2. Thesis and Dissertations
Permanent URI for this communityhttps://idr.nitk.ac.in/handle/1/10
Browse
48 results
Search Results
Item Design of Energy Efficient, Variable Resolution, Hybrid Analog to Digital Converters for Low Frequency Applications(National Institute of Technology Karnataka, Surathkal, 2021) Polineni, Sreenivasulu.; Bhat, M S.; S, Rekha.The advancements in digital technologies made signal processing much easier and provide complex functionalities. The analog to digital converter (ADC) helps to convert the real time data obtained through sensors into digital data. The power consumed by these sensor nodes should be as low as possible in order to improve the battery life. Thus, this research mainly focuses on the design techniques, methodologies, and circuit topologies of ADC with emphasis on minimization of energy consumption and area. Initially, a fully di erential most signi cant bit (MSB) capacitor splitting switching technique for binary weighted capacitive digital to analog con- verter (DAC) in successive approximation register (SAR) ADC is proposed to reduce the energy consumption and area. Also, with this switching technique, one can achieve the same dynamic range as the conventional one, with half of the supply voltage as compared to the existing tech- niques. This makes the proposed switching method suitable for ultra low voltage SAR ADCs, which are widely used in biomedical applications. The proposed method is modelled in MATLAB. The results show that the proposed switching technique reduces energy consumption of DAC by 97% and the capacitance area by 50% over the conventional one. The circuit level implementation of 10-bit SAR ADC is simulated in UMC 90nm CMOS 1P9M process technology with a supply voltage of 0.5 V . It achieved signal to noise and distortion ratio (SNDR) of 55.93 dB, spuri- ous free dynamic range (SFDR) of 77.17 dB. The Walden gure of merit (FoMW) is calculated as 38.67 fJ/conv. Furthermore, this research presents a switched capacitor based SAR ADC using a passive reference charge sharing and charge accumulation. For N- bit resolution, the fully di erential version of this architecture needs only 6 capacitors, which is a signi cant improvement over conventional binary weighted SAR ADC. The proposed SAR ADC is designed and laid out in UMC 180nm 1P6M CMOS technology with a supply voltage of 1.8 V for a target resolution of 11 bit. The total design occupies an area of 568 m x 298 m and consumes a power as less as 0.28 W. It is found that the integral non-linearity (INL) and di erential non-linearity (DNL) of this v ADC are in the range +0.35/-0.84 least signi cant bit (LSB) and +0.1/- 0.6 LSB, respectively. In addition, dynamic performance test shows that the proposed SAR ADC o ers an e ective number of bits (ENoB) of 10.14 and FoMW of 0.12 pJ/conv-step. Finally, a novel switched capacitor integrator based variable resolution hy- brid ADC architecture is proposed. The ADC resolution is programmable from 8-15 bit using a 3- bit control bus (res[2 : 0]). It operates in SAR mode for 8-11 bit resolutions and as the rst-order delta sigma modulator (DSM) with a multi-bit quantizer in 12-15 bit resolutions. A mathematical relationship showing the e ect of mismatch of capacitors on ADC linearity is derived. A fully di erential folded cascode (FC) OTA is designed with programmable unity gain bandwidth (UGB) and slew rate. The proposed ADC, designed and laid out in UMC 180nm standard CMOS technology with a supply voltage of 1.8 V , occupies an area of 0.228mm2. It exhibits SNDR of 45{86 dB and consumes a power of 0.86{98 W across target resolutions (8{15 bits).Item Investigation of Rank Distance Properties of Cyclic, Abelian Codes and Study of their Applicability for Error Correction in Wireless Devices(National Institute of Technology Karnataka, Surathkal, 2021) S, Godkhindi Shrutkirthi.; Acharya, U Shripathi.The useofmultiple-antennaconfigurationscoupledwithsuitablesignalprocessing techniques isoneofthemostpromisingtechniquetoachievehighdataratesandim- provedqualityofserviceinwirelesscommunicationsystems.Overtheyearsmany Multiple-Input Multiple-Output(MIMO)techniqueshavebeenproposedtoachieve high spectralefficiencies,higherdataratesandimprovedreliabilityofcommunica- tion. MIMOtechniqueshavebeenbroadlydividedintotwobroadcategories;Spatial Multiplexingtechniques(SMX)[Foschini(1996)],whichisemployedtoenhancethe capacity ofsystemandSpaceTimeCodingTechniques(STC)[Tarokhetal.(1998)], which havebeenemployedtoimprovethereliabilityofcommunications.InSMX,in- dependent streamsofinformationareradiatedfrommultipletransmitantennas.SMX achieveshighermultiplexinggain,butthistechniqueisnotsuitableforlargescale MIMO, duetoincreaseddecodingcomplexityatthereceiver.SpaceTimeBlockCodes (STBCs) arechannelcodeswhichmaximizespatialdiversity.STBCsarecapableof providingimprovedcodinggainforsamespectralefficiencycomparedwithequivalent diversityschemes.Thisinturnenhancestheoverallreliabilityofthewirelesscom- munication system.STBCscombatchannelimperfectionsandimprovetheintegrityof data transferbycombiningcodinggainwithdiversitygain[Alamouti(1998)]. Space Time(ST)techniquesemployingmultipleantennassufferfromexcessenergy consumption, duetotheactivationofallantennasandtheirpoweramplifierchains.A energyefficientMIMOtechniquenamely,SpatialModulation(SM)wasproposedby Mesleh etal.in[Meslehetal.