Software-Defined Radio (SDR)

On July 1, 2020, this post will be obsolete. See the corresponding post from my 2020 version of No Nonsense Extra Class License Study Guide.

This section is almost entirely new. I’m far from being an expert on digital signal processing (DSP) and software-defined radio (SDR), so if you have some expertise in these areas, please feel free to comment on this section….Dan

E7F – DSP filtering and other operations; Software Defined Radio fundamentals; DSP modulation and demodulation

Some modern radios modulate and demodulate signals entirely in software. This type of radio is called a software-defined radio, or SDR. One type of SDR uses a process called direct digital conversion to convert the analog radio signal into a series of numbers. As applied to software defined radios, direct digital conversion means incoming RF is digitized by an analog-to-digital converter without being mixed with a local oscillator signal. (E7F01)

Analog-to-digital converter specifications are crucial for a software-defined radio. For example, sample rate is the aspect of receiver analog-to-digital conversion that determines the maximum receive bandwidth of a Direct Digital Conversion SDR. (E7F10) An analog signal must be sampled at twice the rate of the highest frequency component of the signal by an analog-to-digital converter so that the signal can be accurately reproduced. (E7F05)

Voltage resolution is also important. The reference voltage level and sample width in bits sets the minimum detectable signal level for an SDR in the absence of atmospheric or thermal noise. (E7F11) The minimum number of bits required for an analog-to-digital converter to sample a signal with a range of 1 volt at a resolution of 1 millivolt is 10 bits. (E7F06)

Modern software defined radios convert an incoming signal into two data streams: I and Q. The letters I and Q in I/Q modulation (and demodulation) represent In-phase and Quadrature. (E7F17). The I and Q data streams are 90 degrees out of phase with one another, and as a result, the two data streams not only show how the amplitude of a signal is changing, but how the phase of a signal is changing.

The digital process that is applied to I and Q signals in order to recover the baseband modulation information is the Fast Fourier Transform. (E7F12) Converting digital signals from the time domain to the frequency domain is the function that a Fast Fourier Transform performs. (E7F07)

Once a signal has been digitized, or converted into a series of numbers, it can be digitally filtered. The kind of digital signal processing audio filter used to remove unwanted noise from a received SSB signal is an adaptive filter. (E7F02) Another type of digital filter, one that is often used in a direct digital conversion receiver, is the finite impulse, or FIR, filter. An advantage of a Finite Impulse Response (FIR) filter vs an Infinite Impulse Response (IIR) digital filter is that FIR filters delay all frequency components of the signal by the same amount. (E7F15)

The FIR filter in a software defined radio might also be a decimating filter. The decimation function reduces the effective sample rate by removing samples when using a digital filter. (E7F08) SDRs perform decimation because the signal of interest will usually have a significantly lower bandwidth than the digitized signal, and reducing the sample rate allows SDRs to use less-powerful processors. One way the sampling rate of an existing digital signal might be adjusted by a factor of 3/4 is to interpolate by a factor of three, then decimate by a factor of four. (E7F16)

An anti-aliasing digital filter is required in a digital decimator because it removes high-frequency signal components which would otherwise be reproduced as lower frequency components. (E7F09)

Taps in a digital signal processing filter provide incremental signal delays for filter algorithms. (E7F13) More taps would allow a digital signal processing filter to create a sharper filter response. (E7F14)

Signals can also be generated using SDR techniques. A common method of generating an SSB signal using digital signal processing is to combine signals with a quadrature phase relationship. (E7F04) The type of digital signal processing filter used to generate an SSB signal is a Hilbert-transform filter. (E7F03)