Traditional Detection Schemes Research Articles

Reinforcement Learning for Intrusion Detection: More Model Longness and Fewer Updates

Several works have used machine learning techniques for network-based intrusion detection over the past few years. While proposed schemes have been able to provide high detection accuracies, they do not adequately handle the changes in network traffic behavior as time passes. Researchers often assume that model updates can be performed periodically as needed, although this is not easily feasible in real-world scenarios. This paper proposes a new intrusion detection model based on a reinforcement learning approach that aims to support extended periods without model updates. The proposal is divided into two strategies. First, it applies machine learning scheme as a reinforcement learning task to long-term learning -maintaining high reliability and high classification accuracies over time. Second, model updates are performed using a transfer learning technique coped with a sliding window mechanism that significantly decreases the need for computational resources and human intervention. Experiments performed using a new dataset spanning 8TB of data and four years of real network traffic indicate that current approaches in the literature cannot handle the evolving behavior of network traffic. Nevertheless, the proposed technique without periodic model updates achieves similar accuracy rates to traditional detection schemes implemented with semestral updates. In the case of performing periodic updates on our proposed model, it decreases the false positives up to 8%, false negatives up to 34%, with an accuracy variation up to only 6%, while demanding only seven days of training data and almost five times fewer computational resources when compared to traditional approaches.

Channel-Aided Transmission Parameter Signalling Detection for DTMB-A

For reliable synchronization and transmission parameter signalling (TPS) detection, the preamble based on distance detection (PBDD) is adopted in the digital terrestrial television multimedia broadcasting - advanced (DTMB-A) standard. However, the conventional signalling detection methods do not take the channel information into consideration and will undergo severe performance loss over frequency-selective channels, especially those with strong and long delay echos. In this paper, a novel two-stage TPS detection method is proposed based on the preamble for DTMB-A system, which utilizes the channel information to enhance the detection accuracy. After coarse distance evaluation, the equalization will be performed on the possible candidate subcarriers to reconstruct the OFDM signalling block, hence a more reliable correlation statistic can be obtained to achieve more accurate signalling detection. As demonstrated in the simulations, the detection accuracy can be improved remarkably by compensating the impact of the transmission channel compared with the traditional detection schemes, while the increment of the computational complexity is acceptable.

Improving the Resolution of Comb-Based Frequency Measurements Using a Track-And-Hold Amplifier

The advent of optical frequency standards with ultimate uncertainties in the low $1\ifmmode\times\else\texttimes\fi{}{10}^{\ensuremath{-}18}$ requires femtosecond frequency combs to support a similar level of resolution in the spectral transfer and the computation of optical frequency ratios. The related experimental challenges grow together with the number of optical frequencies to be measured simultaneously, as in many cases the comb's optical power does not allow reliable beatnote counting or tracking in all the spectral regions of interest. Here we describe the use of a track-and-hold amplifier to implement the gated detection, a previously proposed technique for improving the signal-to-noise ratio of the beatnote between a low-power tooth of the frequency comb and a continuous-wave laser. We demonstrate a 12-dB improvement in the signal-to-noise ratio of beatnotes involving a broadband-spanning optical comb as compared to traditional detection schemes. Our approach enables reliable and cycle-slip-free spectral purity transfer and reduces the system sensitivity to power drops in the comb spectrum. Being based on a single chip, it is robust, versatile, and easily embedded in more complex experimental schemes.

Two Stage Multi Sensor Detection Scheme for Large Scale WSN Using Spatial Proximity

Intuition suggests that in a large-size sensor network, event signal spreads over a sub-region known as event region. The sub-set of sensors inside the event region is located close to each other in the form of a cluster. On the contrary, the reports from the rest of the sensors that are far away from the event will be less credible and prone to faulty detection. In this work, such fact is exploited to improve the detection performance of the detection schemes at the fusion center by proposing a two-stage detection mechanism. In other words, the proposed two-stage detection scheme exploits the spatial proximity of the event region sensors. Numerical results are provided to assess the detection performance of the proposed scheme that shows that the proposed detection scheme has a superior performance over the traditional detection schemes.

