The Design And Implementation Of Information Fusion Anti-Jamming Algorithm For Multi-Mode Compound Guidance Technique

With the increasing complexity of modern battlefield environment,the multi-sensor compound guidance technique has become the development tendency for the precision-guided weapons.Different sensors are combined to complement with each other for the enhanced performance of detection,tracking and recognition,thus the whole performance of the guidance system can be improved.Typical sensors,such as radar,infrared and laser sensors are commonly used as the seeker.Notice that the three typical sensors can be combined to conduct detection and guidance,thus the anti-interference performance of precision-guided weapon can be improved.Based on the information fusion technique in the multi-sensor composite guidance system,an information fusion scheme is designed for a three-sensor composite guidance system,which consists of the radar,infrared,and laser sensors.Furthermore,the corresponding algorithms are designed and developed on the multi-core DSP platform,i.e.TMS320C6678.Finally,the performance of the information fusion scheme is verified with the digital simulation-testing system.First,the basic theory of multi-sensor information fusion technique is briefly introduced.The information fusion structure can be divided according to different criteria.According to the location,it can be divided into four main types,i.e.,the centralized,distributed,hybrid,and multi-level information fusion structures.In terms of the target property,the information fusion structure can be divided into different levels,i.e.data-level,feature-level,and decision-level information fusion structures.Based on the information fusion theory and the application background of air-to-ground missile,the feature-level distributed information fusion structure is constructed for the three-sensor composite guidance system.Accordingly,the data processing scheme is designed for the information fusion structure.Second,based on the designed information fusion scheme,the key algorithms of the information fusion center are studied in detail,including the algorithms for time-space registration,track association and track fusion.And the corresponding algorithm design is provided.The effectiveness of the algorithms can be verified with simulation experiments.Due to the time and space deviations of the acquired data in different sensors,the calibration is necessary for the subsequent processing.The time-space registration can achieve the unification of data in both time and space domain.The space registration is achieved by the translation and rotation of the coordinate system.With respect to the time registration,the interpolation and extrapolation algorithms are utilized to make different state values unified at the same time node.The track association aims to match the target tracks acquired from different sensors by classifying the tracks from the same target as a group.Notice that the accuracy of track association is mainly influenced by the anti-interference capability of the system.After track association,the associated tracks are merged according to the designed track fusion algorithm.Thus the system track is formed.Finally,the data processing hardware platform of the information fusion system,i.e.TMS320C6678,is briefly introduced.Then the whole workflow of the designed information fusion system is provided,with the emphasis on the track association strategy.After any two tracks acquired from different sensors are associated in pairs,the final system track is formed according to the actual association degree of tracks and the designed track fusion algorithm.Besides,the anti-interference capability of the system mainly influence the accuracy of track association.Thus,based on the information fusion theory,the anti-interference scheme is designed for the track association strategy.Furthermore,the anti-interference capability of the information fusion system has been validated by the digital simulation-testing system.