Noise Spectral Density Research Articles

Impact of interface traps and noise analysis on dual material graded channel CGAA FET: A device reliability

This paper explores the effects of changing the device's parameters, such as the gate length (Lg), nanowire radius (r), oxide thickness (tox), and frequency (f), on the noise characteristics of dual material graded channel (DMGC) cylindrical gate all around (CGAA) FET while considering the presence and absence of interface trap charges. The performance analysis of the DMGC CGAA FET with that of the SMGC (single material graded channel) CGAA FET, is carried out while taking into account the impact of trap charges. It's noteworthy that the DMGC FET demonstrates significant advantages, with improvement in Ion/Ioff, reduction in DIBL, and SS when compared to the SMGC CGAA FET, both in the cases of without and with traps. The investigation also extends to analyse the analog/RF performance of the DMGC FET, for different type of trap charges. Furthermore, the power spectral densities of current noise (Sid) and voltage noise (Svg) are examined by varying the device dimensions and temperature. This also addresses the various sources of noise, including flicker noise (1/f), generation-recombination (G-R) noise, and diffusion noise. These types of noise have different prevalence depending on the operating frequency. At low frequencies, flicker and G-R noises dominate, while at higher frequencies, diffusion noise becomes the prominent factor impacting the device's performance. The research findings indicate that the optimal performance of the DMGC CGAA FET is achieved with smaller nanowire radius and thinner oxide thickness, regardless of trap charges. These findings offer valuable insights for improving the design and performance of DMGC CGAA FET in low power applications.

Characteristics Analysis of IGZO TFT and Logic Unit in the Temperature Range of 8–475 K

The effect of high- and low-temperature conditions on the performance of IGZO TFT and logic circuits were investigated in this work. In the temperature range of 250−350 K, the performance of the IGZO TFT did not show significant changes and exhibited a certain degree of high- and low-temperature resistance. When the temperature was below 250 K, as the temperature decreased, the threshold voltage (VTH) of the IGZO TFT significantly increased, the field effect mobility (μFE) and the on state current (ION) significantly decreased. This is attributed to the lower excitation degree of charge carriers at extremely low temperatures, resulting in fewer charge carriers transitioning to the conduction or valence bands, and the formation of defects also limits carrier migration. When the temperature exceeded 350 K, as the temperature increased, more electrons could escape from the bandgap trap state and become free charge carriers, and the IGZO layer was thermally excited to produce more oxygen vacancies, resulting in higher μFE and lower VTH. In addition, the drain current noise spectral density of IGZO TFT conformed to the 1/ƒ noise characteristic, and the degradation mechanism of IGZO TFT over a wide temperature range was confirmed based on the changes in noise spectral density at different temperatures. In addition, an inverter logic unit circuit was designed based on IGZO TFT, and the performance changes over a wide temperature range were analyzed. This lays the foundation for IGZO TFT to be applied in integrated circuits with harsh environments.

Interface engineering by constructing vertical junction for reduced noise and improved sensitivity in 2D photodetector

Two-dimensional (2D) materials have been widely demonstrated as promising candidates for next generation photodetectors, while the noticeable channel current is still a limiting factor for photodetection sensitivity. In this work, the interface engineering has been developed by constructing a vertical pn and Schottky junction in the 2D WS2 channel, resulting in a reduced dark current and noise spectral density, significantly improving the sensitivity. Specifically, the WS2 bottom surface is coupled with p-type tellurium (Te) nanoribbon and gold (Au) stripes, thus a vertical pn and Schottky junction can be constructed at WS2/Te and WS2/Au interface, respectively. In both device architectures, the dark current and electric noise are much suppressed due to the formation of depletion region in WS2 channel. Meanwhile, the out-of-plane built-in electric field at junction can facilitate the separation of photo-excited electron–hole pairs, which subsequently yields a faster temporal response. For the WS2/Au device, the incident light can be reflected by the bottom Au and propagate through the WS2 layer again, further boosting the photo-absorption, thus the photodetection sensitivity. The engineered WS2 photodetectors exhibit the noise spectral density as low as 5.36 × 10−14 A Hz−1/2 and high specific detectivity (D*) up to 1.12 × 1011 Jones, which has one–two orders of magnitude improvement compared to the pristine device. This work provides an effective and universal interface engineering strategy to achieve low noise and high sensitivity in 2D photodetectors.

