the amount of secret information embedding and shortens the time delay of covert communication,but also ensures the detection resistance of covert communication and completely hides the receiver’s identity information.(2)A double steganography model based on blockchain and inter planetary file system(IPFS)is proposed to further improve the amount of secret message embedding while making the secret carrier stealthier.First,the sender uses image steganography and IPFS to convert the secret information twice modally and obtain its identifier;then,it uses plain text steganography to embed the identifier into the normal blockchain transactions and publish them to the blockchain network;finally,the receiver extracts the identifier of the stego-image through the inverse process of plain text embedding,downloads the stego-image from IPFS according to this identifier,and extracts the secret information through the inverse process of image steganography.The experimental results show that the secret information embedding of this model is enhanced from bit-level to KB-level,and the detection resistance of the steganographic channel is further enhanced due to the double steganography method of image steganography and plain text steganography.(3)A blockchain covert communication model hiding the sensitive documents and the identity of the sender is proposed to realize the covert transmission of a large-volume document without initiating a large number of transactions,while achieving the hiding of the identity information of both sides.First,the sender uses ciphertext-policy attribute-based encryption(CP-ABE)and the image steganography based generative adversarial networks(GAN)to modally transform the sensitive documents;then,the sender constructs a special transaction carrying the identifier and publishes it with a ring signature;finally,the receiver looks for the special transaction,extracts the identifier,downloads the encrypted document from IPFS and decrypts it using its own attribute set to obtain the sensitive document.The experimental results show that the secret information embedding of this model is increased from KB-level to MB-level,and the detection resistance and security of the covert channel are greatly enhanced due to the attribute-based encryption method and the GAN-based image steganography.More importantly,the identity information of both sides is truly hidden due to ring signature encryption for transactions,which erases the sender identity information.