專題文章

2019/10/14

PUF: A Crucial Technology for AI and IoT

By PUFsecurity, the subsidiary of eMemory

Introducing AIoT Trends

Artificial intelligence of things (AIoT) is a new trend that combines artificial intelligence (AI) with the internet of things (IoT) to create networks of digital devices that communicate and process data. While IoT creates vast connections, AI makes these devices come alive. Here’s one example: an IP camera system can be used for apartment security. Yet, without AI, people need to monitor video from the system in real-time in order to respond to emergencies. With AI, IP cameras can recognize risks automatically and send alerts. Certainly, AIoT promises to grow rapidly and give rise to new high-value products in the same way that the internet led to the creation of so many huge businesses. Those who enter the production of AIoT devices will be poised to tap a vast new market. IoT devices today number in the billions. These small, connected gadgets include electronic devices and appliances networked together and communicating over internet protocols (IPs).

Yet, adding AI to IoT has created new security challenges.

The Challenges of AIoT

One of the key challenges for AIoT is the protection of AI assets. AI functions often need to detect, evaluate and respond in real time. As a result, a critical security concern is the fact that internal databases and interfaces for AI are not suitable for encryption because such an operation would demand too much time and resources. However, big data and interface designs are all proprietary information that need to be securely protected. The data needed by AI systems is typically so large that it usually be stored in an external non-volatile memory (NVM), thereby exposing it to hacking risks that are increasing worldwide.

Meanwhile, in addition to the “internal” security issues of AIoT systems, the external challenges of AIoT security have also increased. Nearly two million cyberattacks in 2018 resulted in more than $45 billion in losses worldwide as governments struggled with ransomware and other malicious incidents.

The Internet Society’s Online Trust Alliance (OTA), which identifies and promotes security and privacy best practices that build consumer confidence in the internet, said in its Cyber Incident & Breach Trends Report that the financial impact of ransomware rose by 60%, losses from business email compromise (BEC) doubled, and cryptojacking incidents more than tripled in 2018.

It’s clear that while security concerns remain unresolved, the deployment of AIoT devices will increase attack vectors for intrusions. Therefore, we say that for AIoT devices, PUF-based hardware security is the perfect solution. With PUF, the existing tradeoff of security for performance is eliminated.

How PUF Solves AIoT Security Concerns

NeoPUF is a hardware security technology based on the physical unclonable variations occurring in the silicon manufacturing process. The underlying benefit of using a PUF (Physical Unclonable Function) in cryptography is its “uniqueness” and “unpredictability”. With eMemory’s NeoPUF, a chip can generate truly random sequences that can be used in applications with high security requirements. Our innovative technology can enable multi-layered security and resolve PUF-related concerns such as the additional costs of complicated ECC (Error Correction Code). The random number extracted via NeoPUF is so unique and unclonable that it can be used as a silicon “fingerprint” for a wide range of security purposes, including encryption, identification, authentication and security key generation.

The dimension of attacks on AIoT include “data and firmware attacks”, “transmission attacks” and “data integrity attacks”. We mentioned that complex encryption and decryption are impractical for the protection of AI assets. PUF has become a relatively simple and fast solution for security. Below are a few application scenarios that could help you have a clearer picture of how PUF solves AIoT security concerns...more