In the 3rd part of our blockchain blog trilogy we are going to have a look at crypto anchors. Crypto anchors are often overseen in discussions around blockchain and DLT. Although they play a big role in closing the gap between virtual blockchain and the physical world.
When we talk about how blockchain and DLT are going to disrupt almost every industry, authors often draw a vivid picture of eliminating intermediaries and building networks of trust. Ownerships of assets, representing real world objects, are passed and recorded on a blockchain network. However, what this picture is usually missing is an answer to the following two questions:
- How can a real world object be unambiguously identified and mapped to the corresponding asset on the blockchain?
- How (and especially by whom) is an asset registered on the blockchain?
Both questions are tricky. This is why I am going to cover them in two separate posts. In this one we are going to address the first question.
The missing link
Let’s consider an example of a trusted banana food supply chain. Where a box of bananas is registered on the blockchain and every action along its supply chain is registered as well. As a customer I can see that the banana I am eating is a fair trade product, because the blockchain provides a tamper-resistant record of the banana’s supply chain. But how can I make sure that my organic fair trade banana wasn’t replaced with some low-grade one? Well, the answer is: I can’t.
This is where crypto anchors come into play. A crypto anchor is a mechanism to uniquely identify real-world objects in a digital world. It is a mechanism to provide some digital fingerprint or DNA to an object. This fingerprint is registered on the blockchain to connect both the physical real world and virtual blockchain space. Data that relates to the object, like bill of lading, country of origin or harvest date can be tied to the registered crypto anchor. As a customer, in order to validate authenticity of a product, I can simply use the crypto anchor provided with the product and look up the attached data in the blockchain.
Challenges of designing crypto anchors
Unfortunately, designing crypto anchors can be a very daunting task. Mostly, because there is a trade-off between production cost and tamper-resistance.
- Production cost: When applied to billions of goods, manufacturing a crypto anchor must be as cheap as possible. Just imagine anchoring every single banana ever harvested. This might be very expensive. But not only the production cost of the anchor itself is relevant. Also the cost of reading the crypto anchor has to be taken into account. In order for mass adoption of crypto anchors, reading an anchor must be as convenient as possible. Ideally, the process must also not require expensive additional hardware. Otherwise it would render crypto-anchoring infeasible for many convenience goods.
- Tamper-resistance: Despite preferably low production cost crypto anchors also need to be secure, i.e. tamper-resistant. This primarily addresses copy protection of the anchor itself. But also relabeling of anchors, i.e. removing an anchor from an object and placing it onto another one.
Types of crypto anchors
Crypto anchors can be divided into three categories: configured secrets, physical fingerprint and embedded security features.
Configured secrets are a basically digital passwords or crypto keys that are added to a product by its manufacturer. An example could be an integrated circuit (IC), with a unique key and cryptographic functions protecting that key. The price depends on the complexity of the IC. The more advanced security features and crypto operations are supported, the more expensive these systems become. However, simple IC tags like RFID can be easily removed from a product and be placed on another one. Ensuring product and anchor are firmly tied together is key to successfully implementing crypto anchors based on configured secrets.
Crypto anchors based on physical fingerprints are similar to a human fingerprint. Such systems make use of an object’s unique characteristics as a basis to derive a unique key or ID. Despite highly optimized manufacturing processes even highly processed products vary in certain nuances. Given the right technology to measure them, these nuances can be sufficient to identify an item. An example are physical unclonable functions (PUF) as part of semiconductor devices.
An SRAM, for example, consists of a set of registers. On SRAM startup the register bits seem to have a random value (o or 1). However, when the values at startup are repeatedly measured, it shows that they are not random at all. Instead, they tend to be very similar if not identical. The individual pattern however, is unique to each SRAM PUF. This is because exact control over the manufacturing process is infeasible.
Embedded security features
Embedded security features, are embedded into the manufacturing process of an item. That way crypto anchors are intrinsically tied to the object they identify. Removing such a crypto anchor usually breaks either the item or the anchor itself. Examples are micro printing or holograms. This category covers a wide range of different techniques. Often they also require advanced process technology in order to include the anchor into the product.
Blockchain itself is not capable of identifying real world objects. Therefore, crypto anchors play an important role in closing this gap between virtual blockchain and the physical world. They derive a unique ID for every object. This ID is then registered on the blockchain. Once registered, all digital records of a product can be tied to that anchor. However, for mass adoption, crypto anchors must be cheap, tamper-resistant and easy-to-use. This makes crypto anchor research an interesting but challenging research field. Approaches can roughly be divided into the three categories: configured secrets, physical fingerprints and embedded security features.