56.4: Research on Measurement and Evaluation of AR/VR Near‐eye Display

Compared with traditional display devices, such as TVs, mobile phones, and watches, near‐eye display devices have great differences in the optical system, which leads to the difference between the display quality evaluation of near‐eye display devices and traditional display devices, In this paper, we analyze and design a unique optical structure based on the characteristics of the near‐eye display device, and conduct an experimental analysis on the designed imaging system. Then, the experimental research compared the traditional detection scheme, which well shows the advantages of the customized system in imaging quality. We have also developed related algorithms to evaluate the image quality of the NED detection system.

Improvement of detection probability and ranging performance of Gm-APD LiDAR with spatial correlation and adaptive adjustment of the aperture diameter

Gm-APD (Geiger-mode avalanche photodiode) is widely used in the LiDAR (Light detection and ranging) system for its excellent photon detection capability. The triggering probability of Gm-APD is closely dependent on the absolute intensity and the signal-to-noise ratio (SNR) of the echo, which limits the detection results in the daytime. In order to improve target's detection probability under a limited amount of statistics in the daytime, it proposed a detection scheme by using the spatial correlation and adaptive adjustment of the aperture diameter (AAAD) to achieve optimal target detection results in the daytime, besides, a MLE (Maximum Likelyhood Estimation) based ranging walk-error correction method was also put forward for the limited number of measurement. Firstly, considering the strategies of AAAD and spatial correlation, as well as the LiDAR ranging equation and Poisson triggering probability model, the detection probability model and ranging performance model of the scheme of spatial correlation and AAAD were built in this paper. Subsequently, the detection probability and ranging performance of the proposed scheme were theoretically analyzed, besides, a ranging walk-error correction strategy was put forward based on the MLE method. The theoretical analysis results showed that the proposed detection scheme could increase the utilization of echo photons and improve the amount of echo's triggering statistics; moreover, the Monte Carlo simulation results showed that in the case of a limited amount of statistics, the target detection scheme and ranging walk-error correction strategy could effectively improve the target detection probability and obtain a smaller ranging walk-error than the traditional detection scheme. Finally, the Gm-APD LiDAR was used to experimentally verify the proposed scheme, and the results showed that the scheme could effectively improve the echo's triggering probability with 400 measurements; the indoor experiment showed that the triggering probability could be improved by 7.54 times at most for white wall, and the white wall's target restoration ratio could reach 100% which was enhanced by more than 100%; besides, the outdoor experiment in the daytime showed that the target's restoration ratio could reach 75.54% and 99.45% for two investigated regions which were enhanced by 659.2% and 57.4%. Moreover, a high ranging accuracy could be maintained through ranging walk-error correction strategy, ranging walk-errors of -9mm, 12mm and 14mm under 400 measurements were achieved for white wall, concrete slab and black paper respectively. It effectively improved the ranging performance compared with the traditional detection scheme, and it completely satisfied the far-range detection. This paper can provide effective solutions for laser detection and ranging with Gm-APD in the daytime.

Objective identification of potentially damaging tropical cyclones over the Western North Pacific

An impact-oriented objective windstorm identification algorithm (WiTRACK), originally developed and well established for studies on extra-tropical storms, is further developed to identify damage and loss related Tropical Cyclones (TC) over the Western North Pacific. Results based on JRA-55 reanalysis data reveal that WiTRACK is able to detect the majority of strong events with hitrates of about 77%/90%/98% for TCs of the three most severe categories. Using economic loss data for China, it is found that especially events with large losses are associated with a windstorm event identified by WiTRACK. Past loss events successfully tracked by WiTRACK are associated with substantially higher losses than those events, not identified by WiTRACK. Thus, even though less skilful than traditional detection schemes if evaluated for the totality of TC events, WiTRACK is a powerful tool to identify severe, damage and loss-related TC events which additionally benefits from its simplicity and minimal input data demands. The latter may allow to apply WiTRACK to datasets not meeting the data requirements of more complex TC detection schemes.

