Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Practical Evaluation of Quantum Kernel Methods for Radar Micro-Doppler Classification on Noisy Intermediate-Scale Quantum (NISQ) Hardware

Vikas Agnihotri, Jasleen Kaur, Sarvagya Kaushik

TL;DR

This work investigates quantum kernel methods for radar micro-Doppler classification on NISQ hardware by embedding PCA-reduced features into a quantum feature space via a ZZFeatureMap and evaluating a QSVM against a classical SVM. The approach is validated first on a simulator and then on IBM Torino and Fez quantum processors, revealing that QSVM can achieve competitive accuracy with a dramatically reduced feature set, albeit with hardware-dependent performance limits due to noise and measurement shot requirements. Hardware experiments show Fez generally outperforms Torino in fidelity and noise resilience, illustrating the impact of device architecture on quantum kernel estimation. Overall, the study demonstrates the practicality and current constraints of deploying quantum kernel methods for real-world radar tasks, highlighting the need for improved error mitigation and hardware evolution for substantial quantum advantage.

Abstract

This paper examines the application of a Quantum Support Vector Machine (QSVM) for radarbased aerial target classification using micro-Doppler signatures. Classical features are extracted and reduced via Principal Component Analysis (PCA) to enable efficient quantum encoding. The reduced feature vectors are embedded into a quantum kernel-induced feature space using a fully entangled ZZFeatureMap and classified using a kernel based QSVM. Performance is first evaluated on a quantum simulator and subsequently validated on NISQ-era superconducting quantum hardware, specifically the IBM Torino (133-qubit) and IBM Fez (156-qubit) processors. Experimental results demonstrate that the QSVM achieves competitive classification performance relative to classical SVM baselines while operating on substantially reduced feature dimensionality. Hardware experiments reveal the impact of noise and decoherence and measurement shot count on quantum kernel estimation, and further show improved stability and fidelity on newer Heron r2 architecture. This study provides a systematic comparison between simulator-based and hardware-based QSVM implementations and highlights both the feasibility and current limitations of deploying quantum kernel methods for practical radar signal classification tasks.

Practical Evaluation of Quantum Kernel Methods for Radar Micro-Doppler Classification on Noisy Intermediate-Scale Quantum (NISQ) Hardware

TL;DR

This work investigates quantum kernel methods for radar micro-Doppler classification on NISQ hardware by embedding PCA-reduced features into a quantum feature space via a ZZFeatureMap and evaluating a QSVM against a classical SVM. The approach is validated first on a simulator and then on IBM Torino and Fez quantum processors, revealing that QSVM can achieve competitive accuracy with a dramatically reduced feature set, albeit with hardware-dependent performance limits due to noise and measurement shot requirements. Hardware experiments show Fez generally outperforms Torino in fidelity and noise resilience, illustrating the impact of device architecture on quantum kernel estimation. Overall, the study demonstrates the practicality and current constraints of deploying quantum kernel methods for real-world radar tasks, highlighting the need for improved error mitigation and hardware evolution for substantial quantum advantage.

Abstract

This paper examines the application of a Quantum Support Vector Machine (QSVM) for radarbased aerial target classification using micro-Doppler signatures. Classical features are extracted and reduced via Principal Component Analysis (PCA) to enable efficient quantum encoding. The reduced feature vectors are embedded into a quantum kernel-induced feature space using a fully entangled ZZFeatureMap and classified using a kernel based QSVM. Performance is first evaluated on a quantum simulator and subsequently validated on NISQ-era superconducting quantum hardware, specifically the IBM Torino (133-qubit) and IBM Fez (156-qubit) processors. Experimental results demonstrate that the QSVM achieves competitive classification performance relative to classical SVM baselines while operating on substantially reduced feature dimensionality. Hardware experiments reveal the impact of noise and decoherence and measurement shot count on quantum kernel estimation, and further show improved stability and fidelity on newer Heron r2 architecture. This study provides a systematic comparison between simulator-based and hardware-based QSVM implementations and highlights both the feasibility and current limitations of deploying quantum kernel methods for practical radar signal classification tasks.
Paper Structure (26 sections, 9 equations, 9 figures, 12 tables)

This paper contains 26 sections, 9 equations, 9 figures, 12 tables.

Table of Contents

  1. Introduction
  2. Literature Survey
  3. Mathematical Formulation
  4. Classical SVM
  5. Quantum SVM
  6. Methodology
  7. Radar Simulation & Data Preparation agnihotri2019frequencyagnihotri2020automatic:
  8. Feature Extraction:
  9. Classical SVM Setup:
  10. QSVM Setup:
  11. Feature Reduction (PCA and Qubit Selection):
  12. Quantum Feature Map (ZZFeatureMap):
  13. Circuit Design and Entanglement:
  14. Hilbert Space Representation:
  15. Dataset and Quantum Hardware:
  16. ...and 11 more sections

Figures (9)

  • Figure 1: Radar classification pipeline with classical and quantum SVM
  • Figure 2: Micro-Doppler Signatures of Helicopter, Propeller Aircraft and Jet Aircraft.
  • Figure 3: 4-qubit ZZFeatureMap quantum circuit with two repetitions used for feature encoding.
  • Figure 4: Confusion matrix comparison between Classical SVM and Quantum SVM (simulator) for radar target classification
  • Figure 5: Effect of Shot Count on Measurement Precision
  • ...and 4 more figures