(2008)].Thistechniqueachievesincreasedenergyeffi- ciencybyactivatingonlyasingleantennaatatime.Spectralefficiencyisfurtherim- provedbyconveyinginformationthroughthemediumofactiveantennaindices[Mesleh et al.(2008)].ThusSMisanenergyconservingtechniquethatutilizesthespatialdi- mension ofmultipleantennaconfigurationtoconveyadditionalinformation.Thecon- cept ofSMwasfurtherimprovedbyappendingaSTBCwiththeSMscheme.This i arrangement wasintroducedby[Basaretal.(2011)]andisknownasSpaceTimeBlock Coded SpatialModulation(STBC-SM).Mostoftheliteratureonthistopichasbeen devotedtothestudyanddesignofOrthogonalSTBC-SMs.However,ithasbeenob- servedthatseveralNon-orthogonalSTBCsprovideperformance(bothintermsofBit Error Rate(BER)andSpectralefficiency)thanconventionalOrthogonal-STBCs.This has leadustostudytheperformanceofseveralfamiliesofNon-OrthogonalSTBCs when employedoverdifferenttypesofchannelsinthisthesis. The useoferrorcontrolcodeenhancestheintegrityofdatatransfer,especiallywhen the communicationsystemisoperatinginanenvironmentperturbedbychannelinduced distortion andnoise.Thisisdonebyintroducingcontrollednumberofredundantbits into theinformationbitstream.Thedisciplineoferrorcontrolcodeshasevolvedfrom designing binarycodesforbinarysymmetricchannelstodesigningcomplexsymbol oriented codesforvarioustypesofchannelsincludingsemiconductor,opticalandmag- netic memories[Moon(2005)].Acomprehensivestudyoftheliteraturehasinspired us toexploretheuseoffullrankcodesoverSTBCs,STBC-SMsystemsandinother evolvingapplications. In thisthesiswehavedesignedaclassoffullrankNonorthogonalSpaceTime Block Codes(NSTBC)whicharederivedfromfullrankCyclicandAbeliancodes.It is observedthattheseNSTBCsoutperformconventionalSTBCswhenemployedover a varietyofchannelfadingconditions.Thedescriptionofresearchworkconducted commences inChapter3afterabriefdescriptionofthestateofart(literaturesurvey) giveninChapter2.InChapter3,wehaveproposedaclassofSpatiallyModulated Non OrthogonalSpaceTimeBlockCodes(SM-NSTBC)designedfrom nlength non- binary Cycliccodesover GF(qm) where q is aprimeand m is orderoftheextensionfield (m n). TheproposedSM-NSTBCoutperformsconventionalSpatiallyModulated Orthogonal SpaceTimeBlockCodes(SM-OSTBC),STBC-SM,SMschemesandtheir variantsintermsofaveragebiterrorrate(ABER)performance. In chapter4,theconceptofreceivespatialmodulationhasbeenexplored.Thetech- nique ofPrecodedSM-NSTBCisproposedtoattainreceivespatialmodulation.Inthis scheme apartofinformationisconveyedbyreceiveantennaindicesinadditiontothe ii conventionallyradiatedNSTBCsymbols.Thisisachievedbyadoptingsuitableprecod- ing techniqueatthetransmitter.TheperformanceofproposedPrecodedSM-NSTBC scheme issynthesizedforspatiallycorrelatedandspatiallyuncorrelatedRayleighfad- ing environments. The designofacooperativecommunicationsystemwithAmplifyandForwardre- laying forSpatiallyModulatedNon-orthogonalSpaceTimeBlockCode(SM-NSTBC- AF) isdiscussedinChapter5.Asinglerelaywithmultipleantennaelementsisassumed to existbetweensourceanddestination.Theinformationtransmittedbythesourceis receivedbytherelayinfirsttimeslot,inthesecondtimeslotthisinformationisam- plified andforwardedtothedestination.Intheproposedscheme,adirectlinkbetween source anddestinationaswellasthecooperativelink(linkbetweensourceanddesti- nation througharelay)areassumedtoexistandoperateintandem.Itisobservedthat this approachprovidesimprovedABERperformanceoverconventionalschemes.This scheme hasbeencomparedwithequivalentschemesemployingamplifyandforward relaying, namelyCooperativeSTBC-SMschemesandCooperativeSMschemes. In Chapter6,thesystemandchannelmodelofHighAltitudePlatforms(HAPs)has been studied.TheeffectofImperfectChannelStateInformation(Imp-CSI)isinvesti- gatedtoobtaininsightintotheperformanceofHAP-MIMOlinkswhendeployedinreal life situations.TheperformanceoffullrankSM-NSTBC,STBC-SMandSM-OSTBC schemes overaHAP-MIMOenvironmentisanalyzed.ItisobservedthatproposedSM- NSTBC outperformscompetingschemesundertheconditionsofperfect-CSIavailabil- ity aswellasimperfect-CSI. In thelastpartofthisthesiswehaveinvestigatedthedesignofNSTBCsderived from non-binaryAbeliancodes.First,adescriptionofthetransformdomainproper- ties andrank-distancepropertiesofAbeliancodeshasbeenprovided.Thisisfollowed by thedesignandsynthesisofSpatiallyModulatedNon-orthogonalSpaceTimeBlock Codes derivedfrom n-length fullrankAbeliancodesover GF(qm). Itisobservedthat the proposedAbelianSM-NSTBCprovidesaperformanceimprovementwhencom- pared withCyclicSM-NSTBCandotherconventionalSTBCs. In summary,inthisthesiswehaveutilizedtherankdistancecharacterizationof iii Cyclic andAbeliancodes.ThecodewordsderivedfromCyclicandAbeliancodeswhen viewedas (m n) matrices possessfullrank.Byemployingsuitablerankpreserving maps wehavesynthesizedfullrankNon-orthogonalSTBCsfromthesefullrankcodes. SM-NSTBC schemeswhichachievehighreliability,highspectralefficiencywithmod- erate decodingcomplexityhavebeendesigned.Theperformanceofproposedschemes