Conveners
Session 3: Sigma Data and Algorithm Studies
- Neil Gaspar (University of Surrey)
Presentation materials
We will present a summary of current events and opportunities from NTRnet.
The SIGMA data challenge provided the academic community with unprecedented access to 1.5 billion gamma spectra collected in London, made available through NuSec. This extensive dataset enabled research into both detector performance and advanced machine learning methodologies. Multiple academic teams have engaged with the data, with the Surrey campaign supporting a postdoctoral fellowship,...
Detecting the presence of particular radioactive isotopes present for small periods of time in large time series datasets is useful in a number of nuclear security problems. This is a challenging computational task because the number of intervals in a signal quickly becomes large. To tackle it I will combine two mathematical approaches. First, I develop a multivariate likelihood ratio testing...
Sensor networks continue to define measurements across the field of environmental monitoring, including that of radiation detection. NuSec and AWE's SIGMA Data Challenge provides access to measurements of gamma-ray activity recorded by a sensor network of around 100 detectors distributed across central London. We have analysed data from three detectors centred around St Thomas' Hospital and...
AWE is developing a spectroscopic RN detection algorithm GROUSE validated on down-sampled SIGMA data with injected simulated threats and simulated threat templates. Our top-level approach is described and detail provided of our anomaly detection in Poisson stats approach (POODLE) which feeds spectroscopic data to GROUSE, as well as the inject data, and templates. Initial performance is also...
The widespread availability of nuclear and radioactive materials, commonly used in industrial and medical applications, poses a significant risk of misuse in the form of radiological dispersal devices (RDDs) or "dirty bombs." If detonated in densely populated or strategically important locations, such devices could cause widespread panic and necessitate large-scale evacuation and cleanup...
SIGMA data 2. The original SIGMA dataset, made available around two years ago, while both useful and relevant, contains real data with a limited number of threats and lacks ground truth. The priority for iteration of the dataset was identified as injected threats and AWE is in the process of making this happen. Plans and timescales are outlined to provide a hybrid dataset (real background,...
