This thesis describes a novel approach that utilizes a special property of the Hanning window to accurately detect the fundamental frequency of the data signal in the presence of harmonic and interharmonic interference. After obtaining the fundamental frequency, the same procedure is applied to all possible harmonics to be filtered for further analysis of the interharmonics. The proposed approach is validated using numerical tests. In the literature, different authors have shown the difficulty of extracting a synchronized sampling frequency from the analyzed signal and discussed the effect of deviation of the fundamental in the presence of harmonics. This thesis suggests a new approach to overcome the difficulties, even if the interharmonics present are in frequency bins near the fundamental. In order to perform the interharmonic analysis, the author followed the IEC standard draft signal processing recommendations, with the exception of using a weighted Hanning window instead of a rectangular window in order to minimize the effect of the spectral leakage, and to minimize the effect of interharmonics on the main harmonics.

This thesis presents the results of a comprehensive multi-wavelength study of the high mass X-ray binary 2S 0114+650. This enigmatic source has previously been proposed to be the first in a new class of super-slow X-ray pulsars, containing a neutron star revolving once every 2.7 h. The 11.6 d orbital period of this system has been well established in both X-ray and optical wavelengths. During the initial stages of the research presented in this thesis we discovered an additional 30.7 d ???super-orbital??? modulation in the X-ray emission from this source. While similar periodicities seen in other X-ray binaries are commonly attributed to the precession of a warped accretion disc, the observational evidence suggests the absence of such a disc in 2S 0114+650. The purpose of this project is thus to determine the nature of the super-orbital modulation and to better constrain the astrophysical parameters of this system. To investigate the long-term variability we analysed ~8.5 yr of archived data from the Rossi X-ray Timing Explorer space telescope. The problem of the spurious ~24 h periods in this data was solved as a by-product of these studies. Follow-up pointed observations were obtained with this satellite in order to examine the spectral and temporal behaviour over the spin, orbital and super-orbital timescales. Independent confirmation of the super-orbital modulation was performed using ~2 yr of data from the INTEGRAL satellite obtained during a long-term monitoring campaign of the Cassiopeia region. The evolution of the spin, orbital and super-orbital periods over ~10 yr was examined using archived data from the Rossi X-ray Timing Explorer satellite. Radio observations were performed with the Giant Meterwave Radio Telescope to search for any radio emission associated with this source and to determine whether it is variable over the known periodicities. Near infrared observations were performed with the Mt Abu telescope to determine whether a Be star nature can be ruled out for the optical component. Finally, a statistical analysis of the properties of the confirmed super-orbital X-ray binaries was performed in order to search for commonalities between these systems.

This thesis presents the results of a comprehensive multi-wavelength study of the high mass X-ray binary 2S 0114+650. This enigmatic source has previously been proposed to be the first in a new class of super-slow X-ray pulsars, containing a neutron star revolving once every 2.7 h. The 11.6 d orbital period of this system has been well established in both X-ray and optical wavelengths. During the initial stages of the research presented in this thesis we discovered an additional 30.7 d ???super-orbital??? modulation in the X-ray emission from this source. While similar periodicities seen in other X-ray binaries are commonly attributed to the precession of a warped accretion disc, the observational evidence suggests the absence of such a disc in 2S 0114+650. The purpose of this project is thus to determine the nature of the super-orbital modulation and to better constrain the astrophysical parameters of this system. To investigate the long-term variability we analysed ~8.5 yr of archived data from the Rossi X-ray Timing Explorer space telescope. The problem of the spurious ~24 h periods in this data was solved as a by-product of these studies. Follow-up pointed observations were obtained with this satellite in order to examine the spectral and temporal behaviour over the spin, orbital and super-orbital timescales. Independent confirmation of the super-orbital modulation was performed using ~2 yr of data from the INTEGRAL satellite obtained during a long-term monitoring campaign of the Cassiopeia region. The evolution of the spin, orbital and super-orbital periods over ~10 yr was examined using archived data from the Rossi X-ray Timing Explorer satellite. Radio observations were performed with the Giant Meterwave Radio Telescope to search for any radio emission associated with this source and to determine whether it is variable over the known periodicities. Near infrared observations were performed with the Mt Abu telescope to determine whether a Be star nature can be ruled out for the optical component. Finally, a statistical analysis of the properties of the confirmed super-orbital X-ray binaries was performed in order to search for commonalities between these systems.

No description available.

This work focuses on fixing measurement inaccuracies to which models and figures of merit are susceptible in two wireless communication environments: power amplifier and multipath. To emulate or rate the performance of these environments, models and figures of merit, respectively, are often used. The usefulness of a model depends on how accurately and efficiently it emulates its real-world counterpart. The usefulness of a figure of merit depends on how accurately it represents system behavior. Most discussions on the challenges and trade-offs faced in modeling nearly always focus on the complexity of the device or channel of interest and the resultant difficulty in describing it. Similarly, figures of merit are meant only to summarize the performance of the device or channel. At some point, either in generation or verification of a model or figure of merit, there is a dependence on measured data. Though the complexity and performance of the device or channel are challenges by themselves, there are other significant sources of distortion that must be minimized to avoid errors in the measured data. For this work, the unique distortion of power amplifier and multipath environments is considered, and then errors in measurement which would obscure these distortions are eliminated. Specifically, three measurement issues are addressed: 1) identifying measurement setup artifacts, 2) achieving consistent measurement results and 3) reducing variations in the environment. This work contributes to increasing the accuracy of microwave measurements used in the modeling of nonlinear high-power amplifiers and used in figures of merit for power amplifiers and multipath channels.