Plan To Counterfeit Microelectronics? I think Not.

By Aman Ali on ALTCOIN MAGAZINE

Co-Author: Dr. Yong Pei

Cryptocurrency’s market cap may be collapsing, but Blockchain technology is booming. Enterprises such as IBM, Salesforce, Microsoft, Amazon, JPMorgan Chase, EY, SAP, Intel, Walmart, etc are innovating in the blockchain space.

Photo by Laura Ockel on Unsplash

These products promise all the tenets of the blockchain like tamper resistance, decentralized distributed ledger, security, privacy, etc. Of all these principles, we would like to focus on the security features of the blockchain and their applications in ensuring the integrity of various supply chains. For example, Food Quality and Safety, Healthcare Management & Quality Assurance, Microelectronics, etc. In this article, we will focus on the security of the supply chain for computer hardware.

Security of the computer systems can be compromised by exploiting the vulnerability either in its software or hardware. While software security can be dealt with by using antivirus software, hardware security is a different ball game altogether.

Top challenges for hardware security include circuit tampering, counterfeit hardware components & systems, and electronic waste. From the design to manufacturing to distribution, hardware components change many destinations. Malicious hardware components can enter the system at any stage of its supply chain. It is difficult to avoid such counterfeits. And even if we identify such component/shipment, it is strenuous to unmask the entity responsible for it.

In this article, we present a Blockchain-Managed Microelectronic Chip Supply Chain’s Proof of Concept. We believe that this solution will fit together and complement the existing software security practices to provide a more comprehensive strategy to secure electronic systems.

Idea

To have an electronic hardware supply chain based on blockchain which helps detect counterfeit hardware, embraces accountability and assures quality & security across the potentially worldwide distributed product cycle. It is also expected to help further create an ecosystem of electronic hardware design & manufacturing that facilitates healthy market competition and avoids the formation of monopolies.

Introduction to Blockchain

Let us now understand how does a blockchain network work — to avoid malicious tampering of the transactions.

A simple representation of a blockchain network.

1. Let us say the transactions between (A, B), (C, D), (E, F) and so on be labeled as TNX 1, TNX 2, TNX 3, etc.
2. All these transactions are present in the transaction pool.
3. By Metcalfe’s law, the strength of the p2p/telecommunication network is directly proportional to the square of the number of miners/nodes in the network. Thus to maintain a secure network, we must have a good number of nodes. For the sake of simplicity, let us assume that there are 4 miners/nodes in our blockchain network. The job of the miner/node is to pick the transactions from the transaction pool and verify them.
4. The transactions are picked in random. Let us cap the block size to 4 transactions. In the above diagram, Miner 1 has verified 4 transactions. He then propagates the block to the peers via a p2p protocol.
5. This block is then verified by consensus and added to an ever-growing ledger of transactions known as the blockchain. These records of the transaction are now immutable and tamper-resistant.
6. Now, the miners start verifying the next set of transactions from the transaction pool which would be added to the blockchain.

Selection of Blockchain Platform

We will now select the blockchain platform for our POC. There are, in general, two types of blockchain: Public Blockchain and Private Blockchain. We will now explore them from the privacy and network scaling perspective.

Public Blockchain

Privacy:

1. In public blockchains, the transactional data can be viewed by the public. For example, you can view bitcoin’s transactions on Blockchain Explorer.
2. Due to this, the confidential business metrics of an enterprise organization would be known to the public.
3. With trade secrets out, the organizations will lose their competitive edge. This is counterproductive for the enterprises.

