東華大學圖書館 |

The Evolution of Logic Locking: Towards Next Generation Logic Locking Countermeasures.

紀錄類型:	書目-電子資源 : Monograph/item
正題名/作者:	The Evolution of Logic Locking: Towards Next Generation Logic Locking Countermeasures./
作者:	Mardani Kamali, Hadi.
出版者:	Ann Arbor : ProQuest Dissertations & Theses, : 2021,
面頁冊數:	147 p.
附註:	Source: Dissertations Abstracts International, Volume: 83-02, Section: B.
Contained By:	Dissertations Abstracts International83-02B.
標題:	Computer engineering. -
電子資源:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=28548919
ISBN:	9798538128532

The Evolution of Logic Locking: Towards Next Generation Logic Locking Countermeasures.
Mardani Kamali, Hadi.

The Evolution of Logic Locking: Towards Next Generation Logic Locking Countermeasures. - Ann Arbor : ProQuest Dissertations & Theses, 2021 - 147 p.

Source: Dissertations Abstracts International, Volume: 83-02, Section: B.

Thesis (Ph.D.)--George Mason University, 2021.

This item must not be sold to any third party vendors.

The globalization of the design and implementation of integrated circuits has drastically increased, particularly in the past two decades. This is when high-tech companies try (1) to reduce the cost of manufacturing, (2) to access technology that is inclusively available by a limited number of suppliers, (3) to reduce time to market, and (4) to meet the market demand. However, it results in emerging many security threats and trust challenges. Some of these threats include Hardware Trojan insertion, reverse engineering, and IP theft.To combat these threats, numerous Design-for-Trust (DfTr) techniques have been proposed, one of them is logic obfuscation, a.k.a logic locking. In logic locking, the designer adds post-manufacturing programmability into the design controlled by programmable values referred to as the key. The key value is driven from an on-chip tamper-proof non-volatile memory (tpNVM), and it will be initiated after fabrication via a trusted party. The security and the strength of the primitive logic locking techniques have been called into question by various attacks, especially by the Boolean satisfiability (SAT) based attack. To thwart the SAT attack, over the past few years, researchers have investigated different directions, such as point function techniques, cyclic-based locking, and behavioral logic locking. However, many of them are vulnerable to newer attacks.The main aim of this thesis is to open a new direction as a means of logic locking. Unlike almost all previous logic locking solutions that rely on XOR-based locking, we will investigate and evaluate non-XOR-based logic locking solutions, including LUT-based and MUX-based logic locking. We first introduce LUT-Lock as a LUT-based logic locking technique, which relies on some heuristic placement strategies. LUT-Lock is resilient against the existing attacks, especially the SAT attack. However, our comprehensive design space exploration on LUT-based logic locking shows its inefficiency (in terms of overhead) compared to other techniques making this form almost impractical. Then, we introduce Full-Lock as a new MUX-based routing locking solution. We show how MUX-based routing blocks could create SAT-hard instances while the overhead is considerably lower than the LUT-based locking solution. Although Full-Lock guarantees the resiliency against state-of-the-art attacks, we introduce a new attack, called CP&SAT, in which a satisfiability-based routing optimization will be introduced showing how routing-based locking techniques are still vulnerable. With this in mind, we introduce a security-enhanced routing locking technique, called InterLock. Interlock mitigates the weakness of existing routing-based obfuscation techniques against the proposed CP&SAT attack. In InterLock, the routing modules are intercorrelated with actual logic gates. Hence, since the logic is truly twisted with routing all controlled by the key, the adversary cannot convert and model the routing modules using the satisfiability-based routing optimization techniques, and then the CP&SAT attack is no longer applicable to them. We implement InterLock based on three different technologies: (1) transmission-gate (Tgate) CMOS, (2) programmable-via using anti-fuse elements (PVIA), and (3) three-independent-gate field-effect transistors (TIGFET). It helps us to provide a better illustration of the area/delay overhead of routing-based locking. We also show that by implementing in the lower level of abstraction, the area/delay overhead of InterLock could be even below ~10% to make the design resilient against the prevailing attacks at a reasonable area overhead.Since the availability of design-for-testability (DFT) structure, i.e. scan chain pins, is a mandatory requirement of the powerful SAT attack or its derivatives, we also take a step further, and by introducing SCRAMBLE, we evaluate the possibility of using MUX-based routing blocks as a means for locking the DFT. By locking the DFT structure, the SAT attack fails to be applied on SCRAMBLE-locked circuits. We also investigate the modeling/mapping of the logic using small-size memories optimized using the input-multiplexing technique. We will show how the integration of logic in memory and routing blocks will resist different de-obfuscation attacks at low overhead with no test compromising. Apart from locking the DFT structure, we also propose a key-trapped design-for-security (kt-DFS) architecture, which is a DFT blockage mechanism that limits/blocks any unauthorized access to the scan chain. DFT blockage techniques provide resiliency against a wide range of de-obfuscation attacks at lower overhead compared to DFT locking techniques. In kt-DFS, we introduce a new scan chain secure cell, which is designated for safeguarding the logic locking key against any form of key leakage. We will evaluate and compare kt-DFS with other state-of-the-art logic-locking-oriented DFS architectures in terms of overhead, test coverage, and leakage.

