Target Discrimination Research Articles

UPF1 mutants with intact ATPase but deficient helicase activities promote efficient nonsense-mediated mRNA decay.

The conserved RNA helicase UPF1 coordinates nonsense-mediated mRNA decay (NMD) by engaging with mRNAs, RNA decay machinery, and the terminating ribosome. UPF1 ATPase activity is implicated in mRNA target discrimination and completion of decay, but the mechanisms through which UPF1 enzymatic activities such as helicase, translocase, RNP remodeling, and ATPase-stimulated dissociation influence NMD remain poorly defined. Using high-throughput biochemical assays to quantify UPF1 enzymatic activities, we show that UPF1 is only moderately processive (<200 nt) in physiological contexts and undergoes ATPase-stimulated dissociation from RNA. We combine an in silico screen with these assays to identify and characterize known and novel UPF1 mutants with altered helicase, ATPase, and RNA binding properties. We find that UPF1 mutants with substantially impaired processivity, slower unwinding rate, and/or reduced mechanochemical coupling (i.e. the ability to harness ATP hydrolysis for work) still support efficient NMD in human cells. These data are consistent with a central role for UPF1 ATPase activity in driving cycles of RNA binding and dissociation to ensure accurate NMD target selection.

INTERFEROMETRIC PROCESSING OF A DEVELOPED MIMO GBSAR FOR DISPLACEMENT MONITORING

Abstract. This study demonstrates the interferometric processing experiments of our developed multiple-input multiple-output ground-based synthetic aperture radar (MIMO GBSAR) system. GBSAR systems are known as precise noncontact instruments for monitoring earth dynamics. In recent years W band MIMO radars have shown interesting potential in this field due to their low cost, compact size, and high phase sensitivity. MIMO capability enables the angular discrimination of multiple targets in the same range as the radar sensor. In our previous works, we developed a high-resolution MIMO GBSAR system based on the combination of MIMO radar and mechanical rail. Accordingly, this study investigates the developed system’s displacement monitoring capability by presenting a controlled experiment, using fixed and moving corner reflectors and gathering 36 time series of data. We compare and discuss the results obtained from MIMO GBSAR and MIMO radar configurations. The results show that our developed system highly agrees with MIMO radar’s interferometric measurements while providing a better target discrimination capability and higher signal noise ratio.

Consensus based target tracking against deception jamming in distributed radar networks

AbstractIn this study, an anti‐deception jamming target tracking algorithm based on the data‐level fusion is proposed in a distributed radar network. First, the consensus covariance intersection (CCI) algorithm is proposed to improve tracking performance by introducing the consensus theory to the field of covariance intersection fusion. Second, unscented Kalman filter is utilised to implement the preprocessing of distributed filtering. Then, two distributed target discrimination schemes are developed for non‐collaborative and collaborative false targets respectively. For non‐collaborative false targets, according to the characteristics that estimated spatial locations of the false target from different radars are relatively scattered, a false target elimination method based on nearest neighbour algorithm is proposed. For collaborative false targets, after equipped with passive radars that are less vulnerable to interference, the consensus algorithm is introduced to convey the real target information to the unconnected active radars, thus achieving an effect of global anti‐deception jamming. Finally, the CCI algorithm is deployed for distributed fusion of physical target tracking, so as to improve the overall estimation performance.

