Quantum machine learning (QML) and quantum communication (QC) are emerging areas of research for complexity-efficient data processing, information retrieval, system, and simulations. Towards next-generation vehicular technology, quantum circuits and algorithms are critical in providing faster and more reliable solutions. Here, we show a modified quantum K-nearest neighbour (QKNN) algorithm for symbol detection in the communication system. In the proposed algorithm, we embed a quantum swap-test circuit for the distance computation between unknown symbols with known training symbols with a memory-efficient strategy, which eliminated quantum search operations for the small-search space such as 16-QAM symbols. The proposed quantum circuit is simulated on an IBM quantum machine and the numerical results are shown to demonstrate the efficacy of the proposed framework. The classification accuracy as a performance metric is shown and the improvement of quantum resources in terms of gate complexity is discussed.

For next-generation vehicle-to-everything (V2X) communications, there has been a growing emerging technology at the intersection of Radar and Communications, namely RadCom. RadCom provides many benefits such as efficient usage of bandwidth, joint detection and estimation, and improved tracking performance for future intelligent transportation. However, there is a limitation in the computation of parameters for large dimensions (antennas, users, or targets). To provide, a computational complexity advantage for the target detection and parameters, we introduce Quantum-assisted Radar Communication systems, namely QRADCOM. In this work, we have proposed algorithms for the multi-target detection hypothesis based on quantum phase estimation (QPE), and also the quantum inversion method for communication signal detection in presence of observation noise, interference and multi-tap wireless channel. We show numerical results and analysis as well for the QRADCOM framework on both classical and IBM-Quantum computers.

Vandermonde matrix finds a wide range of applications in the digital signal processing domain including antenna array processing. Eigenspectrum estimation remains a challenge in such a system when the matrix dimension increases. This precise problem can be well addressed in a quantum simulation framework based on quantum Hamiltonian simulation (QHS). However, the structural exploitation of the matrix is never considered, which can give benefits in terms of quantum gate complexity. In this work, we exploit the special structural property of a dense Vandermonde matrix by decomposing it into several sparse matrices and then use them for a complexity-efficient eigenvalue spectrum estimation in a quantum framework. In this context, we propose an iterative quantum simulation framework for a Vandermonde-structured Hamiltonian called quantum Vandermonde simulation (QVS). The proposed approach requires fewer quantum gate operations compared to the standard QHS of a dense Vandermonde matrix. In this work, we have considered a large antenna array system, that exploits a delay Vandermonde matrix (DVM) to create a multi-beam beamformer using the proposed low complex quantum architecture. Extensive theoretical and numerical analysis in terms of complexity and other quality parameters have been conducted to show the benefit of the proposed method.

In this work, we propose quantum assisted eigenvalue estimation and target detection algorithms for a large sensor array via Hamiltonian simulation. Quantum algorithms provide complexity advantage of a certain class of problems on a quantum computer with fewer physical resources as compared to their classical counterparts. The proposed algorithms make use of the quantum phase estimation (QPE) as its core computing component. We have introduced an analytical quantum framework to map from classical to quantum in the context of target detection. Target detection involves an appropriate choice of threshold based on the probability of detection or false alarm. We exploited the massive sensor array structure and invoked the random matrix theory to propose an optimal threshold. It also takes into account the quantum measurement noise in the framework. Numerical simulations are performed to ascertain the efficacy of the proposed framework. The results suggest near term applications of the quantum algorithm for large-scale linear systems.

Perturbation in discrete dynamic systems (e.g., a manoeuvring aircraft, an autonomous vehicle having frequent lane change or turning, etc.) makes the linear tracking filter(e.g., Kalman Filter (KF)) sub-optimal. The existing robust filters such as Ridge-based KF (Ridge-KF), Tikhonov-based KF (TRKF), and minimax algorithm perform enhanced estimation under low and moderate perturbations only. The investigation of severe perturbation for multi-degree of freedom (m-DOF) motion of manoeuvres is carried out in this research article. The severity of perturbation due to m-DOF motion makes the observation matrix ill-posed for the tracking filters. To enhance the estimation performance (in terms of root means square error (RMSE)), we propose a novel algorithm called TLSKF. It offers much-improved performance than KF, Ridge-KF, and TRKF under high perturbation for estimating state parameters (position, velocity, etc.). Further, TLSKF approaches to KF at extremely low perturbation. The improvement in the RMSE enhances the tracking resolution of the radar. The comparison with existing literature is also strived to justify the novelty of the proposed algorithm.

We have proposed a machine-learning based classification framework for cognitive radar for target state classification. Based on the estimated frequency of the received signal at the radar receiver, we have classified three rotational movements (yaw, pitch, and roll) of a maneuvering aircraft motion. Direct classification of the data sets for the different rotational movement was found non-separable. It is difficult to find a classier to construct linear boundary for the classification of this data sets. We intended to design an algorithm for this problem. The proposed algorithm is applied on separable, half separable and non-separable data sets. The success rate of the classifier was verified in terms of cross-validation, mean square error, type I and type II error. The algorithm has shown a success rate of approximately 87.28% and 99.15% for not-separable and separable data sets respectively. It also shows that the increment in the accuracy by 6.86% as compared with the conventional approach [12].