Scaling:

1. Public blockchains generally have more than a hundred miners/nodes.
2. Before a node can start working on the next block, the current state/status/snapshot/ledger-entries must be updated on all the nodes. This is known as state replication.
3. In our application, we expect a high number of transactions/transaction throughput. Thus, there is a requirement for the blockchain to be able to scale up to a high volume of transactions per second. To achieve this: (i) one may either increase the block size. (ii) Or use a suitable consensus mechanism.
4. Increasing the block size would mean that the miners have to upgrade their hardware to perform more hash rates. Due to this. the miners have to increase the transaction fees/gas prices. An increase in such fees will fail to retain the existing user weakening the network and making it susceptible to a 51% attack/Sybil attack. This is the reason that block size is capped and cannot be increased arbitrarily.
5. On the other hand, The various existing consensus algorithms that are used in public blockchains have its flaws and still aren’t matured yet for scaling blockchains. For example Proof of Work(POW) consumes a lot of electricity, POS is susceptible to nothing at stake problem and p + epsilon attack, DPOS cost EOS 7.7 million dollars.
6. Therefore, the current public blockchains cannot scale. Though there have been improvements proposed like sharding in POS and Lightening network in Bitcoin’s POW. But they have their flaws. This is probably the reason that the Libra blockchain will initially be permissioned/private, and would transition to a permissionless blockchain when the technology matures.

Private Blockchain

Privacy:

1. Private blockchains enable abstraction of enterprise data, for privacy and confidentiality, with the provision of channels. For example, Hyperledger Fabric gives us this feature.
2. Thus, others can only view the data that an enterprise is willing to share with them. For example in the above diagram, A & B belong to the same channel. Due to this, member C cannot view all the details that are being shared between A & B.
3. With this, enterprises are likely more willing to participate in such an ecosystem.

Scalability:

1. Private blockchain network has a relatively lesser number of nodes and transaction throughput when compared with the public blockchain network.
2. This speeds up the state replication process.
3. Since the nodes are trusted and the blockchain as such is permissioned, there is a little chance of a Sybil attack.
4. As the state replication is now faster, there is no need to increase the block size.
5. The consensus algorithms for private blockchains, though aren’t completely mature, have been adopted in practice. For example Kafka, Raft, PBFT, etc.
6. In our application, there is little requirement to divide the organizations into little quorums. Thus, there is no problem in achieving transactional consistency.

After this analysis, it is logical to incorporate a private blockchain in our supply chain.

System Design

Primarily, microelectronic hardware can be divided into two categories. They are Field-programmable gate array (FPGA) and Application-specific integrated circuit (ASIC). FPGAs are programmable i.e soft processor, while ASICs are an end product solution. Due to this difference, the corresponding supply chains also vary.

ASIC Supply Chain

Let’s divide the whole ASIC hardware supply chain into four primary stages.

1. IP Design: In this stage, the design of the ASICs blueprint is finalized. Let us label this stage as Stage A.
2. Software: Here is where the software for the corresponding design is created. Let us label this stage as Stage B. Tools used in this stage are, for example, ASIC CAD Tools.
3. Manufacture: Once the software and the design are ready, the hardware is now manufactured. Let us label this stage as Stage C.
4. Distribution worldwide: After the previous stage, ASIC is now ready for distribution. Let us label this stage as stage D. There can be multiple distributors before hardware reaches the consumers/customers. This will be explored while explaining the blockchain’s integration.

FPGA Supply Chain

Since FPGAs are programmable by the end-user, its hardware supply chain varies slightly from ASICs’. It primarily consists of three stages.

1. IP Design: In this stage, the design of the FPGAs blueprint is finalized. Let us label this stage as Stage A.
2. Manufacturing and Testing: After the IP design stage, FPGAs are now manufactured according to the blueprint. Finally, various quality assurance tests are performed on the FPGA. Let us label this stage as Stage B.
3. Distribution: FPGAs are then distributed to the customers for its programming. Let us label this stage as Stage C.

Incorporating Blockchain Into The Supply Chain

Blockchain incorporates public key cryptography that has the property of One Way functions which provides security due to the Avalanche effect. This will help us maintain the integrity of the hardware chip when it propagates through its supply chain.

