Mixers
In communication systems transmission is at Radio Frequency(RF), which need to be translated to Intermediate Frequency(IF) to make signal processing easier. Mixer is a non-linear device which translates power at one frequency to another frequency. Functionally it is equivalent to a multiplier.
Let's say, if A1 cos(ω1t)and A2cos(ω2t) are the two input signals then the output is given as,|
O/P =[A1 cos(ω1t)].[A2 cos(ω2t)] =(1/2)A1A2 [cos(ω1t - ω2t) + cos(ω1t+ ω2t)] |
(1.1) |
Out of the sum and difference frequency components, in the down-conversion application the difference components are utilized, whereas the sum components in up-conversion applications. The unutilized other component, in either of the applications, which is to be rejected is called image component and can be filtered using bandpass filter generally.
![]() |
|
| Figure.1. Mixer Block Diagram |
Out of the two input signals one is fed to LO port and other to either RF or IF port. If LO frequency is greater than input frequency it is called High side injection, where as in Low side injection the opposite is true. LO injection side, whether high or low, is often dictated by the spurs and the desired RF and IF band of operation.
LO provides periodic hard limited switching waveform to mixer rather than sinusoidal.
These harmonics cause additional signal energy to be converted into unused spurious component called spurs unless supressed by mixer gain at high frequencies.
While selecting a mixer, the following specifications of the mixer(active)need to be evaluated.Conversion Gain
![]() |
|
| Figure.2. Conversion Gain Vs Lo Drive |
3rd Order Intercept Point (IP3)
1dB Compression Point(P1dB)
![]() |
|
| Figure.3. Gain Compression Vs. Amplitude |
Noise Figure
|
(1.3) |
Dynamic Range
Port Isolation
- LO feedthrough will occur from LO port to IF port due to parasitic capacitance and power supply coupling, etc.,
- Reverse LO feedthrough will occur from LO port to RF input port due to parasitic capacitance, etc., Self-mixing of Reverse LO feedthrough with LO result in DC offset, which poses a problem in zero IF architectures.
- RF to IF leakage determines the fraction of the signal in the RF path that directly appears in the IF path.
Spurs
|
ωIF = | m ωRF ± n ωLO | or ωRF = | m ωIF ± n ωLO | |
(1.4) |
- Spurs : Any signal that is half IF away from LO frequency may be mapped on to the IF channel at the output. For example, consider a down-conversion of 1950MHz signal using a lowside LO to an IF of 70MHz. If an unwanted interfering signal is present at a half-IF below, say 35MHz, the wanted signal, than the 2 x 2 product could result in a interfering product at {2(1950-35) - 2(1880) = 70MHz }. In narrow band system this can be avoided by choosing high IF. In BB systems use balanced mixer topology to overcome this problem.
- Image response : If an undesired signal present at IF away from LO frequency may map on to the top of on-channel IF frequency at output. Remedy for this problem is either to use Image Reject Filter or Image Reject Mixer.


