A demodulation pixel architecture allows for demodulating an incoming modulated electromagnetic wave, normally visible or infrared light. It is based on a charge coupled device (CCD) line connected to a drift field structure. The drift field is exposed to the incoming light. It collects the generated charge and forces it to move to the pick-up point. At this pick-up point, the CCD element samples the charge for a given time and then shifts the charge packets further on in the daisy chain. After a certain amount of shifts, the multiple charge packets are stored in so-called integration gates, in a preferred embodiment. The number of integration gates gives the number of simultaneously available taps. When the cycle is repeated several times, the charge is accumulated in the integration gates and thus the signal-to-noise ratio increases. The architecture is flexible in the number of taps. A dump node can be attached to the CCD line for dumping charge with the same speed as the samples are taken. Different implementations are described herein, which allow for smaller design or faster speed. The pixel structure can be exploited for e.g. 3D time-of-flight imaging. Both heterodyne and homodyne measurements are possible. Due to the highly-efficient charge transport enabled by static drift fields in the photo-sensitive region and small-sized gates in the CCD chain, high frequency bandwidth from just a few Hertz (Hz) up to greater GHz is supported. Thus, the pixel allows for highly-accurate optical distance measurements. Another possible application of this pixel architecture is fluorescence lifetime imaging microscopy (FLIM), where short laser pulses for triggering the fluorescence have to be suppressed.