Go back

Blockchain technology implementation in mussel aquaculture

02/07/2024

A working group in the area of data reliability and security at Università Politecnica delle Marche (UNIVPM) is working on the design and development of a blockchain-based platform to certify data. Special focus is given to the certification of the different processes involved in the case study related to mussel aquaculture.  

The designed architecture is based on a central component, consisting of an off-chain server equipped with storage capacity for the data to be collected and certified, as well as capacity to execute the cryptographic functions required for their certification. Such a central server is expected to perform the following core functions: 

• Expose API (Application Program Interface) functions to enable the collection of data from several possible sources. Such sources may be represented by web interfaces (usable from computers or via smartphone apps) for manual entry, or cyber-physical devices for automated collection. 
• Perform cryptographic functions to compute certification data derived from the data themselves, which enable certification and integrity of collected data to be ensured, without including the data themselves. 
• Interface with a public blockchain (such as Ethereum) to perform writing of certification data to ensure persistence and immutability, without disclosing the collected data. 

Once the collection and blockchain-based certification of data has been carried out, anyone who comes into the possession of the certified data and the related certification information can verify its correctness directly. This can be done by querying the central server again, or in an entirely decentralized manner, via functions executed locally. 

The data collected during the mussel aquaculture case study will be used to test the blockchain-based platform for certifying data. The productive cycle of mussels can be affected by several factors that can represent potential sources of food loss. The acquisition of data from different variables during mussel farming can represent a suitable solution to provide data to monitor the productive cycle and possibly prevent and minimize food loss. However, this process is often endangered by the integrity of the data. In fact, data can be modified or purposely deleted, or crucial information can be lost due to technology malfunctions.. For this reason, there is a crucial need for an efficient and secure way to store and certify data, aiming for data integrity and availability.

The aim is to address the problem of certification and data traceability by identifying an architecture that can exploit the inherent advantages of distributed ledger technologies. 

The proposed innovative architecture is based on the use of Merkle trees for organizing pieces of information. The primary goal of this approach is to limit the number of blockchain transactions to be generated, making the certification process more efficient and cost-effective. The idea is to independently consider data coming from different sources and organize them into data structures called Merkle trees. Merkle trees are data-compressing structures where the data to be certified are the leaves. These leaves are hashed consecutively using a hashing function (e.g., SHA256) until the tree’s root, known as the Merkle root, is reached. The resulting hash digest is a single string representing all the leaves in a compressed way and will therefore be the only data stored on the blockchain. 

Since hashing is not invertible, verifying that information has not been modified after certification requires extracting a set of node values from the tree to test data integrity. Using the so-called Merkle Proof, one can demonstrate that data in a leaf have not been modified.

The certification process is summarized in the next figure. Initially, data that need to be certified are grouped together in a waiting queue and, after a pre-defined amount of time, the certification process begins. Basically, these grouped data are used as leaves to build the Merkle Tree and the Merkle root is computed. While the tree is stored in an off-chain database, the Merkle root is sent to the blockchain through a transaction. Transaction information is stored in the off-chain database as well, together with all the information that allows the verification process (e.g., the Merkle proof). 

When a user needs to verify some data, the proof is extracted from the off-chain database. Then, the transaction that contains the in-chain root needs to be located in order to start the actual verification function, which includes the following operations: 

• Computing the hash of the data, 
• Extracting the root from the blockchain, 
• Locally computing the root through the proof and the hash found in step 1), 
• Comparing the values obtained in steps 2) and 3); in case of equality, the verification is successful, otherwise failed.

Users will have the possibility to upload data and to verify them through a friendly interface that will show, for example, if data are in queue or certified, or if the considered data has maintained their integrity or not.  An example of a preliminary prototype of this application for the mussels aquaculture case study is shown in the next figures.