Discrimination of Target and Mainlobe Jammers With FDA-MIMO Radar

This letter focuses on discriminating true target from mainlobe jammers using a frequency diverse array-multiple-input multiple-output (FDA-MIMO) radar. To this end, a three-stage detection and discrimination architecture is proposed. Specifically, mainlobe jammers caused by large time delays can be discriminated from the true target based on the range degrees-of-freedom (DOFs) of FDA-MIMO radar, while the true target and jammers generated with small time delays are distinguishable in the fast time domain. In particular, a binary hypothesis test is formulated for discrimination, where the <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$H_{1}$</tex-math></inline-formula> hypothesis contains the true target, while the <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$H_{0}$</tex-math></inline-formula> hypothesis contains the unknown mainlobe jammer. The mainlobe jammer can be formulated as either a known subspace model or an unknown rank-one model. The former leads to a subspace-based selective detector (SSD) and the latter generates a rank-one-based SD (ROSD). Then, the formulated tests are handled using the generalized likelihood ratio test (GLRT) criterion. At the analysis stage, the selectivity and detection performance are compared with the conventional GLRT and selective receivers. Simulation results validate that the proposed detectors achieve satisfactory detection performance while providing the selectivity to mainlobe jammers.

Boids Flocking Algorithm for Situation Assessment of Driver Assistance Systems

Driver assistance systems are increasingly becoming part of the standard equipment of vehicles and thus contribute to road safety. However, as they become more widespread, the requirements for cost efficiency are also increasing, and so few and inexpensive sensors are used in these systems. Especially in challenging situations, this leads to the fact that target discrimination cannot be ensured which may lead to false reactions of the driver assistance system. In this paper, the Boids flocking algorithm is used to generate semantic neighborhood information between tracked objects which in turn can significantly improve the overall performance. Two different variants were developed: First, a freemoving flock whereby a separate flock is generated per tracked object and second, a formation-controlled flock where boids of a single flock move along the future road course in a pre-defined formation. In the first approach, the interaction between the flocks as well as the interaction between the boids within a flock is used to generate additional information, which in turn can be used to improve, for example, lane change detection. For the latter approach, new behavioral rules have been developed, so that the boids can reliably identify control-relevant objects to a driver assistance system. Finally, the performance of the presented methods is verified through extensive simulations.

Brain Network Reorganization During Visual Search Task Revealed by a Network Analysis of Fixation-Related Potential.

Visual search is ubiquitous in daily life and has attracted substantial research interest over the past decades. Although accumulating evidence has suggested complex neurocognitive processes underlying visual search, the neural communication across the brain regions remains poorly understood. The present work aimed to fill this gap by investigating functional networks of fixation-related potential (FRP) during the visual search task. Multi-frequency electroencephalogram (EEG) networks were constructed from 70 university students (male/female = 35/35) using FRPs time-locked to target and non-target fixation onsets, which were determined by concurrent eye-tracking data. Then graph theoretical analysis (GTA) and a data-driven classification framework were employed to quantitatively reveal the divergent reorganization between target and non-target FRPs. We found distinct network architectures between target and non-target mainly in the delta and theta bands. More importantly, we achieved a classification accuracy of 92.74% for target and non-target discrimination using both global and nodal network features. In line with the results of GTA, we found that the integration corresponding to target and non-target FRPs significantly differed, while the nodal features contributing most to classification performance primarily resided in the occipital and parietal-temporal areas. Interestingly, we revealed that females exhibited significantly higher local efficiency in delta band when focusing on the search task. In summary, these results provide some of the first quantitative insights into the underlying brain interaction patterns during the visual search process.

Adaptive Spatial Regularization Target Tracking Algorithm Based on Multifeature Fusion

