LARRA

LARRA (LAm Receiver Redundancy Analyzer) is tool for analysis of multilateration coverage, probability of detection and it also simulates degraded states of the system.

Introduction

The LAM (Local Area Multilateration) system needs specific number of receiving stations. The number of stations depends on the covered area, system design and budget of the operator. From the multilateration principle at least 4 stations are necessary, however, small LAM system typically consist of 8 - 20 stations, large system cosists of tens (40 - 60) of stations. Typical system works well in operational, non-degraded state and acceptable in defined degraded state. The degraded state is defined by the manufacturer, typically N-1, N-2. It means the system is still operational with one or two non-functional stations (or more if defined so), the accuracy or probability of detection is reduced but still meets specification.

In order to understand the performance of the LAM system installed at the Prague airport the LARRA analyzer was developed. LARRA analyzes LAM performance in a mesh grid with cell resolution of 50 m over the whole airport. Available metrics calculated by LARRA:

• total number of detections by the whole system
• number of contributing recivers in each cell
• total number of detections by each station
• probability of detection of each station
• degradations simulator - simulator of N-X state

Data used in the analysis

LARRA uses ASTERIX data in Cat020. The item I020/400 in ASTERIX record contains information about receivers contributing to each detection. For the purpose of the analysis the are conditioned:

• Vehicles are filtered out, only aircrafts remain in processing (based on offline ICAO address database).
• Manually selected ICAO adresses are excluded from processing (site monitors, unknown vehicles, test transponders, interrogators).
• Target reports with GSB=1 in ASTERIX I020/020 remain in processing, airborne targets are filtered out.
• Target reports with RAB=0 in ASTERIX I020/020 are filtered out (site monitors).
• Target reports with CNF=0 in ASTERIX I020/170 remain in processing. Unconfirmed reports are filtered out.
• Target reports coming from outer area of the airport are filtered out.

Data processing

There are three stages of data:

1. ASTERIX recoreded data in binary format, raw data recorded by target processsor, hourly rate.
2. Intermediate data that contain: position X, Y, contributing receivers. For future use measured height, geometric height and DOPs from I020/500 are stored as well. Intermediate data are pickled on daily basis.
3. Processed data, stored on daily basis (one day, one file). In each file two matrix object are stored: a. Total number of detections for each square of mesh grid by the whole system. Size of the matrix: (num_cells_x, num_cells_y, 1), 2D matrix. b. Number of detections for each square by each station. Size of matrix: (num_cells_x, num_cells_y, num_of_receiving_stations), 3D matrix.

LARRA dashboard allows to create dataset for given range of dates (eg. to study performance during isolated degraded states). When creating dataset, daily detections by system and stations are merged over the desired period. Output is two files, first contains total number of detections over the whole period, second contains number of detections for each cell by each station accumulated over the whole period.

Probability of detection for receiver

Dashboard

All metrics and analyses are presented on the web gui interface. Let’s go through all possibilities of the web interface.

Preparation

Allows to create a dataset. It is possible to create datasets for separate Target Processors (if data are available) and for given date range. Date range is selectable on two click in double calendar, as default the maximum available data range is prefilled. Already available datasets are displayed in table next to the calendar.

Statistics - Total number of detections

Shows map of the airport with colored layer indicating number of detection in each cell by the whole system. The number is displayed in cursor label on hover over the cell. Total number of detections by each station are also shown in the table. This statistics is useful for identification of low-traffic cells.

Statistics - Number of contributing receivers

For selected dataset the map with colored layer is displayed. Color of the cell corresponds to the number of receivers contributing to position detection. For each cell the number of receivers contributing to position detection is presented also in cursor label. It is possible to click in the figure (which selects the cell under examination) and contributing receivers for selected cell are then added to the map. In orded to decide if receiver contributes or not it is necessary to select desired probability of detection. The probablity of detection is selectable in the range from 0 to 100 %.

Statistics - Number of detections by receivers

Shows map of the airport with colored layer indicating number of detection in each cell by each station - selectable in the radio list. The number of detections is displayed in cursor label on hover over the cell. This statistics is useful for identification of low traffic cells.

Coverage

Shows the coverage of each recieving station. Probability of detection is color-coded in the map. This is useful for identification of cells with poor coverage.

Degradations

Degradation is the main view for analysis of degraded states. In non-degraded state the are no cells with lost coverage of reduced coverage, it is considered as the baseline for other states.

On the left side it is possible un-check stations and “simulate” station’s switch-off. With a station off (degraded or lost for some reason) some cells with lost coverage appear - highlighted by red dot. If “show reduced” option is checked, cells with reduced coverage are highlighted by orange dot.

Lost coverage: there was the coverage by at least 4 stations but now only 3 or less stations can contribute to detection. Reduced coverage: there was the baseline coverage but now less stations can contribute to the cell but it is still more than 4 stations.

When several stations are off, more lost-coverage-cells appear.

It is possible to highlight cells with degraded coverage. In the next figure lost cells are marked by red dot, cells with reduced coverage are highlighted by orange dot.

Since it is quite challenging to recognize high traffic cells just by looking at the map, the option to highlight cells “important” for the analysis is implemented. For our preliminary analysis we need to study converage on runways first. It is possible to highlight cells important for runways by checking “Highlight RWY squares” option. These are cells inside given RWY polygon with significant number of detections.