Metamaterial Engineering for Superior HgTe cQD Photodetector Performance

AbstractHighly responsive, low noise, and inexpensive photodetectors that operate in the mid‐infrared (MIR) wavelength regime are in high demand for applications ranging from fundamental science to large scale industries. However, simultaneously achieving all this in one device architecture is very challenging. In this work, mercury telluride (HgTe) colloidal quantum dot (cQD) based photodetectors are systematically improved by the introduction of new metamaterial designs. The new designs are found by utilizing simulations. Thereby the structures are optimized to increase the responsivity and simultaneously decrease the noise spectral current density. This is achieved by focusing on improving the photogenerated charge carrier collection efficiency while reducing the active material volume without altering the near unity absorption. A standard metamaterial perfect absorber architecture based on disc resonators is used as a starting point for the optimization process. By optimizing the carrier extraction through contact engineering, resulting in a narrow slot metamaterial, an overall ≈13‐fold responsivity and ≈345‐fold detectivity increase is achieved. The final metamaterial design reaches a responsivity of 16.2 A W−1 and detectivity of 6×108 Jones at a wavelength of 2710 nm. The analysis therefore provides a route to improve the responsivity and noise characteristics of mid‐infrared photodetectors based on cost‐efficient colloidal quantum dots.

Gate‐Modulated Polarity Transition and Polarization‐Sensitive Photodetection Enabled by Sandwiching Anisotropic GeSe in vdW Heterojunction

AbstractThe ability to modulate the polarity of carrier transport and photo‐response in assembled heterostructure devices remains a huge challenge for the development of multifunctional optoelectronics. Herein, a polarity‐switchable and polarization‐sensitive photodetector is developed based on a van der Waals (vdW) MoTe2/GeSe/MoS2 sandwich heterostructure. By varying the gate voltage, an anti‐bipolar transfer characteristic is obtained with significant peak‐valley current ratio (PVCR) exceeding 104, showing the extraordinary potential to realize electronic functions in logic circuits. Under 635 nm laser irradiation, the device exhibits a gate‐controlled polarity transition of photocurrent, namely, the sign reversal of photocurrent occurs by changing the gate voltage. Furthermore, the device achieves broadband photovoltaic effect and high photodetection performance with responsivity (R) of 723 mA·W−1, noise spectral density (Sn) of 1 fA Hz−1/2, and specific detectivity (D*) of 2.3×1012 Jones in the absence of external bias, which greatly outperforms the MoTe2/MoS2 device. Leveraging the in‐plane anisotropic structure of GeSe, the device is also endowed with an additional capability of polarization‐sensing for linearly polarized laser, possessing a high polarization ratio (PR) value of 11.2. Thus, this work proposes an effective and facile strategy to realize the gate‐modulated polarity transition and polarization‐sensitive photodetection, offering a broad perspective for multifunctional integrated applications.

Time delay interferometry combination with zero-response

In space-based gravitational wave detection, gravitational wave signals are detected by measuring the Doppler frequency shift of laser exchange between three satellites. Although the laser is pre-stabilized, the laser phase noise is significantly higher than the gravitational wave signal, approximately 7–8 orders of magnitude higher. To reduce the otherwise-overwhelming laser phase noise, the time delay interferometry (TDI) technique is employed by properly time-shifting and linearly combining raw data. Different TDI combinations have various abilities in suppressing laser phase noise, and the response function of gravitational wave signal as well as the power spectral density of noise floor (including test mass acceleration noise and optical path noise) are also different. In this work, we present a unique class of TDI combinations, denoted as [ZX]116,[ZX]216,[ZX]316,[ZX]416, and [ZX]516, which exhibit a distinct geometric space–time diagrams composed of Michelson interferometers as the fundamental constituents. Specifically, when the detector arm lengths are assumed to be equal, these combinations yield response functions and noise power spectral densities of zero. To evaluate the detection performance of these combinations, we investigate their response functions and noise power spectral densities when the detector arm lengths are unequal, and furthermore, derive the analytical expressions for their sensitivity functions. Through our analysis, we have discovered some intriguing results. When the detector arm lengths are unequal, both the response functions and noise spectral densities of these combinations are correlated with the difference in arm lengths L3−L2. However, the sensitivity functions of these TDI combinations remain unaffected by the specific values of the arm length difference, and are consistent with the sensitivity functions of widely-used TDI combinations. For instance, the sensitivity function of [ZX]116 combination is the same as that of the Michelson combination, while the sensitivity functions of [ZX]216 and [ZX]416 combinations match with that of the Sagnac combination. Furthermore, the [ZX]316 and [ZX]516 combinations share the same sensitivity function as the fully symmetric Sagnac combination, enabling the identification of instrument noise and stochastic gravitational wave backgrounds. The discovery of this class of combinations provides support for the diversity of data processing.