The accuracy and robustness of object-tracking algorithms are challenging tasks in the field of artificial intelligence. The discriminative correlation filter has fast tracking speed and target discrimination ability in visual target tracking, but it will be affected by unnecessary boundary effects. Although spatially regularized discriminative correlation filters (SRDCF) effectively solve this problem, it faces a slow tracking speed and cannot meet the real-time requirement. Moreover, the constraints added by this algorithm are fixed during the tracking process, which needs to reflect better the characteristics and appearance changes of a particular object. Therefore, we propose an adaptive spatial regularization target tracking model based on multifeature fusion and learn the spatial regularization weights by adaptive spatial regularization terms, so that the filter can adapt to the changes of the target, which can highlight the target area and suppress the background area more accurately. At the same time, in order to cope with the problem of target loss when the target is obscured by a large area or rotated to a large extent, texture features are introduced to improve the deficiencies of HOG features and CN features, so as to achieve the complementary advantages among different features. In addition, the benchmark algorithm SRDCF uses Gaussian–Seidel iterative method to solve the filter, which makes the tracking speed very slow and cannot be tracked in real time. Therefore, this paper proposes an ADMM (alternating direction method of multiplier) algorithm to optimize the solution of the filter and achieve the tracking real-time requirement. The experimental results on the OTB-2015 dataset show that the accuracy and success rate of the improved algorithm reach 86% and 81.9%, respectively. At the same time, in the OTB-2013 dataset, it has increased by 3.4% and 6.3%, respectively, on the basis of the benchmark algorithm. It is verified that the improved model not only improves the robustness of the algorithm but also can better adapt to the changes of the target, while effectively reducing the computational overhead and meeting the tracking real-time requirements.

9-Cyano-10-telluriumpyronin Derivatives as Red-light-activatable Raman Probes.

Photoactivatable fluorescence probes can track the dynamics of specific cells or biomolecules with high spatiotemporal resolution, but their broad absorption and emission peaks limit the number of wavelength windows that can be employed simultaneously. In contrast, the narrower peak width of Raman signals offers more scope for simultaneous discrimination of multiple targets, and therefore a palette of photoactivatable Raman probes would enable more comprehensive investigation of biological phenomena. Herein we report 9-cyano-10-telluriumpyronin (9CN-TeP) derivatives as photoactivatable Raman probes whose stimulated Raman scattering (SRS) intensity is enhanced by photooxidation of the tellurium atom. Modification to increase the stability of the oxidation product led to a julolidine-like derivative, 9CN-diMeJTeP, which is photo-oxidized at the tellurium atom by red light irradiation to afford a sufficiently stable oxidation product with strong electronic pre-resonance, resulting in a bathochromic shift of the absorption spectrum and increased SRS intensity.

Subset Selection Strategies Based on Target Positioning Characteristics for Anti-Jamming Technology

For the discrimination of false targets, the discrimination probability can be improved by increasing the number of radar stations. However, that may result in a serious waste of equipment resources when too many radars are involved. An asymptotic subset selection strategy based on target positioning characteristics is proposed to address the above issues. Several effective strategies are considered to select some transmitters and receivers to form a radar subset, such as the rapid shrinkage method, global shrinkage method, and predetermined size method, which can guarantee the preset discrimination performance of limited equipment resources and reduce the waste of resources. All of the selected stations have good spatial distribution or strong discrimination capacity in multistatic radar system. Compared with the exhaustive search, the proposed subset selection strategy affords a significant reduction in terms of time complexity. The simulation results show that the radar subset can maintain approximate discrimination performance with the original multistatic radar systems. At the same time, the proposed method optimizes the number of radar stations and reduces data processing time and required communication links, thus effectively saving operating costs.

Differential modulation of visual responses by distractor or target expectations

Discriminating relevant from irrelevant information in a busy visual scene is supported by statistical regularities in the environment. However, it is unclear to what extent immediate stimulus repetitions and higher order expectations (whether a repetition is statistically probable or not) are supported by the same neural mechanisms. Moreover, it is also unclear whether target and distractor-related processing are mediated by the same or different underlying neural mechanisms. Using a speeded target discrimination task, the present study implicitly cued subjects to the location of the target or the distractor via manipulations in the underlying stimulus predictability. In separate studies, we collected EEG and MEG alongside behavioural data. Results showed that reaction times were reduced with increased expectations for both types of stimuli and that these effects were driven by expected repetitions in both cases. Despite the similar behavioural pattern across target and distractors, neurophysiological measures distinguished the two stimuli. Specifically, the amplitude of the P1 was modulated by stimulus relevance, being reduced for repeated distractors and increased for repeated targets. The P1 was not, however, modulated by higher order stimulus expectations. These expectations were instead reflected in modulations in ERP amplitude and theta power in frontocentral electrodes. Finally, we observed that a single repetition of a distractor was sufficient to reduce decodability of stimulus spatial location and was also accompanied by diminished representation of stimulus features. Our results highlight the unique mechanisms involved in distractor expectation and suppression and underline the importance of studying these processes distinctly from target-related attentional control.

