Electronics and Radar Development Establishment

Radome is an integral part of any radar or communication system. The primary function of a radome is to protect the antenna enclosed within from the external harsh environment while having minimal impact on the radiation characteristics of the enclosed antenna and also meet the structural requirements. In this paper, the Electromagnetic design approach for an S-band geodesic radome of diameter 25 meter composed of panels with joints is presented. The optimization of an A-sandwich radome construction of the radome is done using unit cell in 3D EM tool. The skin and core thickness of the radome construction is analyzed to achieve minimum transmission loss across the frequency band and scan angles. Further, an arrangement of radome panels with the optimized radome construction, arranged in a quasi-random manner is analyzed with a horn antenna using SBR technique in 3D EM tool HFSS.

Simulations in quantum chemistry (QC) impose a heavy load on the computing power, and also entail numerous dynamically varying dependant workloads that render the issue of load balancing in a cloud environment rather significant. Although a number of load-balancing methods are usually adequate in a host of high-performance computing tasks, the identical may not hold in the case of QC tasks. This paper draws a hybrid load-balancing strategy that combines reinforcement learning and predictive modeling to find an optimal process in the allocation of resources in QC simulations on the cloud. In particular, RL is used to handle the management of systems resources whereas predictive modeling is used to predict workloads and hence restrain latency and efficiently employ cloud resources. Experiment results indicate that with the aid of ML, the framework could achieve goals in terms of the largest possible load-distribution efficiency, categorical enhancement of tts, efficiency up to 20 percent in terms of load-distribution and reduction of task-completion times. It also helps enhance the effectiveness of the QC simulation that has been conducted on the cloud infrastructure, and provides preconditions to the additional possibility of the application of ML to scientific computing.

Large-scale combinatorial optimization problems (LSCOPs) are ubiquitous in many real-world domains, including but not confined to logistics, scheduling and network design, but are computationally difficult to solve since they belong to the NP-complete problems. Sometimes in high-dimensional discrete spaces, traditional metaheuristic algorithms have difficulty scaling and have the problem of converging too early and show a lack of strategic adaptability. The Hawks-Dove Optimiser (HDO) is presented, a new evolutionary game theory-based metaheuristic employing the conventional Hawks-Dove game that models the behavior of evolutionary conflicts. The idea is to increase exploration- exploitation balance using dynamic agent interactions on the basis of fitness driving pay. It has been applied on benchmark problems of Travelling Salesman Problem (TSP), Quadratic Assignment Problem (QAP), and Job Shop Scheduling Problem (JSP). Findings indicate that HDO provides better solution, converges faster and more stable than other well-known algorithms like Genetic Algorithm (GA), Ant Colony Optimization (ACO), Particle Swarm Optimization (PSO) and Grey Wolf Optimizer (GWO). Statistical tests prove the significance and the reproducibility of the results in several runs. The suggested HDO presents a really powerful but simple model of addressing LSCOPs as it incorporates game theoretical ideas into population based search. Its adaptability and scalability imply that it can become an attractive driver to many applications that deal with optimization tasks. Future work can be improvement through hybridisation by the problem-specific heuristics and theoretical analysis of convergence

Space element determination and monitoring and tracking systems play an important function in Space Situational Awareness (SSA), offering crucial data to approximate the location and trajectory of items in space. This task pursuits to broaden a complete device for area item detection and monitoring with the aid of using leveraging superior information processing techniques, specially specializing in Kalman filtering. The literature survey delves into current methodologies and technology within the discipline of Space Object monitoring and emphasizing the need for classy information processing algorithms. Following this, the device layout phase outlines the structure and additives of the device, inclusive of modules for information acquisition, filtering, and visualization. In the device implementation phase, Kalman filtering algorithms are applied for orbit prediction and conjunction evaluation. Results and discussions provide the assessment of the device’s performance, inclusive of accuracy metrics and comparative evaluation with current methods. In conclusion, the document summarizes the project’s objectives, methodologies, and key findings. Additionally, it proposes destiny upgradations in addition to beautify the competencies of area item dedication and monitoring structures. The summary adheres to the required layout hints and gives a concise evaluation of the task’s scope and objectives.

