Cyber-security is now a critical concern in a wide range of embedded computing modules, communications systems, and connected devices. These devices are used in medical electronics, automotive systems, power grid systems, robotics, and avionics. The general consensus today is that conventional approaches and software-only schemes are not sufficient to provide desired security protections and trustworthiness. Comprehensive hardware-software security solutions so far have remained elusive. One major challenge is that in current system-on-chip (SoCs) designs, processing elements (PEs) and executable codes with varying levels of trust, are all integrated on the same computing platform to share resources. This interdependency of modules creates a fertile attack ground and represents the Achilles’ heel of heterogeneous SoC architectures. The salient research question addressed in this dissertation is “can one design a secure computer system out of non-secure or untrusted computing IP components and cores?”. In response to this question, we establish a generalized, user/designer-centric set of design principles which intend to advance the construction of secure heterogeneous multi-core computing systems. We develop algorithms, models of computation, and hardware security primitives to integrate secure and non-secure processing elements into the same chip design while aiming for: (a) maintaining individual core’s security; (b) preventing data leakage and corruption; (c) promoting data and resource sharing among the cores; and (d) tolerating malicious behaviors from untrusted processing elements and software applications. The key contributions of this thesis are: 1. The introduction of a new architectural model for integrating processing elements with different security and trust levels, i.e., secure and non-secure cores with trusted and untrusted provenances; 2. A generalized process isolation design methodology for the new architecture model that covers both the software and hardware layers to (i) create hardware-assisted virtual logical zones, and (ii) perform both static and runtime security, privilege level and trust authentication checks; 3. A set of secure protocols and hardware root-of-trust (RoT) primitives to support the process isolation design and to provide the following functionalities: (i) hardware immutable identities – using physical unclonable functions, (ii) core hijacking and impersonation resistance – through a blind signature scheme, (iii) threshold-based data access control – with a robust and adaptive secure secret sharing algorithm, (iv) privacy-preserving authorization verification – by proposing a group anonymous authentication algorithm, and (v) denial of resource or denial of service attack avoidance – by developing an interconnect network routing algorithm and a memory access mechanism according to user-defined security policies. 4. An evaluation of the security of the proposed hardware primitives in the post-quantum era, and possible extensions and algorithmic modifications for their post-quantum resistance. In this dissertation, we advance the practicality of secure-by-construction methodologies in SoC architecture design. The methodology allows for the use of unsecured or untrusted processing elements in the construction of these secure architectures and tries to extend their effectiveness into the post-quantum computing era.

A Cyber-Physical System (CPS) is a large-scale, distributed, embedded system, consisting of various components that are glued together to realize control, computation and communication functions. Although these systems are complex, they are ubiquitous in the Internet of Things (IoT) era of autonomous vehicles/drones, smart homes, smart grids, etc. where everything is connected. These systems are vulnerable to unauthorized penetration due to the absence of proper security features and safeguards to protect important information. Examples such as the typewriter hack involving subversive chips resulting in leakage of keystroke data and hardware backdoors crippling anti-aircraft guns during an attack demonstrate the need to protect all system functions. With more focus on securing a system, trust in untrusted components at the integration stage is of a higher priority. This work builds on a red-black security system, where an architecture testbed is developed with critical and non-critical IP cores and subjected to a variety of Hardware Trojan Threats (HTTs). These attacks defeat the classic trusted hardware model assumptions and demonstrate the ability of Trojans to evade detection methods based on physical characteristics. A novel metric is defined for hardware Trojan detection, termed as HTT Detectability Metric (HDM) that leverages a weighted combination of normalized physical parameters. Security analysis results show that using HDM, 86% of the implemented Trojans were detected as compared to using power consumption, timing variation and resource utilization alone. This led to the formulation of the security requirements for the development of a novel, distributed and secure methodology for enhancing trust in systems developed under untrusted environments called FIDelity Enhancing Security (FIDES). FIDES employs a decentralized information flow control (DIFC) model that enables safe and distributed information flows between various elements of the system such as IP cores, physical memory and registers. The DIFC approach annotates/tags each data item with its sensitivity level and the identity of the participating entities during the communication. Trust enhanced FIDES (TE-FIDES) is proposed to address the vulnerabilities arising from the declassification process during communication between third-party soft IP cores. TE-FIDES employs a secure enclave approach for preserving the confidentiality of the sensitive information in the system. TE-FIDES is evaluated by targeting an IoT-based smart grid CPS application, where malicious third-party soft IP cores are prevented from causing a system blackout. The resulting hardware implementation using TE-FIDES is found to be resilient to multiple hardware Trojan attacks. / Ph. D. / The Internet-of-Things (IoT) has emerged as one of the most innovative multidisciplinary paradigms combining heterogeneous sensors, software architectures, embedded hardware systems, and data analytics. With the growth in deployment of IoT systems, security of the sensors and trustworthiness of the data exchanged is of paramount significance. IoT security approaches are derived from the vulnerabilities existing in cyber-physical systems (CPS) and the countermeasures designed against them. An unauthorized penetration due to the absence of safeguards can cripple the system and leak sensitive data. This dissertation studies the vulnerabilities posed due to the presence of hardware Trojans in such IoT-based CPS. FIDelity Enhancing Security (FIDES), named after the Greek Goddess of Trust, is a novel, distributed and secure methodology proposed to address the security requirements and enhance trust of systems developed in untrusted environments. FIDES utilizes a distributed scheme that monitors the communication between the Intellectual Property (IP) cores using tags. Trust Enhanced FIDES (TE-FIDES) is proposed to reduce the vulnerabilities arising from the declassification process of the third-party soft IP cores. TE-FIDES employs a secure enclave approach for preserving the integrity of the sensitive information in the system. In addition, TE-FIDES also uses a trust metric to record snapshots of each IP core’s state during the declassification process. TE-FIDES is evaluated by mapping an IoT-based CPS application and subjecting it to a variety of hardware Trojan attacks. The performance costs for resilient and trustworthy operation of the TE-FIDES implementation are evaluated and TE-FIDES proves to be resilient to the attacks with acceptable cyber costs.

