Department of Electronics and Communication Engineering

Permanent URI for this collection

http://nits.ndl.gov.in/handle/123456789/91

Browse

Browsing Department of Electronics and Communication Engineering by Title

Now showing 1 - 20 of 26
  • Thumbnail Image
    Item
    A Low-Power Analog Front End Circuits for Portable Bio-Medical Devices
    (National Institute of Technology, Silchar, 2026) Prasad, Ronanki Sai
  • Thumbnail Image
    Item
    Application of Generative AI in Automated and LLM-Judged C/SystemVerilog/UVM Test Code Generation
    (National Institute of Technology, Silchar, 2026) Gangrade, Kaustubh
    The engineering of modern VLSI systems-on-chip is becoming increasingly constrained by a single step: verification. This stage is the major cause of schedule delays as well as cost overstretches. Due to the growing use of heterogeneous architecture in chips, the tedious work of writing and maintaining tests has been a major scaling issue. This problem is overcome by the current thesis that produces the pipeline that automates the test case generation. Instead of starting with a blank slate the system will build tests based on human-readable descriptions and structured plans, and will use a small set of so-called seed samples to get a footing. The research aims at the two most frequent pain points, C firmware-level sanity checks at an early stage, and full System Verilog/UVM at an advanced stage of simulation. Its underlying innovation is specification grounding. The code generated is closely linked to the real hardware bottlenecks by requiring the pipeline to retrieve data directly out of register spreadsheet (XLSX) and IP-XACT maps. In fault tolerance, an integrated repair loop is used to detect compiler errors and dynamically patch the code. Further, an LLM-as-Judge model compares the end result with a specially designed rubric,judging such aspects as whether the intent and truth are aligned or not, and whether the value of true verification is real or not. The empirical analysis of the projects proves a trend of equal, steady improvement in quality. Simple, clear data has been used instead of complex academic measures, making it easier for engineers to directly use it in their work.
  • Thumbnail Image
    Item
    Classification of Diabetic Retinopathy and Macular Edema through Segmented Retinal Vessels using Graph Neural Network
    (National Institute of Technology, Silchar, 2026) Mutyala, Sukumar Vamsi
    Every year, countless people go blind from diabetic retinopathy and macular edema and the tragic reality is that most of them didn’t have to. These conditions can largely be stopped in their tracks, but only if they’re caught early enough, before the damage becomes permanent. That early window depends entirely on a doctor’s ability to see what’s actually going on inside the eye. This is why having accurate, detailed maps of the retina’s blood vessels matters so much. The problem is that current deep learning models tend to fall short in exactly the ways that matter most. They produce broken, incomplete vessel maps missing the tiny capillaries and dropping connections between vessels which are precisely the features a doctor needs to identify the earliest warning signs. To solve this problem, this paper introduces a three stage hybrid framework. It takes three steps to operate the pipeline. First, a dual-encoder graph convolutional network (DE-DCGCN-EE) is used to detect the boundaries of the vessels and model the relationships among the colour channels. Second, a novel VessGAT-SAM2 combination of SAM2 and a Graph Attention Network repairs broken vessel segments, ensures vessel continuity in thin capillaries and generates topologically accurate masks. Third, the multi modal classifier combines the fundus image, the refined vessel mask and the SLIC superpixel maps to predict diabetic retinopathy grade (0-4) and the presence of macular edema. The whole pipeline is trained end-to-end with a joint loss function that reflects the correlation between both conditions. For that purpose, experiments were conducted on four datasets which are publicly available: DRIVE, PRIME-FP20, Messidor-2 and IDRiD. The outcomes were impressive. The overall results were superior to the traditional ones like U-Net, standalone SAM2 and GCN based ones in terms of both accurate segmentation of the vessels and classification of the disease. Most crucially, it remained robust in a variety of datasets, and continued to reveal the fine vessel structures that are important to real clinical significance. This is an exciting step towards the development of automated screening systems, which may help to prevent vision loss due to diabetic eye disease.
