No abstract

No abstract

Kernel minimization has already been established as a practical approach to reducing the trusted computing base. Existing solutions have largely focused on whole-system profiling -generating a globally minimum kernel image that is being shared by all applications. However, since different applications use only part of the kernel's code base, the minimized kernel still includes an unnecessarily large attack surface. Furthermore, once the static minimized kernel is generated, it is not flexible enough to adapt to an altered execution environment (e.g., new workload). FACE-CHANGE is a virtualization-based system to facilitate dynamic switching at runtime among multiple minimized kernels, each customized for an individual application. Based on precedent profiling results, FACE-CHANGE transparently presents a customized kernel view for each application to confine its reachability of kernel code. In the event that the application exceeds this boundary, FACE-CHANGE is able to recover the missing code and backtrace its attack/exception provenance to analyze the anomalous behavior.

Abstract-Previous research on virtual machine introspection proposed "out-of-box" approach by moving out security tools from the guest operating system. However, compared to the traditional "in-the-box" approach, it remains a challenge to obtain a complete semantic view due to the semantic gap between the guest VM and the hypervisor.In this paper, we present Process Implanting, a new active VM introspection framework, to narrow the semantic gap by implanting a process from the host into the guest VM and executing it under the cover of an existing running process. With the protection and coordination from the hypervisor, the implanted process can run with a degree of stealthiness and exit gracefully without leaving negative impact on the guest operating system. We have designed and implemented a proof-of-concept prototype on KVM which leverages hardware virtualization. We also propose and demonstrate application scenarios for Process Implanting in the area of VM security.

Real-time microcontrollers have been widely adopted in cyber-physical systems that require both real-time and security guarantees. Unfortunately, security is sometimes traded for real-time performance in such systems. Notably, memory isolation, which is one of the most established security features in modern computer systems, is typically not available in many real-time microcontroller systems due to its negative impacts on performance and violation of real-time constraints. As such, the memory space of these systems has created an open, monolithic attack surface that attackers can target to subvert the entire systems. In this paper, we present MINION, a security architecture that intends to virtually partition the memory space and enforce memory access control of a real-time microcontroller. MINION can automatically identify the reachable memory regions of realtime processes through off-line static analysis on the system's firmware and conduct run-time memory access control through hardware-based enforcement. Our evaluation results demonstrate that, by significantly reducing the memory space that each process can access, MINION can effectively protect a microcontroller from various attacks that were previously viable. In addition, unlike conventional memory isolation mechanisms that might incur substantial performance overhead, the lightweight design of MINION is able to maintain the real-time properties of the microcontroller.

No abstract

Distributed collaborative learning (DCL) paradigms enable building joint machine learning models from distrusting multi-party participants. Data confidentiality is guaranteed by retaining private training data on each participant's local infrastructure. However, this approach to achieving data confidentiality makes today's DCL designs fundamentally vulnerable to data poisoning and backdoor attacks. It also limits DCL's model accountability, which is key to backtracking the responsible "bad" training data instances/contributors. In this paper, we introduce CALTRAIN 1 , a Trusted Execution Environment (TEE) based centralized multi-party collaborative learning system that simultaneously achieves data confidentiality and model accountability. CALTRAIN enforces isolated computation on centrally aggregated training data to guarantee data confidentiality. To support building accountable learning models, we securely maintain the links between training instances and their corresponding contributors. Our evaluation shows that the models generated from CALTRAIN can achieve the same prediction accuracy when compared to the models trained in non-protected environments. We also demonstrate that when malicious training participants tend to implant backdoors during model training, CALTRAIN can accurately and precisely discover the poisoned and mislabeled training data that lead to the runtime mispredictions.

No abstract