The Israeli Ministry of Defense Directorate of Defense Research and Development (DDR&D) is excited to announce the launch of the “MAFAT challenge” - a series of prize competitions in the field of Data Science that are fully open to the general public, academic and industry sectors.
The goal of the challenge is to explore the potential of advanced data science methods to improve and enhance the IMOD current data products. The winning method may eventually be applied to real data and the winners may be invited to further collaborate with the IMOD on future projects
The participant’s goal is to infer how many people are present in a particular room by analyzing the RSSI signal received from various devices.
In this challenge, MAFAT’s DDR&D (Directorate of Defense Research & Development) would like to tackle the challenge of inferring room occupancy, based on machine learning methods applied to WiFi Received Signal Strength Indicator (RSSI) data. Typical WiFi communication runs at 2.4GHz - 6GHz. In these frequency ranges, objects such as the human body constitute a reflector / obstacle for the electromagnetic wave. Hence, the presence of people and their movement can affect the medium, resulting in changes to the received signal in the presence of a person.
The participants’ goal is to detect the presence of people in a certain room during various time frames. The solution will be based on the analysis of the RSSI signals as they were received from different devices. In this task, the input will be the RSSI value and the output should describe how many people are present in the room: 0 (empty room), 1, 2, or 3 people.
The competition has two different tracks, both using the same dataset and resources.
Track 1: Participant’s goal is to infer whether a room is empty (0) or occupied (1). The results in this track should be a continuous number between 0 and 1 that describes the model’s confidence that the room is empty or occupied.
Track 2: Participant’s goal is to classify how many people are in the room, this is a discrete regression task. Predictions could be empty room (0) , 1, 2 or 3 people.
The data is RSSI signals that were received from different devices: smartphones, tablets, laptops, etc. All the devices were connected to a standard commercial off-the-shelf router and the RSSI signals were recorded from it. The router was located in a room and all of the endpoints (devices) were placed in its surroundings: that is in the same room or adjacent rooms. The data was collected throughout the time the devices were connected to the router, at different times and sampling rates.
Fine grained classification of objects in high resolution aerial imagery. ($30,000)
Can you distinguish between humans and animals in radar tracks? ($40,000)
Future competition - stay tuned for updates
Future competition - stay tuned for updates
Future competition - stay tuned for updates
Future competition - stay tuned for updates