Psychophysics in the ophthalmological practice—II. Contrast sensitivity

: Contrast is the differential luminance between one object and another. Contrast sensitivity (CS) quantifies the ability to detect this difference: estimating contrast threshold provides information about the quality of vision and helps diagnose and monitor eye diseases. High contrast visual acuity assessment is traditionally performed in the eye care practice, whereas the estimate of the discrimination of low contrast targets, an important complementary task for the perception of details, is far less employed. An example is driving when the contrast between vehicles, obstacles, pedestrians, and the background is reduced by fog. Many conditions can selectively degrade CS, while visual acuity remains intact. In addition to spatial CS, “temporal” CS is defined as the ability to discriminate luminance differences in the temporal domain, i.e., to discriminate information that reaches the visual cortex as a function of time. Likewise, temporal sensitivity of the visual system can be investigated in terms of critical fusion frequency (CFF), an indicator of the integrity of the magnocellular system that is responsible for the perception of transient stimulations. As a matter of fact, temporal resolution can be abnormal in neuro-ophthalmological clinical conditions. This paper aims at considering CS and its application to the clinical practice.

Feedback-Assisted Automatic Target and Clutter Discrimination Using a Bayesian Convolutional Neural Network for Improved Explainability in SAR Applications

In this paper, a feedback training approach for efficiently dealing with distribution shift in synthetic aperture radar target detection using a Bayesian convolutional neural network is proposed. After training the network on in-distribution data, it is tested on out-of-distribution data. Samples that are classified incorrectly with high certainty are fed back for a second round of training. This results in the reduction of false positives in the out-of-distribution dataset. False positive target detections challenge human attention, sensor resource management, and mission engagement. In these types of applications, a reduction in false positives thus often takes precedence over target detection and classification performance. The classifier is used to discriminate the targets from the clutter and to classify the target type in a single step as opposed to the traditional approach of having a sequential chain of functions for target detection and localisation before the machine learning algorithm. Another aspect of automated synthetic aperture radar detection and recognition problems addressed here is the fact that human users of the output of traditional classification systems are presented with decisions made by “black box” algorithms. Consequently, the decisions are not explainable, even to an expert in the sensor domain. This paper makes use of the concept of explainable artificial intelligence via uncertainty heat maps that are overlaid onto synthetic aperture radar imagery to furnish the user with additional information about classification decisions. These uncertainty heat maps facilitate trust in the machine learning algorithm and are derived from the uncertainty estimates of the classifications from the Bayesian convolutional neural network. These uncertainty overlays further enhance the users’ ability to interpret the reasons why certain decisions were made by the algorithm. Further, it is demonstrated that feeding back the high-certainty, incorrectly classified out-of-distribution data results in an average improvement in detection performance and a reduction in uncertainty for all synthetic aperture radar images processed. Compared to the baseline method, an improvement in recall of 11.8%, and a reduction in the false positive rate of 7.08% were demonstrated using the Feedback-assisted Bayesian Convolutional Neural Network or FaBCNN.

The Neural Signature of Impaired Inhibitory Control in Individuals with Heroin Use Disorder.

