Telecoms, Datacoms, Wireless, IoT


Accessible machine learning for cellular IoT solutions

27 October 2021 Telecoms, Datacoms, Wireless, IoT

Together with design partner Edge Impulse, Nordic Semiconductor has introduced TinyML support for both the nRF9160 DK (development kit) and the Nordic Thingy:91 multi-sensor cellular IoT prototyping platform. Following the introduction, both Nordic low-power cellular IoT development tools are integrated into Edge Impulse’s ‘Edge Impulse Studio’.

The collaboration allows developers without TinyML programming expertise to quickly get started on fully supported, standalone ‘inferencing’ (applied machine learning (ML) at the device level) example projects based on Nordic’s nRF Connect SDK (software development kit). The example projects can then be run on the nRF9160 DK and Thingy:91.

TinyML is a scaled-down form of ML suitable for Internet of Things (IoT) edge devices such as wireless sensors. When using TinyML, cloud-based ML training and model synthesis is made simpler and faster compared with conventional techniques that rely on server training of increasingly large and complex models. The resultant lightweight models can then run efficiently on the optimised compute resources of the target edge devices.

nRF9160 SiP-based solution

The nRF9160 DK and the Thingy:91 are based on the Nordic nRF9160 low-power System-in-Package (SiP) with integrated LTE-M/NB-IoT modem and GPS and are used for the prototyping and development of low-power cellular IoT products. Edge Impulse Studio is a development platform used to collect data from sensors, train a TinyML model and then deploy that model to the target device.

The nRF9160 SiP is certified for global cellular IoT applications and enables IoT data to be sent over distances of kilometres. In addition to a multimode LTE-M/NB-IoT modem, 1 MB Flash and 256 KB RAM, the nRF9160 features a dedicated Arm Cortex-M33 applications processor – making it an ideal edge device for supporting TinyML applications.

Fast route to TinyML

All that’s needed for developers to get started with TinyML projects is access to Edge Impulse Studio, an nRF9160 DK and some sensors, or a Thingy:91 (which incorporates built-in sensors for motion, impact, air quality and more). Once the nRF9160 development tools are linked to Edge Impulse Studio it is straightforward to build and deploy a ready-to-go binary file of an industrial-grade TinyML algorithm for a low-power cellular IoT application.

To help developers get started with TinyML projects on the nRF91 Series, Edge Impulse has published a couple of blogs with links to tutorials for nRF9160 DK and Thingy:91 (www.edgeimpulse.com).

For more information contact Andrew Hutton, RF Design, +27 21 555 8400, andrew@rfdesign.co.za, www.rfdesign.co.za


Credit(s)



Share this article:
Share via emailShare via LinkedInPrint this page

Further reading:

Connectivity for IoT devices with SoftSIM
Otto Wireless Solutions Telecoms, Datacoms, Wireless, IoT
IoT device vendors are facing more and more complex cellular connectivity issues these days. Manufacturing is often outsourced to third-party factories and finished devices are being distributed to many ...

Read more...
High-power GaN RF amplifier
RFiber Solutions Telecoms, Datacoms, Wireless, IoT
The CHA7618-99F is a three-stage GaN high-power amplifier (HPA) in the frequency band 5,5-18 GHz. This United Monolithic Semiconductors HPA typically provides 10 W of output power allied with 20% of power-added ...

Read more...
ExpressLink modules for secure connectivity to AWS Cloud
RF Design Telecoms, Datacoms, Wireless, IoT
u-blox has announced two new modules designed to enable Amazon Web Services (AWS) cloud services for device and fleet management out of the box: the NORA-W2 AWS IoT ExpressLink Wi-Fi module and the SARA-R5 ...

Read more...
Cellular antenna for 4G/LTE and 5G NR
iCorp Technologies Telecoms, Datacoms, Wireless, IoT
Antenova’s new ‘Affini’ (part number SRFL064) is a flexible printed circuit (FPC) antenna for the LTE, 4G and 5G NR networks in global markets. This antenna covers all main 4G bands plus 617-698 MHz which ...

Read more...
Passive MMIC diplexer
RF Design Telecoms, Datacoms, Wireless, IoT
The MDPX-0710 from Marki Microwave is a passive MMIC diplexer with passbands from DC to 7 GHz (low band) and 10 to 26,5 GHz (high band). It has an insertion loss of less than 0,9 dB within its passbands ...

Read more...
Automotive-grade GNSS signal splitter
RF Design Telecoms, Datacoms, Wireless, IoT
Tallysman Wireless has added the TW162A automotive-grade ‘Smart Power’ GNSS signal splitter to its line of GNSS accessories. The device supports the full GNSS spectrum: GPS/QZSS-L1/L2/L5, QZSS-L6, GLONASS-G1/G2/G3, ...

Read more...
u-blox extends ‘bring your own SIM’ approach to MQTT
RF Design Telecoms, Datacoms, Wireless, IoT
u-blox has introduced a service that gives IoT sensor network developers the flexibility to combine any cellular connectivity with all the benefits of MQTT communication. The u-blox MQTT Flex service ...

Read more...
K-band passivated VCO
RF Design Analogue, Mixed Signal, LSI
The K-VCO-2324 from Arralis is a GaAs PHEMT MMIC voltage-controlled oscillator (VCO) that operates from 23,25 to 24,25 GHz. It provides 15 dBm output power over tuning voltages between 0 V and 1 V and ...

Read more...
GNSS correction service receivers for cm-level accuracy
RF Design Telecoms, Datacoms, Wireless, IoT
For designers who require reliable centimetre-level positioning accuracy, e.g. for the industrial navigation and robotics markets, a new suite of products and feature additions have been rolled out by ...

Read more...
4xMIMO cross-polarised antenna
RF Design Telecoms, Datacoms, Wireless, IoT
The TGX.04 from Taoglas is a 5G/4G 4xMIMO dipole antenna that operates from 450 to 6000 MHz. It supports worldwide sub-6 GHz cellular bands including 5G NR, LTE, NB-IoT, etc. This cross-polarised ...

Read more...