In-Field Sensor Defects

• In-field sensor defects are deviations arising during sensor deployment due to environmental, fabrication, or operational stresses that alter calibration, noise, and signal fidelity.
• Detection methods employ temporal analysis, statistical residuals, and multimodal evidence to differentiate true environmental events from sensor malfunctions.
• Mitigation strategies include event-aware detection, design refinements, and calibration protocols to manage anomalies while sometimes exploiting defects for enhanced sensing.

Searching arXiv for relevant papers on in-field sensor defects across environmental monitoring, semiconductor sensors, piezoelectric sensors, and defect-based sensing. In-field sensor defects are deviations in sensing hardware or sensor response that arise during deployment, operation, or environmental exposure, and that alter measurement fidelity, calibration, readout, or interpretation. Across environmental monitoring networks, piezoelectric ceramics, silicon strip and pixel sensors, optical power-meter coatings, and solid-state defect platforms, the term encompasses short faults, noise bursts, drift, microcracks, pinholes, coating damage, metal-stack shorts, and field-induced distortion mechanisms that become observable only under realistic operating conditions. A unifying feature is that such defects are not merely fabrication anomalies: they interact with system architecture, readout electronics, physical stimuli, and data-analysis pipelines, often creating ambiguity between genuine phenomena and malfunction signatures (Gupchup et al., 2019, Affolder et al., 30 Sep 2025, Eberwein et al., 17 Mar 2026).

1. Defect classes and physical manifestations

In field deployments, defect modes span both abrupt and slowly evolving failure signatures. In environmental monitoring networks, the documented modes include spikes (SHORT faults), zero-or-constant readings, high-variance bursts (NOISE faults), and drift and offset; the underlying causes listed are battery depletion, corrosion of electrical contacts in wet soils, transducer hysteresis, and software glitches in motes’ sampling loops (Gupchup et al., 2019). In PZT sensors, several surfaces and subsurface micro defects develop due to delamination, corrosion, and huge temperature fluctuation, producing a decline in performance that can be interrogated by ultrasonic structural-health-monitoring workflows (Bhattacharya et al., 2022). In strip silicon detectors, a “pinhole” is a microscopic electrical short bridging the metal readout electrode to the underlying implant through the dielectric of an AC-coupled strip sensor, with causes including wire-bond wedge errors, thermal-cycling-induced cracking, and scratches or tool marks during module assembly (Affolder et al., 30 Sep 2025).

Large-area CMOS pixel sensors exhibit a different but related defect class: recurrent short-type defects in the top copper layers of the back-end-of-line stack. In the MOSS study, shorts dominate, parametric degradations are not observed beyond those shorts, and no opens are detected in the BEOL nets; the failure-analysis chain localizes the defect mechanism to M7–M8 copper crossing regions (Eberwein et al., 17 Mar 2026). Optical sensor coatings add yet another in-field defect category. For laser power-meter sensors, the identified coating defects include thermal damage and scratches, and the associated problem is degradation of laser energy measurement accuracy (Zheng et al., 25 Sep 2025).

Not all relevant “defects” are discrete failures. In photon sensors, non-ideal sensor details such as edge distortions, field-free regions, lithography errors, fringing, doping variation, and charge-accumulation-induced field distortions alter astrometry, diffusion, point-spread functions, and flats, and are therefore operationally defect-like even when they arise from systematic electrostatic nonuniformity rather than catastrophic damage (Peterson et al., 2020). By contrast, in 4H-SiC the deep defects themselves—divacancies and silicon vacancies—are intentionally exploited as sensing elements rather than treated as failures; their charge-state conversion becomes a probe of high-frequency electric fields (Wolfowicz et al., 2018). This contrast is important because it separates “defect as malfunction” from “defect as engineered sensing center.”

2. Detection as an inference problem: faults, events, and latent defect states

A central difficulty of in-field defect analysis is that anomalous measurements need not be faulty. In environmental monitoring networks, the first raindrop on dry soil can produce the same rising-edge spike as a short-circuit-to-ground-like signature, and Gupchup et al. explicitly show that standard fault detectors may discard scientifically interesting data rather than true failures (Gupchup et al., 2019). The reported misclassification rate for event-period samples labeled faulty reaches up to $45\%$ for the first half-hour of rain-event soil-moisture samples under the SHORT rule, and no choice of the NOISE-rule allowance simultaneously drives event misclassification and false negatives toward zero (Gupchup et al., 2019).

