S. M. Kayser Azam

This paper introduces three novel nonlinear feature extraction techniques for ultra-high-frequency (UHF) partial discharge (PD) signals: time-domain Recurrence Plot Feature Extraction (RPFE), frequency-domain Recurrence Plot Feature Extraction (FRPFE), and State Space reconstruction (SS). To evaluate their robustness, clean PD signals are contaminated with white noise, narrowband radio frequency interference (RFI), and impulsive noise at varying magnitudes (10%-90%). This paper uses a novel process where the noise type and its level are classified to enable a more effective denoising. For each noise condition, wavelet-based denoisers are first applied with and without fine parameter tuning. Subsequently, the three proposed methods are employed on both denoised and raw noisy signals. Performance is quantified using Fisher’s Discriminant Ratio (FDR) as a measure of class separability among different PD patterns. Results demonstrate that RPFE consistently achieves the highest average FDR (10.1 with fine tuning), followed by FRPFE (9.7) and SS (9.2), significantly outperforming advanced wavelet methods across all noise types and magnitudes, even when the latter are optimally tuned. The superior noise resilience and discriminative power of the proposed recurrence- and state-space-based approaches make them highly suitable for reliable PD diagnosis in harsh measurement environments.

The concept of Chipless Radio Frequency IDentification (CRFID) multi-parameter tag-sensors is fairly new and nowadays has gained importance among the researchers for their ability to identify objects and sense multiple parameters at once without requiring integrated circuits (ICs), on-board power sources and active components. This paper presents a comprehensive review of such tag-sensors, emphasizing its current stage of development, performance comparisons, and existing research gaps. This study provides a detailed classification of resonator structures, identification (ID) & sensing mechanisms, sensing functions & factors, and smart/sensitive materials (SMs), establishing a new unified taxonomy. Further, it provides details on the sensing mechanism and physics behind it, maps resonator types to their design principles with their sensing functions, alongside their corresponding SMs and reveals their multi-parameter decoupling and selectivity techniques. It also comments on the substrates used & the adopted printing process. A critical quantitative evaluation follows based on key ID (coding capacity, spectral and spatial density) and sensing metrics (sensitivity, dynamic range, and read range). Finally, the paper discusses current challenges such as limited coding capacity, short read range, absence of dedicated readers and standards, and issues with selectivity and suggests future directions including AI integration and scalable deployment for IoT, smart logistics, environmental monitoring, and industrial automation.

This paper comprehensively reviews advanced signal processing methods for partial discharge (PD) analysis, covering traditional time-frequency techniques, wavelet transform, Hilbert–Huang transform, and artificial intelligence-based methods. This paper critically examines the principles, advantages, limitations, and applicable scenarios of each method. A key contribution of this review is the systematic comparison of these methods, highlighting their evolution and complementary roles in processing non-stationary and noisy PD signals. However, a significant gap in current research remains the lack of standardized, explainable, and embeddable AI solutions for real-time, fine-grained PD classification. Future trends point to hybrid approaches and edge AI systems that combine physical insights with lightweight deep learning models to improve diagnostic accuracy and deployability.

This paper comprehensively reviews advanced signal processing methods for partial discharge (PD) analysis, covering traditional time-frequency techniques, wavelet transform, Hilbert-Huang transform, and artificial intelligence-based methods. This paper critically examines the principles, advantages, limitations, and applicable scenarios of each method. A key contribution of this review is the systematic comparison of these methods, highlighting their evolution and complementary roles in processing non-stationary and noisy PD signals. However, a significant gap in current research remains the lack of standardized, explainable, and embeddable AI solutions for real-time, fine-grained PD classification. Future trends point to hybrid approaches and edge AI systems that combine physical insights with lightweight deep learning models to improve diagnostic accuracy and deployability.

This article investigates advanced signal processing methodologies, with a focus on wavelet-based techniques, for the analysis of time-domain partial discharge (PD) signals captured using ultra-high frequency (UHF) sensors. The raw signals are systematically processed through a sequence of operations including bandpass filtering, wavelet-based denoising, DC offset removal, and pulse extraction. Each processing stage is critically examined in both time and frequency domains to ensure signal integrity and noise suppression. Emphasis is placed on the optimization of wavelet parameters alongside extraction of key temporal and amplitude-based features such as time difference, charge difference, pulse height, rise time, and pulse width. To address the challenge of identifying the most discriminative features, this work integrates advanced feature ranking algorithms, namely, auto-encoder-based ranking and the ReliefF method. Their effectiveness is evaluated through criteria including training convergence, reconstruction error, and relevant quality metrics. The methodological novelty lies in the systematic fusion of optimized wavelet-based signal conditioning with robust feature selection frameworks, enabling a comprehensive assessment of feature importance and clarity. A total of 7 wavelet transformations with various decomposition levels, 17 wavelet families with various sub-categories and 2 non-wavelets are involved in this study. Comparative analysis between wavelet-based and conventional non-wavelet methods demonstrates the superior performance of the former in terms of feature extraction fidelity and signal enhancement. The findings establish that the proposed clarity-importance ranking framework significantly advances the accuracy and efficiency of UHF PD signal processing. This contributes to enhanced interpretability and reliability in PD diagnostics, thereby supporting more effective monitoring and maintenance of high-voltage insulation systems in real-world applications.

