This post is also available in: heעברית (Hebrew)

IoT connected devices such as sensors have a wide range of applications in the military and security realms. The US Defense Advanced Research Projects Agency (DARPA) intends to take the connected electronic devices technologies into the sea with their new Ocean-of-Things (OoT) initiative. 

The OoT forms part of DARPA’s ongoing “mosaic warfare” initiative, which takes individual sensor and fighting platforms as independent “tiles” that the US defense force can build into an on-the-fly fighting ready unit. The idea is that these sensors would contribute to the data that self-piloting drones and other craft can use to navigate to a problem hotspot.

The agency will have to take into account the environmental limitations of current IoT devices and make adjustments.

IoT devices are unique in the way they can interact with their environment and the other IoT devices next to them, forming a smart web. First and foremost, they are electronic devices and have certain limitations due to their construction. They are subject to environmental constraints such as heavy rain or fog, something that will invariably affect them out at sea. By increasing the coverage area, DARPA believes it can account for the missing data from obstructed sensors by using nearby functional units to offer a “big picture” view of the situation.

As it stands, DARPA can invest money in small, inexpensive sensors and opt for a “scattershot” to process data, filling in the incomplete data through interpolation. 

The agency envisions for the OoT a system of floats that have these embedded IoT devices in them. They can then beam real-time updates about weather and other local phenomena to the agency remotely. DARPA can then catalog and analyze the data it receives to gain insight about conditions in that particular area of the sea. 

According to, the floats are made of environmentally friendly materials to lower the pollution footprint of the network. The real challenge DARPA will have is using analytics to uncover insights from incomplete data.