  • Thumbnail Image
    Item
    Classification of Diabetic Retinopathy through Segmented Microaneurysm Features Using Deep Learning Algorithm
    (Department of Electronic and Communication Engineering, NITS, 2024-06-06) S., Muralitharan
    Diabetes Retinopathy (DR) is a consequence of diabetes that can result in blindness. Elevated blood sugar levels cause retinal capillaries to shrink, resulting in the production of tiny, blob-like formations known as microaneurysms (MA). Detecting these little MAs manually is challenging and usually requires specific training from ophthalmologists. As a result, there is an urgent need for automated MA detection and DR categorization approaches to speed up diagnosis and treatment for a greater number of patients. This thesis is divided into two chapters, both of which focus on DR classification utilizing Machine Learning and Deep Learning methods. The MA detection procedure typically consists of three stages: preprocessing, augmentation, and classification. Preprocessing is very important for improving retinal characteristics and eliminating imaging artifacts, which are a difficult element of retinal fundus image analysis In this paper, we provide a new Improved Non Local Mean filter (INLMF) that successfully eliminates image artifacts. Furthermore, we present an improved U-Net architecture for MA segmentation, in which traditional U-Net blocks are substituted with customized Res-Net blocks. These Res-Net blocks are made up of three residual blocks connected by a dropout layer, which provides increased performance. Our goal is to detect DR early in the presence of MAs. To do this, we provide a unique deep CNN model that diagnoses DR severity based on the number of MAs present in fundus pictures, categorizing it as normal, mild, or moderate. Both strategies are carefully tested on publicly available datasets such as e-ophtha, with our segmentation strategy outperforming state-of-the-art measures.
  • Thumbnail Image
    Item
  • Thumbnail Image
    Item
    DESIGN & DEVELOPMENT OF AN UWB MIMO SYSTEM FOR IOT APPLICATIONS
    (National Institute of Technology, Silchar, 2026) PATHAK, SAKSHI
    With the rapid growth of electronics in IoT field, there is need of system which has high data rate and has compact size.4-port MIMO system with UWB technology fulfils this demand and also provides solution for multi-path fading and high data rates. This thesis provides a complete solution for new electronics technology. It contains complete analysis of 4-port system including design, geometry, parametric optimization and re- sults. Proposed design of 4-port MIMO array has a size of 60 × 60 mm2.To enhance the bandwidth, two rectangular stubs added into the feedline and it successfully increase the bandwidth of 4-port design from 4.3 GHz to 7.8 GHz. Also, to maintain the iso-lation between all the antenna elements, an orthogonal arrangement is used. This tech-nique effectively suppressed electromagnetic coupling and maintain isolation without the need of any external decoupling structures. Simulation of this design demonstrates excellent diversity parameters includes ECC strictly below 0.005, diversity gain (DG) of 9.99 dB, TARC below -10dB and channel capacity loss (CCL) below 0.4 bps/Hz. These results proves that the proposed 4-port MIMO array offers broad-band matching, isola-tion and good diversity performance makes this system reliable for modern high speed UWB communication systems.
  • Thumbnail Image
    Item
    Design and Analysis of a Low-Power, Low-Jitter LC Oscillator with Capacitor Bank Tuning for Bluetooth, WiFi Applications
    (National Institute of Technology, Silchar, 2026) Bharti, Anupama
    Almost every modern device from smartphone and wi-fi router to Bluetooth gadgets we use is based on wireless communication. These devices works on stable and accurate frequency generation which is local oscillator (LO) frequency. At the heart of this local oscillator frequency is the Phase-Locked Loop circuit. The overall performance of PLL is strongly influenced by Voltage controlled Oscillator (VCO), acting as the main component of PLL circuit. The VCO directly impacts various parameters such as phase noise, tuning range, lock time, power consumption, jitter etc. A poorly designed VCO can easily limit the performance of the whole system, while a carefully optimized VCO design can significantly improve the signal quality and robustness of the system. Various techniques are developed to address these challenges while achiving optimized circuit design. This project focuses on design and implementation of a differential LC VCO for Bluetooth and wifi applications in 65 nm CMOS technology using Cadence Virtuoso tool. Starting from