In conventional chipless RFID systems, antennas and separate resonator tags work interdependently—resonator tags rely on antennas for data transmission, while antennas alone cannot perform RFID operations without them. This research introduces, for the first time, a standalone identification antenna (SIDA), eliminating the need for supplementary resonator tags to independently perform chipless RFID functions in the far-field. SIDA is designed as a planar monopole antenna in the ultra-high frequency (UHF) range, employing a new mechanism of notch element concentration on its patch. Intermingled coupling resonators (ICRs) are utilized as notch elements to generate multi-resonating coding bits in far-field radiation. SIDA prototypes are fabricated on Rogers RO4003C substrate with distinctive 6-bit multi-resonating codes (111111 and 111101) within the 0.78–1.38 GHz range. By eliminating the need for separate resonator tags, SIDA significantly improves conventional retransmission (ReTx)- and received signal strength (RSS)-based chipless RFID operations. Both in ReTx and RSS systems, SIDA is applied as resonator tag-free far-field chipless RFID sensor for the simultaneous detection of partial discharge (PD) defect in high-voltage (HV) systems and RF identification of PD-affected HV equipment. Experimental performances prove SIDA’s suitability as far-field chipless RFID sensor and its prospect in multi-purpose smart sensing uses by eliminating the need for additional resonator tags.

Insulations at high voltage (HV), whether in HVAC or HVDC systems, often encounter an unwanted partial discharge (PD) phenomenon. PD poses a potential danger to HV insulation and can eventually lead to equipment failures. Out of all electromagnetic (EM) sensing techniques for PD diagnosis, the ultra-high frequency (UHF) method has gained popularity as it allows non-invasive detection at varying distances from PD defects. The EM behavior of PD is significantly affected by the distance between a PD defect and a UHF sensor. This effect varies for different PD defects under different HV conditions. So, considering variable distances, there is a need to analyze EM characteristics of different PD phenomena at different HV conditions. To the best of our knowledge, for the first time, electromagnetic characteristics of different PD defects are experimentally investigated in this work by capturing PD signals from variable distances in both HVAC and HVDC conditions and individually validating their characteristics with EM radiation theories. For wirelessly capturing PD signals, a new UHF sensor is designed with a modified elliptical-shaped antenna on an FR-4 material. The fabricated sensor provides an average realized gain of about 3.66 dBi while covering more than 97% of the total UHF range for PD detection. Applying HVAC, HVDC+, and HVDC–, PDs from three defects (surface, void, free wire) are captured from 1–4 m distances. Interpreted results show that EM radiation in the UHF range from a PD defect is heavily impacted by its detection distance and defect formation despite applied voltage and other conditions being unchanged for the defect. This investigation is particularly beneficial to the variable distance-based PD diagnosis, such as PD localization and handheld PD detection at HVAC/HVDC open substations.

In this article, we present a low-cost 2×2 metasurface-based rectenna array operating at the 2.45 GHz industrial, scientific, and medical (ISM) band for powering wearable medical sensors (WMS). The proposed antenna is a novel textile-based metasurface interlayer patch (meta-patch) antenna fabricated on a felt fabric substrate, replacing the conventional ground plane with a grounded metasurface consisting of a 3×3 unit cell array. A single excitation port is used for both the patch antenna and the metasurface to improve gain, efficiency, and bandwidth. The meta-patch antenna achieved a high gain of 7 dB, an efficiency of 77%, and an enhanced bandwidth of 120 MHz. The rectifier adopts a seven-stage Cockcroft-Walton voltage multiplier (CWVM) topology fabricated on a rigid substrate. To evaluate the performance of the meta-patch rectenna array and the effect of adding more elements, we tested the proposed meta-patch rectenna in single-element, 2×1 array, and 2×2 array configurations under continuous RF power. The single-element, 2×1 rectenna array, and 2×2 rectenna array achieved power conversion efficiencies (PCE) of 52%, 53%, and 56%, with DC output powers of 414 μW, 429 μW, and 450 μW, respectively, at an input power of -1 dBm. The results demonstrate the potential of the proposed rectenna arrays for efficiently energizing low-power WMS.