Search time optimization for random pulse sources with given accuracy

Author(s):  
А.Л. Резник ◽  
А.В. Тузиков ◽  
А.А. Соловьев ◽  
А.В. Торгов ◽  
В.А. Ковалев

Исследуются вопросы построения быстродействующих алгоритмов обнаружения и локализации точечных источников, имеющих случайное распределение и обнаруживающих себя в случайные моменты времени генерацией мгновенных дельта-импульсов. Поиск осуществляется системой, включающей одно либо несколько приемных устройств, и выполняется с соблюдением требований по точности локализации. Предполагается, что все приемные устройства имеют произвольно перестраиваемые во времени окна обзора. Оптимальной считается процедура, которая в статистическом плане (т.е. по ансамблю реализаций) минимизирует среднее время локализации. Показано, что даже при сравнительно невысоких требованиях к точности локализации оптимальная процедура состоит из нескольких этапов, каждый из которых заканчивается в момент регистрации очередного импульса. Вполне допускается ситуация, когда в процессе оптимального поиска часть генерируемых источником импульсов может быть пропущена приемной системой. В работе рассчитаны и систематизированы параметры оптимального поиска в зависимости от количества приемных устройств и требуемой точности локализации. Для случая предельно высоких требований к точности локализации рассчитаны параметры асимптотически оптимальных поисковых алгоритмов. Показана возможность использования полученных результатов в многомерном случае. Purpose. The main goal of the research is to develop time-optimal algorithms for the localization of point sources that have a random spatial distribution and indicate themselves by generating instantaneous delta pulses at random time points. Methods. In many practically important problems requiring the highest reduction in the average time of localization of signal objects, the complexity of constructing optimal search algorithms forces researchers to resort to various kinds of simplifications or to the use of methods of numerical and simulation modelling. The mathematical apparatus used in the article belongs to probabilistic-statistical and non-linear programming methods. In a number of sections of the study (in particular, when constructing optimal control algorithms for multi-receiving search engines), traditional methods of discrete analysis and applied programming were used. Results. The solution of the variational problem is found, which minimizes the average localization time in the class of one-stage search algorithms with a known distribution density and the simultaneous absence of a priori information about the intensity of a random pulse source. For random point sources with a priori known intensity of the instantaneous generation of pulses, physically realizable multistage search algorithms have been constructed that have a significant gain in speed over single-stage algorithms, especially with increased requirements for localization accuracy. For a uniform distribution of a random source, an optimal strategy of multi-stage search was calculated, depending on the required localization accuracy and the number of receivers used. Findings. A distinctive feature of the studies is their universality, since in mathematical terms, the discussed problems and algorithms for the time-optimal search of random point-pulse objects arise in many scientific and technical applications. In particular, such studies are needed when developing methods for intermittent failures troubleshooting in the theory of reliability, in mathematical communication theory and in problems of technical diagnostics. Scientifically equivalent problems appear in the problems of detection, localization and tracking of radiation targets for eliminating malfunctions that manifest themselves in the form of intermittent failures. Scientifically equivalent problems arise in the problems of detecting, localizing and tracking radiation source targets.

2019 ◽  
Vol 43 (4) ◽  
pp. 605-610 ◽  
Author(s):  
A.L. Reznik ◽  
A.V. Tuzikov ◽  
A.A. Soloviev ◽  
A.V. Torgov ◽  
V.A. Kovalev

The article describes methods and algorithms related to the analysis of dynamically changing discrete random fields. Time-optimal strategies for the localization of pulsed-point sources having a random spatial distribution and indicating themselves by generating instant delta pulses at random times are proposed. An optimal strategy is a procedure that has a minimum (statistically) average localization time. The search is performed in accordance with the requirements for localization accuracy and is carried out by a system with one or several receiving devices. Along with the predetermined accuracy of localization of a random pulsed-point source, a significant complicating factor of the formulated problem is that the choice of the optimal search procedure is not limited to one-step algorithms that end at the moment of first pulse generation. Moreover, the article shows that even with relatively low requirements for localization accuracy, the time-optimal procedure consists of several steps, and the transition from one step to another occurs at the time of registration of the next pulse by the receiving system. In this case, the situation is acceptable when during the process of optimal search some of the generated pulses are not fixed by the receiving system. The parameters of the optimal search depending on the number of receiving devices and the required accuracy of localization are calculated and described in the paper.


2021 ◽  
Author(s):  
A.L. Reznik ◽  
A.A. Soloviev ◽  
A.V. Torgov

In this paper, we describe algorithms for the optimal search for pulsed-point sources, and the information on their distribution is limited to single-mode functions with a stepped probability distribution density, which makes it possible to physically implement the algorithms.


