Sunday, December 1, 2013

What is M2M Internet of Things Impact on Mobile Networks?

The eventual creation of many large sensor networks will have some impact on network traffic, but the precise nature of the new traffic load is a reasonable question.

One reasonable assumption most might make is that large networks of sensors will increase demand for mobile network connections, which might be the easiest way to activate and use large networks of sensors, many of which are related to logistics, automobiles and health care.

That will place a premium on mobile or untethered access, though not necessarily huge amounts of bandwidth from any single sensor.

Some networks, though, might combine collection and transmission of sensor data including some video and audio feeds, with obvious implications for the amount of new bandwidth load.

Security networks are one obvious example, although auto cameras or digestible cameras for health care provide other use cases.

Still, any predictions about the eventual volume of new data, or its impact on networks, are conditional. Nobody knows yet precisely how the volume, velocity or variety of data will shape networks.

Clearly there is a volume impact, but unless the traffic is video, it will be the multiplication of devices that is key, and the locations where those devices are used, not the actual bandwidth load that is crucial.

It will matter whether most transmitted data consists of web logs, RFID sensor readings,
unstructured social networking data, streamed video or audio.

Sampling frequency (how often the data is generated) will affect transmission load as well.  Real-time data need to be processed real-time, with implications for total traffic demand.

But signaling network activity might have the more notable impact. The sheer number of additional devices, requiring relatively little bandwidth but transmitting at discontinuous times, will likely stress the signaling network more than the bearer traffic network.

No comments:

Many Winners and Losers from Generative AI

Perhaps there is no contradiction between low historical total factor annual productivity gains and high expected generative artificial inte...