You are hereTopological And Processing Issues
Topological And Processing Issues
This document briefly discusses topological and processing issues in information fusion.
Introduction - Topological Issues - Processing Issues - References
A fusion system can be a very complicated system. It is composed of sources of information, of means of acquisition of this information, of communications for the exchange of information, of intelligence to process the information and to issue information of higher content. The issues involved may be separated in topological and processing issues. Despite the interconnection between both issues in an integrated fusion system design, they can be decoupled from each other in order to facilitate the development of a systematic methodology of analysis and synthesis of a fusion system according to Thomopoulos (1990, 1991). Recent advances in technology and in the modeling of complex systems may render this separation useless or unrealistic (e.g., UML language).
The topological issues address the problem of the spatial distribution of sensors, the communication network between sensors and places of processing and decision-making, the bandwidth and the global architecture. Also at stake are issues for the exchange of information, the availability and reliability of information at the time of the fusion. The cost of acquiring the information may also be relevant to the topological issues. In non-military applications, these issues are partly addressed by the vendors / distributors of information. They are also partly addressed by the customer, given its objectives and constraints, including the financial budget.
The processing issues address the question of how to fuse the data, i.e. select the proper measurements, determine the relevance of the data to the objectives, select the fusion methods and architectures, once the data are available, and according to the specifications issued by the project under concern. There is no specific processing general techniques in data fusion. All mathematical tools may apply.
Hall proposed taxonomy of algorithms for sensor fusion (Hall 1992). The first category of techniques deals with the positional fusion, i.e. the assessment of the state vector from the observations. The second category, identity fusion, seeks to combine data to establish the identity of an entity. The third category includes ancillary techniques to support the processing in the level 1 of the JDL model. This taxonomy is not efficient. Hall himself wrote that positional and identity fusions may occur in a simultaneous or interleaved fashion, using similar algorithms.
In military applications, three stages of processing often appear, which may perform independent of the level of information being fused (DSTO, 1994). Correlation applies a metric to each of the redundant parameters on which association is dependent to measure the degree to which that data is related, or associated, to an entity (e.g. a target track). If these parameters cannot be obtained from the source data then there is no way to fuse that data with the entity. Association combines all of the correlations together and thresholds the result to decide if association exists between the source data and an entity. If they are associated then the combination stage occurs. Combination estimates the new state of an entity. It may use intermediate results from the preceding stages, particularly correlation, by aggregating and then merging the parameters. The multiple values for each redundant parameter are aggregated to form the single new updated value of that parameter. This results in a set of complementary parameters, which are then merged into the one unified representation of that entity.
S. C. A. Thomopoulos. Sensor integration and data fusion. Journal of Robotic Systems, vol. 7, pp. 337-372, 1990.
S. C. A. Thomopoulos. Decision and evidence fusion in sensor integration. In Advances in Control and Dynamic Systems, Ed. C. T. Leondes, vol. 49, part 5, pp. 339-412, Academic Press, 1991.
D. Hall. Mathematical Techniques in Multisensor Data Fusion. Artech House, Boston, London, 1992.
DSTO (Defence Science and Technology Organization) Data Fusion Special Interest Group, Data fusion lexicon. Department of Defence, Australia, 7 p., 21 September 1994.