The questions of creating high-speed algorithms for detecting and localizing point sources having a random distribution and manifesting themselves by generating instantaneous delta pulses at random times are described. The search is carried out by a system including one or more receiving devices, and is performed taking into account the requirements for localization accuracy. It is assumed that all receivers have freely tunable viewing windows. The optimal procedure is one that minimizes (in a statistical sense) the average localization time. It is established that even with relatively low requirements for localization accuracy, the optimal procedure consists of several stages (each such stage ends at the moment of the next pulse registration). In this case, it is possible receiving system to miss some pulses generated by the source during the optimal search. In the work, the optimal search parameters are calculated depending on the number of receiving devices and the required localization accuracy. The possibility of using the results in a multidimensional case is shown.


10.29007/v7zc ◽  
2018 ◽  
Author(s):  
Justin Lovinger ◽  
Xiaoqin Zhang

In 1992, Stuart Russell briefly introduced a series of memory efficient optimal search algorithms. Among which is the Simplified Memory-bounded A Star (SMA*) algorithm, unique for its explicit memory bound. Despite progress in memory efficient A Star variants, search algorithms with explicit memory bounds are absent from progress. SMA* remains the premier memory bounded optimal search algorithm. In this paper, we present an enhanced version of SMA* (SMA*+), providing a new open list, simplified implementation, and a culling heuristic function, which improves search performance through a priori knowledge of the search space. We present benchmark and comparison results with state-of-the-art optimal search algorithms, and examine the performance characteristics of SMA*+.


2021 ◽  
Author(s):  
Sebastian Wolff ◽  
Friedemann Reum ◽  
Christoph Kiemle ◽  
Gerhard Ehret ◽  
Mathieu Quatrevalet ◽  
...  

<p>Methane (CH<sub>4</sub>) is the second most important anthropogenic greenhouse gas (GHG) with respect to radiative forcing. Since pre-industrial times, the globally averaged CH<sub>4</sub> concentration in the atmosphere has risen by a factor of 2.5. A large fraction of global anthropogenic CH<sub>4</sub> emissions originates from localized point sources, e.g. coal mine ventilation shafts. International treaties foresee GHG emission reductions, entailing independent monitoring and verification support capacities. Considering the spatially widespread distribution of point sources, remote sensing approaches are favourable, in order to enable rapid survey of larger areas. In this respect, active remote sensing by airborne lidar is promising, such as provided by the integrated-path differential-absorption lidar CHARM-F operated by DLR. Installed onboard the German research aircraft HALO, CHARM-F serves as a demonstrator for future satellite missions, e.g. MERLIN. CHARM-F simultaneously measures weighted vertical column mixing ratios of CO<sub>2</sub> and CH<sub>4</sub> below the aircraft. In spring 2018, during the CoMet field campaign, measurements were taken in the Upper Silesian Coal Basin (USCB) in Poland. The USCB is considered to be a European hotspot of CH<sub>4</sub> emissions, covering an area of approximately 50 km × 50 km. Due to the high number of coal mines and density of ventilation shafts in the USCB, individual CH<sub>4</sub> exhaust plumes can overlap. This makes simple approaches to determine the emission rates of single shafts, i.e. the cross-sectional flux method, difficult. Therefore, we use an inverse modelling approach to obtain an estimate of the individual emission rates. Specifically, we employ the Weather Research and Forecast Model (WRF) coupled to the CarbonTracker Data Assimilation Shell (CTDAS), an Ensemble Kalman Filter. CTDAS-WRF propagates an ensemble realization of the a priori CH<sub>4</sub> emissions forward in space and time, samples the simulated CH<sub>4</sub> concentrations along the measurement’s flight path, and scales the a priori emission rates to optimally fit the measured values, while remaining tied to the prior. Hereby, we obtain a regularized a posteriori best emission estimate for the individual ventilation shafts. Here, we report on the results of this inverse modelling approach, including individual and aggregated emission estimates, their uncertainties, and to which extent the data are able to constrain individual emitters independently.</p>


2003 ◽  
Vol 209 ◽  
pp. 412-412
Author(s):  
J. R. Walsh ◽  
L. B. Lucy

Long slit spectra of astronomical objects either contain point sources, characterized by a known Point Spread Function (PSF), which is often wavelength dependent, and extended sources, such as nebulae, whose spatial extent is not a priori known. The analysis of long slit spectra consists in separating the spectrum into either: the point source(s), free of the background (“extraction”); or the extended source(s), free of contaminating point source spectra. Depending on the scientific aim, one or both of these data are of interest, such as the spectrum of the central star of a planetary nebula AND the line and continuum spectrum of the nebula with the star removed. In the simple case of a point source with a background gradient, the spectrum of the point source can be simply extracted by subtracting a background fit by a low order function and summing (perhaps with weights, as in optimal extraction) the point source signal at each spectral element in the cross-dispersion direction. When the background is complex or there are many point sources, there is no guide as to how to fit the extended source spectrum beneath the point sources. Simple methods can give a poor estimate of the spectra of point sources and the spectrum of the background in the vicinity of the stars. The application of image restoration algorithms to the spatial component of long slit spectra offers a potential solution.


Sign in / Sign up

Export Citation Format

Share Document