the basic concepts of PLL and VCO operation, the report briefly reviews different oscillator types and explains the importance of LC-based VCO which are preferred for RF applications. A concise literature review summarizes several state of the art of PLL and VCO designs, highlighting their frequency range, tuning range, phase noise, power consumption and implementation technology. The VCO uses a complementary cross coupled NMOS–PMOS pair to generate the negative resistance required to cancel tank losses and sustain oscillation. The Channel length optimization technique is employed to improve the vco transconductance ‘gm’ making it sufficient to compensate for the losses in tank. The LC tank is designed around a target frequency around which Bluetooth, wifi, zigbee etc works and the device sizes are choosen to meet the required transconductance and output swing as swing below 600mV leads the oscillation to die out. The design is implemented and simulated in Cadence, and key results such as oscillation frequency, tuning behavior, output swing, power consumption, senstivity ‘KVCO’ variation, and phase noise are analyzed. In this work, the systematic design and optimization of a low-power, low-phase-noise, and low jitter LC based voltage-controlled oscillator (LC VCO) is implemented in 65nm CMOS technology. A dual path tuning topology is implemented to achieve a wide tuning xii range as well as low phase noise. Capacitor bank is designed for coarse tuning path by utilizing a digitally controlled capacitor bank to achieve the required bandwidth of approximately 400MHz with 15 % tuning range and the fine tuning path is employed by by keeping the varactor varactor size small for achieving minimal sensitivity ‘KVCO’. The design is compared with the existing LC VCOs to highlight its advantages. Simulation results demonstrate the improved performance, with a figure of merit(FoM) of 189. Process, Voltage, and Temperature (PVT) variations and Monte Carlo simulations are also performed to check the robustness of the device. Thus, the result conform the design reliability across different conditions. The LC VCO design successfully illustrate the critical trade-off between various performance parameter. This results established a practical methodology for high performance RF synthesizer design with 180nm channel length making it ready for full PLL integration.
  • Thumbnail Image
    Item
    Design and Analysis of Eddy Current-Based Non-Contact MEMS Sensor for Detection of Pesticides
    (National Institute of Technology, Silchar, 2026) Chhajer, Shivang
  • Thumbnail Image
    Item
    Design of Low Power, Low Noise with High Output Swing Preamplifier for Cochlear Implant
    (Department of Electronics and Communication Engineering, NITS, 2022) Kumar, Navin
    A cochlear implant is a tiny, intricate electronic device that can enable a person who is profoundly deaf or very hard-of-hearing hear. Cochlear implantation is a procedure that uses electrical stimulation in the auditory nerve system to restore partial hearing. The development of a hearing aid devices for cochlear implant at a cheap cost is now essential. The entire cochlear system is made up of various components, including a microphone that detects sound waves, a preamplifier to raise the signal level, and a signal processing unit for producing electrical pulses for the electrode in order to trigger the auditory nerve. In this work, a low-power, high-gain with high output swing preamplifier has been designed for cochlear implant. It is critical for the incoming signal to get processed further. Balanced operational transcendence amplifier (OTA) has high swing amplifier. Here, splitting input transistors of balanced OTA reduces the power consumption as one pair of transistors operates in saturation region while the other pair operates in subthreshold region (for low power). Because the current gets split into two different branches, the gain of the proposed OTA increases as well as input referred noise reduces increasing the overall transconductance. Capacitive feedback topology is used along with current scaling resistor to maintain the noise under limit and to overcome the trade-off between noise and power. Simulations and post-layout analyses for the proposed preamplifier were done in Cadence Virtuoso using UMC 180 nm technolοgy. In proposed OTA, we have obtain maximum gain of 44.23 dB, total average power of 1.69, phase margin of 45 degree, and noise at 1 Hz, 20 Hz, and 20 kHz are 2.71 V/sqrt(Hz), 2.71 V/sqrt(Hz) and 2.71 V/sqrt(Hz), respectively. Also, we have analysed the capacitive feedback closed loop OTA in which we have obtained maximum mid-band gain of 43.75 dB, and bandwidth of 22.55 Hz to 18.55 kHz.
