Drexel ICE Lab

ICCD 2025 Tutorial on Machine Learning for Automated Physical Design

2025-11-18T00:00:00+00:00

Prof. Ioannis Savidis presented the tutorial Machine Learning for Automated Physical Design at the 2025 IEEE International Conference on Computer Design (ICCD) in Dallas, Texas.

The tutorial highlighted ML-driven methodologies for modern VLSI physical design flows developed in the ICE Lab.

ISCAS 2025 Tutorial on ML-AI Techniques for Analog EDA

2025-05-24T00:00:00+00:00

Prof. Ioannis Savidis presented the tutorial Emerging ML-AI Techniques for Analog EDA at the 2025 IEEE International Symposium on Circuits and Systems (ISCAS) in London, UK.

The tutorial covered machine learning approaches for analog electronic design automation, building on the ICE Lab’s ongoing work in ML-guided physical design and analog circuit sizing.

Vaibhav Venugopal Rao Completes Ph.D. on Analog Circuit Security

2024-03-15T00:00:00+00:00

Vaibhav Venugopal Rao successfully defended his Ph.D. dissertation Enhancing Analog Circuit Security Through Obfuscation at Drexel University.

His thesis advances parameter obfuscation and security analysis techniques for analog integrated circuits developed in the ICE Lab.

TVLSI Paper on Hidden Costs of Analog Deobfuscation Attacks

2023-11-01T00:00:00+00:00

The paper Hidden Costs of Analog Deobfuscation Attacks by V. V. Rao, K. Juretus, and Prof. Ioannis Savidis was published in the IEEE Transactions on Very Large Scale Integration (VLSI) Systems.

The work analyzes the security vulnerabilities and hidden costs associated with deobfuscation attacks on analog intellectual property protection techniques.

Two Papers Presented at ISCAS 2019

2019-05-28T00:00:00+00:00

The papers entitled Increasing the SAT Attack Resiliency of In-Cone Logic Locking and Mesh Based Obfuscation of Analog Circuit Properties were presented at ISCAS 2019.

Prof. Ioannis Savidis Awarded 2018 NSF CAREER Award

2018-06-18T00:00:00+00:00

Prof. Ioannis Savidis has been awarded a prestigious NSF CAREER Award for “Parameter Obfuscation: A Novel Methodology for the Protection of Analog Intellectual Property.”

The Joseph and Shirley Carleone Endowed Fellowship is awarded to Divy Pathak for the academic year 2017-2018.

2018-05-29T00:00:00+00:00

The Joseph and Shirley Carleone Endowed Fellowship for the academic year 2017-2018 is awarded to Divy Pathak by the College of Engineering, Drexel University.

Prof. Ioannis Savidis Wins IEEE Philadelphia Section Award

2018-03-28T00:00:00+00:00

Prof. Ioannis Savidis wins the 2018 IEEE Philadelphia Section Delaware Valley Young Electrical Engineer of the Year Award. Citation: “For contributions in hardware security and trust. Specifically, for the obfuscation of digital and analog circuits through novel algorithms and methodologies that secure intellectual property from reverse engineering, cloning, and counterfeiting.”

Two Papers Presented at GOMACTech 2018

2018-03-28T00:00:00+00:00

Prof. Ioannis Savidis presented two ICE Lab papers at the Government Microcircuit Applications & Critical Technology Conference in Miami, Florida held on March 12-14, 2018: 1) Kyle Juretus and Ioannis Savidis, “Enhanced Circuit Security Through Hidden State Transitions.” Abstract: A technique is developed that uses the state space of an integrated circuit (IC) to increase security. Timing path dependencies and coupling capacitance are utilized to create hidden state transitions that are not observable after netlist extraction. Temporal based transitions are also introduced to shift key dependencies into the time domain. The state space dependencies of the IC serve as a means to counter the SAT attack. Implementing temporal based transitions increases the area of the circuit by 68.42%, the power by 43.17%, and does not impact circuit delay. However, an increased circuit size significantly reduces the overhead of implementing state space encryption. For example, encrypting two registers in the s15850 ISCAS89 benchmark circuit resulted in an area overhead of 0.026%, providing a means to secure sequential logic with minimal overhead. 2) Vaibhav Venugopal Rao and Ioannis Savidis, “Dynamic Current Mode Inverter for Ultra-Low Power Near-Threshold Computing.” Abstract: A technique to enhance the security of analog circuits using Satisfiability Modulo Theory (SMT) based design space exploration is described. The analog satisfiability (aSAT) technique takes as inputs generic circuit equations and performance constraints and, by exhaustively exploring the design space, outputs transistor sizes that satisfy the given constraints. The aSAT methodology is applied to parameter biasing obfuscation, where the width and length of a transistor are obfuscated to mask circuit properties. The proposed methodology was used in the design of a differential amplifier and an operational amplifier, where the widths and lengths determined through aSAT analysis were shown to meet the target circuit specifications. For the operational amplifier, transistor dimensions determined through aSAT analysis for a set of performance constraints were characterized and were found to meet the performance targets, however, there was a 7 MHz reduction in the gain bandwidth product. e simulated results indicate that the developed design methodology achieves a fast and accurate determination of transistor sizes for target specications.

United States Patent Number 9,912,325 awarded on March 6, 2018

2018-03-06T00:00:00+00:00

The patent “ Power supply voltage detection and power delivery circuit ” is issued by the USPTO on March 6, 2018 as United States Patent Number 9,912,325. Prof. Ioannis Savidis and Divy Pathak are the co-founders of the patent.