Do you truly understand the SDR concept?

SDR Concept articleA lot of magazine and articles have already covered the principles of software defined radio. Most of them contain mathematical equations which require a greater level of math. Certainly, without math there would be no SDR, but I believe, that the concept of SDR can also be explained without. Let me show you the SDR concept from a different point of view!

Fundamentals: The three radio components

Independent of what kind of radio we are talking, a radio can always be broken down into three fundamental components:

  1. RF Part
  2. Signal Processing
  3. User Interface

the three radio components, RF, Signal Processing and User Interface

All three components interact with each other.

RF component iconThe RF component transforms high frequency signals into baseband signals or vice versa. Parts of the RF component are filters, mixers, amplifiers, oscillators, etc.

Signal Processing IconThe baseband signal is processed / generated in the Signal Processing component. In this component we will find functionality taking charge of modulation, demodulation, NF filtering, noise reduction, etc.

UI component IconThe User Interface (UI) is the command and control unit of the radio. It’s the frontend to the (human) operator. The user interface might consist of buttons, knobs, LEDs or just a PC screen.

The classic radio

The classic radio is the kind of transceiver which we have in our shacks. It might be a FT1000MP, an IC706 or a Drake TR-7, just to mention a few of them. Over the last decades, the features and user interfaces have slightly changed, but the concept has remained the same: All-in-one-box.

All components are integrated in a single housing. With the increasing level of integration (and use of digital technologies) the physical dependency between the components increased as well. Even if some transceivers come with detachable control units (user interfaces), they are not interoperable with interfaces from other vendors. The schematic below illustrates the dependencies which come with this “All-in-one-box” concept of our today’s transceivers.

schematic of a classic radio

The SDR approach

As the classic radio, the software defined radio consists also of the three fundamental components RF, Signal Processing and User Interface. However, the tremendous difference is the physical and logical independency between the three components.

SDR principal concept of independent components

Physical Independency means that each of the three components can be realized in an individual housing. There is no reason why the components would need to be located at same site. Geographically they could be easily separated by 10.000 miles or more without any degradation in performance.

Logical Independency means that each particular component can be designed and changed without affecting the other components. Especially the digital domain allows us to have various independent instances of the same component.

It must be admitted that the enabler of the Software Defined Radio concept are standardized interfaces and protocols (USB, Firewire,TCP/IP..etc) as well as our speedy PCs with Internet connection. Do you remember that just a decade ago we were dreaming about having all that available once a day?

Maybe due to the unavailability of these standards and technologies commercial radio manufactures couldn’t produce what we were dreaming about 10 years ago – but on the other hand I suspect that they never had (and still don’t have) the interest to adopt to common (open) standards. The best examples are the various transceiver interfaces (CAT) and microphone jacks.

Enough theory – let’s have a look onto some real world examples how the SDR concept is already applied to!

Softrock with PC

Softrock with QSD

At the moment, this is worldwide the most common SDR setup. The antenna is connected directly to a Quadrature Sampling Detector (QSD) like the famous Softrock. The QSD mixes the RF signal into a baseband signal. The complex baseband (I/Q) signal is fed into the soundcard of the PC where the signal is digitalized. Signal Processing is performed by the PC. The user selects through the user Interface (here: Keyboard, Mouse and Screen) the settings like frequency, mode, filter bandwidth, etc.

Using our three components, the schematic would look like this:

Softrock with QSD schematic

The “radio” consists of two physical components, the QSD and a desktop PC. The only connection between the QSD and the PC is a stereo audio signal. There is no physical dependency. The QSD could be exchanged (e.g. a softrock receiver against a Genesis) without affecting the signal processing or the user interface at all. On the other hand, if a new digital mode (e.g. PSK31) shall be decoded, this software update can be installed without affecting the RF component at all.