  • Thumbnail Image
    Item
    Digital Image Lossy Encryption and Decryption
    (Department of Electronics and Communication Engineering, NITS, 2022) Yellapu, Girish Ram Pradeep
    In today's environment, image encryption and decryption are extremely We a medium to transport image data from one location to another. However, the chances of data theft are higher in the medium range. To get around this, Image encryption has to be used. As the usage of digital systems for sending and storing photos grows, the question of how to safeguard the confidentiality, integrity, and authenticity of photographs becomes increasingly crucial. Various approaches for encrypting images to make images more secure are from time to time. This thesis two methods: the mathematical cube roots approach and the mathematical logarithms method. Images will encrypted as well as compressed using these algorithms. To represent digital images, a considerable number of bits were necessary, and their canonical representations typically contain significant re- dundancy. By using these redundancies, image algorithms lower the number of bits to describe a picture. The lossy compression method utilised here avoids these redundancies. This thesis included all of these approaches and progressions. Using these picture quality in less area, compression quantity, and compression are accomplished, resulting in a more reliable image transmission with information security.
  • Thumbnail Image
    Item
    Ferroelectric Field Effect Transistor (FeFET) for Embedded Non - Volatile Memory Application
    (National Institute of Technology, Silchar, 2026) Kumar, Aman
  • Thumbnail Image
    Item
    Handwritten Character Recognition using Deep Convolutional Neural Networks
    (Department of Electronics and Communication Engineering, NITS, 2022) Saidulu, Naragoni
    Human perception takes the help of size information to recognize the upper- and lower-case alphabets. Recognizing handwritten characters is difficult mainly because of the variation in the character pattern, handwriting style, preferred size, and larger dictionary. To handle this, researchers have opted for designing and developing customized networks like CapsuleNet, TextCaps, etc. Also, the performance of these networks on case-sensitive characters is mini mal, because of the inability to distinguish characters that are differentiable only by size and not by their pattern. The results of the handwritten character recognition (HCR) based open source state-of-art (SOTA) deep convolutional neural networks reveal that the performance is improved on testing NIST SD19 (image:128x128) compared to EMNIST (image: 28x28). On equivalent, to human perception, the developed SOTA-DRBAM-Net (state-of-art deep residual bottleneck attention convolutional neural network) proved that channel and spatial attention at the bottleneck of each residual block helps to improve the performance further. Also, SOTA DRBAM-Net can (i) have a recognition performance more than the existing state-of-the-art network; (ii) resolve and handle the complexity of case-sensitiveness in handwritten characters provided peripheral information is being preserved. The performance of SOTA-DRBAM-Net on (i) EMNIST (By-class: accuracy (AR)- 88.05%, By-merge: AR– 91.2%, Balanced: AR 89.6%, Letters: 94.8%, Digits: 99.6%) (ii) NIST SD19 (By-class: AR- 91.70%, By-merge: AR– 92.86%, Balanced: AR-90.89%, Letters: 90.88%, Digits: 99.69
  • Thumbnail Image
    Item
    Investigation of Device-to-Device Communication in 5G mmWave Network using an Uplink Channel
    (Department of Electronics and Communication Engineering, NITS, 2022) Sachan, Atul
    The exponentially growing need for higher data rates and the ubiquitous usage of smartphones has arisen unprecedented issues in providing seamless communication while satisfying band width constraints. Enabling device-to-device (D2D) communication in the mmWave network enhances throughput and bandwidth but at the cost of higher pathloss attenuation. Also, D2D communication introduces interference in the traditional cellular network, thereby lowering system performance. Thus, the proposed work applies dynamic mode selection criteria for effi cient D2D communication in 5G mmWave cellular network. Radius of coverage of D2D users are derived for switching of inner mode to outer mode. It switches to the outer mode when the D2D users suffer from extensive pathloss attenuation along with an increase in radius of cover age. Furthermore, power optimization is done for obtaining optimal D2D transmit power which in turn enhances the data rate, solved by Lagrangian dual optimization technique. The perfor mance metric, coverage probability is derived which depicts the capability of the proposed scheme at higher pathloss attenuation by maintaining the quality-of-service (QoS) constraint. Simulation results prove the efficacy of the proposed scheme. Furthermore, the work also con siders a multicell scenario-based resource allocation for interference mitigation for D2Ds in the edge region. Thus, based on our system model and assuming that only outer region D2Ds can share the resources of inner region cellular users found, the signal-to-interference-ratio (SINR) calculation is done. Furthermore, the coverage probability is calculated using cumulative dis tribution function (CDF) and also the mathematical analysis of area spectral efficiency (ASE) is calculated to maximize the throughput
