Technology, Risks and Outlook
Analytics, machine learning and edge computing for device data.
Start here · Essential guide
How artificial intelligence turns streams of raw sensor readings into decisions, with plain examples from homes, factories, farms and health, and the limits.
How unusual sensor readings are flagged, from simple thresholds and averages to learned models, plus why false alarms and bad sensors complicate the job.
A forward look at learning models running on small devices and in the cloud, what may change, and the open challenges around updates, power, bias and trust.
Edge versus cloud processing for connected devices, compared on latency, bandwidth, privacy and cost, with guidance on when each approach makes sense.
How streams of sensor readings are collected, stored, cleaned and analyzed to find patterns, from simple summaries to trend and comparison methods.
A plain comparison of wireless connection types for IoT by range, power use, data rate and typical uses, plus a simple way to match a link to a job.
Related topics