The Research On Key Technologies Of Distributed Computing In Passive Position

In recent years,with the development of electronic and wireless communication technology,illegal/unknown radiation sources are increasing.The passive positioning system uses multiple terminal monitoring data to perform positioning calculations,and can achieve functions such as positioning of radiation sources and spatial electromagnetic state analysis.It is widely used in electronic reconnaissance and confrontation,radio frequency spectrum monitoring and other fields.In addition,passive positioning systems can also be used for passive detection and location acquisition of typical targets such as drones to meet major operational radio security needs.This paper studies the architecture and key technology of distributed computing module in passive location system,including:1.In-depth analysis of system requirements.Considering that the positioning calculation needs to parse a large number of data files and the complexity of the algorithm is high,resulting in a long computation time,a distributed computing scheme is proposed to improve the real-time performance of the passive location calculation.Then,the topological model and basic algorithm of task scheduling of distributed computing module are selected.Then the design of distributed computing module is proposed.According to the functional requirements,the distributed computing module is divided into three parts: task access layer,data floor layer and real-time computing sub-module,and a detailed design scheme is proposed for each part.2.For the needs of passive positioning systems,an optimized compression escheduling algorithm based on Min_Min algorithm is proposed.Since the key actor of the Min_Min algorithm to allocate tasks is whether the expected ompletion time is the minimum,the problem of unbalanced load is prone to occur.n this paper,two optimization points are put forward based on the Min_Min lgorithm: The first is the optimization of the expected completion time calculation ethod.Aiming at the characteristics of the distributed computing module,the alculation method of the expected completion time is reset,and the concept of riority is added to the requirements of the passive positioning system.The second s to solve the problem of load imbalance and propose a method based on fuzzy set lassification and compression rescheduling.Compression rescheduling is erformed for unbalanced nodes until an optimal scheduling scheme is found.Finally,a series of test analysis was performed.3.For the distributed computing module requires the server to run unattended hroughout the day,a fault-tolerant control scheme of the distributed computing odule--heartbeat detection.The use of heartbeat detection can detect abnormal odes in time and process them.This article optimizes on the fixed heartbeat cycle f the traditional heartbeat mechanism.Referring to the design of the accelerated eartbeat,a scheme for gradually reducing the heartbeat period and gradually educing the degree of confidence of the node is used to comprehensively etermine whether the node is invalid.Reducing the false positive rate also hortens the discovery delay of abnormal nodes.