  • Thumbnail Image
    Item
    Millimeter-Wave MIMO Antenna For 5G Application
    (Department of Electronics and Communication Engineering, NITS, 2022) Manirathnam, Chinthapalli
    A circularly polarized, two-port, multiple input and multiple output (MIMO) antenna having high gain for 5G wireless communication systems is proposed in this study. The proposed unit cell utilises a Microstrip patch antenna is made up of Rogers RT/Duroid 5880 substrate material, which has a loss tangent of 0.0009, a dielectric constant of 2.2, and a substrate thickness of 0.8 mm, with a full ground plane and H-shaped slot to obtain a resonant fre quency of 28 GHz in the millimeter-wave (mm-wave) region. The dimension of the designed patch antenna is 14 mm x 10 mm x 0.8 mm.To realize circular polarization in operating band width a cross-shaped slot is etched from its radiating patch. The circularly polarized antenna maintains an appropriate gain of 4.47 dB, the antenna has a larger operational bandwidth of 2.25 GHz and operates in the frequency range from 26.36 GHz to 28.61 GHz. Initially, to de sign MIMO configuration, the proposed single element antennas are conventionally arranged with an edge-to-edge separation of 0.7λ. The dimension of the designed MIMO system is 26 mm x 14 mm x 0.8mm. TheProposed MIMO antenna maintains an appropriate gain of 4.95 dB,both the antennas are resonating at 28 GHz frequency in the range of 27.16 GHz to 28.7 GHz having bandwidth of 1.54 GHz. Further to reduce the mutual coupling between radiating elements, meta material(MTM) based SRR is introduced in between two antennas. By using the SRR the isolation between antenna elements are improved by 10 dB at resonant frequency of 28 GHz. An artificial magnetic conductor (AMC) is placed under the MIMO antenna to enhance the gain of the system by about 75% over the operating frequency band. The dimension of the designed AMC based MIMO antenna is 36 mm x 36 mm. Initially The proposed AMC based MIMO antenna is verified by the simulation results in terms of S-parameters, isolation, gain, Axial ratio and MIMO diversity parameters. The isolation between two radiating elements is also significantly higher than 37 dB. The circularly polarized AMC based MIMO antenna maintains A high gain of 8.6 dB, the antenna has a bandwidth of 1.4 GHz and operates in the frequency range from 27.3 GHz to 28.7 GHz and the Axial ratio bandwidth is 1.52GHz in the operating band 26.86 GHz to 28.38 GHz. The proposed AMC based MIMO antenna system also features high diversity gain, low channel capacity loss, and good envelope correlation coefficient. Which makes the proposed AMC based MIMO antenna is suitable for 5G wireless communication systems.
  • Thumbnail Image
    Item
    Miniaturized High-Gain Conformal Ku-Band MIMO Antenna for Biomedical Imaging Applications
    (National Institute of Technology, Silchar, 2026) Eshwar, Velavalapalli Hanuman Naga Sai
  • Thumbnail Image
    Item
  • Thumbnail Image
    Item
    Optimal Deployment of IRS in Fresnel Region for QoS Aware Terahertz Communications
    (Department of Electronics and Communication Engineering, NITS, 2022) Sankar, Nagineni Siva
    To compensate for the substantial propagation losses, an Intelligent Reflecting Surface (IRS) assisted with Terahertz bands is projected to feature a large number of reflecting elements. The usual far-field assumption becomes erroneous when the IRS rises in size, then the spherical waveform model of radiated waves should be considered. We look at a spherical wave channel model in an elliptical path with a single antenna transmitter (TX) and a receiver (RX) at the focal points in this work. We develop an approximate but accurate close-form equation for system throughput and normalised gain under beam focusing for this purpose. Using the golden search (GS) technique, we show the optimal distance between the BS and the IRS based on the generated model. Numerical results back up our findings and reveal new information about the Performance and designing of IRS-assisted THz communications.