Analyzing the phase and amplitude noise of RZ, NRZ and RF DACs

This article considers a multi-mode RF DACs phase and amplitude noise for non-return-to-zero (NRZ), return-to-zero (RZ), and mix/RF reconstruction waveforms under the assumption that two switch pairs (or two channels) in each current switch cells are alternately active in every clock cycle, Analysis shows that each channel of multimode DAC is completely characterized with RZ mode, and all other modes can be derived from this mode, in particular NRZ mode is not a complete analog of conventional DAC performance when reconstruction pulse duration and period are equal. When determining phase and amplitude noise of multi-mode RF DAC, one should keep in mind the phase shift between two switch pairs output pulses spectral components in NRZ and RF modes. This phase shift depends upon DAC mode (NRZ or RF) and output oscillation frequency. As consequence an analysis indicates that the 3 dB improvement of 1/F phase noise can be achieved in 1st Nyquist zone when using NRZ mode, the same in 2nd Nyquist zone with RF mode if amplitude noise of DAC switch pairs is negligible. The independence of 1/F noise from clock frequency has simple physical interpretation. As correlation period of 1/F noise is significantly greater than output oscillation period, the delay of this oscillation caused by phase fluctuations in DAC is the same for any mode, and analysis of noise spectral density for conventional DAC can be extended to multi-mode RF DACs.

Transponder-type laser interferometer prototype for spaceborne gravitational wave detectors.

Transponder-type laser interferometry is essential in spaceborne gravitational wave detection missions. This paper presents a transponder-type laser interferometer prototype for potential noise calibration of spaceborne gravitational wave detectors. Using a digital optical phase-locked loop, we successfully locked the phase of the slave laser to the master laser (∼200p W). Once the link between the master laser and the slave laser is established, the two satellites (essentially two lasers) form a transponder-type laser interferometer. We carefully analyze the measurement stability and noise characteristics of the interferometer, and the results show that the Allan deviation of the zero drift can reach 243.2pm at t=0.429s, while the noise spectral density has a typical 1/f line shape with a floor of 21p m/H z 1/2 at 1Hz. The coherence analysis shows that the temperature drift is an important factor limiting the performance of the interferometer below 2mHz, while the frequency noise of the master laser is not dominant in the experiment. Transponder-type laser interferometers have a wide range of applications in intersatellite communication and measurement. Our design can serve as a valuable reference for gravitational wave detection missions such as LISA.

Exceptional points enhance sensing in silicon micromechanical resonators

Exceptional points (EPs) have recently emerged as a new method for engineering the response of open physical systems, that is, systems that interact with the environment. The systems at the EPs exhibit a strong response to a small perturbation. Here, we show a method by which the sensitivity of silicon resonant sensors can be enhanced when operated at EPs. In our experiments, we use a pair of mechanically coupled silicon micromechanical resonators constituting a parity–time (PT)-symmetric dimer. Small perturbations introduced on the mechanically coupled spring cause the frequency to split from the EPs into the PT-symmetric regime without broadening the two spectrum linewidths, and this frequency splitting scales with the square root of the perturbation strength. The overall signal-to-noise ratio is still greatly enhanced, although the measured noise spectral density of the EP sensing scheme has a slight increase comparable to the traditional counterpart. Our results pave the way for resonant sensors with ultrahigh sensitivity.

Self-powered wireless structural sensors for long-term monitoring of bridges

The present work introduces the design of a self-powered wireless static and dynamic structural monitoring system developed by Wisepower Srl and deployed on some bridges along the S.S. 675 "Umbro-Laziale" (former "Civitavecchia - Orte" junction) since 2018. The most significant aspect of the developed monitoring system lies in the fact that it consists of a wireless network of MEMS accelerometers (noise spectral density in all axes: 22.5 μg/√Hz, and sensitivity: 3.9 μg/LSB) powered by energy harvesting systems, which can convert environmental energy (i.e. vibrations and light) into electrical energy collected in batteries, so extending the life and minimizing the maintenance cycles of the devices. Vibration-based energy harvesting is achieved through a non-linear resonator, which utilizes a wider spectrum of frequencies compared to traditional linear vibration energy harvesters owing to its non-linear dynamics. The efficiency of the system is further enhanced by the combination of solar panels, overcoming the limitations associated with traditional wiring or batteries, widening the lifespan of the electronics, and reducing the maintenance costs. The paper reports the analysis method and the data elaboration of some ambient acceleration records, validating the system for natural frequencies identification.

