Image of RF-mixere: sådan flytter modtageren et signal til en anden frekvens

RF mixers: how a receiver moves a signal to another frequency

  • September 29, 2026
  • |
  • eBits

A radio mixer is not an audio mixing desk. It is a frequency-conversion circuit. By combining an incoming signal with a local oscillator, it produces components at new frequencies. Filtering selects the useful result. This lets a receiver process information at a frequency that suits the rest of its circuitry.

RF, LO and IF

The signal arriving from the radio path is commonly called RF. The local oscillator provides LO. A converted signal is often called the intermediate frequency, IF. In a simple ideal model, multiplication produces sum and difference frequencies: fRF + fLO and |fRF − fLO|.

For an illustrative example, an 868 MHz signal mixed with an 858 MHz oscillator produces a difference at 10 MHz and a sum at 1,726 MHz. A suitable filter selects the desired output. These are educational numbers, not a proposed transmitter configuration. Analog Devices’ mixer exercise explains the principle.

Frequency conversion can preserve the useful modulation when designed correctly. It does not automatically decode the message, remove encryption or translate Meshtastic packets into Reticulum. Modulation, decoding, network framing and application interpretation remain separate steps.

Why convert frequency at all?

A superheterodyne receiver uses conversion so amplification and filtering can operate around an intermediate frequency. Direct-conversion receivers instead move the desired signal toward baseband, often using in-phase and quadrature paths. Modern integrated radios hide much of this inside the chip, so a board user rarely needs an external mixer for ordinary LoRa operation.

The architecture determines which errors and unwanted signals need managing. No single block diagram represents every receiver. Use the actual radio’s datasheet when tracing a real product rather than assuming an educational sketch is its schematic.

Illustration: two signal sources and a mixer. The precise frequency calculation is in the explanation and interactive example.
Illustration: two signal sources and a mixer. The precise frequency calculation is in the explanation and interactive example.

The image-frequency problem

In the example with an 858 MHz LO and 10 MHz IF, an 848 MHz input also produces a 10 MHz difference. If the receiver admits both inputs without another way to distinguish them, an unwanted signal can land in the same output band. This is the basic image problem.

Input filtering, image-reject architectures and I/Q processing are ways of managing it. A narrow filter after the mixer cannot always undo two signals already translated to the same place. The position of each filter in the chain matters, not just whether a filter is present somewhere.

Real mixers produce more than two perfect outputs

Non-ideal circuits can leak LO or RF into the output and generate unwanted mixing products. Conversion gain or loss, port isolation, linearity, noise and required LO drive all affect the result. Active and passive mixers have different trade-offs; a device with conversion gain is not automatically superior.

The oscillator is also imperfect. Its phase noise can interact with a strong unwanted signal and degrade reception near the desired signal. Raising gain after the mixer does not remove that problem. See Analog Devices on local-oscillator noise.

A practical receive chain

Think of the functions as antenna, selection/filtering, low-noise amplification, frequency conversion, further filtering, digitisation and demodulation. A real design can combine or reorder functions. The LNA tries to preserve weak-signal performance, the mixer changes frequency and the demodulator recovers information from the waveform. Each stage has a separate job.

This helps diagnose misleading assumptions. An antenna problem cannot usually be repaired by changing an application channel name. A strong blocker may overload an early stage before software sees anything. A packet decoder cannot fix an absent radio path. Work from the physical input toward the application, testing one layer at a time.

Learning without building an RF transmitter

You can explore the sum/difference arithmetic on paper or in a simulation first. Change one input frequency and follow both outputs. Then add an unwanted input and see where its converted signal lands. This makes filter placement and image rejection much easier to understand than memorising abbreviations.

For bench experiments, use equipment and signal levels suitable for the instruments, with the required attenuation and terminations. Do not connect a transmitter output directly to an unprotected receiver or measurement input. Follow a documented laboratory exercise rather than improvising RF connections.

Continue with LNA and PA amplifiers, antenna mismatch and LoRa fundamentals. These articles explain the radio’s building blocks; they do not imply that eBits stocks every laboratory component or that extra RF stages are needed for a standard Meshtastic setup.

Reviewed 28 September 2026. The frequencies and block descriptions are educational examples, not measured specifications for a particular radio.

Leave a comment

Please note, comments need to be approved before they are published.