Heroin addiction imposes a devastating toll on society, with little known about its neurobiology. Excessive salience attribution to drug over nondrug cues/reinforcers, with concomitant inhibitory control decreases, are common mechanisms underlying drug addiction. Although inhibitory control alterations generally culminate in prefrontal cortex (PFC) hypoactivations across drugs of abuse, patterns in individuals with heroin addiction (iHUDs) remain unknown. We used a stop-signal fMRI task designed to meet recent consensus guidelines in mapping inhibitory control in 41 iHUDs and 24 age- and sex-matched healthy controls (HCs). Despite group similarities in the stop-signal response time (SSRT; the classic inhibitory control measure), compared with HCs, iHUDs exhibited impaired target detection sensitivity (proportion of hits in go vs false alarms in stop trials; p = 0.003). Additionally, iHUDs exhibited lower right anterior PFC (aPFC) and dorsolateral PFC (dlPFC) activity during successful versus failed stops (the hallmark inhibitory control contrast). Lower left dlPFC/supplementary motor area (SMA) activity was associated with slower SSRT specifically in iHUDs and lower left aPFC activity with worse target sensitivity across all participants (p < 0.05 corrected). Importantly, in iHUDs, lower left SMA and aPFC activity during inhibitory control was associated with shorter time since last use and higher severity of dependence, respectively (p < 0.05 corrected). Together, results revealed lower perceptual sensitivity and hypoactivations during inhibitory control in cognitive control regions (e.g., aPFC, dlPFC, SMA) as associated with task performance and heroin use severity measures in iHUDs. Such neurobehavioral inhibitory control deficits may contribute to self-control lapses in heroin addiction, constituting targets for prevention and intervention efforts to enhance recovery.SIGNIFICANCE STATEMENT Heroin addiction continues its deadly impact, with little known about the neurobiology of this disorder. Although behavioral and prefrontal cortical impairments in inhibitory control characterize addiction across drugs of abuse, these patterns remain underexplored in heroin addiction. Here, we illustrate a significant behavioral impairment in target discrimination in individuals with heroin addiction compared with matched healthy controls. We further show lower engagement during inhibitory control in the anterior and dorsolateral prefrontal cortex (key regions that regulate cognitive control) as associated with slower stopping, worse discrimination, and heroin use measures. Mapping the neurobiology of inhibitory control in heroin addiction for the first time, we identify potential treatment targets inclusive of prefrontal cortex-mediated cognitive control amenable for neuromodulation en route to recovery.

Target discrimination in the extended shortwave infrared band (2 to 2.5μm) compared with visible, near-infrared, and SWIR in degraded visual environments

Long-range target identification is well studied in the visible (Vis) and near-infrared (NIR) bands and more recently in the shortwave infrared (SWIR). The longer wavelength of SWIR (1 to 1.7 μm) improves target detection for both long ranges and under challenging atmospheric conditions because it is less limited by scattering and absorption in the atmosphere. For these reasons, SWIR sensors are proliferating on military platforms. The extended shortwave infrared (eSWIR) band spanning from 2 to 2.5 μm is not typically limited by diffraction, and as a result, the band benefits target acquisition both at long ranges and for degraded visual environments (DVEs). Theoretical and experimental data compare eSWIR with Vis, NIR, and SWIR for atmospheric transmission, reflectivity, illumination, and sensor resolution and sensitivity. The experimental setup includes two testbeds, each with four cameras. The first is a wide field of view (FOV) testbed matching FOV at 20 deg for each camera. The second is a narrow FOV telescope testbed to match instantaneous FOV for consistent resolution across all four bands at long ranges. Both the theory and experiment demonstrate advantages of using eSWIR for long-range target identification under DVEs.

UPF1 mutants with intact ATPase but deficient helicase activities promote efficient nonsense-mediated mRNA decay.