Effect of heavy ion radiation on low frequency noise characteristics of AlGaN/GaN high electron mobility transistors

AlGaN/GaN high election mobility transistor (HEMT) has important application prospects in satellite communication, radar, nuclear reactors and other extreme environments, owing to its excellent electrical performance and strong radiation resistance. Heavy ion radiation mainly causes single-event effect and displacement damage effect in AlGaN/GaN HEMT device. In this work, the displacement damage defects introduced by heavy ion radiation are analyzed in detail. With the increase of heavy ion radiation influence, more defects are introduced by displacement damage. These defects reduce the two-dimensional electron gas (2DEG) concentration through carrier capture and removal effect, and reduce the carrier mobility through scattering mechanism, resulting in gradual degradation of the electrical characteristics of the device. In this work, AlGaN/GaN high electron mobility transistors are irradiated by <sup>181</sup>Ta<sup>32+</sup> ions with fluences of 1×10<sup>8</sup> ions/cm<sup>2</sup>, 1×10<sup>9</sup> ions/cm<sup>2</sup> and 1×10<sup>10</sup> ions/cm<sup>2</sup>. The electrical characteristics, EMMI and low-frequency noise characteristics of the device before and after heavy ion radiation are measured. The results show that heavy ion radiation can lead to the degradation of electrical parameters. When the heavy ion radiation dose reaches 1×10<sup>10</sup> ions/cm<sup>2</sup>, the electrical characteristics of the device deteriorate seriously, the threshold voltage shifts forward by 25%, and the drain saturation current deteriorates obviously. The defect locations introduced by irradiation are analyzed by EMMI test, and it is found that the number of “hot spots” increases significantly after the having been irradiated by heavy ions with a fluence of 1×10<sup>10</sup> ions/cm<sup>2</sup>, indicating that the radiation leads to the increase of defect density and serious damage to the device. Through the noise test, it is found that with the increase of the radiation fluence, the current noise power spectral density gradually increases. When the fluence reaches 1×10<sup>10</sup> ions/cm<sup>2</sup>, the defect density increases to 3.19×10<sup>18</sup> cm<sup>–3</sup>·eV<sup>–1</sup>, and the Hooge parameter increases after having been irradiated by heavy ions. We believe that the radiation leads to the defect density and parasitic series resistance of AlGaN/GaN HEMT device to increases, resulting in larger Hooge parameters. Through analyzing the variation of the normalized power spectral density of the drain current noise with bias voltage, it is found that the defects caused by heavy ion radiation will cause the parasitic series resistance to increase.

Spectroscopic effects of distributed-line phenomena in integrated feedback resistors for charge-sensitive pre-amplifiers

The distributed capacitive coupling of high-valued resistors with ground planes modifies their effective impedance and power spectral density of noise, especially at high frequency. This effect is clearly visible in poly-silicon resistors inside integrated circuits. When such resistors are used in the feedback network of Charge Sensitive Pre-amplifiers an excess noise arises with a non-white noise spectral density. The effect is a worsening of the spectroscopic resolution. The relevance of such effect with respect to the conventional noise sources is discussed in different practical cases.

New method of fabrication of suspended metallic single electron transistor (SET)

Charge sensing applications utilizing single electron transistors (SETs) as electrometers face challenges due to nearby background charge movements. In this study, we present an innovative fabrication method for creating suspended Al-AlOx-Al SETs positioned above a cavity. These suspended SETs exhibit significantly reduced flicker noise with 1fα noise spectral density when compared to their substrate-based counterparts. This noise reduction can be attributed to the elimination of the substrate beneath the SET island. Consequently, our fabricated suspended SETs are highly suitable for demanding charge sensing applications and provide a promising platform for in-depth investigations into the sources of charge noise in such devices.

