Worldwide communication and networking
© AdobeStock - Victoria
Worldwide communication and networking

Communication and Networking

Communication interfaces on embedded systems with AI allow data exchange with other devices and connection to networks

Communication and Networking

In most cases Embedded Systems are connected devices and many of them even have several interfaces for communication with external units and for general networking. Many embedded systems are components of a larger network and fulfill their actual function only through their networking. The networking of a large number of units creates the so-called "Internet of Things" (IoT), in industrial applications it is called "Industrial Internet of Things" (IIoT). Examples for networked systems are distributed smart sensors which collect process data in industrial plants and send it to the cloud for further evaluation, for example for condition monitoring or predictive maintenance.

Another example are sensor/actuator components in smart home systems or telemedicine devices that monitor patients at home as part of aftercare and provide vital data to the attending physician.

Especially for embedded systems with embedded AI, communication is essential for connecting external data sources, e.g. sensors, and likewise for distributed learning ,where training of complex tasks is performed on multiple small embedded systems.

Further communication requirements may exist for the connection to external operating or configuration units and for the installation of software updates.

Numerous technologies and standards are available for the realization of communication interfaces for the networking of devices which differ widely in their technical data and are individually established in their areas of application.

When selecting a suitable communication interface for a new product, a number of aspects must be considered which can be clarified by answering the following questions:

  • Should the networking be wired or wireless?
  • What data rate (bit/s) must be transmitted or received?
  • What distance must be bridged?
  • How much energy is available for the data communication?
  • Has a standard been established in the application?
  • Is the communication predetermined by an already existing network?
  • Are there special requirements for transmission security?
  • Are there any costs for using existing networks?
  • Is the available computing power sufficient for a complex communication protocol?

Fraunhofer IMS can competently support its R&D partners in clarifying these questions. There are experiences from numerous development projects from the most different fields of application and the associated networks as well as the associated interfaces and communication protocols:

  • Smart Home:  LON, Konnex (wired / wireless / Powerline), LAN, WLAN, ZigBee, EIB, BACnet, EEBus
  • Industry:        CAN, LAN, WLAN, LoRaWAN, NB-IoT, IOLink, IOLink Wireless
  • Consumer:     LAN, WLAN, USB, Bluetooth
  • Automotive:  CAN, FlexRay, LIN, Ethernet

A special type of communication between a sensor and the evaluation unit is RFID technology in which not only data exchange but also energy transfer to the sensor is possible. For this purpose, there are several technology variants that can be used depending on the respective requirements for range, power consumption, media penetration and data rate.

Of course, Fraunhofer IMS is also happy to develop an individual proprietary solution for communication and networking if none of the established standards is suitable for the solution of the task at hand.

Our technologies - Innovations for your products

Sensor Networking for Condition Monitoring

Depending on the specific application (industry, buildings, infrastructure), sensors are networked with each other and their data is sent wirelessly or via cable to a central evaluation unit.

Sensor Connection for Teleservices

Depending on the application area of the teleservice (medicine, trade, logistics) and the respective design, sensors are connected to an information infrastructure either wirelessly or by cable.

Our technology areas - Our technologies for your development

User Interfaces

User interfaces as interface between device and user allow the configuration and operation of a product

Machine Learning for Embedded Systems

Artificial intelligence on resource-limited systems can be used to extract higher quality information from raw sensor data

Computer Vision

Computer Vision methods extract the maximum amount of information from image data

 

Embedded Software and Artificial Intelligence (Home)

Here you can get back to the overview page of the core competence Embedded Software and Artificial Intelligence (ESA).