The conserved RNA helicase UPF1 coordinates nonsense-mediated mRNA decay (NMD) by engaging with mRNAs, RNA decay machineryand the terminating ribosome. UPF1 ATPase activity is implicated in mRNA target discrimination and completion of decay, but the mechanisms through which UPF1 enzymatic activities such as helicase, translocase, RNP remodeling, and ATPase-stimulated dissociation influence NMD remain poorly defined. Using high-throughput biochemical assays to quantify UPF1 enzymatic activities, we show that UPF1 is only moderately processive (<200 nt) in physiological contexts and undergoes ATPase-stimulated dissociation from RNA. We combine an in silico screen with these assays to identify and characterize known and novel UPF1 mutants with altered helicase, ATPase, and RNA binding properties. We find that UPF1 mutants with substantially impaired processivity (E797R, G619K/A546H), faster (G619K) or slower (K547P, E797R, G619K/A546H) unwinding rates, and/or reduced mechanochemical coupling (i.e. the ability to harness ATP hydrolysis for work;K547P, R549S, G619K, G619K/A546H) can still support efficient NMD of well-characterized targets in human cells. These data are consistent with a central role for UPF1 ATPase activity in driving cycles of RNA binding and dissociation to ensure accurate NMD target selection.

Deficits in visuospatial attentional cueing following mild traumatic brain injury

Visual attentional deficits are frequently reported in patients with mild traumatic brain injury (TBI). In the present study, the ability to orient visual attention (i.e., the use of endogenous and exogenous visual cues) was investigated using a modified Posner visual search task, in which the participant was required to search for a target shape (radial frequency patterns) amongst distractor shapes. Participants were required to determine whether a target radial frequency pattern was present or absent from an array of distractors. Attention to the target location was cued using central or peripheral cueing procedures to investigate endogenous or exogenous attention allocation. Predictability was not manipulated between central and peripheral cues. Search difficulty was varied by systematically changing the radial frequency difference between target and distractors (and thereby shape difference), and cues could be valid or invalid in that they correctly or incorrectly indicated the position of the target shape. Both target discriminability (i.e., identifying the presence or absence of the target) and reaction times were measured. Thirteen patients with chronic mild TBI and 21 age-, sex-, and IQ -matched healthy controls participated in the study. For control participants, both discrimination accuracy and reaction times improved with visual search efficiency, and they were sensitive to the type of cue, with performance worst for cue invalid conditions than valid conditions. However, the results for TBI patients were strikingly different; we find that discrimination accuracy slightly improved with visual search difficulty (compared to controls), but not reaction times, and TBI patients were largely insensitive to the type of visual cue, and did not show a selective deficit for central or peripheral cues, suggesting an impairment in both endogenous and exogenous visual attention. In conclusion, patients with mild TBI exhibit a poor ability to orient visual attention.

Allocation of Visuospatial Attention Indexes Evidence Accumulation for Reach Decisions.

Visuospatial attention is a prerequisite for the performance of visually guided movements: perceptual discrimination is regularly enhanced at target locations before movement initiation. It is known that this attentional prioritization evolves over the time of movement preparation; however, it is not clear whether this build-up simply reflects a time requirement of attention formation or whether, instead, attention build-up reflects the emergence of the movement decision. To address this question, we combined behavioral experiments, psychophysics, and computational decision-making models to characterize the time course of attention build-up during motor preparation. Participants (n = 46, 29 female) executed center-out reaches to one of two potential target locations and reported the identity of a visual discrimination target (DT) that occurred concurrently at one of various time-points during movement preparation and execution. Visual discrimination increased simultaneously at the two potential target locations but was modulated by the experiment-wide probability that a given location would become the final goal. Attention increased further for the location that was then designated as the final goal location, with a time course closely related to movement initiation. A sequential sampling model of decision-making faithfully predicted key temporal characteristics of attentional allocation. Together, these findings provide evidence that visuospatial attentional prioritization during motor preparation does not simply reflect that a spatial location has been selected as movement goal, but rather indexes the time-extended, cumulative decision that leads to the selection, hence constituting a link between perceptual and motor aspects of sensorimotor decisions.