havebeenanalyzedunderRayleighfadingenvironments.Further,wehaveexplored the conceptofreceivespatialmodulationandhavesynthesizedschemeswhichemploy full rankNon-orthogonalSTBCs.Thisisfollowedbythederivationofatechnique which employscooperativeSM-NSTBCwithamplifyandforwardrelayingtoimprove the overalllinkperformance.WehavealsoproposedaMIMOschemewhichcanbe employedonaHighAltitudePlatformcommunicationsystem.Theperformanceofthis scheme overHAP-MIMOenvironmenthasbeendetermined.Thedesignofallsystems has beendonekeepinginviewofthe3GPPcommunicationstandards. It isexpectedthattheresultscontainedinthisthesiswillbeusefulforthedesigns of moderncommunicationsystemssuchas5Gandbeyond-5G.Item Automatic Estimation of Personal Characteristics using Speech Data(National Institute of Technology Karnataka, Surathkal, 2021) Babu, Kalluri Shareef.; Vijayasenan, Deepu.Many paralinguistic speech applications demand the extraction of information about the speaker's characteristics from as little speech data as possible. In this work, we explore the estimation of the speaker's multiple physical parameters from the short duration of speech in monolingual (English) and multilingual settings. This has applications in forensics as well as ecommerce. We explore di erent feature streams derived from the speech spectrum at di erent resolutions. Short-term log-mel spectrogram, formant features, and harmonic features are extracted for age and body build estimation (height, weight, shoulder size, and waist size) of the speaker. The statistics of these features accumulated over the speech utterance are used to learn a support vector regression model for speaker age and body build estimation. The experiments performed on the TIMIT dataset show that each of the individual features can achieve results that outperform the default predictor (prediction of the mean of test samples by blindly predicting the mean of training data without looking at the features) in height and age estimation. Furthermore, the estimation errors from these di erent feature streams are complementary, allowing the combination of estimates from these feature streams to improve the results further. The combined system from short audio snippets achieves a performance of 5:2 cm, and 4:8 cm in Mean Absolute Error (MAE) for male and female, respectively, for height estimation. Similarly, in age estimation, the MAE is 5.2 years and 5.6 years for male and female speakers. We extend the same physical parameter estimation system to other body build parameters like shoulder width, waist size, weight, and height. We created two datasets for the speaker pro ling task in a multilingual and multi-accent setting. Speech data is collected along with speaker parameter details (like height, age, shoulder size, waist size, and weight). A pilot dataset Audio Forensic Dataset (AFDS) with 207 speakers across 12 di erent native Indian languages has around 8 hours of native languages speech and around 9 hours of English speech data. Later, a bigger dataset NITK-IISc Multilingual Multi-accent Speaker Pro ling (NISP) dataset has collected, and it has 345 speakers across ve Indian languages as well as English. NISP dataset has around 25 hours of native languages speech data and 32 hours of English speech data. The system can estimate all the physical parameters and showed better improvement than the default predictor in the multilingual and multi-accent setting. The duration analysis shows that the state-of-the-art results can be achieved using short utterances(around 12 seconds) of speech data. To the best of our knowledge, i this is the rst attempt to use a common set of features for estimating the di erent physical traits of a speaker from short utterances. An integrated end-to-end deep neural network architecture is proposed for joint prediction of all the physical parameters. A novel initialization scheme for deep neural architecture is introduced, which avoids a large training dataset requirement. On the TIMIT dataset, the system achieves an RMSE error of 6:85 and 6:29 cm for male and female height prediction. In the case of age estimation, the RMSE errors are 7:60 and 8:63 years for male and female, respectively. Analysis of shorter durations of speech reveals that the network only degrades around 3% at most with only 1 second of the speech input. Also, the performance saturates around 3seconds in predicting the height and age of a speaker using the TIMIT dataset. In the multilingual setting using collected datasets, the predicted error metrics are less than the default predictor except for female age prediction in both AFDS and NISP datasets. In male speakers, the system performance is less than the default predictor in height estimation of the NISP dataset.Item Algorithms for Color Normalization and Segmentation of Liver Cancer Histopathology Images(National Institute of Technology Karnataka, Surathkal, 2021) Roy, Santanu.; Lal, Shyam.With the advent of Computer Assisted Diagnosis (CAD), accuracy of cancer detection from histopathology images is significantly increased. However, color variation in CAD system is inevitable due to variability of stain concentration and