Tools and frameworks for computer security alternatives

点击翻译按钮获取中文摘要

Our society is heavily reliant on our computers, devices, and networks, but there are still any number of problems making our computers secure, even­ after decades of computer security research. Given the sheer volume of research in computer security, relatively little of it has translated into new, improved defenses that are deployed for the protection of end users. Arguably, computer users have seen stagnation or at best marginal improvements in their defenses, while at the same time the number of threats has increased dramatically. The long-­term goal of my research program is to find novel methods for security that break from this status quo, radically in some cases, to find entirely new alternatives to the current state of computer security that users experience, and make users more secure. Practically speaking, I will be realizing this long­-term goal through the construction of software tools and frameworks, meaning that the research work is not just seen in academic publications, but will result in software usable by others.     In the shorter term, for the tenure of this grant, I will be looking at advancing my long-­term goal in three areas: code obfuscation (protecting computer code), program execution monitoring and sampling (watching computer code), and computer systems (new ways to create and express computer code). These areas and the initiatives within them are not arbitrarily chosen. One guiding principle that underlies the security alternatives I'm proposing is how to employ the abundance of resources (e.g., computing power, network bandwidth, Internet-­connected users) we either have presently, or will have in the next 5­-10 years. A number of initiatives are being undertaken as a deliberate strategy of risk mitigation; this is a necessary consideration because what I am proposing is higher-risk work and not just incremental research.     One novelty here is in identifying and challenging deeply­ engrained assumptions related to security, e.g., anti-­virus software must work a certain way, a solution must work on x86/ARM/Windows/Mac..., computer programming must be done a certain way. Proffering self-same defenses is simply not working, and effort needs to be expended on seeking revolutionary approaches to security rather than incrementally less-bad ones. The security of our computers must be improved, and leading the way with new approaches instead of following others is a good investment for Canada to make.