ISSNIP

Information Processing: ISSNIP Research Themes

Background

Information processing encompasses the data related processing that aims to enhance and interpret the collected data and maximize the efficiency of the system, through signal conditioning, data reduction, event detection and decision making. This may involve a collection of filtering and other data manipulation techniques together with advanced learning techniques for feature extraction and classification in order to provide the most relevant data in an efficient representation to the communications interface.

The development of new techniques is required to process the large volumes of information produced by sensor networks and adaptively implement the necessary response.

This may encompass:
  1. Sensor scheduling
  2. Decision theory
  3. Feedback theory
  4. State estimation
  5. Tailored supervisory control systems control systems
  6. Optimal sensor location
  7. Pattern recognition
  8. Data mining
  9. Network flow control
  10. Multi-resolution data transmission integrated with data fusion and reconstruction.
  11. Mathematical and hybrid system tools for monitoring distributed networks of large arrays of sensors and actuators
  12. Statistical modelling for sensor networks

Challenges

Machine learning research provides some of the most the effective and efficient techniques for information processing. Along with neural network technology, machine learning offers the potential for online adaptive processing of immense volumes of information without explicit knowledge of the underlying data. Supervised and unsupervised learning techniques continue to improve in these tasks. Further optimizations are required to improve operational efficiency. Tools such as support vector machines can offer potential alternatives in this regard. These areas are well established in respective research communities, yet still attract considerable attention due to significant potential for improvement.

Applications

Surveillance and Monitoring: Surveillance and monitoring is an application which places an ever increasing demand on the advancement and development of new information processing techniques.

Data analysis: The operation of surveillance and monitoring systems implies the detection and tracking of an event or target. Based on the sensor type and resulting data output, appropriate algorithms and processing techniques are required to extract this information in real time, in order for other systems to be able to react in some way, either by providing and alert to a sub-system or modifying the behaviour of the surveillance system, such as tracking a given target following positive identification.
Data reduction: With the continual introduction of new sensor technologies and the influx of information previously unavailable, considerable processing of the gathered data is required to enable efficient dissemination of the data.

Data interpretation:
In order to obtain meaningful interpretations of the large volumes and various forms of data obtained from various sensor type and networks, additional processing of the information is required.

top of page