manual tissue sectioning. Small variation in color may lead to misclassification of cancer cells. Therefore, color normalization is the first step of Computer Assisted Diagnosis (CAD), in order to reduce the inter-variability of background color among a set of source images. In this thesis, first a novel color normalization method is proposed for Hematoxylin and Eosin (H and E) stained histopathology images. Conventional Reinhard algorithm is modified in our proposed method by incorporating fuzzy logic. Moreover, mathematically it is proved that our proposed method satisfies all three hypotheses of color normalization. Furthermore, several quality metrics are estimated locally for evaluating the performance of various color normalization methods. Experimental result reveals that our proposed method has outperformed all other benchmark methods. The second step of CAD is nuclei segmentation which is the most significant step since it enables the classification task computationally efficient and simple. However, automatic nuclei detection is fraught with problems due to highly textured nuclei boundary and various size and shapes of nuclei present in histopathology images. In this thesis, a novel edge detection technique is proposed for segmenting the nuclei regions in liver cancer Hematoxylin and Eosin (H and E) stained histopathology images, based on the notion of computing local standard deviation value. Moreover, the edge-detected image is converted into a binary image by using local Otsu thresholding and thereafter, it is refined by an adaptive morphological filter. The experimental result indicates that proposed segmentation method overcomes the limitations of existing unsupervised methods and subsequently its performance is also comparable with deep neural models. To the best of our knowledge, our proposed method is the only unsupervised method iii which achieves nuclei detection accuracy closest to 1 (0.9516). Furthermore, two more quality metrics are computed in order to measure the performance of nuclei segmentation methods quantitatively. The mean value of quality metrics reveals that our proposed segmentation method outperforms other existing methods both qualitatively and quantitatively.Item Development of CeO2 based High Performance MEMS Oxygen Gas Sensor(National Institute of Technology Karnataka, Surathkal, 2021) N, Ramshanker.; Mohan, S.; Bhat, M S.A gas sensor is a device that is used to measure the concentration of gas in its vicinity. It can also be used as a leak detector to detect a gas leak or other emissions. Extensive research is being carried out on gas sensor in designing miniaturized and cost e ective sensors that possess the required characteristics of high sensitivity, selectivity and stability with respect to a speci c application. Fast and unambiguous analysis of human surroundings will be in the near future inseparable part of public health, security and life quality control. Semiconductor metal oxide gas sensors stand out among the other types of sensors because of their simplicity and low cost. In the present work, we developed scalable, high sensitivity, fast response and low operating temperature CeO2 thin lm based oxygen sensors. A systematic investigation has been carried out to develop the high performance oxygen sensor which includes the optimization and integration of sensor lm and micro-heaters. CeO2 thin lms of di erent thicknesses ranging from 90 nm to 340 nm have been deposited at 400oC using RF magnetron sputtering on Al2O3 substrates. Characterization techniques such as Ellipsometry, XRD, XPS and AFM have been used to characterize the CeO2 lms for their thickness, structural, compositional/chemical and surface morphological properties. From XRD and XPS data, it has been observed that all the lms are polycrystalline and with thickness more than 195 nm are stoichiometric. It has also been observed that the resistivity of the lms depends on the texture coe cient of (200) plane of CeO2. It has been found that 260 nm thick lm has high textured coe cient of (200) plane which shows minimum electrical resistivity and maximum sensitivity towards the oxygen gas. The CeO2 lm with an optimum thickness of 260 nm has shown very high sensitivity (12.6), fast response time ( 10 s) and recovery time (15 s) at a low operating temperature of 400oC, which are the best values reported till date in case of undoped CeO2 thin lm based sensors. The response time of CeO2 based sensor may be reduced further by increasing the conductivity of the CeO2 lms with appropriate dopants. v A novel technique was used for the synthesis of CeO2-HfO2 mixed oxide thin lms using RF sputtering. The mixed oxide lms showed better sensing performance in comparison with pure CeO2 lms. The Hf atomic concentration was controlled varying the size and number of HfO2 pellets to achieve the best sensing performance. The CeO2-HfO2 mixed oxide sensor with 10-11% of Hf concentration showed best sensitivity ( 15), response time (8 s) and recovery time (10 s) at a low operating temperature of <400oC reported till date. From XRD and XPS data, it was understood and concluded that the