Multirate Signal Research Articles

Predicting Oil Temperature in Electrical Transformers Using Neural Hierarchical Interpolation

Effective electricity consumption planning is critical for power distribution. Ensuring the distribution network aligns with expected demand fluctuations is a challenging task influenced by various time‐related and seasonal variables. This study focuses on improving transformer oil temperature forecasting, an indicator of transformer health, using the neural hierarchical interpolation for time series (NHITS) model. The NHITS model’s architecture is designed to handle long‐term forecasting efficiently, making it ideal for capturing extended trends in transformer oil temperature. It incorporates multirate signal sampling via MaxPool layers and hierarchical interpolation to merge predictions across different time scales. The proposed methodology involves two key phases: data preparation and model development. In the data preparation phase, the electricity transformer temperature (ETT) datasets are used, normalized with a standard scaler, and essential features such as oil temperature and external power load are selected. During the model development phase, the proposed NHITS model is trained and its hyperparameters are optimized for optimal performance. The study evaluates the model’s performance under various conditions, including the comparison of multivariate and univariate time series, the effects of short and long‐term forecasting horizons, and the impact of temporal resolution. The model was validated using the ETT dataset, and our results were benchmarked against a previous study that employed the same dataset and used the Informer model. The results indicate that the NHITS model outperforms the Informer model, showing an average decrease of 51.37% in mean squared error (MSE) and 37.83% in mean absolute error (MAE). These findings highlight the model’s ability to capture both long‐term and short‐term characteristics of time series data, making it a promising solution for forecasting transformer oil temperatures.

Research on future 6G green wireless networks

Research on future 6G green wireless networks

Filter-Bank-Enabled Leaky-Wave Antenna Array Technique for Full-Band-Locked Radar System in Stitched Frequency-Space Domain

Inspired by the filter-bank (FB) concept that is normally used for multi-rate signal processing, an FB-enabled array technique of leaky-wave antennas (LWAs) is proposed and studied for creating full-band-locked frequency-scanning radar (FSR) systems in a stitched frequency-space domain. This is mainly for addressing the coupling dilemma between the range and angle resolutions, which is historically and naturally inherited from a conventional FSR. Firstly, the typical frequency-modulated continuous-wave system architecture is selected to exemplify and recall the characteristics of a conventional FSR with emphasis on its resolution coupling. Then, a radar solution featuring a stitched frequency-space domain and depending on an FB-enabled array of LWA channels is introduced for realizing the desired resolution decoupling. With the radar equation, FB-related conditions for obtaining the critical “frequency-space stitching” are analytically derived and converted into relevant design specifications of an LWA array, including engineered beam-scanning functions, beam-crossovers, and phase alignments. To facilitate the practical implementation of such a front-end array technique, a detailed and generalized design flow is developed. Finally, for a simple proof of concept, an FB-enabled two-channel LWA array is modeled, fabricated, and measured. Simulated and measured results are in reasonable agreement; both demonstrate the desired “frequency-space stitching” behavior, i.e., enhanced spectrum bandwidth and widened radiation beamwidth. The proposed array solution may be potentially deployed for FSR systems with a full-band-locked beam illumination and a decoupled range-angle resolution, which may present a highly competing scheme against the phased array radar techniques.