Spin-Current Detection via an Interfacial Molecular Paramagnet

Measuring and understanding the properties and interactions of spin currents at interfaces are important for the design of future low-dissipation devices for logic and data storage. Traditional detection schemes are indirect, using the inverse spin Hall effect to convert the spin current into a voltage. Taking a different approach, the authors present a contact-free probe of spin current, based on paramagnetic adsorbates, that combines electron-spin-resonance detection with ferromagnetic-resonance excitation to monitor spin-pumping effects at an interface $d\phantom{\rule{0}{0ex}}i\phantom{\rule{0}{0ex}}r\phantom{\rule{0}{0ex}}e\phantom{\rule{0}{0ex}}c\phantom{\rule{0}{0ex}}t\phantom{\rule{0}{0ex}}l\phantom{\rule{0}{0ex}}y$, with atomic precision.

IMPROVING THE SPACE SURVEILLANCE TELESCOPE'S PERFORMANCE USING MULTI-HYPOTHESIS TESTING

The Space Surveillance Telescope (SST) is a Defense Advanced Research Projects Agency program designed to detect objects in space like near Earth asteroids and space debris in the geosynchronous Earth orbit (GEO) belt. Binary hypothesis test (BHT) methods have historically been used to facilitate the detection of new objects in space. In this paper a multi-hypothesis detection strategy is introduced to improve the detection performance of SST. In this context, the multi-hypothesis testing (MHT) determines if an unresolvable point source is in either the center, a corner, or a side of a pixel in contrast to BHT, which only tests whether an object is in the pixel or not. The images recorded by SST are undersampled such as to cause aliasing, which degrades the performance of traditional detection schemes. The equations for the MHT are derived in terms of signal-to-noise ratio (S/N), which is computed by subtracting the background light level around the pixel being tested and dividing by the standard deviation of the noise. A new method for determining the local noise statistics that rejects outliers is introduced in combination with the MHT. An experiment using observations of a known GEO satellite are used to demonstrate the improved detection performance of the new algorithm over algorithms previously reported in the literature. The results show a significant improvement in the probability of detection by as much as 50% over existing algorithms. In addition to detection, the S/N results prove to be linearly related to the least-squares estimates of point source irradiance, thus improving photometric accuracy.

UV Absorption Cross Sections of ClOOCl Are Consistent with Ozone Degradation Models

Recently, discrepancies in laboratory measurements of chlorine peroxide (ClOOCl) absorption cross sections have cast doubt on the validity of current photochemical models for stratospheric ozone degradation. Whereas previous ClOOCl absorption measurements all suffered from uncertainties due to absorption by impurities, we demonstrate here a method that uses mass-selected detection to circumvent such interference. The cross sections of ClOOCl were determined at two critical wavelengths (351 and 308 nanometers). Our results are sufficient to resolve the controversial issue originating from the ClOOCl laboratory cross sections and suggest that the highest laboratory estimates for atmospheric photolysis rates of ClOOCl, which best explain the field measurements via current chemical models, are reasonable.

Biosensors in chemical separations.

Identification of biomolecules in complex biological mixtures represents a major challenge in biomedical, environmental, and chemical research today. Chemical separations with traditional detection schemes such as absorption, fluorescence, refractive index, conductivity, and electrochemistry have been the standards for definitive identifications of many compounds. In many instances, however, the complexity of the biomixture exceeds the resolution capability of chemical separations. Biosensors based on molecular recognition can dramatically improve the selectivity of and provide biologically relevant information about the components. This review describes how coupling chemical separations with online biosensors solves challenging problems in sample analysis by identifying components that would not normally be detectable by either technique alone. This review also presents examples and principles of combining chemical separations with biosensor detection that uses living systems, whole cells, membrane receptors, enzymes, and immunosensors.