best sensing characteristics of CeO2-HfO2 mixed oxide lm with 10-11% atomic concentration of Hf can be attributed to the existence of a highly reactive plane (200) with the highest surface energy and a strongly reduced surface with oxygen vacancy formation due to the presence of Ce3+ ions and HfOx, x<2 on the surface of the mixed oxide lm. The sensor performance is reproducible without any drift in the base line resistance. Microheaters play a crucial in MEMS gas sensor technology. Several microheater designs have been studied, however new heater patterns and designs are required to achieve excellent temperature uniformity and low power consumption. Here in this work, the area of the heater is optimized in order to increase the resistance by adopting novel designs / geometries. The single meander shape was taken as a reference design. After several modi cations, iterations and optimizations, two di erent geometrical structures namely Perforated Type 1 and Type 2 Platinum microheaters of dimension 500 m x 500 m were designed and analyzed using FEM based software COMSOL. The simulated results show the temperature being distributed uniformly across the entire structure in both the designs. The designed microheaters were fabricated and characterized thermally and electrically and showed excellent temperature uniformity and the power consumed to obtain the temperature of 400oC is nearly between 1.14 to 1.44 W which is considerably lower than reported values in the literature. The fabricated heaters were integrated into a gas sensor and the device was tested for oxygen gas. The sensing results were found to be in good agreement with the results obtained using a conventional heater. Readout circuits are circuits used to convert the sensed signal, such as vi voltage, current, resistance etc. or changes in it into a more convenient form of the same or di erent type of signal for further processing. A highly e cient 3-stage op-amp based readout circuit is designed to measure the dynamic change of the sensing lm resistance. The three stages are namely : constant current source, bu er ampli er and feedback ampli er. The 3 ampli er con guration with a constant current source is used to measure the change in resistance and voltage is measured across the resistance under test. The real-time simulation results show that the circuit is highly e cient and linear.Item Modelling of Gamakas for Karnatic Flute Music Synthesis(National Institute of Technology Karnataka, Surathkal, 2021) M, Ragesh Rajan.; Vijayasenan, Deepu.In this work, we propose a spectral model to efficiently synthesise Karnatic bamboo flute music from the notes, duration, and raga information of a song. Karnatic flute music synthesis from basic notations is a challenging problem due to two major reasons. The first one is that the gamakas are generally omitted from the musical notations in the tradition of KM. Hence, for the automatic synthesis of KM, the gamakas associated with every note need to be predicted from the musical notations. The second reason is the continuously varying pitch contour of a note in the presence of gamakas. We propose a method to detect the presence and type of gamakas associated with each note in a data-driven manner, from the annotated symbolic music alone. In this regard, we propose features based on the notes of the song. These features are used as inputs to a Random Forest Classifier (RFC). From our experiments, the accuracy values obtained for predicting the presence and type of gamakas are 77% and 70%, respectively. These are significantly better than random classification accuracies. We also analyse the importance of neighbourhood of notes for the detection and classi cation of gamakas. It is observed that the best accuracy is obtained for gamaka presence detection when a both-sided neighbourhood of size three is considered; and the best accuracy for gamaka type prediction is obtained with a both-sided neighbourhood of size one. The analysis performed on the training data reveals that there is information contained in these neighbourhoods for distinguishing between gamaka and non-gamaka notes. For synthesising Kar _ n at.ic ute music, we model three di erent components of the ute sound, namely, pitch contour, harmonic weights, and time domain amplitude envelope. Cubic splines are used to parametrically represent these components. Subjective analysis of the results shows that the proposed method is better than the existing popular methods in terms of tonal quality as well as the propriety of rendering gamakas. Hypothesis test results show that the observed improvements over other methods are statistically signi cant at 95% con dence interval.Item Design and Analysis of Reliable Wireless Optical Communication System for the Underwater Channels(National Institute of Technology Karnataka, Surathkal, 2021) Naik, Ramavath Prasad.; Acharya, U Shripathi.Internet of Things (IoT) is a technology that deals with devices and protocols employed to interconnect smart devices that are deployed in benign and hazardous environments to collect or exchange information. In recent years, attempts have been