Analysis of ENF Signal Extraction From Videos Acquired by Rolling Shutters

Electric network frequency (ENF) analysis is a promising forensic technique for authenticating multimedia recordings and detecting tampering. The validity of the ENF analysis heavily relies on the capability of extracting high-quality ENF signals from multimedia recordings. This paper analyzes and compares two representative methods for extracting ENF signals from visual signals acquired by cameras using the rolling-shutter mechanism. The first method proposed in prior work, <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">direct concatenation</i> , ignores the idle period of each frame. The second method proposed in this paper, <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">periodic zeroing-out</i> , inserts zeros to missing sample points instead of ignoring the idle period. Our theoretical analyses of using multirate signal processing reveal and experiments confirm that while the first method can extract ENF signals without knowing the exact value of camera read-out time, there exists some mild distortion to extracted ENF signals. In contrast, the second method taking the read-out time as the additional input is capable of extracting distortion-free ENF signals, and its frequency component of the highest strength is always located at the nominal frequency. Additionally, we examine aliased DC and negative ENF components caused by the two methods and show that their impact on the accuracy of frequency estimation is minimum. This paper facilitates the fundamental understanding of extracting ENF signals from videos. The research findings imply that the periodic zeroing-out method offers more accurate frequency estimates, but the performance improvement is not significant.

Design of Decimation Filters for Wireless Local Area Network Applications

Multirate signal processing is critical to realizing the digital frequency converter in WLAN technologies. In this paper, we focus on designing and analyzing the different structures of decimators that support WLAN-b applications to reduce the frequency by 12 for an IEEE. The structure modeling of the decimator used Simulink. Implementing a single-stage decimator required a higher-order filter, extra storage space, and a long simulation time. Results showed that the necessary storage elements for 2-stage design are 55% and for 3-stage design is 65% of single stage. For 133 MHz WLAN-b application, a two-stage decimator is proved to be efficient.

Universal photonics tomography.

3D imaging is essential for the study and analysis of a wide variety of structures in numerous applications. Coherent photonic systems such as optical coherence tomography (OCT) and light detection and ranging (LiDAR) are state-of-the-art approaches, and their current implementation can operate in regimes that range from under a few millimeters to over more than a kilometer. We introduce a general method, which we call universal photonics tomography (UPT), for analyzing coherent tomography systems, in which conventional methods such as OCT and LiDAR may be viewed as special cases. We demonstrate a novel approach (to our knowledge) based on the use of phase modulation combined with multirate signal processing to collect positional information of objects beyond the Nyquist limits.

Design of Multichannel Multirate Signal Acquisition System Based on FPGA

With the continuous advancement of science and technology, the fields of national defense, astronomy, industry, agriculture, and so on often require real-time acquisition of voltage, current, temperature, pressure, flow, and other signals, and the real-time requirements for signal acquisition are getting higher and higher. The traditional temperature acquisition method is to use the single chip or DSP as the control core to design the temperature monitoring system. However, this method has certain defects and cannot meet many high-demand signal acquisition occasions. On the one hand, most functions of this system are realized by software, and the operation of software will occupy part of the system sampling time, thus reducing the system sampling rate; on the other hand, the logic of the system to complex peripheral circuits cannot be well controlled. And Field Programmable Gate Array (FPGA) has characteristics such as parallel processing, controllability, and processing speed; therefore, the signal acquisition system with FPGA as the control core can realize high-speed signal collection. This paper studies the signal acquisition based on FPGA. In the signal acquisition system, FPGA is the control core of the whole system, and the relevant function modules of the control system complete the signal acquisition task. The acquisition control module is mainly to control the channel selection switch and the A/D converter to work, that is, to generate the logic control signal of multichannel selection and the control sequence of A/D conversion, so as to complete the acquisition and control of the multichannel analog input signal. The FPGA-based signal acquisition control system has the characteristics of short development cycle, flexible design, low cost, and high reliability. FPGA is an ideal architecture for signal acquisition, and the FPGA-based signal acquisition technology has very broad application prospects.

Methods of multi-rate signal processing in the problem of heart rate variability analysis

Methods of multi-rate signal processing in the problem of heart rate variability analysis

Multirate Adaptive Equalization

Finite Impulse Response (FIR) filter model emulates the Inter Symbol Interference (ISI) in a wireless communication channel. An equalizer, typically an Infinite Impulse Response (IIR) filter, behaves as an inverse filter to the FIR filter to remove the effects of the ISI. IIR filters are generally avoided due to tractability issues, and an FIR filter, with an adaptive signal processing algorithm to minimize the error due to the ISI, is deployed at the receiver. However, the filter is observed to quickly reach a steady state where further iterations do not yield a reduction in the error. This can be attributed to relatively slow variations in the steady state error which prevent further reduction of the errors. This work focuses on converting the low frequency error variations to high frequency variations by the use of multirate signal processing. As such, the steady state error can be damped as well, providing further reduction in the error and an enhanced adaptive filter performance.