WebSDR with Various WebClients

WebSDR schematic

Have you already seen Pieter-Tjerk, PA3FWM’s WebSDR which he operates from the University of Twente in the Netherlands? Pieter built up a Direct Sampling Receiver consisting of a very fast Analog Digital Converter (ADC), followed by a Field Programmable Gate Array (FPGA) where several Direct Digital Down Converters (DDC) are implemented to (basically) filter out the 7 desired bands. The receiver sends the I/Q signal of 7 bands, with a bandwidth between 150kHz and 1,2MHz via Gigabit Ethernet to a local Server. This server is the interface between the receiver and the users.

A user connects through the web browser to the server where the server provides a spectrum scope for each band. Within the spectrum, the user can select the desired frequency, mode and filter bandwidth, and receives instantly the audio signal. The I/Q to Audio conversion is performed on the server with the parameters (frequency, mode..etc) requested by the user. In theory it would be possible to stream the I/Q Signal as well, but this would overload today’s Internet connections.

The WebSDR became already quite popular. Pieter reported that during Christmas time (2009) more than 100 user were by today (February 2010), every weekend (peak time) more than 150 users are listening on his receiver simultaneously.

So how does this WebSDR scenario look like with our three radio components?

WebSDR Diagram

The Direct Sampling Receiver is the RF component. He transforms the RF signal in a digital I/Q stream. The I/Q (seven band) stream is send via Gigabit Ethernet to the WebSDR server. In the case of sufficient bandwidth, the direct sampling receiver could be located anywhere (physical independency).

The WebSDR Server is used as an interface server between the receiver and the user. The server is in charge of the signal processing. The server generates out of the I/Q stream a spectrum view and audio streams for each connected user. Any improvement (e.g. better filters) can be realized by the means of a software update.

The user connects through the web browser to the WebSDR server. The Web browser is the User Interface. The user selects the desired signal on the band with a single mouse click. A few milliseconds later he can hear the signal through his loudspeakers. The user can be located everywhere in the world. There is no geographical / physical dependency between the server and the user (client). There is also no logical dependency. A high number of instances of the User Interface (the persons who are listening) can access at the same time on the same receiver without interfering each other.

Could you imagine such an application a few years ago?


A radio is a big box with a ton of knobs, even more LEDs and occupying the majority of our desks? Yes, that’s still the case. But we are moving – towards an exciting future. Therefore it is crucial to understand the SDR approach. In contrary to the classic radio, the three components (RF, Signal processing and User Interface) of a SDR are independent of each other. This is mainly enabled by interfaces and protocols which have been established as industry standards during the last decade. Powerful semiconductors and broadband Internet allows us the design of a whole new range of applications. The two presented examples illustrate how powerful this technology already is.

A radio is a big box with a ton of knobs, even more LEDs and occupying the majority of our desks? Yes, that’s still the case. But we are moving – towards an exciting future. Therefore it is crucial to understand the SDR approach. In contrary to the classic radio, the three components (RF, Signal processing and User Interface) of a SDR are independent of each other. This is mainly enabled by interfaces and protocols which have been established as industry standards during the last decade. Powerful semiconductors and broadband Internet allows us the design of a whole new range of applications. With the two presented examples you can see how powerful this technology already is.

About Tobias (DH1TW)

Self-confessed Starbucks addict. Loves to travel around the globe. Enjoys the technical preparations of Amateur-Radio contests as much as the contests themselves. Engineer by nature. Entrepreneur. For more, follow him @DH1TW


  1. David heath says:

    I have seen the future and it is now. Thanks

  2. good work, tobias. except there is a real latency with web based receivers that cannont match real
    time signal sampling from MY antenna, hi. WHEN i can hear the sigs anyway…

  3. Hello Tobias

    Tks for contact about SDR. I made two peaces of SDR receiver this year. First one ZETA SDR for 40m and second is DR 2++ from YU1LM but I have a few days to finish double symmetric supply for it HI.
    For this moment I experiment with one receiving wire antenna made from old small transformer about 10 m long to the nearest three HI. It works. I have old 16 bit sound Creative Labs and for this moment software is Rocky and M0KGK decoder becouse my old 1.1 GHz computer.I just try to connect signal from Creative Labs receiving sound card to second on board receiving sound card reserved for MixW2 couse I heard yesterday afternoon good RTTY signal.Excuse me about my website because I made it with old tools and is on Croatian language only but you can see our radio club IOTA expeditions and my small hobbies electronic area. The SDR pages are still waiting to be fulfilled.