The system will work as follows:

1. From the IP Design to the Distribution, Blockchain will track the destinations/stages that Hardware Chip changes through the entire supply chain.
2. Let Each Stage will have its unique digital signature which is represented by A1, A2, A3… and so on.
3. Hardware is manufactured individually and is transported/shipped in bulk. Hence, it is necessary to have proper security mechanisms implemented both on a unit level and the overall shipment/pack level. On the shipment level — While the hardware is changing places, there will be two QR codes that would be printed on the packaging that contains the intermediate products. The first QR code, Q1, will contain the signature of the agent who just finished his task on the product. And the second QR code, Q2, generated with a function which taken input as the cumulative sum of the signature of the current stage and all the previous stages in the supply chain. For example, when the stage with signature A6 completes his job then the second QR code will have the signature as sig(A1+A2+A3+A4+A5+A6). Where sig() is the signature function. On the unit level — To further enhance the security of the hardware, there will also be a unique electronic signature, E, that would be added on the hardware circuit itself. This electronic signature is programmable only once. As and when the hardware reaches a stage, an electronic signature which is the function of the agent’s signature and the timestamp of manufacturing of the hardware is added to the microelectronic chip. This is essential for the integrity of the hardware. All this information, both unit and the shipment level, will be recorded on a Blockchain. And when the signature is scanned, the blockchain can confirm the provenance of the hardware. In the case that the information doesn’t match with the data stored on the blockchain, the hardware is rejected. With this, the chances for counterfeiting the hardware can be significantly reduced.
4. After A1 does his job on a batch of shipment. Q1 = sig(A1) and Q2 = sig(A1).
5. Now when A2 receives this batch from A1, he will verify the Q1 and Q2 with the help of a scanner. After confirming its authenticity, A2 will do the required changes/modifications on this shipment. Now the values of Q1 and Q2 are Q1=sig(A1) and Q2 = sig(A1+A2).
6. And the process continues. When it reaches the final agent “AL”, only one QR code Q will be attached to the final hardware product. Q is defined as : Q = sig(A1+A2+A3 …+AL+Batch_Number), where batch number is the batch of the hardware.
7. This is now distributed to the various final destinations and their integrity can be verified from the data recorded in the blockchain.

Incentive Design

There is a good chance for an agent to go selfish and add impurities to the shipment. We intend to avoid this by giving incentive points to the agent after completion of every batch. If the agent goes selfish and the batch is rejected because of him(we will know this when the next agent in the chain rejects this batch), the blockchain decreases his points.

This process can be automated with the help of Autonomous Smart Contracts. These incentive points can be used as a metric to get further contracts with other agencies in other hardware supply chains. For example: In the above diagram Agent B, a member of Supply Chain A, can use his points to be a member of Supply Chain B.

This will make us trust even the new entrants thus democratizing the market and avoids the formation of monopolies. This will also help reduce the wealth gap.

Hence, the agents/agencies will be working towards the benefit of the network.

Implementation

1. This ecosystem can be offered as “Blockchain As A Service” with the help of the cloud. Eventually, the service would have multi-cloud integration for worldwide availability of the network.
2. Agents and Stores can participate in the governance of this ecosystem by paying a fee. This fee is required for maintaining the cloud architecture.
3. A website that will display the incentive points of agents. This will help “The Agents to trust the Agents” and “The Customers to trust the Hardware”.
4. For the sake of simplicity, a cross-platform mobile app will also be developed. With the help of the mobile application, the track record of the stores can be checked on the fly.

This will help us have an open market where the quality of the hardware will not be compromised.

Thank you for reading our article.

Note:
This article comes under research work done at Wright State University’s SMART Lab under the guidance of Dr. Yong Pei.

Wright State University: https://www.wright.edu/
Department of Computer Science and Engineering: https://www.wright.edu/degrees-and-programs/profile/computer-science
Dr. Yong Pei: https://people.wright.edu/yong.pei

If you have any questions, please feel free to send me an email. You can also contact me via Linkedin. You can also follow me on Twitter.

In order to get started with Blockchain read my article published on Hackernoon.

Reference:

[1] Xiaolin Xu, Fahim Rahman, Bicky Shakya, Apostol Vassilev, Domenic Forte, and Mark Tehranipoor. 2019. Electronics Supply Chain Integrity Enabled by Blockchain. ACM Trans. Des. Autom. Electron. Syst. 24, 3, Article 31 (May 2019), 25 pages. DOI: https://doi.org/10.1145/3315571

Altcoin Magazine (@Alt__Magazine) | Twitter