made to interconnect devices deployed in underwater environments. These devices have been employed in underwater environments to monitor physical phenomena such as climate and ocean monitoring and enable activities such as underwater surveillance and ocean exploration. This technology is referred to as the Internet of Underwater Things (IoUT). IoUT is a technology, which requires a high degree of information integrity, high data transfer rates and energy e ciency for e ective deployment. Optical signaling gives the exibility of providing high data-rates than that of acoustic and RF signaling for medium link-ranges. Hence, optical signaling is an excellent candidate to enable high speed IoUT communication link between the underwater devices. The devices that employ optical signaling to enable communication between underwater vehicles and the underwater channel together constitute the Underwater Wireless Optical Communication (UWOC) system. The transmitted optical data experiences beam attenuation, turbulence and pointing errors, all of which can contribute to introduce errors in the received data stream. We initiated the work in this thesis by conducting experimental work to determine various parameters a ecting the propagation of light in an underwater channel and determining the optimum wavelength for UWOC communication. This is followed by a discussion of channel models that are appropriate descriptors of weak and strong turbulence in the underwater medium. Analytic models describing strong and weak turbulence have been derived, and simulation studies (Monte- Carlo simulations) that determine the accuracy of these analytic models have been carried out. The performance of UWOC system is mainly dependent on the underwater turbulence, beam attenuation and pointing errors, to mitigate these e ects we have introduced multiple input multiple output (MIMO), forward error control codes and Space-Time Block Codes (STBCs) to the proposed UWOC system. In many instances, the UWOC link operating under the surface of the water has v to be linked with a RF system operating over the water surface. Such a link is referred to as a cooperative RF-UWOC system. Channel models for the combined RF-UWOC system have been drawn up. It has been recognized that in addition to channel induced impediments, a major cause of link outage is the introduction of pointing errors due to the physical displacement of the transmitter-receiver pair, which causes the Line of Sight (LOS) requirement to be disturbed. Loss of LOS has the potential to severely compromise the working of the UWOC/ RFUWOC system. The performance of the RF-UWOC (co-operative IoT and IoUT) system in the presence of pointing errors has been studied and various outage probabilities have been determined. We have concluded the technical contributions of the thesis by studying underwater image transmission through the turbulent oceanic medium and suggesting various remedial techniques for proper image reception and enhancement. The thesis has been concluded by drawing conclusions from the research work conducted in the thesis and suggesting avenues for further research.Item Object Extraction from Remotely Sensed Aerial Images(National Institute of Technology Karnataka, Surathkal, 2021) Eerapu, Karuna Kumari.; Lal, Shyam.; Narasimhadhan, A V.The topographical map of the Earth is recorded by capturing high spatial resolution (HSR) aerial images from higher altitudes using Aircraft, Helicopters, and Unmanned Aerial Vehicles (UAVs). The object extraction from HSR remotely sensed aerial imagery data is a prerequisite in a large number of applications such as planning urban cities, accessing disasters, managing traffic congestion, and providing up-to-date road maps. The development of automated methods to extract objects accurately is highly required for the applications, as mentioned above. However, this is a notoriously challenging task in the field of remote sensing. The deep learning field has gained massive interest due to its ability to learn after the availability of high-volume data and computational resources. This thesis investigates evolutionary optimization based framework for quality enhancement of remotely sensed aerial images and various Convolutional Neural Network (CNN) based approaches of deep learning and proposes an enhancement and object segmentation techniques for HSR remotely sensed aerial imagery data. In the first part of the thesis, the contrast enhancement technique to improve the visual quality of remotely sensed aerial images is presented. The visual quality of captured aerial images is impaired due to the atmospheric effects and limitations of sensors. The visual quality of aerial images needs to improve to extract the hidden object details by increasing the pixel intensity ranges. Most of the techniques in the literature do not consider multi-objective function optimization, either computationally complex or less stable. There is a high demand to introduce a framework to find stable optimum values for multi-objective function with reduced computational complexity. The new framework is introduced to restore the visual