Transmit Receive Modules (TRMs) form the main building blocks of an Active Phased Array Radar (APAR) system. TRMs are low-voltage DC pulsed loads drawing a pulse current of typically 5 to 10 times the average current during the pulse ON time. Furthermore, depending on radar requirements, the pulse ON time can vary up from $1 \mu s$ to $500 \mu s$. Power converters supplying the TRM pulsed load need to have high power density ($\gt 6 \mathrm{~W} / \mathrm{cm}^{3}$ typically), high efficiency ($\gt 90 \%$), provide isolation and decouple pulse power from input. This paper discusses typical TRM power supply requirements and a proposed Series Resonant Converter (SRC) and Interleaved Synchronous Buck Converter (ISBC) scheme for APAR systems.

Beam squint occurs when the beam angle varies for wide bandwidth signals in phased array radars when steering is achieved with the help of phase shifters. The advancements in the digital technologies enable the mitigation of this effect with the help of Field Programmable Gate Arrays (FPGAs). Most of the modern active phased radars use digital beam forming for the formation of multi beams. In this work, we report on a FPGA based solution for beam squint correction by providing suitable corrections in the frequency domain using real time echo samples before forming the beams in the digital domain. Floating point operations are considered in this method to obtain better accuracy results with multiple computations. The beam squint effect is briefly explained by considering wide bandwidth signals. Further, the digital implementation of the beam squint mitigation method is proposed. The work is supported with various results obtained at different stages of analysis.

In certain communication systems, single-ended signal transmission over fiber links with Erbium Doped Fiber Amplifiers (EDFA) undergoes heavy distortions due to transient effects, caused by the slow gain dynamics of EDFA. This distortion occurs when the input digital signal frequency is comparable with the lifetime of Erbium ions, leading to data misinterpretation at receivers. This is very critical during control and synchronization signal transmissions. Conventionally, long-distance communication prefers differential signals due to its common mode noise rejection capability. In this work, we report a method utilizing differential signals and explains how transients are nullified due to its inherent signal characteristics.

Synthetic aperture radar (SAR) imaging has numerous uses in surface mapping, civil infrastructure, remote sensing, and terrain monitoring. Despite the benefits of multiple input multiple outputs (MIMO) in automotive radars, they are primarily used to provide range, azimuth, and elevation information for automotive applications. Obtaining acceptable angular resolution for automotive radar is a recurring difficulty due to vehicle-to-vehicle, vehicle-to-ground, vehicle-to-guardrail, and vehicle-to-tunnel discrimination. The purpose of this work is to demonstrate MIMO-SAR for finer angular resolution utilizing the 77-GHz Texas Instruments (TI) frequency-modulated continuous wave (FMCW) AWR1642 radar. SAR and MIMO radar topologies are used to increase synthetic or virtual aperture while maintaining adequate angular resolution. SAR is used to rebuild images from experimental data, and the images are created using a backpropagation algorithm. The findings are presented for SAR, MIMO, and MIMO-SAR. Furthermore, the experimental demonstration of MIMO-SAR using 77 GHz automobile radar verifies the prior modeling results. In addition, MIMO-SAR has been shown to provide better angular resolution than SAR and MIMO approaches. This algorithm's superior performance makes it appropriate for the automotive industry to perform SAR imaging on ego-corner deployed short-range radars (SRR) to picture guard rails, crossing vehicles, and VRUs.

This paper presents a python-based (Graphical User Interface (GUI) to extract the permittivity, permeability, impedance and refractive index of a dielectric material from the S-parameters available from the measurements. The Nicolson-Ross-Weir (NRW) algorithm is used to extract the electromagnetic parameters taking into consideration the branching effect. The GUI is user-friendly and users with no background in the electromagnetics can use this extractor. For validating the results CST simulation is performed for homogenous dielectric slabs and inhomogeneous metamaterial. The CST simulated results match with the parameters extracted using the proposed GUI extractor.