Home » Blog » Radio » Software Defined Radio » Radio Spectrum Processor – Good Name for SDRPlay

Radio Spectrum Processor – Good Name for SDRPlay

radio spectrum processor
RSP2 Diagram from RadComm Review

Radio Spectrum Processor is a good name for the SDRPlay boxes. Because they are not your normal kind of radio. 

RSP1 and RSP2 differ from what we would normally call a “radio” because three things are missing: demodulation, integrated gain control, and effective HF preselection filters. These are indeed radio spectrum processors which need a lot of support from software for the Mirics MSi3101 Chipset to perform well.

The absence of demodulation is no big deal. That is easier and perhaps better to do in software once the processed I/Q data arrives over the USB port. Applications such as SDRUno and SDR-Console do a great job of demodulating a variety of signal modes and displaying spectrum. This all happens at zero or low IF. However, in a normal radio, feedback is provided back into the signal chain in the form of AGC derived from the IF or audio stages.

Which brings us to the second missing element: integrated gain control. A normal radio is designed to optimize noise and amplification at the various stages of the signal chain. While you can adjust RF gain and attenuation manually, most of the work is done automatically by AGC. With the Radio Spectrum Processor, however, you need to make adjustments to front end, mixer and baseband gain within and before the MSi001 tuner.

The good news is that this form of gain control in the Mirics chipset enables superb reconfiguration to optimize the receiver over a nearly 2 GHz frequency range. The bad news is that this approach is not what we are used to in “normal” radios. Instead, we need to adjust digital attenuators to increase or decrease gain. Now a lot of this work is done through look-up tables. In particular, SDRPlay have done a good job in normalizing signal strength measurement in the face of different levels of amplification.

Keep in mind that the main purpose of gain control in these RSP devices is to get radio signals to the proper level for analog to digital conversion in the MSi2500 chip. You can learn more about this process in their Technical Note.

Radio Spectrum Processor Preselection Filters – Not

Finally, the absence of HF preselection filters is the last difference. Yes, RSP2 has preselection filters for broad swaths of its 2GHz range. But for HF and lower, the only front end filtering is a 12 MHz lowpass filter.

Many radios these days are direct sampling at RF. Most have preselection to limit the range of signals arriving at the analog to digital converter. This can be especially important if you have strong signals nearby. Too much power going into the ADC can result in clipping/overloading. Mostly, it depends on the quality of the ADC circuit. Also, problems can occur in the front end RF amplifier. Both of these issues relate to the quality of these circuits. Receivers such as Perseus, ICOM IC7300 and RFSpace NetSDR have banks of 10-15 bandpass filters for preselection over 0-30 MHz.

But essentially the RSPx is wideband up to 30 MHz, and that can cause some overload problems, as well as susceptibility to nearby inter modulation artifacts from strong signals.  This is true especially if you are using the Hi-Z antenna port. The good news, though, is that RSP2 has incorporated a MW/FM notch filter to attenuate local strong broadcast signals by 40-60 dB. This helps.

I also find that I can sometimes improve HF performance by inserting 6 dB attenuation at the antenna input. Perhaps the RSP2 is a little bit “too hot” when used with a tuned ham band antenna.

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.