Multirate Audiometric Filter Bank for Hearing Aid Devices.

The frequency-dependent nature of hearing loss poses many challenges for hearing aid design. In order to compensate for a hearing aid user's unique hearing loss pattern, an input signal often needs to be separated into frequency bands, or channels, through a process called sub-band decomposition. In this paper, we present a real-time filter bank for hearing aids. Our filter bank features 10 channels uniformly distributed on the logarithmic scale, located at the standard audiometric frequencies used for the characterization and fitting of hearing aids. We obtained filters with very narrow passbands in the lower frequencies by employing multi-rate signal processing. Our filter bank offers a 9.1× reduction in complexity as compared to conventional signal processing. We implemented our filter bank on Open Speech Platform, an open-source hearing aid, and confirmed real-time operation.

A Robust Facial Expression Recognition Algorithm Based on Multi-Rate Feature Fusion Scheme.

In recent years, the importance of catching humans’ emotions grows larger as the artificial intelligence (AI) field is being developed. Facial expression recognition (FER) is a part of understanding the emotion of humans through facial expressions. We proposed a robust multi-depth network that can efficiently classify the facial expression through feeding various and reinforced features. We designed the inputs for the multi-depth network as minimum overlapped frames so as to provide more spatio-temporal information to the designed multi-depth network. To utilize a structure of a multi-depth network, a multirate-based 3D convolutional neural network (CNN) based on a multirate signal processing scheme was suggested. In addition, we made the input images to be normalized adaptively based on the intensity of the given image and reinforced the output features from all depth networks by the self-attention module. Then, we concatenated the reinforced features and classified the expression by a joint fusion classifier. Through the proposed algorithm, for the CK+ database, the result of the proposed scheme showed a comparable accuracy of 96.23%. For the MMI and the GEMEP-FERA databases, it outperformed other state-of-the-art models with accuracies of 96.69% and 99.79%. For the AFEW database, which is known as one in a very wild environment, the proposed algorithm achieved an accuracy of 31.02%.

Implementation of optimal multi-rate cardiac signal processing on a 1967BH028 DSP from “Milandr” design center in order to analyze heart rate variability

The analysis of heart rate variability in recent years has become very widespread as a tool for versatile diagnostics of the functional state of the human body. The analysis of heart rate variability is associated with processing procedures characterized by high requirements for the speed of the computational element base. These procedures, however, must be performed in real time in an embedded computing system. The article deals with the problem of reducing the number of computational operations and an implementation of heart rate variability on modern processor elements offered by the domestic industry. The aim of the work is to model the optimal structure of multi-rate signal processing in the analysis of heart rate variability and to implement this structure on a digital signal processor with an estimate of processing time and `memory costs. By modeling, it is shown that the developed optimal structure of multi-rate processing allows getting a reliable processing result while reducing computational costs by several hundred thousand times compared to the implementation at the original sampling frequency. The optimal structure is constructed as a two-stage filtering-decimation structure, followed by passing the signal at a reduced sampling rate through a set of analysis filters. The end-to-end decimation factor is 500. The simulation results allow us to proceed to the implementation on the signal processor. The program codes of the main processing stages, including filtration and filtration decimation, have been developed. It is shown that the processing time with high-quality optimization of program codes can reach 10 million clock cycles, which corresponds to 23 ms and fully satisfies the real-time processing requirement, leaving a large margin for implementing additional more complex analysis algorithms on the same processor. The practical significance of the results is that in addition to the proposed method of reducing computational and memory costs, a prototype of a possible device based on one of the most popular domestic signal processors is obtained.