Temporal changes in ERP amplitudes during sustained performance of the Attention Network Test

This study aimed to use Event-Related Potential (ERP) data to investigate how multiple attentional networks might contribute to vigilance decrement, based on Petersen and Posner's (2012) model of networks for executive control, alerting, and spatial orienting. The networks may differ in their sensitivity to effects of time on task. Based on the theory of attentional networks and previous findings, it was hypothesized that temporal decrements would be found in executive control and alerting networks. 102 participants (53 females) performed a version of the Attention Network Test (ANT) that was modified to require a prolonged period of continuous performance of approximately 70 min during electroencephalogram (EEG) recording. ERP amplitudes were measured in task conditions associated with executive control, alerting, and spatial orienting networks were measured across three consecutive task stages. N100 and P300 amplitudes were used to assess early selective attention and later target discrimination, respectively. Results largely replicated previous findings on ERP responses to the ANT. Amplitudes of the N100 wave and P300 declined over time in multiple conditions, consistent with increasing vulnerability to vigilance decrement. Evidence from cue × stage interactions suggested a temporal decrement in alerting processes, indicated by N100 amplitude, but behavioral data did not show any impairment specific to alerting. By contrast, there was a correspondence between a modest decline in frontal P300 amplitude seen in a no cue condition and slowing of response in this condition. A full explanation for vigilance decrement in terms of attention networks remains elusive. However, the current data suggest that parietal N100 and frontal P300 index two distinct processes that may contribute to loss of alertness and vigilance, depending on task demands.

Towards Uncovering the Intrinsic Data Structures for Unsupervised Domain Adaptation Using Structurally Regularized Deep Clustering.

Unsupervised domain adaptation (UDA) is to learn classification models that make predictions for unlabeled data on a target domain, given labeled data on a source domain whose distribution diverges from the target one. Mainstream UDA methods strive to learn domain-aligned features such that classifiers trained on the source features can be readily applied to the target ones. Although impressive results have been achieved, these methods have a potential risk of damaging the intrinsic data structures of target discrimination, raising an issue of generalization particularly for UDA tasks in an inductive setting. To address this issue, we are motivated by a UDA assumption of structural similarity across domains, and propose to directly uncover the intrinsic target discrimination via constrained clustering, where we constrain the clustering solutions using structural source regularization that hinges on the very same assumption. Technically, we propose a hybrid model of Structurally Regularized Deep Clustering, which integrates the regularized discriminative clustering of target data with a generative one, and we thus term our method as H-SRDC. Our hybrid model is based on a deep clustering framework that minimizes the Kullback-Leibler divergence between the distribution of network prediction and an auxiliary one, where we impose structural regularization by learning domain-shared classifier and cluster centroids. By enriching the structural similarity assumption, we are able to extend H-SRDC for a pixel-level UDA task of semantic segmentation. We conduct extensive experiments on seven UDA benchmarks of image classification and semantic segmentation. With no explicit feature alignment, our proposed H-SRDC outperforms all the existing methods under both the inductive and transductive settings. We make our implementation codes publicly available at https://github.com/huitangtang/H-SRDC.

Cooperative Anti-Deception Jamming in a Distributed Multiple-Radar System under Registration Errors.

A distributed multiple-radar system has natural advantages in anti-deception jamming. However, most of the anti-jamming methods are proposed in full spatial registration. In practice, the registration error is difficult to eliminate, which will seriously degrade the performance of cooperative anti-jamming. Therefore, it is of great significance to consider the problem of cooperative anti-deception jamming under registration error. In this paper, the cooperative anti-deception jamming method is proposed in a distributed multiple-radar system under registration errors. On the premise of the known registration error, target received signal vectors are estimated from an uncertainty region in each channel by maximum likelihood (ML) algorithm. With the estimated received signal vectors, a target discrimination algorithm is introduced based on the difference in target spatial scattering characteristics, which calculate the correlation coefficient between different target received signal vectors and discriminate a false target with a designed threshold. Furthermore, since the registration error depends on the radar site errors, theoretical derivation for the registration error is given as a function of the transmitter and receiver site errors. Finally, simulation results verify the feasibility of the proposed discrimination method, and its performance due to the influence of the jamming-to-noise ratio (JNR), the registration error, the target size, and the discrimination threshold are considered.