Using virtualization to implement a scalable trusted execution environment in secure SoCs
Static-based approaches for embedded system security which define secure and non-secure zones by partitioning separate hardware subsystems for each zone have been effective so far. However, more scalable and cost-effective approaches are required to address the needs of newer devices running multiple applications over several secure zones. This paper examines the use of virtualization for creating the requisite scalable trusted execution environment for secure embedded systems. It also provides an overview of MIPS’ existing and forthcoming solutions for virtualization-based security.
1. The Need for Better Security in Connected Embedded Devices
Throughout the evolution of the internet, we have seen ongoing innovation as new services and applications drive growth in network capacity, and the resulting excess capacity then spurns the development of new services and applications. This trend continues in alternating cycles. It was only a few years ago that the telecom industry was talking about the “bandwidth glut” and the lack of the killer app that could consume the tremendous unused capacity in dark fiber. These discussions are well behind us, and we are clearly once again in the “services driving bandwidth growth” phase, with a plethora of killer apps including video on demand, network gaming, video conferencing, e-commerce, telecommuting, cloud storage, social media and others. Broadband connections to the home have risen dramatically, and the number of connected devices, including smart phones, tablets, gateways and set top boxes (STBs), continues to grow at a blistering pace. Adding to this evolution and bandwidth consumption will be connected cars and the “internet of things” where even municipal and industrial machines will be increasingly connected to ubiquitous broadband IP networks.
With this growth in the number of applications and connected devices comes a need for increased security. Security related concerns are skyrocketing among service providers, consumers and even governments as usage increases. Digital Rights Management (DRM) is a critical concern for film and TV studios in protecting their content from piracy—a problem that already leads to billions of dollars in losses every year. Traditional STBs have been the primary means for delivering video to consumers, but the risks have increased with more people streaming video to their smartphones, tablets, gaming systems and non-traditional over-the-top (OTT) STBs, which connect directly to broadband links.
Telecommuting has also increased with the availability of fat pipes on both wired and wireless broadband connections, which make working from home practically seamless from an IT standpoint. Several enterprises even support the use of mobile enterprise applications deployed on remote laptops or tablets. Obviously, the risk of the leak of confidential enterprise information is heightened as users mix work and personal use on their connected devices. We are also seeing increased deployment of smart gateways, which provide not only a broadband connection for the consumer, but also provide utility information—such as gas and electricity consumption—to a public utility company.
Governments are increasingly concerned about homeland security, as would-be hackers may be able to access public utility IT systems if the customer premise equipment (CPE) is not secure. There are many other examples, but the point is that there has never been more risk for compromised security than today, with the widespread use of connected devices. Therefore, embedded devices must be able to effectively and reliably isolate secure applications from non-secure applications while meeting the appropriate levels of functionality, performance, cost, and power consumption.