DARS: Distributed Amateur Radio Station

NOTE: This subject is in the early stages of development. Expect it to change substantially and often.

• SDR Primer – the basics of Software Defined Radio for Amateurs
• SDR in action – my experience with Software Defined Radio

DARS – Digital Amateur Radio Station – is the name I have given my effort to build an all-digital amateur radio station. The station is to be a distributed computer network, with each component (radios, mics, speakers, keys, computers, etc.) connecting through a digital interface.

Amateur Radio is going digital with software defined radios and digital communication modes. A DARS station should carry that process to its logical conclusion.

There are a number of goals that I have for this project.

• To solve problems I've encountered while building my the station.
• To do something new that I can't do now.
• To advance the state of the art of Amateur Radio.
• To simply have fun working on interesting things.

Let's take a closer look at these things.

I'm relatively new to Amateur Radio, having earned my license in 2016. I like technology, and get pleasure from exploring how things work and what they can do. I want to experience as much of our hobby as possible, which means building a station that will do lots of things. And that, of course, leads to technical problems the have to be overcome.

The complexity of my station seems to increase exponentially as I pursue more bands and modes. The result is a bewildering assortment of wires and adapters, CW keys and mics and computers. Making it all work together is a significant challenge. Life would be sweeter if I could simplify the station so as to eliminate some of the problems I face.

I came to radio from the world of computers, which benefits immensely from the use of standard protocols and interfaces. I was deeply disappointed to learn that no such thing applies to Amateur Radio equipment.

Incompatible equipment interfaces abound, in no small part stemming from brand lock-in promoted by the manufacturers. Icom gear won't work with Yaesu, Yaesu won't work with Kenwood, and so on. I want to get away from this.

• Ham software is heavily MS Windows-centric, and I'm a Linux user who avoids Windows as far as possible. I want to whatever I can to promote interoperability.

The challenge here is setting up a station that that:

• Is implemented as a distributed computer network, where the network is at the core and all the station components are connected at the edges.
• Follows the general model of cloud computing, such that:
  • RF components (radios, amplifiers, antennas) are exposed as services that can be accessed over the network.
  • Control stations are network clients that use the remote services.

This involves development of appropriate software and interfaces needed to make it work.

There are many possibilities for things to do here.

• Be able to configure and reconfigure the station through software, without rewiring.
• Develop open protocols for data transfer between station components.
• Develop open APIs (Application Programming Interfaces) for connecting and using components.
• Configure open network interfaces for station hardware.
• Employ wireless capability to give the station Internet-of-Things-style integration.

I get great pleasure and satisfaction from working solutions to problems such as this. That's what hobbies do for us.

The object here is to build a Distributed Amateur Radio Station (DARS). In a conventional analog station, the components that make up the station are connected together by electricity flowing through physical conductors. In a distributed station, the components are connected together by data sent over a computer network. A telegraph key, for example, would have a network interface in place of wires that plug into a rig. So would a microphone, a headset, a computer – and a radio.

DARS is an extrapolation of Software Defined Radio (SDR) to produce a software defined station. Until recently, computers have effectively been peripherals of the radio, processing signals for digital modes and rig control, which are fed to the rig. With SDR, that relationship is reversed: the computer performs all the radio signal-processing functions, while the rig is effectively a transducer that sends and receives RF radiation on one side and digital data on the other. In short, the computer becomes the radio, with the rig as one of its peripheral devices.

The reasoning behind DARS is:

1. In the not-too-distant future, all new amateur radio sets will be software-defined; analog design will go the way of vacuum tubes. (This trend has already begun.)
2. By its nature, software can be loosely coupled. Whereas analog circuit components (stages) are tightly coupled by electrical currents of specific magnitude and frequency, software componants are coupled through digital data passed via a communications medium.
3. Communications media can take many forms. One thing they all have in common is that they can work over variable distances, ranging from nanometers to kilometers.
4. By using suitable communications media with the necessary speed and capacity, the components of a software defined radio can be located at a distance from each other. That is, components can be distributed across a communications network.
5. Distribution means that, given appropriate protocols and interfaces, the components of a radio can be located wherever it's convenient to do so, and can be assembled via software into a working radio.
6. The actual physical form a radio takes will be dictated mostly by the circumstances in which it is to be used – portable, base, etc. The “circuits-in-a-box” paradigm becomes optional.
7. The concept can be extended to encompass an entire Amateur radio station.
8. In this scenario, a station would be a networked computer system with analog peripherals working together virtually through data transfer, rather than a collection of electronic components connected together physically by electrical conductors.

