In the paper, the variation of the intensity of the geomagnetic field force is analysed in time and space. For the research, the data from measurements of the intensity of the geomagnetic field force at four airports (Kaunas, Klaip˙eda, Palanga andVilnius) and 6 geomagnetic field repeat stations aswell as the data from Belsk Magnetometric Observatory (Poland) were used. For the data analysis, the theory of covariance functions was applied. The estimates of the cross-covariance functions of the measured intensity of the geomagnetic field force or the estimates of auto-covariance functions of single data were calculated according to the random functions created from the force intensity measurement data arrays. The estimates of covariance functions were calculated upon varying the quantization interval on the time scale and applying the software created using Matlab package of procedures. The impact of radars of airports on the intensity of geomagnetic field variation and on changes of their covariance functions was established.

Time domain analysis is used to determine whether A/D converters that employ higher order sigma-delta modulators, widely used in digital acoustic systems, have superior performance over classical synchronous A/D converters with modulators of first order when taking into account their important metrological property which is the magnitude of the quantization error. It is shown that the quantization errors of delta-sigma A/D converters with higher order modulators are exactly on the same level as for converters with a first order modulator.

Audio data compression is used to reduce the transmission bandwidth and storage requirements of audio data. It is the second stage in the audio mastering process with audio equalization being the first stage. Compression algorithms such as BSAC, MP3 and AAC are used as standards in this paper. The challenge faced in audio compression is compressing the signal at low bit rates. The previous algorithms which work well at low bit rates cannot be dominant at higher bit rates and vice-versa. This paper proposes an altered form of vector quantization algorithm which produces a scalable bit stream which has a number of fine layers of audio fidelity. This modified form of the vector quantization algorithm is used to generate a perceptually audio coder which is scalable and uses the quantization and encoding stages which are responsible for the psychoacoustic and arithmetical terminations that are actually detached as practically all the data detached during the prediction phases at the encoder side is supplemented towards the audio signal at decoder stage. Therefore, clearly the quantization phase which is modified to produce a bit stream which is scalable. This modified algorithm works well at both lower and higher bit rates. Subjective evaluations were done by audio professionals using the MUSHRA test and the mean normalized scores at various bit rates was noted and compared with the previous algorithms.

The contradiction between the restriction of grating manufacturing technology and high-resolution measurement requirements has been the focus of attention. The precision requirement of angle calculation during the digital subdivision processing of a Moiré signal is focused on, the causes of errors in the solution of arcsine function are analysed, and an improved coordinate rotation digital computer (CORDIC)with double-rotation iteration is proposed by discussing the principle of the conventional CORDIC in detail herein. Because the iterative number and data width of the improved CORDIC are limited by the finite digital circuit resources and thus determine the calculation accuracy directly, subsequently the overall quantization error (OQE) of the improved CORDIC is analysed. The approximate error and rounding error of the algorithm are deduced, and the error models of iterative number and data width are established. The validity and application value of the improved CORDIC are proved through simulations and experiments involving a subdividing circuit. The corresponding relation between the approximate error, rounding error and iteration number, as well as the bit width are proved by quantization. The error of subdivision with the improved CORDIC, obtained through a calibration experiment, is within ±0.5′′ and the mean variance is 0.2′′. The results of the research can be applied directly to a digital subdivision system to guide the parameter setting in the iterative process, which is of crucial importance in the quantitative analysis of error separation and error synthesis.

This paper explores cost-effective alternatives for resource-constrained environments in the context of language models by investigating methods such as quantization and CPUbased model implementations. The study addresses the computational efficiency of language models during inference and the development of infrastructure for text document processing. The paper discusses related technologies, the CLARIN-PL infrastructure architecture, and implementations of small and large language models. The emphasis is on model formats, data precision, and runtime environments (GPU and CPU). It identifies optimal solutions through extensive experimentation. In addition, the paper advocates for a more comprehensive performance evaluation approach. Instead of reporting only average token throughput, it suggests considering the curve’s shape, which can vary from constant to monotonically increasing or decreasing functions. Evaluating token throughput at various curve points, especially for different output token counts, provides a more informative perspective.

Digital speech copyright protection and forgery identification are the prevalent issues in our advancing digital world. In speech forgery, voiced part of the speech signal is copied and pasted to a specific location which alters the meaning of the speech signal. Watermarking can be used to safe guard the copyrights of the owner. To detect copy-move forgeries a transform domain watermarking method is proposed. In the proposed method, watermarking is achieved through Discrete Cosine Transform (DCT) and Quantization Index Modulation (QIM) rule. Hash bits are also inserted in watermarked voice segments to detect Copy-Move Forgery (CMF) in speech signals. Proposed method is evaluated on two databases and achieved good imperceptibility. It exhibits robustness in detecting the watermark and forgeries against signal processing attacks such as resample, low-pass filtering, jittering, compression and cropping. The proposed work contributes for forensics analysis in speech signals. This proposed work also compared with the some of the state-of-art methods.

The paper describes the construction, operation and test results of three most popular interpolators from a viewpoint of time-interval (TI) measurement systems consisting of many tapped-delay lines (TDLs) and registering pulses of a wide-range changeable intensity. The comparison criteria include the maximum intensity of registered time stamps (TSs), the dependency of interpolator characteristic on the registered TSs’ intensity, the need of using either two counters or a mutually-complementing pair counter-register for extending a measurement range, the need of calculating offsets between TDL inputs and the dependency of a resolution increase on the number of used TDL segments. This work also contains conclusions about a range of applications, usefulness and methods of employing each described TI interpolator. The presented experimental results bring new facts that can be used by the designers who implement precise time delays in the field-programmable gate arrays (FPGA).

This paper deals with the amplitude estimation in the frequency domain of low-level sine waves, i.e. sine waves spanning a small number of quantization steps of an analog-to-digital converter. This is a quite common condition for high-speed low-resolution converters. A digitized sine wave is transformed into the frequency domain through the discrete Fourier transform. The error in the amplitude estimate is treated as a random variable since the offset and the phase of the sine wave are usually unknown. Therefore, the estimate is characterized by its standard deviation. The proposed model evaluates properly such a standard deviation by treating the quantization with a Fourier series approach. On the other hand, it is shown that the conventional noise model of quantization would lead to a large underestimation of the error standard deviation. The effects of measurement parameters, such as the number of samples and a kind of the time window, are also investigated. Finally, a threshold for the additive noise is provided as the boundary for validity of the two quantization models

The aim of this work is to present problems related to tinnitus symptoms, its pathogenesis, hypotheses on tinnitus causes, and therapy treatment to reduce or mask the phantom noise. In addition, the hypothesis on the existence of parasitic quantization that accompanies hearing loss has been recalled. Moreover, the paper describes a study carried out by the Authors with the application of high-frequency dither having specially formed spectral characteristics. Discussion on preliminary results obtained and conclusions are also contained.