1,721,076 research outputs found
Sensor Network Tomography: The Revenge of the Detected
No full textA sensor network is deployed to detect the presence of a moving object (a target) in a surveyed region. Sensors make decisions about the presence of the target. Let us assume the target is aware of the detections it has caused, but has no idea which sensor has made which call. Can the target infer the positions of the detecting sensors? Since this is an inverse problem (of prey locating its predators), we shall refer to it as tomography. Maximum likelihood (ML) offers a solution, but it is combinatorial and therefore not of great practical interest. Here we propose several alternatives and investigate their performances. One class of estimators looks for a nexus of detection activity: the peak, Fourier, and ESPRIT estimators fall into this class. But the best tradeoff between complexity and performance seems to be trellis-based and of philosophy similar to the multi hypothesis tracker (MHT) idea for disambiguation of measurement-origin uncertainty (MOU) in target tracking
Polynomial-Time Algorithms for the Exact MMOSPA Estimate of a Multi-Object Probability Density Represented by Particles
No full text
PMHT Approach for Underwater Bearing-Only Multisensor–Multitarget Tracking in Clutter
No full textIn this work, we apply the probabilistic multihypothesis tracker (PMHT) for the problem of underwater bearing-only multisensor-multitarget tracking in clutter. The PMHT is a batch tracking algorithm that can efficiently process a large number of measurements from multiple sensors. We investigate both the extended Kalman filter (EKF) and the unscented Kalman filter (UKF) for dealing with the high degree of nonlinearity in the measurement model. Due to multiple sensors, the unobservability of single sensor bearing-only target tracking is avoided. Simulation results show that the PMHT works very well in a highly cluttered environment and its computational load is low
Maritime Anomaly Detection Based on Mean-Reverting Stochastic Processes Applied to a Real-World Scenario
No full text
One plus two may not equal two plus one in a social sensing network with unknown parameters
No full textParametric estimation for the generative social sensing model proposed in [19,20] is addressed. First, we provide a detailed analysis of the estimation performance bounds, in terms of the Fisher information matrix, with emphasis on the fundamental scaling laws as the number of network agents and/or the number of monitored agents' activities is large. Then, we examine two viable estimation procedures that can be useful even in such large dataset applications: the Expectation-Maximization and the Fisher scoring algorithms, which both achieve the aforementioned performance bounds
Parsimonious Kernel Fisher Discrimination
No full textBy applying recent results in optimization transfer, a new algorithm for kernel Fisher Discriminant Analysis is provided that makes use of a non-smooth penalty on the coefficients to provide a parsimonious solution. The algorithm is simple, easily programmed and is shown to perform as well as or better than a number of leading machine learning algorithms on a substantial benchmark. It is then applied to a set of extreme small-sample-size problems in virtual screening where it is found to be less accurate than a currently leading approach but is still comparable in a number of cases
- …