The heuristic and estimation-based formulations in that work make the defect-detection problem explicit. The SHORT rule flags a sample when

$\bigl|X(t)-X(t-1)\bigr|>\delta,$

while the NOISE rule compares the windowed sample standard deviation against a training-derived interval. The LLSE method instead exploits spatial correlation using

$\hat s_{ij}(t)=\beta_{0j}+\beta_{1j}s_j(t),$

followed by residual thresholding and voting across neighbors (Gupchup et al., 2019). These methods formalize a generic in-field pattern: a detector uses anomaly relative to temporal continuity, local variance, or inter-sensor consistency, but the same criteria may be triggered by real exogenous events.

A related inference problem appears in distributed sensor networks with unknown defective nodes. The observation model

$y_i=h_i\theta+w_i,$

with $h_i\in\{0,1\}$, encodes the latent distinction between valid and invalid sensing modes (Zhou et al., 2015). The mixed detection-estimation (MDE) algorithm alternates mode learning and target estimation using iterative local decisions and consensus+innovation recursions. Its significance for in-field defects lies in the explicit treatment of defect status as a hidden variable rather than a one-shot anomaly label. In the high-SNR regime, the estimation error converges to that of an ideal centralized estimator with perfect knowledge of the node sensing modes, whereas naive average consensus without mode learning has estimation error that grows linearly in SNR (Zhou et al., 2015).

This body of work suggests that “defect detection” in deployed sensors is often underdetermined unless event structure, sensor modality, or system-level priors are embedded into the inference procedure. A plausible implication is that the boundary between fault diagnosis and scientific signal interpretation is itself modality dependent, not universal.

3. Electrical and electromechanical defect mechanisms in deployed hardware

Electrical shorting defects illustrate how in-field conditions couple microscopic damage to macroscopic readout anomalies. In AC-coupled strip sensors, a pinhole bypasses the coupling capacitor and directly connects implant and metal electrode. From the HV-return point of view, this places sensor and front-end in parallel, with effective impedance

$R_h=\frac{R_S+R_b}{N},$

and yields an AMAC output

$$V_{\text{out}}=\frac{R_f}{R_h}(HV_{\text{ref}}-V_S)+HV_{\text{ref}}.$$

The defect therefore injects a DC error current through the current-measurement amplifier, shifting or saturating the measured leakage current (Affolder et al., 30 Sep 2025). In practice, modules with pinholes show $45$–$55$ ADC counts at $0$ V bias with $\bigl|X(t)-X(t-1)\bigr|>\delta,$0, instead of approximately $\bigl|X(t)-X(t-1)\bigr|>\delta,$1 counts, and reducing $\bigl|X(t)-X(t-1)\bigr|>\delta,$2 removes saturation (Affolder et al., 30 Sep 2025).

The operational origin of such pinholes is also distinctly in-field. The documented causes include off-center or wireless wedge bonding, rebonding damage to neighboring strips, mechanical handling during stave loading, and thermal cycling down to $\bigl|X(t)-X(t-1)\bigr|>\delta,$3, with at least one case of a new pinhole forming at crack onset as indicated by a sudden HV-return offset shift (Affolder et al., 30 Sep 2025). Radiation damage does not directly punch through the dielectric but can exacerbate the visibility of existing pinholes via increased bulk currents (Affolder et al., 30 Sep 2025).

In wafer-scale stitched CMOS pixel sensors, the short defects are characterized by low-impedance net pairs, ohmic turn-on under power ramps, burn-through events, and localized thermal hotspots. Impedance scans classify candidate shorts using an empirical threshold $\bigl|X(t)-X(t-1)\bigr|>\delta,$4, and controlled power-ramping localizes the defect spatially with thermal imaging (Eberwein et al., 17 Mar 2026). Over $\bigl|X(t)-X(t-1)\bigr|>\delta,$5 tested wafers, the mean number of short events per wafer is $\bigl|X(t)-X(t-1)\bigr|>\delta,$6 with $\bigl|X(t)-X(t-1)\bigr|>\delta,$7, and the empirical distribution is well approximated by a Poisson law with $\bigl|X(t)-X(t-1)\bigr|>\delta,$8 (Eberwein et al., 17 Mar 2026). At the half-unit level, $\bigl|X(t)-X(t-1)\bigr|>\delta,$9 exhibit a burn-through transient and $\hat s_{ij}(t)=\beta_{0j}+\beta_{1j}s_j(t),$0 remain in persistent over-current; after burn-through, $\hat s_{ij}(t)=\beta_{0j}+\beta_{1j}s_j(t),$1 pass register tests and pixel scans (Eberwein et al., 17 Mar 2026).