Robust online video super-resolution using an efficient alternating projections scheme

Video super-resolution reconstruction (SRR) algorithms attempt to reconstruct high-resolution (HR) video sequences from low-resolution observations. Although recent progress in video SRR has significantly improved the quality of the reconstructed HR sequences, it remains challenging to design SRR algorithms that achieve good quality and robustness at a small computational complexity, being thus suitable for online applications. In this paper, we propose a new adaptive video SRR algorithm that achieves state-of-the-art performance at a very small computational cost. Using a nonlinear cost function constructed considering characteristics of typical innovation outliers in natural image sequences and an edge-preserving regularization strategy, we achieve state-of-the-art reconstructed image quality and robustness. This cost function is optimized using a specific alternating projections strategy over non-convex sets that is able to converge in a very few iterations. An accurate and very efficient approximation for the projection operations is also obtained using tools from multidimensional multirate signal processing. This solves the slow convergence issue of stochastic gradient-based methods while keeping a small computational complexity. Simulation results with both synthetic and real image sequences show that the performance of the proposed algorithm is similar or better than state-of-the-art SRR algorithms, while requiring only a small fraction of their computational cost.

Gaussian Lifting for Fast Bilateral and Nonlocal Means Filtering.

Recently, many fast implementations of the bilateral and the nonlocal filters were proposed based on lattice and vector quantization, e.g. clustering, in higher dimensions. However, these approaches can still be inefficient owing to the complexities in the resampling process or in filtering the high-dimensional resampled signal. In contrast, simply scalar resampling the high-dimensional signal after decorrelation presents the opportunity to filter signals using multi-rate signal processing techniques. Cis work proposes the Gaussian lifting framework for efficient and accurate bilateral and nonlocal means filtering, appealing to the similarities between separable wavelet transforms and Gaussian pyramids. Accurately implementing the filter is important not only for image processing applications, but also for a number of recently proposed bilateralregularized inverse problems, where the accuracy of the solutions depends ultimately on an accurate filter implementation. We show that our Gaussian lifting approach filters images more accurately and efficiently across many filter scales. Adaptive lifting schemes for bilateral and nonlocal means filtering are also explored.

Performance analysis of CIC multirate filter in cognitive radio for efficient pulse shaping in wireless domain

&lt;span&gt;In this paper, the Performance Analysis of Cascade Integrator Comb Filter in context of Filter Bank Multicarrier Transmission has been presented for Cognitive radio. A benefit of the chosen technique is that, a CIC filter can be designed with a slight adjustment in parameters of interest. The entire performance of the filters designed is analyzed and evaluated by analyzing Normalized Amplitude versus Normalized Frequency plots at typical K and N values. The roll off factor plays a significant role in performance analysis of CIC filter. The results shown are a useful advance for rf design engineers working in the domain of multirate signal processing in wireless communication. To ensure the acceptable performance of Enhanced FBMC, computational complexity, and transmission burst length need to be reduced. The effect of Stop band attenuation on the edges of Magnitude and Frequency responses has been studied under constraints such as Lp, K and M during different simulation runs&lt;/span&gt;.

Decimator systolic arrays design space exploration for multirate signal processing applications

This study presents a new systolic array structure for a decimator that merges the antialiasing finite impulse response (FIR) filter with the downsampler. The development of the structure is based on a systematic methodology. Using this methodology, a dependence graph for the decimator was obtained that combined the antialiasing filter and the downsampler. Different data scheduling and projection operations were developed to obtain different proposed designs. Six systolic array design options were obtained and evaluated. The fastest design was selected for hardware implementation and compared with the other two well known decimator designs; namely, conventional design, in which the antialiasing filter is followed by a downsampling and the polyphase design, in which a commutator is followed by the polyphase antialiasing filter. Field-programmable gate array implementations for the proposed and the other two designs confirm that the proposed decimator implementation outperforms in terms of area, speed, and power as the decimation factor increases regardless of the number of FIR filter coefficients.