  • Thumbnail Image
    Item
    Patch-based Cascaded Underwater Image Enhancement via HSV-Domain Patch Processing and Full-Resolution RGB Refinement with Heterogeneous Skip Connections
    (National Institute of Technology, Silchar, 2026) Gautam, Aditi
    Underwater images often suffer from severe degradation due to light scattering, color attenuation, and non-uniform illumination. This paper proposes a cascaded two-stage deep learning framework that combines localized enhancement in the HSV color space with global refinement in the RGB domain. In Stage-1, overlapping 256 × 256 patches are enhanced independently to correct region-specific color and illumination distortions, followed by cosine-based blending for seamless reconstruction. Stage-2 performs full-resolution refinement using a residual learning-based network with multi-scale feature extraction, deep supervision, and channel attention to enforce global consistency and improve perceptual quality. Experimental results demonstrate that the proposed method achieves superior performance in terms of PSNR, SSIM, UIQM, and UCIQE, producing visually coherent images with improved color fidelity and structural preservation. The framework provides an effective and robust solution for underwater image enhancement across diverse degradation conditions.
  • Thumbnail Image
    Item
    Physical Design Synthesis and Implementation using Synopsys Fusion compiler
    (National Institute of Technology, Silchar, 2026) Vishal
    The high rate of semiconductor technology development has greatly complicated the modern integrated circuits and efficient Physical Design methodologies can never be of less importance. As technology nodes scale to a larger scale, issues of interconnect delay, congestion, power integrity and timing violations have emerged to dominate the overall performance of the chips. Conventional ASIC design processes (individually synthesize and physically implement) tend to exhibit poor stage-correlation, resulting in multiple design cycles and longer design cycles. Physical Design synthesis and implementation flow with Synopsys Fusion Compiler, which is a single platform of logical and physical design synthesis in the same optimization engine. It includes the entire ASIC design chain, such as starting with the generated gate-level netlist to the ultimate layout that is to be fabricated. Floor planning, power planning, placement, clock tree synthesis (CTS), routing, and timing closure are examined in detail to learn how they affect the quality of designs. Considerable focus is on physical-aware synthesis and early synthesis optimization, which are useful to minimize timing difference between pre-layout and post-layout phases. Several optimization schemes, such as cell sizing, buffering and the congestion-based placement, are investigated to balance timing closure and performance improvements. The integrated solution also quite reduces the repetition of Engineering Change Orders (ECOs), the design cycles and enhances the general productivity as well.The outcomes show that with Fusion Compiler, Quality of Results (QoR) in terms of timing, power and area is improved and convergence.
  • Thumbnail Image
    Item
    PIN Diode Based Frequency Reconfigurable U-Slot Microstrip Patch Antenna for Multiband Microwave Communication
    (National Institute of Technology, Silchar, 2026) Shankar, Priyanka
    This thesis shows the design and analysis of a frequency reconfigurable microstrip patch antenna with a U-slot and an adjusent vertical slot with PIN diode switching. The antenna is fabricated on Rogers RT/duroid 5880 substrate and simulated using ANSYS HFSS. The introduction of the U-slot along with the vertical slot creates multiple current paths on the patch, resulting in the excitation of multiple resonant modes. The effective current distribution can be tuned by switching elements i.e. by equivalent RLC circuits, to obtain the frequency reconfigurability. This antenna operates at multiple resonant frequencies in the range of 6 GHz to 14 GHz covering bands of C-, X- and Ku-bands. The antenna has good impedance matching at these operating bands with return loss values below −10 dB and attained a maximum gain between 5–9 dB. A parametric study has also been performed to study the effect of the dimensions of the slots and patch parameters on the performance of the antenna. This allows to control the tuning of the resonant frequencies. The proposed antenna is suitable for satellite communication applications, such as fixed satellite service (FSS) uplink in the C-band (6.9–7.1 GHz) and Ku-band (13–14 GHz) frequency ranges. Besides the main application in satellite systems, the antenna can also be used in X-band radar and terrestrial microwave backhaul links due to the multi-band operation. The design exhibits a balance between compactness, reconfiguration and practical usage in modern RF communication systems