RANWALK: An Electromagnetic Simulation Methodology that Redefines Integrated Circuit Design

In today’s fast-paced tech landscape, innovation is everyone’s constant pursuit. From 5G networks and artificial intelligence to the Internet of Things (IoT) and autonomous vehicles, the demand for smaller, faster, and more efficient integrated circuits (ICs) is skyrocketing. Enter RANWALK, a groundbreaking research project born in Greece, offering a revolutionary solution to one of the most pressing challenges in modern microelectronics: electromagnetic crosstalk.

What is Crosstalk and Why Should We Care?

As ICs become increasingly compact and densely packed, especially in technologies below 40 nanometers, unwanted electromagnetic interactions between conductors begin to emerge. These interactions, known as crosstalk, can disrupt circuit performance, delay product launches, and cost manufacturers millions in lost revenue. Crosstalk is particularly problematic in high-frequency applications, where even minor interference can lead to significant performance degradation.

The Solution: Pharos and Its Evolution Through RANWALK

Ansys, part of Synopsys, has developed pioneering technology that identifies high-risk areas for crosstalk during the IC design phase, enabling engineers to focus their layout optimization or debugging efforts where it matters most.

But as technology advances, so must the tools. The Greek state-funded project RANWALK supported the development of a new stochastic methodology for accurately and efficiently extracting parasitic capacitances, a critical step in predicting crosstalk. This advancement significantly enhances the capabilities of our on-chip electromagnetic simulation tools, making them more robust and scalable for next-generation IC design.

The Innovation Behind RANWALK

RANWALK introduced a novel random walk method to solve the electrostatic problem in high-density circuits. In simple terms, this method enables precise prediction of unwanted interactions even in complex 3D structures. The new modeling engine developed through RANWALK is:

• Extremely fast and parallelized, capable of handling massive datasets.
• Geometry-agnostic, meaning it adapts to any circuit shape or structure.
• Highly accurate, even accounting for manufacturing-induced distortions in circuit geometry.

This approach enables designers to simulate realistic operating conditions and identify potential issues before fabrication, saving time and resources.

Real-World Applications and Testing

As part of the project, high-frequency circuits were designed and fabricated, including:

• Voltage-controlled oscillators (VCOs) for frequency synthesis.
• Low noise amplifiers (LNAs) for wireless receivers.
• Mixers for frequency conversion.
• Passive transformers (baluns) for signal balancing.

These circuits were tested in lab environments, and the results were benchmarked against simulations from the new modeling engine, validating its precision and effectiveness. The ability to accurately model parasitic effects in these components is crucial for achieving optimal performance in real-world applications.

As the demand for high-performance, energy-efficient electronics continues to grow across sectors such as telecommunications, automotive, IoT, and AI, the technologies born from RANWALK will play a pivotal role in enabling smarter, more resilient, and future-ready innovation.

Who Benefits from RANWALK?

RANWALK is more than a technological achievement. It is a catalyst for transformation in the IC industry. The markets that benefit directly include:

1. 5G and telecommunications

5G networks require circuits operating at millimeter-wave frequencies, with ultra-low power consumption and high reliability. RANWALKs tools reduce design risk and accelerate product deployment, enabling faster and more reliable communication infrastructure.

2. Internet of Things (IoT)

With billions of devices connecting daily, the need for compact, efficient, and reliable ICs is immense. RANWALK helps create such circuits, optimizing power and performance while minimizing interference.

3. Automotive and electrification

Modern vehicles integrate numerous ICs for sensors, communication, and data processing. RANWALKs technology enhances the reliability of these systems, supporting safer and smarter mobility solutions.

4. High-performance computing (HPC) and AI

AI applications demand high-speed processors and data transfer. Accurate circuit modeling of high-speed interfaces is essential, and RANWALK is designed to deliver just that, enabling breakthroughs in machine learning and data analytics.

Commercial Impact and Future Outlook

RANWALK results are already embedded in our simulation software portfolio focusing on electromagnetic simulation for on-die applications:

• Synopsys Exalto electromagnetic-aware parasitic extraction signoff
• Synopsys VeloceRF inductor and transformer synthesis tool
• Synopsys RaptorQu high-capacity electromagnetic simulation for quantum computing applications
• Ansys HFSS-IC chip-to-system electromagnetic simulation platform

These results significantly enhancing their ability to model complex, high-frequency integrated circuits with greater accuracy and speed. These advancements are expected to expand market reach and drive substantial growth in the adoption of advanced modeling solutions.

The RANWALK benefits are clear: enhanced simulation accuracy, faster design cycles, and improved reliability of complex chip architectures. As the demand for high-performance, energy-efficient electronics continues to grow across sectors such as telecommunications, automotive, IoT, and AI, the technologies born from RANWALK will play a pivotal role in enabling smarter, more resilient, and future-ready innovations.