Using deep learning to combine static and dynamic power analyses of cryptographic circuits

Xu, Jiming; Heys, Howard M.

doi:10.1002/cta.2623

Cited by 9 publications

(8 citation statements)

References 46 publications

(81 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…LSTM has three type of gates called forget gate (

{normalΓ}_{f}^{t}

), update gate (

{normalΓ}_{u}^{t}

), and output gate (

{normalΓ}_{0}^{t}

), which are calculated using Equations (1)–(3). Forget gate has been used to clear the memory which is no longer use, and update gate is used to update the memory with new information, whereas output gate has been used to decide which output needs to be used 29,30 . Output of LSTM model has been calculated using Equations (4)–(7).…”

Section: Methodsmentioning

confidence: 99%

See 1 more Smart Citation

Short‐term electric power load forecasting using factor analysis and long short‐term memory for smart cities

Veeramsetty¹,

Chandra²,

Salkuti

2020

Circuit Theory & Apps

View full text Add to dashboard Cite

Electric load estimation is an important activity for electrical power system operators to operate the system stably and optimally. This paper develops a machine learning model with a long short-term memory and a factor analysis to predict the load at a specific hour of the day on an electrical power substation. Historical load data from the 33-/11-kV substation near Kakatiya University in Warangal are taken at each hour of the day for the period from September 2018 to November 2018. A new long short-term memory architecture with factor analysis is being designed based on the approach used to predict substation loads by simulation in Microsoft Azure Notebooks. Based on the study, it was found that the proposed design predicts loads with good accuracy.

show abstract

“…LSTM has three type of gates called forget gate (

{normalΓ}_{f}^{t}

), update gate (

{normalΓ}_{u}^{t}

), and output gate (

{normalΓ}_{0}^{t}

Section: Methodsmentioning

confidence: 99%

“…Forget gate has been used to clear the memory which is no longer use, and update gate is used to update the memory with new information, whereas output gate has been used to decide which output needs to be used. 29,30 Output of LSTM model has been calculated using Equations (4)- (7).…”

Section: Long Short-term Memorymentioning

confidence: 99%

Short‐term electric power load forecasting using factor analysis and long short‐term memory for smart cities

Veeramsetty¹,

Chandra²,

Salkuti

2020

Circuit Theory & Apps

View full text Add to dashboard Cite

show abstract

“…It can be used to combine two different neural networks which are trained for the same task but on different datasets. Two fusion techniques are commonly used in existing works, one is called early fusion and another one is late fusion [24], [25]. Early fusion merges layers of different neural networks at an early stage while late fusion merges layer lately.…”

Section: Construction Of Multi-input Modelmentioning

confidence: 99%

Multi-Leak Deep-Learning Side-Channel Analysis

Wang

2022

IEEE Access

View full text Add to dashboard Cite

Deep Learning Side-Channel Attacks (DLSCAs) have become a realistic threat to implementations of cryptographic algorithms, such as Advanced Encryption Standard (AES). By utilizing deeplearning models to analyze side-channel measurements, the attacker is able to derive the secret key of the cryptographic algorithm. However, when traces have multiple leakage intervals for a specific attack point, the majority of existing works train neural networks on these traces directly, without a appropriate preprocess step for each leakage interval. This degenerates the quality of profiling traces due to the noise and nonprimary components. In this paper, we first divide the multi-leaky traces into leakage intervals and train models on different intervals separately. Afterwards, we concatenate these neural networks to build the final network, which is called multi-input model. We test the proposed multi-input model on traces captured from STM32F3 microcontroller implementations of AES-128 and show a 2-fold improvement over the previous single-input attacks.INDEX TERMS AES, deep learning, multiple leakage, multi-input model, side-channel attacks.

show abstract

“…In addition, the LFDRNN evaluation from input to output is very fast 23 . Moreover, LFDRNN can better approximate the complex nonlinear device input–output relationships compared to the conventional RNN structure 17,46 . In this paper, we have implemented a boost converter via an EH system, and LFDRNN is investigated and used to model its nonlinear behavior.…”

Section: Local Feedback Deep Recurrent Neural Networkmentioning

confidence: 99%

Modeling and implementation of nonlinear boost converter using local feedback deep recurrent neural network for voltage balancing in energy harvesting applications

Noohi

Mirvakili

Sadrossadat

2021

Circuit Theory & Apps

View full text Add to dashboard Cite

Balancing the voltage of series connected supercapacitors is a necessity. Various passive and active balancing techniques are reported for alleviating the problems of leakage and overvoltage. In this paper, a novel active balancing approach based on boost converter is presented leading to the implementation of a piezoelectric energy harvesting (EH) system. Besides, this nonlinear boost converter is designed, implemented, and modeled using a new macromodeling approach. In this regard, data measured by the implemented boost converter passed through local feedback deep recurrent neural networks (LFDRNNs), in order to model the nonlinear behavior of this converter, and this model can be used to design the EH system. LFDRNN can be trained directly using the input–output waveform samples of the main circuit without knowing its internal details, and the obtained model has similar accuracy compared to the original circuit. The main focus of this paper is the new LFDRNN macromodeling method which is associated with the boost converter‐based active balancing technique. Our experimental results show that LFDRNN extends the ability of conventional neural network‐based models to express the dynamic behavior of nonlinear circuits while increasing the accuracy. Additionally, LFDRNN‐based models are much faster than existing models in simulation tools.

show abstract

Using deep learning to combine static and dynamic power analyses of cryptographic circuits

Cited by 9 publications

References 46 publications

Short‐term electric power load forecasting using factor analysis and long short‐term memory for smart cities

Short‐term electric power load forecasting using factor analysis and long short‐term memory for smart cities

Multi-Leak Deep-Learning Side-Channel Analysis

Modeling and implementation of nonlinear boost converter using local feedback deep recurrent neural network for voltage balancing in energy harvesting applications

Contact Info

Product

Resources

About