LOOPLock is the state-of-the-art cyclic logic locking method in hardware security. LOOPLock is able to invalidate SAT Attack, Removal Attack, and CycSAT simultaneously by introducing two types of cycle pairs in a circuit. In this work, we analyze LOOPLock’s locking mechanism and propose an attacking approach based on locking structure analysis. Furthermore, to defend the new attack, we propose LOOPLock 2.0, which strengthens the original cyclic logic locking method—LOOPLock. Experimental results show the efficiency and effectiveness of the proposed attacking approach to LOOPLock and the high defense capability of LOOPLock 2.0.

Semantic relations are core to how humans understand and express concepts in the real world using language.Recently, there has been a thread of research aimed at modeling these relations by learning vector representations from text corpora.Most of these approaches focus strictly on leveraging the co-occurrences of relationship word pairs within sentences.In this paper, we investigate the hypothesis that examples of a lexical relation in a corpus are fundamental to a neural word embedding's ability to complete analogies involving the relation.Our experiments, in which we remove all known examples of a relation from training corpora, show only marginal degradation in analogy completion performance involving the removed relation.This finding enhances our understanding of neural word embeddings, showing that co-occurrence information of a particular semantic relation is the not the main source of their structural regularity.

SAT Attack, CycSAT, and Removal Attack have demonstrated their abilities to break most existing logic locking methods. In this article, we propose a new cyclic logic locking method to invalidate these attacks simultaneously. Our main intention is to create noncombinational cycles to lock a circuit. Specifically, the noncombinational behavior in the noncombinational cycles that is unobservable at the primary outputs (POs) needs to be preserved when the correct key-vector is fed to resist CycSAT, and the noncombinational behavior in the noncombinational cycles affecting POs needs to be preserved when the incorrect key-vector is fed to invalidate SAT Attack. Furthermore, some nodes will be removed when applying our locking method, which is able to defend Removal Attack. The experimental results show the effectiveness and low area overhead of the proposed method.