Low-frequency noise and impedance measurements in Auger suppressed LWIR N+p(π)P+n+ HgCdTe detector

The paper reports on the electrical performances of LWIR N+p(π)P+n+ HgCdTe photodiodes optimized for λcut-off(50%) ≈ 10.6 μm at T = 230 K. Presented work mainly focuses on the region where negative resistance is observed due to Auger suppression effect. Previous reports showed that for such detectors it is usually observed at temperature conditions above ∼120 K for devices designed to operate in HOT conditions. The studies were carried out over a wide temperature range using few measurement methods: dark current, low-frequency noise, and impedance components. The results were also investigated as a function of the applied bias. A detailed analysis of the results were presented to show the influence of operating conditions when the detector works in negative resistance region. In these regions, we observed a dynamic change of the complex impedance from both parallel capacitance and resistance. The low-frequency noise measurements showed that in the negative resistance region, the noise spectral density is still rising with the reverse bias voltage despite reduction of the dark current by Auger suppression and the noise-current related intensity coefficient α is also still tends to rise. For temperatures above 160 K the noise-bias voltage interdependence near Si ∼ V2 was noticed. Moreover, the studies showed that the noise-temperature behaviour depends on the applied bias. The results obtained in experiments were discussed based on the current knowledge of these devices to verify and explain the behaviours of the dark current, noise and impedance-related parameters.

Vertical Graphene/Pentacene Single Crystal/Graphene Transistors for Self‐Powered Weak Light Detection and High‐Speed Imaging

AbstractOrganic semiconductors are promising candidates for next‐generation photodetectors owing to their unique properties including large absorption coefficient and intrinsic mechanical flexibility. However, limited by high defect density, large exciton binding energy, and short exciton diffusion length, many challenges remain for fabricating high‐performance organic photodetectors, especially for scenarios where weak light sensing and high‐speed imaging are required. Herein, by coupling single‐crystal pentacene with monolayer graphene, a vertical organic phototransistor with excellent zero‐bias photon detecting capacity is demonstrated. Under self‐powered weak light condition, the responsivity and specific detectivity are 2.86 AW−1 and 3.3 × 1011 Jones (based on experiment spectral noise density), respectively. It is demonstrated that weak light down to 25 nWcm−2 can be detected. When operating with an external bias, the responsivity of the device increases dramatically (up to 105 AW−1), and a better 3 dB bandwidth (≈74 kHz) is obtained. Finally, imaging functionality is demonstrated by employing this vertical transistor as a single‐sensing pixel. This work suggests organic single crystals exhibit enormous potential in advanced optoelectronic systems, and may provide a viable route for weak light detecting and high‐speed imaging.

Analysis of noise behavior and reliability of pocket doped negative capacitance FET under the impact of trap charges and temperature

This paper represents an in-depth investigation of low to high frequency noise performance of highly doped double pocket double gate NCFET (HDDP-DG-NCFET) and single gate NCFET (SG-NCFET) in the existence of Uniform and Gaussian trap charges at various frequencies and temperatures (200 K–400 K). The drain current noise spectral density (Sid) and Gate voltage noise spectral density (Svg) at different frequencies in existence of both type of trap charges are explored. In addition, the Sid as a function of frequency for various ferroelectric thicknesses (Tfe) is examined in the existence of Uniform trap charges. The findings show that, at lower frequencies flicker noise dominates, while at higher frequencies, the dominant noise element is diffusion noise. For the whole range of considered frequencies, the Generation-Recombination (G-R) noise is more substantial in existence of trap charges. It has been found that, the net Sid value of the HDDP-DG-NCFET for Uniform and Gaussian types of trap distribution is 6.98 × 10−14A2/Hz and 6.17 × 10−14A2/Hz, respectively. Moreover, the examination of Sid for various Tfe in presence of traps illustrates the significance of the ferroelectric layer in obtaining lower value of net Sid, particularly in the higher frequency region and the minimal net Sid value for Tfe = 7 nm as compared to other Tfe values. Finally, to examine the device's reliability, the temperature investigation for the noise in the non-existence and existence of trap charges demonstrates that, the impact of an increase in noise caused by the existence of trap charges is much pronounced at lower temperatures and smaller gate voltages, where G-R noise is observed to be the dominant noise component.

Ultra-Low Noise Level Infrared Quantum Dot Photodiodes with Self-Screenable Polymeric Optical Window.