Electromechanical defect mechanisms dominate the PZT case. Point-contact Coulomb coupling launches ultrasonic phonons via the piezoelectric constitutive relation

$\hat s_{ij}(t)=\beta_{0j}+\beta_{1j}s_j(t),$2

and subsurface defects are inferred from wave-defect interactions in the received current signal $\hat s_{ij}(t)=\beta_{0j}+\beta_{1j}s_j(t),$3 (Bhattacharya et al., 2022). The study reports that defects up to $\hat s_{ij}(t)=\beta_{0j}+\beta_{1j}s_j(t),$4 in diameter could be successfully distinguished and localized, with $\hat s_{ij}(t)=\beta_{0j}+\beta_{1j}s_j(t),$5 localization accuracy and hole diameter detected to within $\hat s_{ij}(t)=\beta_{0j}+\beta_{1j}s_j(t),$6 (Bhattacharya et al., 2022). Here the defect signature is not direct electrical shorting but altered ultrasonic energy distribution and wavelet-band energy loss.

4. Measurement signatures, statistical features, and localization methods

The observable signature of an in-field defect is specific to the sensor modality and readout chain. In environmental networks, the key observables are temporal discontinuities, abnormal variance, and cross-node residuals. The paper defines the misclassification rate

$\hat s_{ij}(t)=\beta_{0j}+\beta_{1j}s_j(t),$7

and reports for LLSE that soil moisture yields $\hat s_{ij}(t)=\beta_{0j}+\beta_{1j}s_j(t),$8 and $\hat s_{ij}(t)=\beta_{0j}+\beta_{1j}s_j(t),$9, while box temperature yields $y_i=h_i\theta+w_i,$0 and $y_i=h_i\theta+w_i,$1 (Gupchup et al., 2019). These results quantify the threshold trade-off between event preservation and fault capture.

For PZT monitoring, the diagnostic observable is extracted from Haar discrete wavelet transform coefficients. After baseline correction and bandpass filtering, the method computes detail-band energies

$y_i=h_i\theta+w_i,$2

and total wavelet energy

$y_i=h_i\theta+w_i,$3

A damage index is then defined as

$y_i=h_i\theta+w_i,$4

with defect declaration when $y_i=h_i\theta+w_i,$5 for $y_i=h_i\theta+w_i,$6 (Bhattacharya et al., 2022). In baseline trials over $y_i=h_i\theta+w_i,$7 positions and $y_i=h_i\theta+w_i,$8 repeated scans, $y_i=h_i\theta+w_i,$9 never exceeded $h_i\in\{0,1\}$0, corresponding to an empirical false-alarm rate below $h_i\in\{0,1\}$1 (Bhattacharya et al., 2022).

Strip-sensor pinhole localization uses electrical perturbation of the front-end bias rather than waveform decomposition. The described methods include DCDC on/off offset comparison, per-ASIC BVREF scans, light-induced leakage with channel-gain testing, and IV scans with negative-voltage offset (Affolder et al., 30 Sep 2025). In the BVREF scan, the ABC input bias is varied between approximately $h_i\in\{0,1\}$2 and $h_i\in\{0,1\}$3, and a chip whose setting produces a large upward offset shift in AMAC counts is identified as carrying pinholes (Affolder et al., 30 Sep 2025). Under white-light-induced leakage of approximately $h_i\in\{0,1\}$4–$h_i\in\{0,1\}$5 total sensor current, channels with pinholes show a characteristic gain drop at high leakage current $h_i\in\{0,1\}$6, allowing defective channels to be pinpointed (Affolder et al., 30 Sep 2025).

Thermal localization is central in the MOSS failure-analysis workflow. The hotspot extraction algorithm compares each thermal frame against a reference image, thresholds the difference inside a transformed region of interest, and identifies the hotspot center from the enclosed-contour maximum (Eberwein et al., 17 Mar 2026). That localization is then cross-correlated with CAD layers and subsequently verified by FIB-SEM and EDS. For $h_i\in\{0,1\}$7 single-event cases, $h_i\in\{0,1\}$8 agree with the M7–M8 crossing hypothesis within a $h_i\in\{0,1\}$9 resolution window, rising to $R_h=\frac{R_S+R_b}{N},$0 after manual realignment, while compatibility with alternative single-layer hypotheses remains below $R_h=\frac{R_S+R_b}{N},$1 (Eberwein et al., 17 Mar 2026).

