Cross-platform networking & information exchange in embedded systems

Author : Nick Pridham, Managing Director of Hamersham

05 June 2018

Middleware communications are not new, yet system complexity issues persist. Including a cover piece on ROV wireless communication, this piece outlines the benefits of using DDS technology to enable efficient P2P communication in many embedded applications.

For the digital issue of this piece, please visit this link – or click here to register for EPDT's magazine.

Huge complexity differences exist in systems, particularly where ease of deployment, network security and efficiency are concerned.

When the embedded environment is being considered, network resources are critical to the operation and performance of software systems. Mismanaged software system resources can lead to poor stability, reduced performance and increased deployment costs. Scalability is also impacted, as it becomes harder to add new features, users or connections to an existing system that is close to its memory capacity.

Gaining insight into memory usage is key to enabling architects and engineers to develop and deploy embedded systems: a factor which is particularly challenging when commercial software is used. The question arises: how do you control the CPU utilisation, or examine the resource utilisation of a ‘black box’ component?

Often, a variety of technologies, operating systems and programming languages are used, and integrating them becomes a complex task – depending on the size of the different software systems, the number of devices connected, and the distance over which they need to communicate.

Increasingly, more real-world, complex distributed systems are turning to Data Distribution Service (DDS) middleware technology to achieve a data-centric, publish-subscribe approach to their digital communication strategy.

Flexibility across the board

DDS is a standardised communication middleware technology that offers flexibility and configurability when supporting large-scale, high-performance, constrained-bandwidth systems on embedded platforms. It is suitable for a wide range of uses, including embedded applications in defence, space, medical, consumer electronics and energy, to name but a few.

Despite not being as widely deployed as some middleware systems, such as JMS (Java Message Service), its configurability is achieved through a powerful QoS (Quality of Service) feature set.

DDS’s purpose is to simplify the design, programming and managing of software applications by streamlining the way they receive and process data. Suitable for use in a wide range of embedded software systems, it can connect disparate system configurations (including chip architectures such as ARM, X86 and PPC) on operating systems such as Windows, Linux and QNX.

A wide range of transports for DDS are available including UDP (default), TCP, Serial and Zigbee. DDS has also been successfully bridged to other protocols, such as CAN. This means that DDS can truly be the communication ‘glue’ that holds diverse systems together.

A wide number of embedded systems configurations are currently in use that, thanks to DDS technology, can all communicate with one another in real time, regardless of whether their programming language is Java, C, C++ or C#. This is also regardless of whether it is operating at the edge or in the cloud.

In summary, a very wide range of embedded devices can be configured to support DDS, enabling system designers to enjoy a truly integrated approach to their digital data communication strategy across all device types.

How DDS works

In essence, DDS facilitates the transfer of information between publishers (producers and senders of messages) to subscribers, or those who read messages, again, no matter which operating platform or system is used.

When a publisher wants to share information, it is shared via a data writer. Recipients subscribe to information through a data reader, and the writer and the reader exchange information on a topic. The discovery of publishers and subscribers is automatic, meaning IP addresses and port number configuration are unnecessary.

DDS automatically ‘knows’ how to send and receive messages between users, concluding who should receive the relevant messages, where they are located, and how to proceed if the receiver is unavailable. This simplifies data distribution, reducing the code required to deliver messages – which in turn makes data delivery faster and more efficient.

Neither a message broker nor a cloud server is needed for information to be exchanged, reducing service costs. This is regardless of the scale in which such exchanges can take part. Tens of thousands of devices can be connected at the same time without affecting either the reliability or speed at which messages are sent and received.

Swift and secure

Security is obviously a key consideration, whether the application be in aerospace and defence technology or in consumer electronics. Leading DDS implementations such as CoreDX DDS, delivered by Hamersham and produced by partner company Twin Oaks Computing, provide an end-to-end, state-of-the-art security solution that meets the requirements of all military, Industrial IoT, healthcare and financial systems.

The major components of the security system concern authorisation, permissions and encryption. This includes identification checks for a system participant: controlling what is allowed to be read and written, and introducing an algorithm controlled by the system designer to encrypt the data stream in question.

Crucially, the CoreDX DDS implementation boasts an industry leading small code footprint of just 500KB, enabling it to be run on deeply embedded devices with limited flash memory. A typical application requires less than 250KB of RAM, making it highly desirable for use in consumer electronics, where limited resource microprocessors are deployed.

While the performance of most DDS implementations is positive, particularly when compared with other XML-based middleware technology, CoreDX DDS takes this to the next level. For example, when operating on a 1000Mbps network, throughputs of over 900 Mbps within a maximum stack timeframe latency of 60 microseconds can be achieved. Whether devices are located at the edge or the cloud, CoreDx DDS will process data onto the wire extremely quickly.

Case study: networking underwater

For many years, the use of remotely operated vehicles (ROVs) underwater has been associated with both substantial expenditure and considerable hazards. Divers and physically-connected robots have often had to be used.

The objective of the SWARMs (Smart and Networking Underwater Robots in Co-operation Meshes) project was to develop a group of connected Autonomous Underwater Vehicles (AUVs) that could communicate wirelessly with each other and with surface vehicles – in order to plan and execute maritime and offshore operations.

Activities such as sub-sea mapping, asset inspection and pollution monitoring can be carried out more accurately and efficiently through teams of AUVs, rather than using multiple single units. This is also a more cost-effective use of AUVs, as costly sensors need only be fitted to one unit, rather than all those involved.

Wireless communication between vehicles is essential, to enable each AUV to share the information. However, subsea conditions are very different from a land environment; therefore, the most effective method of information exchange uses acoustic modems.

Incorporating CoreDX DDS communications middleware into the design enabled engineers to develop intelligent, integrated embedded platforms, with the result that the underwater robots can communicate and share functionalities.

Tests were carried out off Gran Canaria, where it was found that the system could handle turbulent sub-sea conditions reliably while retaining its powerful networking functionality through its publish-and-subscribe methodology. The small footprint of the CoreDX application allowed it to be used on a variety of AUVs, even those with limited memory capacity.

The SWARMs project will reduce operational costs and improve safety, through allowing subsea projects to be controlled remotely, from the workplace – ultimately making autonomous maritime and offshore operations a viable option for new and existing markets.


Contact Details and Archive...

Print this page | E-mail this page