Quantum dot photodiodes (QPDs) have garnered significant attention because of their unparalleled near-infrared (NIR) detection capabilities, primarily attributable to their size-dependent bandgap tunability. Nevertheless, the broadband absorption spectrum of QPD engenders substantial noise floor within superfluous visible light regions, notably hindering their use in several emerging applications necessitating the detection of faint micro-light signals. To overcome these hurdles, a self-screenable NIR QPD featuring an internal optical filter with a thick polymeric interlayer to reduce electronic noise is demonstrated. This effectively screens out undesirable visible light regions while reducing the ionized defect owing to decreased density of state, yielding an extremely low dark current (≈1010 A cm-2 at V = -1V). Consequently, the electronic noise spectral density is attained at levels below ≈10-27 -10-28 A2 Hz-1 , and responsivity (R) dropped to 92% within the visible light spectrum.

Impact of Gaussian Traps on the Characteristics of L-shaped Tunnel Field-effect Transistor

Background: Internet of Things (IoT) applications require high-performance TFET devices that can be efficiently integrated with the cyber world and physical world. Objectives: The impact of introducing Gaussian traps in hetero-junction tunneling-field-effecttransistors (TFET) with an L-shaped gate is presented. Methods: The 2-D TCAD study of different characteristics, like input, output characteristics, and noise spectral density with trap and without trap, has been performed. Results: The simulation results showed that in L-shaped TFET (L_TFET), the high on-current of 1.93×10-5 A/μm, low off-current/leakage current of 1.09×10-13 A/μm, and steep sub-threshold slope (SS) of 24 mV/dec without traps and on-current of 8.46×10-6 A/μm, off-current of 2.86×10- 11 A/μm, and degraded SS with traps are observed. They also indicated that the presence of traps reduces gate-drain capacitance (Cgd), while gate-source capacitance (Cgs) remains unaffected. In addition, in L_TFET, the drain current noise spectral density (SID) of 7.63 E-21 (A2/Hz) at LF and 2.69 E-26 (A2/Hz) at HF while the noise voltage spectral density (SVG) of 7.33 E-4 (V2/Hz) at LF and 2.59 E-15 (V2/Hz) at HF without traps have been investigated in this study. The inverse dependence of drain current noise spectral density on frequency has been observed to lower the effect of noise at HF. Conclusion: It can be concluded that the proposed L_TFET device is free from ambipolarity conduction and can be well-suited for low-power applications.

Two-dimensional synchronous motion modulation MEMS structure for suppressing 1/f noise in magnetoresistive sensors

Magnetoresistive (MR) sensors have great application prospects in the field of weak magnetic field detection due to their high sensitivity, small size, and low power consumption. However, 1/f noise greatly limits the low-frequency detectivity of MR sensors. In order to suppress 1/f noise, this paper proposes a two-dimensional synchronous motion modulation (TDSMM) structure based on microelectromechanical systems (MEMS). This structure can effectively reduce 1/f noise by modulating the frequency of the measured magnetic field in the high-frequency band. Theoretical analysis and finite element simulation were conducted on three different modulation models: TSDMM, magnetic flux concentrators motion modulation, and MR components longitudinal motion modulation. The results showed that the modulation efficiency of the TDSMM reached as high as 127%, which is currently the highest value in MR-MEMS sensors. The TDSMM MEMS structure has been successfully manufactured, and the resonant frequency of the transverse resonator is twice that of the longitudinal resonator, enabling extremely high modulation efficiency. The noise spectral density of giant-magnetoresistive components on a silicon-on-insulator substrate was tested, and the noise level in the high-frequency band was three orders of magnitude lower than that in the low-frequency band. These results position MR-MEMS sensors with TDSMM structures as highly competitive candidates in the field of ultra-weak magnetic field detection.

Drone audition: improved Gaussian mixture model Wiener filtering approach for audio signal enhancement

Audio signal enhancement is an integral function of drone audition systems which emerges intelligent audio applications for military and civil environments. Signal enhancement using a drone is challenging due to its emission of significant noise from drone motors and propellers resulting in an extremely low signal-to-drone noise ratio level. A method for signal enhancement from an on-board microphone array on a drone for static flight conditions in adverse noisy environments is proposed. In this paper, we present an improved Gaussian mixture model Wiener filter approach for a noisy drone platform in the presence of interfering noise sources such as wind, and traffic. The key to this method is to exploit the spatial properties of the noise sources by using a set of Gaussian components to derive an accurate estimate of the power spectral densities of both residual drone noise and interfering noise sources to obtain the enhanced target sound source signal. We demonstrate the method's general applicability for both speech, and bird calls using an extensive experimental study, including an experimental recording from a drone acoustics dataset using measured impulse responses, and an outdoor measurement from a hovering drone for a bioacoustic application, respectively.