Digital Frequency-Selective Filters Based on Spectra of Atomic Functions

A method for synthesis of digital low-pass filter (LPF) based on spectra of atomic functions $${{{\text{h}}}_{a}}(x)$$ is first proposed. Compactly supported function $${{{\text{h}}}_{a}}(x)$$ is a solution to the problem of partition of unity, which makes it possible to consider a sum of its $$S$$ shifts $${{h}_{{S,a}}}(x)$$ as a perfect frequency response of LPF. Truncation of the spectrum of function $${{{\text{h}}}_{{S,a}}}(x)$$ is used to obtain digital filters with rapidly decaying coefficients. Formulas for estimation of deviations of frequency responses in the passband and stopband are derived. Application of new filters in multirate signal processing is considered.

Link adaptation strategies for IEEE 802.15.4 WPANs: Protocol design and performance evaluation

This paper proposes two link adaptation strategies for IEEE 802.15.4 wireless personal area networks (WPANs), using a multi-rate signaling set. In the proposed link adaptation strategies, the most adequate modulation and coding scheme (MCS) satisfying the target bit error rate (BER) of the end device is selected based on the signal-to-interference-plus-noise ratio (SINR) of either a beacon or an acknowledgement (ACK) frame. The beacon-based link adaptation scheme has low complexity and overhead, given that it performs link adaptation only once per superframe. In contrast, the ACK-based strategy performs link adaptation at every ACK frame, and therefore provides a faster and more effective link adaptation, but at the expense of a larger overhead. The specific protocol design for the proposed link adaptation strategies is developed by constructing the signal flow based on the service primitives between the protocol stack layers. The network simulator OPNET is used to implement an accurate IEEE 802.15.4 WPAN protocol stack and the simulation environment required for performance evaluation. The simulation results show that the received throughput of IEEE 802.15.4 WPANs can be improved by exploiting the proposed link adaptation strategies instead of auto-rate fallback, the conventional WPAN strategy.

Design of sparse cosine modulated filter banks based on Hopfield neural network

Cosine modulated filter bank (CMFB) is the most basic and important module in multi-rate signal processing system. Sparse FIR linear-phase CMFB can not only reduce the complexity of hardware design, but also ensure good performance. In this paper, a new design method of CMFB is proposed. First, the set of sparse coefficients of the prototype filter is found by using the orthogonal matching pursuit (OMP) algorithm. Then the Hopfield neural network (HNN) is employed to optimize the non-zero coefficients. The analytical and simulation results demonstrate that the new approach not only ensures that the sparse CMFBs satisfy near perfect reconstruction (NPR), but also improves the efficiency of hardware operation.

FFT-Based Multirate Signal Processing for 18-Band Quasi-ANSI S1.11 1/3-Octave Filter Bank

For characteristic analysis of the 24 KHz audio, the 10-ms, 18-band quasi-ANSI filter bank has been proposed and designed for advanced hearing aids. To greatly reduce the computation complexity, several time-domain multirate signal processing techniques, such as the up- and down-sampling rate conversions and the interpolated finite impulse response (IFIR) filters, were investigated. However, it is well known that using fast Fourier transform (FFT) to perform the linear convolution can dramatically reduce the computational complexity. Taking both advantages, in this brief we investigated the FFT-based multirate signal-processing technique and explored their efficient architecture. To demonstrate the success of the proposed architecture, we implemented a real-valued FFT-based 10.7-ms, 18-band quasi-ANSI 1/3-octave filter bank using TSMC 90-nm CMOS high-VT technology. We found that, for each input sample, the proposed FFT-based quasi-ANSI 1/3-octave filter bank used approximately 77% fewer multiplications than the previous time-domain design. The proposed FFT-based quasi-ANSI filter bank was operated at 13 MHz to process the 24-KHz audio in real time, and it consumed only $14~ \mu \text{W}$ (@0.9V) of dynamic power.