ISBN: 9798538128532Subjects--Topical Terms:

621879
Computer engineering.
Subjects--Index Terms:

Design for Trust

The Evolution of Logic Locking: Towards Next Generation Logic Locking Countermeasures.
LDR:06162nmm a2200373 4500 001 2348065
005 20220906075156.5
008 241004s2021 ||||||||||||||||| ||eng d
020 $a 9798538128532
035 $a (MiAaPQ)AAI28548919
035 $a AAI28548919
040 $a MiAaPQ $c MiAaPQ
100 1 $a Mardani Kamali, Hadi. $3 3687385
245 1 4 $a The Evolution of Logic Locking: Towards Next Generation Logic Locking Countermeasures.
260 1 $a Ann Arbor : $b ProQuest Dissertations & Theses, $c 2021
300 $a 147 p.
500 $a Source: Dissertations Abstracts International, Volume: 83-02, Section: B.
500 $a Advisor: Sasan, Avesta.
502 $a Thesis (Ph.D.)--George Mason University, 2021.
506 $a This item must not be sold to any third party vendors.
520 $a The globalization of the design and implementation of integrated circuits has drastically increased, particularly in the past two decades. This is when high-tech companies try (1) to reduce the cost of manufacturing, (2) to access technology that is inclusively available by a limited number of suppliers, (3) to reduce time to market, and (4) to meet the market demand. However, it results in emerging many security threats and trust challenges. Some of these threats include Hardware Trojan insertion, reverse engineering, and IP theft.To combat these threats, numerous Design-for-Trust (DfTr) techniques have been proposed, one of them is logic obfuscation, a.k.a logic locking. In logic locking, the designer adds post-manufacturing programmability into the design controlled by programmable values referred to as the key. The key value is driven from an on-chip tamper-proof non-volatile memory (tpNVM), and it will be initiated after fabrication via a trusted party. The security and the strength of the primitive logic locking techniques have been called into question by various attacks, especially by the Boolean satisfiability (SAT) based attack. To thwart the SAT attack, over the past few years, researchers have investigated different directions, such as point function techniques, cyclic-based locking, and behavioral logic locking. However, many of them are vulnerable to newer attacks.The main aim of this thesis is to open a new direction as a means of logic locking. Unlike almost all previous logic locking solutions that rely on XOR-based locking, we will investigate and evaluate non-XOR-based logic locking solutions, including LUT-based and MUX-based logic locking. We first introduce LUT-Lock as a LUT-based logic locking technique, which relies on some heuristic placement strategies. LUT-Lock is resilient against the existing attacks, especially the SAT attack. However, our comprehensive design space exploration on LUT-based logic locking shows its inefficiency (in terms of overhead) compared to other techniques making this form almost impractical. Then, we introduce Full-Lock as a new MUX-based routing locking solution. We show how MUX-based routing blocks could create SAT-hard instances while the overhead is considerably lower than the LUT-based locking solution. Although Full-Lock guarantees the resiliency against state-of-the-art attacks, we introduce a new attack, called CP&SAT, in which a satisfiability-based routing optimization will be introduced showing how routing-based locking techniques are still vulnerable. With this in mind, we introduce a security-enhanced routing locking technique, called InterLock. Interlock mitigates the weakness of existing routing-based obfuscation techniques against the proposed CP&SAT attack. In InterLock, the routing modules are intercorrelated with actual logic gates. Hence, since the logic is truly twisted with routing all controlled by the key, the adversary cannot convert and model the routing modules using the satisfiability-based routing optimization techniques, and then the CP&SAT attack is no longer applicable to them. We implement InterLock based on three different technologies: (1) transmission-gate (Tgate) CMOS, (2) programmable-via using anti-fuse elements (PVIA), and (3) three-independent-gate field-effect transistors (TIGFET). It helps us to provide a better illustration of the area/delay overhead of routing-based locking. We also show that by implementing in the lower level of abstraction, the area/delay overhead of InterLock could be even below ~10% to make the design resilient against the prevailing attacks at a reasonable area overhead.Since the availability of design-for-testability (DFT) structure, i.e. scan chain pins, is a mandatory requirement of the powerful SAT attack or its derivatives, we also take a step further, and by introducing SCRAMBLE, we evaluate the possibility of using MUX-based routing blocks as a means for locking the DFT. By locking the DFT structure, the SAT attack fails to be applied on SCRAMBLE-locked circuits. We also investigate the modeling/mapping of the logic using small-size memories optimized using the input-multiplexing technique. We will show how the integration of logic in memory and routing blocks will resist different de-obfuscation attacks at low overhead with no test compromising. Apart from locking the DFT structure, we also propose a key-trapped design-for-security (kt-DFS) architecture, which is a DFT blockage mechanism that limits/blocks any unauthorized access to the scan chain. DFT blockage techniques provide resiliency against a wide range of de-obfuscation attacks at lower overhead compared to DFT locking techniques. In kt-DFS, we introduce a new scan chain secure cell, which is designated for safeguarding the logic locking key against any form of key leakage. We will evaluate and compare kt-DFS with other state-of-the-art logic-locking-oriented DFS architectures in terms of overhead, test coverage, and leakage.
590 $a School code: 0883.
650 4 $a Computer engineering. $3 621879
650 4 $a Design. $3 518875
650 4 $a Integrated circuits. $3 584067
650 4 $a Computer science. $3 523869
650 4 $a Computer security. $3 540555
650 4 $a Manufacturing. $3 3389707
650 4 $a Investigations. $3 3561159
650 4 $a Semiconductors. $3 516162
650 4 $a Logic. $3 529544
650 4 $a Trust. $3 536251
653 $a Design for Trust
653 $a Hardware security
653 $a Reverse engineering
653 $a VLSI design
690 $a 0464
690 $a 0389
690 $a 0984
690 $a 0395
710 2 $a George Mason University. $b Electrical and Computer Engineering. $3 3193956
773 0 $t Dissertations Abstracts International $g 83-02B.
790 $a 0883
791 $a Ph.D.
792 $a 2021
793 $a English
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=28548919