EEVblog Electronics Community Forum
Electronics => Projects, Designs, and Technical Stuff => Topic started by: eecook on September 02, 2014, 03:29:36 pm
-
Hi all,
I am trying to calculate the optimum distance between the main_ant and div_ant for a 3G/4G (700-2100Mhz) modem I am working on. This modems (ideally) will be on a fixed position attached to light poles. From what I've found, multipath fading will appear either
1. As a standing wave, in which case a half wave separation would be ok, or.,
2. As signals coming from different channels (different multipaths), in which case it is desired to minimize the cross correlation.
If 1. was the case, what is a good trade off for multiband scenario?
If 2. was the case, I cannot seem to find any pair of references that say the same thing. Some talk about a minimum and a maximum distance, some talk only about a minimum (sory I am not referencing, it's been a bit of a messy search), and in any case, this distances are not always the same value (expressed as fraction of wave-length).
I found this document http://www3.alcatel-lucent.com/bstj/vol50-1971/articles/bstj50-6-1859.pdf (http://www3.alcatel-lucent.com/bstj/vol50-1971/articles/bstj50-6-1859.pdf) which seems useful, except for the part were it says that I should be using a 70lambda spacing, which is way bigger than the enclosure (ants should go inside). Plus, cell phones have the div_ant only a couple of centimeters seperated from the main one.
Could anyone shed some light on the matter please?
Is there any document that explains this at a more fundamental level?
Regards.