One way to think about this is that it is the architecture of a distributed computing system applied to an Amateur radio station. The system could be fully self-contained, open to the Internet, or some combination of the two.

Of course, a distributed station still must have hardware: RF components (transmitters, receivers, antennas), and user interfaces (keys, mics, computers, speakers). But its heart is composed of software running on computers.

The big idea here is that the pieces of a distributed station are connected together not by electrical voltages, but rather by digital data. The essential requirement of systems connected by digital data is that both systems understand the form and format of the data being conveyed between them. That means some form of data standards that both systems can apply independently in order to communicate.

Think of cell phones. They are, at base, radio transceivers that convey digital data. The cellular network works because phones and towers are programmed to use standard protocols. That lets you buy a new phone and immediately start using the network. The Internet works the same way, with an unlimited number of unique systems bound together by standard protocols.

No such standards exist for Amateur Radio equipment – as anyone who has tried to connect a rig to a computer can testify. That means a system of standards would have to be developed. The object would be to enable station components to communicate with each other regardless of make and model.

To my mind, the possibilities of such a system are boundless. Think of some of that already are being realized:

• FlexRadio, Apache Labs, and Expert Electronics offer transceivers that have no built-in controls. They act as radio servers on a network, with user control via software running on a client machine connected to the network. All three, along with Elecraft, offer transceivers that have built-in controls, but also may be operated over a network by software.
• RemoteRig makes a system that connects a radio's control head to its base units via a TCP/IP network, making it possible to locate them any distance apart.
• USB SDR dongles (such as RTL-SDR) take RF from an antenna, digitize it, and place it on the USB bus of a computer for additional processsing – which may include forwarding to a TCP/IP port.
• WebSDR is a Software-Defined Radio receiver connected to the Internet, allowing many listeners to listen and tune it simultaneously.
• Some external speakers incorporate digital signal processing (DSP) in the form of audio-frequency filters that operate independently of the radio.

Flex, Expert, and Elecraft provide proprietary client software, with varying support for Windows, Linux, Mac, Android, and iOS devices. All but Expert offer public APIs that can be used to develop custom third-party client software. All this contributes to the radios being part of a distributed network – though the lack of API standards substantially complicates things.

The SDR dongles and online SDR receivers also illustrate the potential of a distributed station, in that they can be combined with any external system that can handle the data they produce (a computer, in this case).

The external speakers with DSP are somewhat less clear-cut. Viewing the DSP section as three blocks – analog-to digital coverter, digital processor, digital-to-analog converter – reveals its distributed nature. Those blocks could be implemented independently and be physically separated, connected via data transfer over a network. Here is an example.

• A receiver could convert analog audio to digital and output that data directly to the network. (An SDR receiver could output digital audio directly.)
• A DSP unit could receive the data and process it, placing the resulting data on the network.
• A loudspeaker with a digital interface could receive the processed data, convert it to analog, and send it to its drivers to produce sound waves.

The essential aspect of DARS is that it separates station operator functions from the hardware that makes up a station. That is, station hardware is controlled not by manipulating direct physical controls (knobs, buttons) that are part of the electrical circuit, but indirectly, through computer software. A second aspect is that some of the analog hardware functions can be performed by software instead – not just its controls.

The station operator has these primary functions:

• Information transfer: what we send and receive over the air when making contacts.
• Contact control: how the information is sent and received during a contact.
• Station operation: setting up and running the station.

[NOTE: Partial entry; still under development]