Vision-based inspection provides a complementary route for surface defects. The laser power-meter framework segments the region of interest using Laplacian edge detection, circle localization, CLAHE, and K-means, then performs unsupervised anomaly detection with a UFlow architecture trained only on good images (Zheng et al., 25 Sep 2025). The anomaly score is aggregated across scales as

$R_h=\frac{R_S+R_b}{N},$2

and image-level thresholding uses the F1-optimal threshold

$R_h=\frac{R_S+R_b}{N},$3

On $R_h=\frac{R_S+R_b}{N},$4 real sensor images, the reported image-level AUROC is $R_h=\frac{R_S+R_b}{N},$5, pixel-level AUROC is $R_h=\frac{R_S+R_b}{N},$6, defective accuracy is $R_h=\frac{R_S+R_b}{N},$7, and good-sample accuracy is $R_h=\frac{R_S+R_b}{N},$8 (Zheng et al., 25 Sep 2025).

5. Defect-induced measurement distortion versus defect-enabled sensing

A technically important distinction separates defects that corrupt sensing from defects that act as sensing transducers. In 4H-SiC, divacancies and silicon vacancies undergo optical charge conversion between bright and dark charge states, and the conversion rate depends on RF electric-field energy density. The measured rate shift obeys

$R_h=\frac{R_S+R_b}{N},$9

with low-field scaling $$V_{\text{out}}=\frac{R_f}{R_h}(HV_{\text{ref}}-V_S)+HV_{\text{ref}}.$$0, $$V_{\text{out}}=\frac{R_f}{R_h}(HV_{\text{ref}}-V_S)+HV_{\text{ref}}.$$1, and $$V_{\text{out}}=\frac{R_f}{R_h}(HV_{\text{ref}}-V_S)+HV_{\text{ref}}.$$2 for the VV sample (Wolfowicz et al., 2018). The resulting all-optical high-frequency electrometer achieves $$V_{\text{out}}=\frac{R_f}{R_h}(HV_{\text{ref}}-V_S)+HV_{\text{ref}}.$$3 at $$V_{\text{out}}=\frac{R_f}{R_h}(HV_{\text{ref}}-V_S)+HV_{\text{ref}}.$$4 for an estimated ensemble of approximately $$V_{\text{out}}=\frac{R_f}{R_h}(HV_{\text{ref}}-V_S)+HV_{\text{ref}}.$$5 VV centers and supports three-dimensional mapping of surface acoustic wave electric fields in an AlN/SiC resonator (Wolfowicz et al., 2018).

This use of deep defects as sensors contrasts sharply with non-ideal sensor distortions in imaging devices. In photon Monte Carlo simulations, non-ideal sensor details modify the effective electric field and charge transport, producing edge flat-field roll-off, PSF broadening, tangential ellipticity near edges, lithography-induced flat stripes, dead-layer brick-wall patterns, fringing, tree-ring astrometric shifts, and brighter-fatter behavior (Peterson et al., 2020). For example, edge surface charge of $$V_{\text{out}}=\frac{R_f}{R_h}(HV_{\text{ref}}-V_S)+HV_{\text{ref}}.$$6 produces PSF FWHM increases of approximately $$V_{\text{out}}=\frac{R_f}{R_h}(HV_{\text{ref}}-V_S)+HV_{\text{ref}}.$$7–$$V_{\text{out}}=\frac{R_f}{R_h}(HV_{\text{ref}}-V_S)+HV_{\text{ref}}.$$8 within $$V_{\text{out}}=\frac{R_f}{R_h}(HV_{\text{ref}}-V_S)+HV_{\text{ref}}.$$9 pixels of the edge and astrometric shifts up to approximately $45$0 (Peterson et al., 2020). Doping variation can yield flat-field ring amplitudes of approximately $45$1 in DECam-like conditions and PSF-size variation up to approximately $45$2, while charge accumulation produces approximately $45$3 FWHM growth near saturation and sub-linear signal variance in flats (Peterson et al., 2020).

These examples show that “defect” is not a single ontological category. In one setting, a point defect is the active quantum sensor; in another, electrostatic nonuniformity constitutes a distortion source that must be calibrated out. A plausible implication is that the operational meaning of in-field sensor defect is best defined by its effect on the measurement task rather than by its microscopic form alone.

6. Mitigation, monitoring, and design principles

Mitigation strategies in the cited literature are generally procedural, modality-specific, and closely tied to measurable signatures. For environmental monitoring networks, the core recommendation is event-aware detection. The proposed principles are to buffer suspicious readings, analyze short buffer-history before assigning a fault label, exploit asynchronous but correlated responses across spatially distributed motes, and benchmark detectors on datasets that interleave real environmental events with ground-truth fault injections (Gupchup et al., 2019). The broader lesson is that purely anomaly-based rejection is insufficient in sensing modalities where rare events are the primary scientific targets.

