Adaptive Chosen-Plaintext Collision Attack on Masked AES in Edge Computing

Abstract

Edge computing handles delay-sensitive data and provides real-time feedback, while it brings data security issues to edge devices (such as IoT devices and edge servers). Side-channel attacks main threaten to these devices. Collision attack represents a powerful category of side-channel analysis in extracting security information from embedded cryptographic algorithms. Since its proposition in 2003, plenty of collision detection algorithms are presented, most of which enumerate all the values of target plaintext byte to find a collision. In this paper, we establish a relation between ''Euclidean distance between traces'' and ''Hamming distance between values,'' and take advantage of the distance information leaked from the power traces of encrypting an adaptively chosen plaintext to reduce the candidate plaintext space. Consequently, the collision is detected at a high pace. Moreover, this improvement is fault-tolerant, and its self-correction feature promotes the efficiency of attacks based on our method significantly. We take AES implemented with masks, which is usually employed in edge computing devices, for instance, to introduce our method and conduct experiments to verify its efficiency. According to the experimental results, for whole key recovery attacks, our method requires only 26.5% plaintexts, 32.2% traces, and much less than 10% computations of the collision-correlation attack launched by Clavier et al. INDEX TERMS Adaptive chosen-plaintext collision attack, edge computing, masking, the least square method. The associate editor coordinating the review of this manuscript and approving it for publication was Chien-Ming Chen. the framework of side-channel attacks on edge computing devices. Therefore, it is necessary to evaluate their security with side-channel analysis. Since Kocher [11] proposed timing attacks in 1996, cryptanalysts have considered various kinds of side-channel information and presented plenty techniques such as differential power analysis [12], collision attack [13], correlation power analysis [14], template attack [15], fault sensitivity analysis [16]-[18], etc. In this paper, we focus on improving the efficiency of collision attack when applied to masked AES in edge computing. Collision attack was firstly proposed by Schramm et al. [13] in 2003. Subsequently, a similar attack on AES was proposed to detect collisions in the first MixColumn operation [19]. In 2007, Bogdanov proposed

CMXsafe: A Proxy Layer for Securing Internet-of-Things Communications

“…Whole key recovery attack. As a result, the traffic from IoT devices using outdated security implementations for communications is subject to efficient key recovery attacks [44] [45].…”

Secure File Operations: Using Advanced Encryption Standard for Strong Data Protection

“…This importance is further underpinned by the incorporation of inventive data reading protocols and compression methods, all working collectively to attain this essential goal. A sophisticated approach to mitigate side-channel attacks, specifically collision attacks, in the context of edge computing was introduced by Ding et al [13]. Their approach, which capitalizes on the correlation between distance metrics and the reduction of plaintext candidate space, leads to a substantial boost in the efficiency of their proposed technique.…”

Secured Optimized Resource Allocation in Mobile Edge Computing

“…(1) Device RAM Memory (Rm), Result of Offloadies (Rs), ( 2) Energy (E), FCaompatibleDevice () is a function in Android to check for RAM, (3) Compatible Device (Cd), CPU, and Device minimum requirements. (4) Input: (Cr) (5) for each MEC Server do (6) Output: {Rm, E, CPU} (7) end for (7) for each Device d do (8) Calculate the Compatibility check using FCompatibleDevice (), add the device to the Cd list (9) end for (10) for each Cd do (11) Rs � Output {task} (12) end for ND.add (D {i}) (7) end if (8) if (i < DL.size() then (9) Move to Step 3 (10) Output {ND} (11) end if (12) MES Input {NCR} (13) end for (14) for each ND do (15) if ND {i}.ram>3 GB && ND {i}.CPU>2.4 GHz then (16) CD.add (ND {i}) (17) end if (18) if i< CD.size() then (19) Move to Step 9 (20) end if (21) Output {CD} (22) Input {T} (23) Calculate Result using function Compute_Result () (24) Output {Result} (25) end for ALGORITHM 3: Algorithm for optimization.…”

“…e ploy of [23] is that it does not consider the offloading task in the network easily and more securely. e depraved thing of [24] is that it does not consider more instructions to make the system more secure. e downside of the study is that it does not consider cryptography techniques to make the network more secure [25].…”