quality of images by adjusting saturation, color values, and finally enhanced the contrast of the images through Particle Swarm Optimization (PSO). The experimental and visual quality results showed that the proposed framework outperformed over other state-of-the-art quality restoration techniques. Next, in the thesis, two deep learning-based semantic segmentation architectures are introduced to extract diversified objects from aerial images. The deep learning-based approaches achieved better results as compared with conveni tional techniques. The conventional techniques involve manual feature extraction in multiple stages, which demands to maintain the accuracy of each stage to get overall high accuracy. On the other hand, the Convolutional Neural Network (CNN) based architectures in deep learning field provide object segmentation by learning the image features from a higher volume of imagery data. Plentiful of semantic segmentation architectures are present in the literature to perform object segmentation. However, there is still scope to improve accuracy while segmenting small objects in high-resolution aerial images. In aerial images, objects appear irregularly shaped and tiny in size and also present in dominant background scenarios. Further, class-wise pixels and background pixels are in the ratio of ones-to-tens, which leads to class imbalance problem. Therefore it is very challenging to obtain better prediction accuracy and completeness by preserving the connectivity without any gap between successive objects. In this thesis, a Dense Refinement Residual Network (DRR Net) is proposed for road extraction from aerial imagery data. The DRR Net is introduced based on dense convolutions for feature learning, residual connections to guide the learning path, and provides refinement through the stacking of DRR modules in the network. In the final part of the thesis, the robust encoder and decoder architecture, namely O-SegNet for objects segmentation from high-resolution aerial imagery data, is introduced. The O-SegNet provides emphasis to relevant object details and extracts global context through the selfattention mechanism and multi-level pooling. Both of these proposed architectures are trained with composite loss function to focus more on small object instances. The proposed semantic segmentation architectures have achieved significant quantitative and qualitative results compared with the other existing semantic segmentation architectures.Item Design and Analysis of Leakywave Antennas for Novel Functionalities(National Institute of Technology Karnataka, Surathkal, 2021) R, Karthik.; Krishnamoorthy, K.This thesisdealswiththeeffectivedesignofplanarleakywaveantennas(LWA)for novelfunctionalities.Theworkdescribesamethodologybasedonsingle-layertrans- mission basedGoubaulineandinvestigationofasymmetryinHalf-modesubstratein- tegratedwaveguide(HMSIW)periodicLWAstoaddressverydifferentantennaconfig- urations. FournovelLWAsareproposedtoachievewideanglebeamscanning,single layer basedEnd-Fireradiation,dual-bandandsimultaneousdual-beamscanningfunc- tionalities. WidebeamscanningisdesignedusingGoubauline.Incontrasttotheconventional LWAwithasmallscanninganglerangeandlowefficiency,thisworkemployedape- riodically bendingGoubauline,whichnotonlybringsinaperiodicperturbationfor leakywaveradiation,butalsoenhancesthescanningrangeduetotheincreaseddelay for eachlineelement.Theproposedantennafeaturesgoodradiationperformanceand has acompactandlowprofileconfiguration.Also,asimpleandlow-profilePlanar Goubau linebasedEndfireantennaisdesigned.Endfireradiationisachievedbymod- ifying theGoubaulineintoinvertedperiodicarrangementofV-shapedunitcells.The proposed endfireantennahasasinglemetalliclayerwithasimpleconfigurationwhich is easytofabricateandalsoeasytointegratewithotherelectronicscircuits. Most oftheexistingLWAsaregenerallydesignedforasinglebandapplication.In this work,byetchingsimplespiraltypeslotsontheHMSIWcavity,anoveldual-band dual-polarized LWAisdesigned.Theantennaradiatesthelinearlypolarized(LP)wave in thelowerbandandcircularlypolarizedwaveinthehigheroperatingband.Inthe higher operatingbandantennaexhibitsbothright-handcircularpolarization(RHCP) as wellasthelefthand(LHCP)circularpolarizationinthenearmainbeamdirection. Also, asymmetryisinvestigatedinHMSIWLWA.Basedonasymmetry,twodifferent configurations ofdual-bandLWAsaredesigned.ThefirstconfigurationradiatesLP v wavesinthelowerbandandCPwavesintheupperbandwhentheunitcellsarecas- caded sequentially.Thesecondconfiguration,wherethesidewallviaoftheHMSIW unit cellsareconnectedalternatingly,providessimultaneousdual-beamswithdiffer- ent polarizationintheupperoperationalbandinadditiontotheLPbeaminthelower operational band.ThesenovelfunctionalitiesofLWAcanbeeffectivelyusedformulti- objectivetrackingapplicationsandforenhancingthechannelcapacityofautomotive and satellite-basedwirelesscommunicationsystems.Sincebothantennaconfigurations had dualfrequencybandswithdifferentpointingdirectionsinscanning,itcanalsobe useful forvehicularapplicationstotrackdifferentobjectsinitsnearbylocations Proposed designsarefabricatedusingLPKFprotomatmachineS103andexperi- mentally