As I explored in a previous #askIoT post, nearly every IoT system needs some way to connect sensors/devices to the cloud so that data can be sent back-and-forth between them. IoT gateways can be essential in making this connection possible, but what is a gateway?

Gateways act as bridges between sensors/devices and the cloud. Many sensors/devices will “talk” to a gateway and the gateway will then take all that information and “talk” to the cloud (what’s the cloud?).

But why the extra step?

Now that you know what a gateway is, you may be wondering what benefit there is to taking that extra step between the sensors/devices and the cloud. Turns out there are several benefits:

Battery life

If a sensor/device is in a remote area, it may need a long-range connection such as satellite connectivity to talk to the cloud. As explained here, longer range typically means increased power consumption (and costs); this can be a problem for small sensors/devices with limited battery life.

If you’re doing Smart Agriculture, you want your field sensors to last years, not months or weeks. By using an elevated gateway installed near the top of an outbuilding or grain silo, the sensors/devices only have to send data a short distance to the gateway and the gateway can backhaul the data to the cloud through a single higher bandwidth connection.

Gateways allow sensors/devices to communicate over shorter distances, boosting battery life.

Varying protocols

A complete IoT application might involve many different kinds of sensors and devices. Using Smart Agriculture again, you might want sensors for temperature, moisture, and sunlight and devices such as automated irrigation and fertilizer systems.

All of the different sensors and devices can use varying transmission protocols (basically, the rules and format for the information being transmitted). Protocols include LPWAN, Wi-Fi, Bluetooth, and Zigbee, among many others.

Gateways can communicate with sensors/devices over varying protocols and then translate that data into a standard protocol such as MQTT to be sent to the cloud.

Unfiltered data

Sometimes, sensors/devices can generate so much data that it’s overwhelming to the system or extremely costly to transmit and store. Often in such cases, only a small fraction of the data is actually valuable. For example, a security camera doesn’t need to send video data of an empty hallway.

Gateways can pre-process and filter the data being generated by sensors/devices to decrease transmission, processing, and storage requirements.

High latency

In last week’s #askIoT post, I explained that time can be critical for certain IoT applications; the sensors/devices can’t afford to transmit data to the cloud and wait to get a response before taking action. This is true for life-or-death situations in the medical realm or for fast-moving objects like cars.

Higher latency can be avoided by processing the data on the gateway and giving commands locally. However, many sensors/devices in IoT applications are too small and too low-battery to do the processing themselves.

Gateways can reduce latency in time-critical applications by performing processing on the gateway itself rather than in the cloud.

Security

Every sensor/device that is connected to the internet becomes vulnerable to being hacked. Hacked sensors/devices are bad. Not just for the owner, but for everyone else too.

A few weeks ago, malware called Mirai was used to hack and control thousands of IoT devices. This “bot-net” of devices was then used to take down major parts of the internet (more about Mirai).

Gateways reduce the number of sensors/devices connected to the internet because the sensors/devices are only connected to the gateway. However, this makes gateways themselves targets and also the first line of defense. This is why security needs to be a priority for any gateway.