Computing at the Edge of IoT – Google Developers – Medium

https://medium.com/google-developers/computing-at-the-edge-of-iot-140a888007b
We’ve seen that demand for low latency, offline access, and enhanced machine learning capabilities is fueling a move towards decentralization with more powerful computing devices at the edge.

Nevertheless, many distributed applications benefit more from a centralized architecture and the lowest cost hardware powered by MCUs.

Let’s examine how hardware choice and use case requirements factor into different IoT system architectures.

310 Comments

  1. Tomi Engdahl says:

    Automation at the edge

    Read seven unique industry examples and use cases for automation at the edge in this enriched e-book.
    https://red.ht/3dJr6es

    Reply
  2. Tomi Engdahl says:

    Koneoppimista verkon reunalla
    https://etn.fi/index.php/13-news/14243-koneoppimista-verkon-reunalla

    MACHINER LEARNING AT THE EDGE
    https://etn.fi/index.php/tekniset-artikkelit/14242-machiner-learning-at-the-edge

    Many customers fail to assess and demonstrate the benefits AI will bring to their application. To jumpstart applications on the right foot, STMicroelectronics ́ Edge AI Sprint brings a whole support system of experts that can guide developers through the minefields inherent to their application and use case.

    Traditionally, large companies looking to benefit from machine-learning must hire one or more data scientists to collect a massive amount of data for months, clean them, and create AI models. Embedded developers then port the implementation on microcontrollers or use dedicated tools to convert neural networks into optimized code for MCUs.

    When a company wrestles with tight budget constraints, hiring one or more data scientists may be out of the question. Additionally, it may not be possible to outsource the job. Some situations are sensitive, while others require someone to be constantly on staff.

    Even with the right people and all the time in the world, obtaining quality data is still an issue. Despite all the advances in machine learning, getting reliable training samples can be a severe problem. For instance, if an application tries to detect abnormal behaviors, data may be unavailable. Indeed, while many datasets work for classification problems, such as anomaly detection, they’re useless when trying to detect new situations. It is also critical to obtain good quality data, which is far from obvious. When samples aren’t plagued by typos or missing information, recording clean sets and precisely labeling them can demand serious investments.

    NanoEdge AI Studio is a utility that speaks to embedded developers, even to those with no data-science expertise. The magic lies in running the training phase that learns a complex nominal behavior and the inference on the same device. The entire process can thus run on the same STM32 microcontroller.

    Reply
  3. Tomi Engdahl says:

    Powering the Next Generation of Edge-Compute Capable Satellites
    A November 30th Electronic Design-hosted webinar sponsored by Renesas Electronics
    https://www.electronicdesign.com/resources/webinars/webinar/21253838/powering-the-next-generation-of-edgecompute-capable-satellites?sti=ED2&pk=ED2&utm_source=EG+ED++Webcasts&utm_medium=email&utm_campaign=CPS221118158&o_eid=7211D2691390C9R&rdx.identpull=omeda|7211D2691390C9R&oly_enc_id=7211D2691390C9R

    There’s no question that the amount of data that needs to be processed only increases with time and the space industry is no exception. While the current generation of satellites act mostly as relays between ground stations with very little on-board processing, the projected increase in the amount of data in space puts a strain on the already limited uplink and downlink bandwidth. The solution is to make satellites edge-compute capable such that the data transmitted can be limited to more relevant processed data as opposed to raw data.

    Satellites with such capabilities are more of a reality these days thanks to the new FPGAs and ASICs that have come into the space market. The increased compute capability provided by these new processors open many new possibilities for satellites but also comes at the cost of power. These new processors utilize much smaller process nodes that have much lower core voltages and tolerances.

    Reply
  4. Tomi Engdahl says:

    Korona kiihdytti koneoppimista verkon reunalla
    https://etn.fi/index.php/tekniset-artikkelit/14357-korona-kiihdytti-koneoppimista-verkon-reunalla

    Covid-19-pandemialla on ollut oma vaikutuksensa sulautettujen järjestelmien markkinakehityksessä ja se on merkittävästi lisännyt koneoppimiseen liittyvän automaation, ilmaisun ja edge-prosessoinnin suosiota.

    Siinä missä nopeat ethernet-verkot ovat yhä yleisemmin käytössä teollisissa edge-sovelluksissa, reunalaitteessa oleva FPGA-ratkaisu mahdollistaa, että suunnittelijat voivat konfiguroida järjestelmän eri teollisuusverkoille kuten Profibus- tai Hart-verkolle. Verkkoprotokollien joustavasti tapahtuva lisääminen FPGA-suunnitteluun vähentää verkkosolmujen ja yhdyskäytävien tarvitseman koon ja monimutkaisuuden määrää, mikä onkin ollut pääasiallinen suuntaus nyt ja ensi vuonna.

    Koneoppiminen on myös mahdollista toteuttaa laajassa valikoimassa mikro-ohjaimia, joita käytetään esimerkiksi reunalaitteiden ennakoivan huollon ja ylläpidon sovelluksissa.

    Mitä useampia laitteita ollaan yhdistämässä IoT-järjestelmään, tulee suunnittelijoiden ottaa huomioon sovellusten lisääntyvä haavoittuvuus vihamielisille kaappauksille ja se, että edge-solmut ovat yleisesti varsin alttiita turvallisuutta uhkaaville hakkeroinneille. Tästä syystä on ilmeinen tarve langattomasti (OTA) suoritettaville päivityksille, jotta edge-järjestelmän IoT-laitteiden turvallisuusominaisuudet pysyvät ajan tasalla.

    OTA-päivitykset ovat jo nyt pääosin käytössä turvallisuudeltaan kriittisissä järjestelmissä, muutoinhan verkkoon olisi helppo ujuttaa haitallista koodia edge-solmusta käsin. Microchip on yhteistyössä kaikkien merkittävien pilvipalveluiden toimittajien kanssa takaamassa, että noudatetaan viimeisimpiä turvallisuusstandardeja. Tällä toimialalla markkinat kasvavat nopeasti tällä hetkellä.

    Reply
  5. Tomi Engdahl says:

    Explaining an AI Edge-Computing Solution—Plus Live IoT Demos
    Dec. 20, 2022
    The two solution demonstrations by STMicroelectronics, which took place at electronica 2022, focused on AI computing and the new Matter IoT protocol.
    https://www.electronicdesign.com/technologies/iot/video/21256566/electronic-design-explaining-an-ai-edge-computing-solution-and-live-iot-demonstrations?utm_source=EG+ED+Connected+Solutions&utm_medium=email&utm_campaign=CPS221212111&o_eid=7211D2691390C9R&rdx.identpull=omeda|7211D2691390C9R&oly_enc_id=7211D2691390C9R

    Reply
  6. Tomi Engdahl says:

    FPGAs for the Intelligent Edge – Intelligent Embedded Vision
    FPGAs and SoCs are enabling IoT devices to analyze image and video without the help of a central server.
    https://community.element14.com/learn/learning-center/the-tech-connection/w/documents/28303/fpgas-for-the-intelligent-edge—intelligent-embedded-vision

    Reply
  7. Tomi Engdahl says:

    A growing number of companies are turning to edge computing for their organizational needs, with worldwide spending on edge expected to reach $208 billion in 2023, according to IDC. This is an increase of 13.1% over 2022 and IDC has also identified more than 400 named use cases across various industries and domains.

    Reply

Leave a Comment

Your email address will not be published. Required fields are marked *

*

*