verifiedusingradiationpatternmeasurementset-up.Impedancebandwidth, peak gain,axialratiobandwidthandradiationefficiencyaremeasuredandcompared with thesimulatedresults.Theperformanceoftheproposedworksarecomparedwith the previousliterature.Item Low Power Nonbinary Weighted Successive Approximation Register Analog to Digital Converters(National Institute of Technology Karnataka, Surathkal, 2021) Bhat, Kalpana G.; T, Laxminidhi.Modern instrumentation systems and data acquisition (DAQ) systems demand low to medium resolution, medium speed analog to digital converters (ADC). For DAQ systems with dedicated ADCs per channel, ADC core area is of prime concern to minimize cost/chip. For DAQ systems with a single ADC for multiple channels, it may be required to operate the ADC at different resolutions depending on the channel it is digitizing. In such a case, programmable resolution ADC is an added advantage. Even though the traditional successive approximation register (SAR) ADCs are popular in these applications, they consume large chip area. To optimize ADC over area, nonbinary weighted capacitive (NBWC) digital to analog converter (DAC) architectures are proposed in this work, which makes use of a fixed number of nonbinary weighted capacitors, low power transconductors and switches. Initially, as a proof of concept, a DAC architecture using minimum number of unit size capacitors and buffers is designed for the SAR ADC. In order to enhance sampling speed, a 2 bit per cycle algorithm is developed instead of the conventional 1 bit per cycle evaluation. The designed 8 bit ADC is simulated in 180 nm technology and supply voltage of 3.3V. The ADC designed could achieve a conversion speed of 500 kS/s with a power dissipation of 1.8mW. Integral non linearity (INL) error and differential non linearity (DNL) error are less than 0.5 LSB. The simulated signal to noise and distortion ratio (SNDR) at the input frequency of 56.64 kHz is 48.14 dB and at 232.42 kHz is 47.03 dB. Though the proof of concept ADC simulation results were encouraging, the voltage dependent nature of the parasitic capacitance posed by the switches were found to be the bottleneck in achieving a low value of area efficiency (AE). Therefore, the switched capacitor DAC architecture used in the proof of concept ADC is modified so that the reference generation circuit for the 2 bit per cycle evaluation is least sensitive to parasitics as well as the ADC resolution is made programmable. A programmable 4 to 8 bit DAC is further designed for the SAR ADC with increased sampling rate of 1.2MS/s at 8 bit mode. Passive charge sharing and charge recycling through low power buffers is adopted in the NBWC DAC. A mathematical analysis has also been carried out to justify the parasitic insensitive nature of the reference generation scheme. The proposed idea has been validated by designing a 4 to 8 bit ADC in 90 nm CMOS technology for operations on 1V supply. The post layout simulation of the designed ADC has 1.2MS/s sampling rate at 8 bit mode with a power consumption of 185 μW achieving an effective number of bits (ENOB) of 7.6. The active area of the designed ADC is 0.06mm2, resulting in an AE as low as 295.66 μm2/code. The DAC resolution scaling and the use of variable sampling rate maximizes efficiency at lower resolutions. Therefore figure-of-merit (FOM) degrades only by a factor 4.7 for resolution scaling from 8 bits to 4 bits. This is a significant improvement over the 16x degradation expected from 8 bit to 4 bit resolution scaling by truncating the bits. Another work, as part of this thesis, is exploring switched capacitor integrator (SCI), NBW-CDAC based SAR ADC. The objective is to develop resolution independent, programmable resolution architecture using minimum number of NBW capacitors and switches for fully differential approach. A compact SC integrator based nonbinary weighted 10 bit SAR ADC is designed for low power portable devices and data acquisition systems. This technique requires an operational transconductance amplifier (OTA), a comparator and six equal sized capacitors of moderate value for fully differential approach. The reference voltage is generated by charge sharing between a reference capacitor and the input capacitor of an switched capacitor (SC) integrator. The DAC voltage for comparison is generated by accumulating the charges on the integrating capacitor. ADC being fully differential, has a wide input range and is largely parasitic insensitive. As a stand alone data converter it has a small capacitance spread and hence its input capacitance is easy to drive. A 10 bit 0.9MHz sampling rate SAR ADC is designed using 180 nm CMOS technology, operating at 1.8V supply, has ENOB of 9.5 at Nyquist frequency. The area occupied by the designed ADC is 0.05mm2, resulting in an AE as low as 69.32 μm2/code. It is found that the designed ADCs are having better AE as compared to state of the art binary weighted capacitive (BWC) and NBWC SAR ADCs in the literature. These designs are also useful to build high resolution, low power, high speed ADCs such as pipelined, sub-ranging, folding and interpolating ADCs.