Attch_2_SOO_FINAL_CALL_001_dtd_20181101.docx

Statement of Objectives (SOO)

Topic 3 a) A: Solar, Solar Wind and Ionospheric Effects: Oblique Ionogram Data Scaling and Data Quality Control Tool Dated 01 Nov 2018

1.0 Scope

To more efficiently incorporate ionogram data into regional ionospheric models, with the rationale of improving said models. In terms of the current capability to assimilate ionogram data, regional ionospheric models can only automatically ingest vertical ionogram data. Currently, the only means of incorporating oblique ionogram data is by manually extracting ionogram features from the data set. The ability to incorporate quality data from a variety of sources at a near real-time interval will enable models to create the best possible picture of the present ionospheric environment. The scope of this call is to create a process which automates the extraction of these oblique ionogram features, as well as an interface that allows for a person-in-the-loop to perform quality control on flagged data. AFRL would like this tool to do the following: (1) scale OI for data and flag difficult-to-scale OI (2) provide a graphic user interface (GUI) for a user to perform in-the-loop quality control on flagged ionogram data; (3) develop a quality metric from the automated tool that can be used to assess whether data should be excluded; (4) assess the quality of the tool’s outputs against a regional ionospheric model.

2.0 Performance Objectives

2.1 Overall Objective:

As part of the research mission requirements of AFRL/RVBX, there is a demand to ingest real-time data into regional ionospheric models that perform ionospheric inversions. These regional models inform our overall picture of the space environment, and ultimately possible impacts to Air Force systems. The ability to ingest a wider variety of data types into ionospheric models will improve the model’s performance, and provide better situational awareness of the space environment. While the models are able to automatically ingest a variety of data types, the ability to extract data from Oblique Ionograms (OI) is currently a purely manual process requiring an operator to detect and select ionogram features to be ingested into ionospheric models. Oblique ionograms are a data source of interest, as OIs are another source that can provide insight into the bottomside structures forming in the ionosphere. Under ideal ionospheric conditions and in situations where the signal to noise ratio is 30 dB or more, it may be possible to extract data from OIs automatically. For these cases, AFRL would like to develop a process that can detect ionogram features to be ingested into ionospheric models. However, ionogram quality often degrades due to the common occurrence of ionospheric disturbances such as D region absorption, sporadic E layers (which can include blanketing E), and spread F. Such disturbances can often result in weak ionogram traces or spread traces that may make automatic algorithms less than robust. In the harshest cases, ionospheric disturbances can render an ionogram unusable. Under these circumstances, we wish the operator to be notified for potential intervention and be given the ability to flag or withhold both vertical and oblique ionogram data from ionospheric model ingestion. AFRL/RVBX is interested in developing a tool that scales OI features, but allows for quality control on ionograms that the automatic process flags as difficult.

2.2 Technical Objectives:

a) Automated Extraction Process and Data Quality Control

· Develop an algorithm that can extract features from an Oblique Ionogram (OI) when ionospheric conditions are ideal (ie. no Sporadic E or Blanketing E, D region absorption, Spread F) and when the signal to noise ratio is greater than 30 dB.

· Extract features such as the apparent range, maximum usable frequency, uncertainty in the range and frequency for the E, F1 and F2 regions at the first hop.

· The extracted ranges and MUFs must be within the reported uncertainty of the actual trace.

· The tool will assign a quality metric to both vertical and oblique ionograms which will essentially serve as a confidence factor. The higher the confidence in the data, the better the quality metric. Instances where the tool might have high confidence are cases where the features extracted correlate well with predicted values from ionospheric models or if the data matches well with other validated data points.

· The tool should be able to flag and detect when Sporadic E occurs.

· The tool should be able to track changes in the ionosphere over time and flag data where there appears to be a non-physical change in the ionosphere.

· The tool should be able to extract oblique ionogram features from a global spectrum of sources. For example, the tool should not be constrained to merely evaluating mid-latitude data.

b) Graphic User Interface (GUI) for Person-in-the-Loop Quality Control

· The GUI should allow a user to view flagged data (for both vertical and oblique ionograms) in the case where the automated tool is unable to extract ionogram features such as range and frequency for the E, F1 and F2 regions at the first hop.

· The purpose of this tool would be to serve as a “sanity check” when data is flagged, where the operator would evaluate the confidence metric.

· The purpose of this tool is for the operator to check results when the quality metric assigned by the automated tool is low. The operator can then determine from this whether to include or exclude the data.

c) Quality Assurance of Tool Outputs via Comparison with a Model

· This metric should be able to assess from a statistics standpoint whether the data correlates well with the predictions of an assimilative regional ionospheric model.

· The data assimilated into the model for comparison should already be validated by another means. The model predictions will serve as a means of checking the scaling done by the tool, to make sure the measurements are within bounds.

SOO to CALL 001 FA9453-19-S-0001 1