Hello readers!
This week, we’re talking about crowd counting using a 5G signal, SGP.32 explained, inventory intelligence, and more!
A 5G network just did a security camera’s job

Every wireless network picks up information about the space it runs through. A person walking between a transmitter and a receiver bends, blocks and scatters the signal, and for most of radio’s history engineers filtered that disturbance out as noise. Now it is becoming a data product. Operators and chipmakers on both the cellular and Wi-Fi sides are teaching radios already installed for connectivity to report on who is moving through their coverage. As a result, a growing share of presence sensing in factories, offices, venues and homes will come from infrastructure nobody bought as a sensor. That changes the economics of occupancy and safety monitoring, because the hardware is already paid for and powered. It also changes the privacy conversation. A network that can sense is harder to spot than a camera, easier to overlook in a procurement review, and governed by rules written either for cameras or for connectivity, with nothing yet written for the space between them.
Turkcell supplied the clearest data point yet last week. Working with Tiami Networks, the operator ran integrated sensing and communication (ISAC) trials on private 5G networks at the Istanbul Expo Center and its Küçükyalı Plaza offices. It reported 90% accuracy in estimating crowd density from C-band 5G signals, with a person’s position resolved to within about 3 meters. No cameras were involved; AI models read changes in live communication signals. The ITU has written ISAC into its IMT-2030 roadmap as a 6G capability, so this is an early look rather than a commercial service. Even so, it ran on today’s 5G radios inside a working building. The same week, Ericsson, Polish operator Play and Asseco demonstrated drone detection over mobile network infrastructure, combining radio-frequency sensing with acoustic detection. Vehbi Çağrı Güngör, Turkcell’s deputy GM for network technologies, presented the camera-free design as a privacy-friendly option. On the narrow question of capturing faces, he has a point.
Wi-Fi got to this idea first, and more cheaply. The IEEE published 802.11bf, the Wi-Fi sensing standard, in September 2025, and the underlying technique of reading channel state information off a router’s subcarriers predates it by years. Hobbyist builds show how low the hardware bar already is. One open-source project uses a roughly $10 ESP32 to detect motion through walls with signal processing alone, reports 98% recall in testing, and has drawn more than 4,000 GitHub stars. It still can’t tell people from pets, count occupants or recognize specific activities. Newer radios should narrow those gaps. Wi-Fi 7’s higher resolution and lower latency are expected to make sensing accurate enough for commercial uses like elder care monitoring, fall detection and energy optimization. Turkcell has worked this side as well, launching a Wi-Fi sensing service with Huawei for real-time presence detection in February. The 5G trials take the same principle from a single room to a whole building. Güngör’s description of future networks as systems that detect their environment suggests operators want to go further still.
A network that never records a face can still build a record of where people go. Smart city privacy concerns have long centered on aggregation: a trail of locations assembled from many individually harmless sensors can reveal a visit to an HIV clinic or a psychiatrist. Radio sensing that counts people and places them within a few meters generates that kind of trail as a byproduct of carrying traffic. Even the ESP32 project’s documentation advises users to get consent and follow laws like GDPR and CCPA, though CSI data is technically anonymous. Consent is harder to arrange when the sensor is the access point everyone already joins, or the private 5G network a venue installed for its staff. Visitors can see a camera on the ceiling. They have no way of knowing whether the network carrying their phone traffic is also counting them, and the disclosure habits the industry built around cameras, such as signage and posted notices, don’t map neatly onto a radio.
📖 Top Articles

A smart meter installed today will probably still be on the wall in 2040. An EV charger, an asset tracker on a shipping container or a sensor on a power line often lives just as long. The mobile contract that connects it rarely does. Prices change, operators merge, networks get switched off, and a deal that made sense on day one can look very different five years later.

Every connectivity vendor with a blog has published the same RedCap post. You know the one: “Goldilocks tier,” a chart comparing megabits per second, a market-size stat pulled from an analyst report, and a promise to “future-proof” your deployment. But there’s a more useful way to look at RedCap.

Warehouses and Distribution Centers have spent years automating the hands and legs of their operations, with conveyors, AS/RS, AMRs and robotic arms moving boxes, pallets and parcels faster than manual labor ever could. Many have also added the eyes, in the form of barcode scanners, RFID, sensors and machine vision that identify items and record where they are.
🔥 Rapid Fire
Northern Tech simplifies connected product security and management with Mender component orchestration
Kroger taps RFID, computer vision data to boost inventory management
IoT sensors and hybrid machine learning model predict solar panel faults
Hackers seize on vulnerable IoT devices to spy, pivot and route traffic
🎙 The IoT For All Podcast
In this episode of the IoT For All Podcast, Alper Yegin, CEO of the LoRa Alliance, joins Ryan Chacon to discuss the state of LoRaWAN in 2026. The conversation covers what's driving LoRaWAN adoption, how LoRaWAN compares to other connectivity options, new LoRaWAN capabilities, LoRaWAN deployments in smart buildings and other verticals, moving from a pilot to large-scale deployment, misconceptions about LoRaWAN, the role of LoRaWAN in physical AI, and the future of LoRaWAN.
📆 Events & Webinars

Hosted by akenza
Pressac Communications and akenza show you how to connect IoT devices to the cloud without writing code, and how to make sense of the data once it’s there.