For strip-sensor pinholes, mitigation begins during assembly. The documented controls are strict wire-bond QC, avoiding rebonding when the wedge touches the sensor without wire, reducing ultrasonic energy or bonding time when pad deformation is excessive, improving stave-loading tooling, performing IV scans with front-end ASICs powered off, using a small negative bias such as $45$4 for the reference point, and lowering $45$5 during suspect scans (Affolder et al., 30 Sep 2025). In operation, the guidance is to monitor per-module $45$6 offsets and early IV data for sudden offset jumps or saturation signs and to perform periodic light tests on a sample of modules (Affolder et al., 30 Sep 2025).

For MOSS-like wafer-scale sensors, the mitigation chain extends to foundry process changes. The reported corrections are revised BEOL design rules to eliminate tight M7–M8 overlaps, modified dual-damascene etch and copper deposition parameters, and a two-stage Cu anneal to suppress hillock formation (Eberwein et al., 17 Mar 2026). The recommended in-field early-detection protocol retains pre-power impedance scans of all net pairs, controlled voltage ramping with current limits, thermal imaging at at least $45$7 Hz, and SPC-style tracking of defect counts through Poisson and control-limit monitoring (Eberwein et al., 17 Mar 2026).

PZT field deployment recommendations focus on stabilizing the measurement interface rather than redesigning the sensor. The cited measures are spring-loaded fixtures to maintain contact force $45$8, portable Faraday shielding or differential probe pairs to suppress ambient electromagnetic coupling, solvent wiping to control contact impedance, built-in temperature sensing, adaptive thresholding in the damage-index algorithm, and on-site calibration on a reference PZT coupon before each campaign (Bhattacharya et al., 2022).

Optical inspection systems favor computational mitigation. In the laser power-meter framework, compute-heavy training is performed offline, while on-device inference consists of approximately $45$9 s preprocessing, $55$0 s UFlow forward pass, and $55$1 s post-processing for a total of approximately $55$2 s per image (Zheng et al., 25 Sep 2025). Recommended adaptations include tuning Laplacian and K-means parameters, collecting approximately $55$3 new normal crops under novel lighting for retraining, domain-randomized augmentation, quantization or pruning, and occasional human-in-the-loop threshold recalibration (Zheng et al., 25 Sep 2025).

7. Conceptual synthesis and open technical issues

Across the surveyed work, three themes recur. First, in-field defects are system-level phenomena. A pinhole in a strip detector matters because it perturbs HV current measurement circuitry (Affolder et al., 30 Sep 2025); a copper short matters because it triggers power-ramp burn-through and yield loss (Eberwein et al., 17 Mar 2026); a soil-moisture spike matters because it may be either rainfall or malfunction (Gupchup et al., 2019). The defect is therefore inseparable from the architecture that renders it observable.

Second, localization requires multimodal evidence. Electrical offsets, gain collapse under induced leakage, thermal hotspots, layout correlation, wavelet-band energy changes, cross-sensor residuals, and anomaly maps are not interchangeable, but they perform the same epistemic function: they reduce ambiguity between competing explanations (Bhattacharya et al., 2022, Affolder et al., 30 Sep 2025, Eberwein et al., 17 Mar 2026, Zheng et al., 25 Sep 2025). This suggests that robust in-field defect analysis benefits from layered evidence rather than single-threshold diagnostics.

Third, the distinction between degradation, distortion, and useful defect state remains technically consequential. Deep defects in SiC enable electrometry (Wolfowicz et al., 2018), whereas non-ideal electrostatic structure in photon sensors distorts images and must be modeled or calibrated (Peterson et al., 2020). A plausible implication is that future taxonomies of in-field sensor defects will increasingly be framed in terms of function under deployment: whether a defect reduces sensitivity, biases inference, saturates readout, modifies spatial transfer characteristics, or can itself be exploited as a transduction mechanism.

Open issues identified in the cited literature remain specific rather than generic. Environmental fault detectors still need event-aware mechanisms customized to sensing modality (Gupchup et al., 2019). V$55$4 EOCC contrast in 4H-SiC vanishes above approximately $55$5–$55$6 K, whereas VV contrast persists to room temperature (Wolfowicz et al., 2018). Pinhole formation can occur during thermal cycling, so continuous monitoring is required even after assembly QC (Affolder et al., 30 Sep 2025). Surface-inspection pipelines can degrade under glare, color shifts, or extreme shadows (Zheng et al., 25 Sep 2025). Large stitched CMOS sensors require continued yield surveillance because recurrent BEOL fault signatures may vary wafer-to-wafer with substantial dispersion (Eberwein et al., 17 Mar 2026). Taken together, these results establish in-field sensor defects as a broad research domain at the intersection of device physics, metrology, statistical inference, structural health monitoring, and deployment-aware quality control.