    73 Davor

  4. A new species of high frequency (HF) radio is evolving and its technological maturity is at least a decade or two down the road. Anyone, like yourself, who envisions a software defined radio (SDR) future is already ahead of the curve, in fact, pioneering our HF frontier.

    I’m building out my HF station, slowly, ever so slowly and I see SDR technology in the shack. Perhaps, in the beginning, as a means to improve receive capabilities, additionally, I’m hedging a future when SDR will significantly impact and improve HF low power operations.

    You are blazing the trail toward a new frontier where a new species of HF radio already exists.


  5. Gary S says:

    SDR is the future and yes its getting much closer ;-)

    Myself, I am coming from 10 years out of amateur radio except for listening with a ft-817 for last few.

    I’m tired of buying “appliances” and want the pleasure of doing some building and it looks like SDR is definately the way to go. The HPSDR project has the “Hermes” coming soon (for rx/tx) but the “funkamateur” will probably be cheaper and sooner.

    I still am looking for the one I never had, a Drake TR4c which was the “item” when I became a ham, back in the mid ’70′s.


  6. DL5RMH says:

    Hi Tobi, thanks for this very interesting article. I hope to get my Softrock SDR working as soon as possible to expore SDR by myself.
    73s de Martin, DL5RMH

  7. Mike N4GU says:

    Much work has been done (and continues to be done) in the amateur community focusing on the RF and signal processing portions of the SDR. The user interface is lagging behind. Keyboards as input control devices are not optimal. As a UI for tuning in a radio signal, you can’t beat a knob as a man-machine interface. Yes, there are some knobs out there, but the potential, and acceptance, of the SDR will occur when we can customize the UI as easily as the other elements.

    Very good piece, Tobias.

    73, Mike N4GU

  8. Bob K7YB says:

    Interesting site and I will bookmark it for future reference and education. I’m amazed by the excellent signal to noise (to my ear) characteristics of SDR on PA3FWM’s receiver. Got a shack full of old stuff (Drake 2b/Collins KWM2/etc) and newer gear (IC756/FT857)…but the NEW stuff (SDR) has me excited about the future of the hobby and the advances technology can bring.

  9. Greg Overkamp says:

    The article states “The QSE mixes the RF signal into a baseband signal” when in fact a QSE does just the opposite. A Quadrature Sampling Exciter takes I and Q baseband as its inputs, and produces RF as its output. With the lofty title of “Do you truly understand the SDR concept?”, I would expect the author to do so.

    • Hey Greg!
      Thanks for your comment! Indeed, I mixed up QSE with QSD. I just corrected it in the post. Thanks again for the notice!

  10. Victor RN4CW /UR7GU/ says:

    Bad not see this page half year early.

  11. Edwin G Delgado says:

    Excellent description, without even a single mathematical equation…… Good step by step procedural (and easy to follow) of what really happens inside an RF/DSP system to detect and reproduce ham (and several others) communication. The diagrams and description can be followed by even a kid. Thanks to the description …. and thanks for the few links for further info………

    WP3WC…. Electrical Engineer with Master RF and Power Electronics.


  1. [...] Pensez-vous vraiment comprendre le concept du SDR? PMSDR – Un interview avec IW3AUT [...]

  2. [...] Before we go one let’s recall the differences between a classic radio(as for example the Icom IC-756 ProIII) and a software defined Ra…. [...]

  3. [...] that I do not want to develop my own SDR Software but instead, making maximum reuse of existing building blocks, namely the RF and Signal processing building blocks. Since I couldn’t find an SDR framework or software with proper Interfaces I had to look for an [...]

Speak Your Mind