# Why bother to have a high power base station when mobile units are generally low-power?

I want to find out why, say in a GSM/cellular system, a base station can be up to 50 watts however the mobile units can be only 100mw (for example).

Surely if the base station ever uses 50w to reach a mobile at long range, the mobile will have no chance to transmit back.

I understand that on a base station receiver tech is perhaps more sophisticated / bigger and better then on a mobile unit, but why would there be such a large difference in max power? - why not limit the base stations to 5 watts or such?

EDIT

Note: Only using GSM as an example, but question stands for any technology. Here is what I think I know already:

• Difference between a 1W mobile and 50W base station is 17 dB.
• Base Station Diversity can add up to 5-6 dB gain, which would balance a 1W mobile to a ~4w basestation
• MIMO is on both sides, so this would be roughly equal gain and so would probably not be relevant in this topic (although the BTS with its larger antennas + better spacing may be able to get a few more dB gain).

Is the antenna technology in the base station that much better - they are larger, higher, and directional...

• I believe this is also true of FM repeaters as used in amateur radio — the repeater is generally higher power (though not as large a ratio). So there must be a general principle that isn't specific to GSM technology. I'd love to see that answer (I tried to write it, but couldn't justify the claim that the higher transmit power is useful given that you need bidirectional communication and the better antenna positioning improves transmission as well as reception). Oct 6 '15 at 20:58
• @Kevin Reid AG6YO♦ I also thought about this, but was unable to give a good answer. As for the repeaters, I don't know how it's in USA, but for example here in Serbia, maximum power is 5 W for local, 10 W for regional and 15 W for backbone repeaters. On the other hand, maximum power for repeater users is 75W. Now, I'm not saying that our authorities were thinking when they made such regulations, but it might be a hint that there is something there pointing in the other direction? Maybe it's not just the raw power? Oct 7 '15 at 4:44
• @AndrejaKo Certainly the majority of the repeater's advantage is that it has a good antenna in a good location. (In the US, there are no power limits on repeaters. But also I don't actually know what power typical repeaters use.) Oct 7 '15 at 5:02
• So, based on people comments/answers I have added an edit to my post with further information. Oct 7 '15 at 7:44

The link is actually balanced, because the base station receiver has diversity gain.

2G and 3G cellular systems use two antennas at the base station (for each coverage direction). They are either spaced a few metres apart, or more commonly, +45 and -45 degree polarisation, in the same housing. With two antennas and two receivers, the base station has a much greater probability of receiving the handset, above some threshold. If one antenna is in a null, the other might not be.

The base station also transmits from one of the antennas. Your phone doesn't have two antennas, so can't use diversity to increase its probability of receiving the signal. But the base station transmits a lot more power, and this compensates somewhat for the lack of diversity.

The phone is about 1W, but if it's close to the base station, it may be instructed to turn down its power, to save battery and to balance all the mobile signals at the base station receiver.

Diversity is related to MIMO, but is not the same thing. A diversity receiver selects the best signal from the two antennas. A MIMO receiver combines the two signals in the optimum way.

• Your phone doesn't have two antennas, so can't use diversity to increase its probability of receiving the signal. Are you quite sure this us up to date? I know that today's smart phones have MIMO antennas for WiFi. I wasn't able to quickly find info about number of antennas for cell networks, but I'd be surprised if it isn't also 2x2. If I remember correctly, LTE-A supports up to 4x4 antenna configurations in uplink and up to 8x8 for downlink. Oct 6 '15 at 15:35
• Yes, my answer is all pre-LTE. Regular 2G and 3G phones only have one cellular antenna (per band). Wifi, perhaps more. LTE certainly uses 2x2 and perhaps more, but this is more for data modems with well separated external antennas. In an LTE handset, I wonder if you'd get much better than 1 antenna, 2 polarisations absolutely the best you could do. Oct 6 '15 at 21:37
• Your extended question starts to mix up cause and effect. Some links are asymmetrical the other way, for example satellite ground stations can be more powerful than the satellite. You can't analyse that in isolation and wonder why, the original engineer considered all the constraints and did the best he could. The biggest one is probably power - fixed things on earth have more power available, so they generally use it. Handheld things are weak, so other tricks are invented, like diversity. Higher power is not a natural property of base stations, it's a solution to a broader problem. Oct 7 '15 at 12:18
• You also need to account for the fact that a base station is generally handling a large number of mobiles simultaneously, and it needs more-or-less to devote an equal amount of power to each mobile. (Give or take power controls, etc.) So if it's handling 100 conversations, it needs ~ 20 dB more power overall. Each mobile unit is only handling one conversation. It's also possible that the mobile unit's receiver is less sensitive than the base station's. That demands more transmit power at the base. Oct 10 '15 at 1:05
• @martinewing I can give an example, from 2G GSM which is TDMA. There are 8 time slots, the base station transmits its 30-40 W in all of them, while the mobile transmits its 2 W in only one. So the ~12 dB difference remains, if you look at power in the time slot used to communicate to/from that phone. In CDMA, 3G etc, the mobile transmits for a longer period, and there are more phones per base station, so I think you're onto something. Oct 10 '15 at 5:37