The basis of today’s security systems is the trust and confidence that the system will behave as expected and are in a known good trusted state. The trust is built from hardware and software elements that generates a chain of trust that originates from a trusted known entity. Leveraging hardware, software and a mandatory access control policy technology is needed to create a trusted measurement environment. Employing a control layer (hypervisor or microkernel) with the ability to enforce a fine grained access control policy with hyper call granularity across multiple guest virtual domains can ensure that any malicious environment to be contained. In my research, I propose the use of radium's Asynchronous Root of Trust Measurement (ARTM) capability incorporated with a secure mandatory access control policy engine that would mitigate the limitations of the current hardware TPM solutions. By employing ARTM we can leverage asynchronous use of boot, launch, and use with the hypervisor proving its state and the integrity of the secure policy. My solution is using Radium (Race free on demand integrity architecture) architecture that will allow a more detailed measurement of applications at run time with greater semantic knowledge of the measured environments. Radium incorporation of a secure access control policy engine will give it the ability to limit or empower a virtual domain system. It can also enable the creation of a service oriented model of guest virtual domains that have the ability to perform certain operations such as introspecting other virtual domain systems to determine the integrity or system state and report it to a remote entity.

<p>This thesis presents a comparative study of four Network Access Control (NAC) technologies; Trusted Network Connect by the Trusted Computing group, Juniper Networks, Inc.’s Unified Access Control, Microsoft Corp.’s Network Access Protection, and Cisco Systems Inc.’s Network Admission Control. NAC is a vision, which utilizes existing solutions and new technologies to provide assurance that any device connecting to a network policy domain is authenticated and is subject to the network’s policy enforcement. Non-compliant devices are isolated until they have been brought back to a complaint status. We compare the NAC technologies in terms of architectural and functional features they provide.</p><p>There is a race of NAC solutions in the marketplace, each claiming their own definition and terminology, making it difficult for customers to adopt such a solution, resulting in much uncertainty. The NAC paradigm can be classified into two categories: the first category embraces open standards; the second follows proprietary standards. By selecting these architectures, we cover a representative set of proprietary and open standards-based NAC technologies.</p><p>This study concludes that there is a great need for standardization and interoperability of NAC components and that the four major solution proposals that we studied fall short of the desired interoperability. With standards, customers have the choice to adopt solution components from different vendors, selecting, what is commonly referred to as the best of breed. One example for a standard technology that all four NAC technologies that we studied did adopt is the IEEE’s 802.1X port-based access control technology. It is used to control endpoint device access to the network.</p><p>One shortcoming that most NAC architectures (with the exception of Trusted Network Connect) have in common, is the lack of a strong root-of-trust. Without it, clients’ compliance measurements cannot be trusted by the policy server whose task is to assess each client’s policy compliance.</p>

This thesis presents a comparative study of four Network Access Control (NAC) technologies; Trusted Network Connect by the Trusted Computing group, Juniper Networks, Inc.’s Unified Access Control, Microsoft Corp.’s Network Access Protection, and Cisco Systems Inc.’s Network Admission Control. NAC is a vision, which utilizes existing solutions and new technologies to provide assurance that any device connecting to a network policy domain is authenticated and is subject to the network’s policy enforcement. Non-compliant devices are isolated until they have been brought back to a complaint status. We compare the NAC technologies in terms of architectural and functional features they provide. There is a race of NAC solutions in the marketplace, each claiming their own definition and terminology, making it difficult for customers to adopt such a solution, resulting in much uncertainty. The NAC paradigm can be classified into two categories: the first category embraces open standards; the second follows proprietary standards. By selecting these architectures, we cover a representative set of proprietary and open standards-based NAC technologies. This study concludes that there is a great need for standardization and interoperability of NAC components and that the four major solution proposals that we studied fall short of the desired interoperability. With standards, customers have the choice to adopt solution components from different vendors, selecting, what is commonly referred to as the best of breed. One example for a standard technology that all four NAC technologies that we studied did adopt is the IEEE’s 802.1X port-based access control technology. It is used to control endpoint device access to the network. One shortcoming that most NAC architectures (with the exception of Trusted Network Connect) have in common, is the lack of a strong root-of-trust. Without it, clients’ compliance measurements cannot be trusted by the policy server whose task is to assess each client’s policy compliance.