16

It seems confusing that my phone receives -87 dbm LTE signal and shows a full 4 bars with the speeds of

UPLOAD: 20Mbps

DOWNLOAD:13.6Mbps

But my WiFi is showing 1 bar at -89 dbm and I get disconnected as soon as I move slightly away, and speeds are very low. Why does this happen? It happens with all my phones.

psmears
  • 678
  • 4
  • 6
ObsessionWithElectricity
  • 349
  • 3
  • 12
  • 7
    Just to make sure you are clear, -89dBm is only 67% of the power of -87dBm, which is quite a difference. It's not the answer to your question, but worth noting. – Tom Carpenter Sep 06 '18 at 08:34
  • 1
    But at -92 dbm LTE shows full bars on my phone. – ObsessionWithElectricity Sep 06 '18 at 09:52
  • 13
    The "bars" are basically a nice fiction that bears only a very loose relationship to actual reception. – pjc50 Sep 06 '18 at 14:15
  • But my friend, I know the bars aren't that indicative. But surely data speeds are. – ObsessionWithElectricity Sep 06 '18 at 15:45
  • 1
    @ObsessionWithElectricity No, data speeds have nothing to do with connection quality on WiFi. – Mast Sep 06 '18 at 16:05
  • 1
    Ok, come on, my connection drops by just moving a few steps away from the router. Remember, I just wanna know why WiFi has poor connection at -89dbm while LTE seems to be fine even at -95dbm – ObsessionWithElectricity Sep 06 '18 at 16:10
  • @mast while technically true, connection quality does affect error rates, and error rates affect the PERCEIVED connection speed. – barbecue Sep 06 '18 at 21:57
  • 1
    The main problem has to do with not having a large enough transmitter/receiver. If you look on youtube, you'll see that people have established wifi *miles* apart, but it requires rather significant work on both ends, proper line-of-sight, and proper aiming. The larger the antenna, the more sensitive it is, and the less power is required overall. – phyrfox Sep 07 '18 at 00:20
  • Eye damage is one reason to keep the power levels low, at the ground level. – analogsystemsrf Sep 07 '18 at 04:02
  • How are higher radio power and eye damage related? Can you explain it a bit.? – ObsessionWithElectricity Sep 07 '18 at 09:36
  • (I think the answer to your underlying question "why is my wifi so bad" is "your AP is broken", by the way. All sorts of failures are possible with cheap APs.) – pjc50 Sep 07 '18 at 12:27

3 Answers3

27

As with any radio receiver, if it can handle a higher data rate, then it is usually burdened with having a higher RF bandwidth and this inevitably means more received background noise i.e. a wider BW lets in more noise and hence, you need a higher received signal level to operate with a decent SNR (signal to noise ratio).

Therefore WiFi is at a significant disadvantage because it has a wider RF bandwidth than LTE (normally) and needs a higher signal level to operate at a decent bit-error-rate (BER). This is embodied in the following empirical but commonly-found relationship.

Power (dBm) needed by a receiver is -154 dBm + \$10log_{10}\$(data rate)

For example, if the WiFi data rate is ten times your LTE data rate, then you need 10 dB more signal to operate at the same SNR. Basically if you double the RF bandwidth you "collect" 3 dB more noise. This means that WiFi is usually the first to suffer as signal levels drop (compared to LTE data rates).

Why WiFi has shorter range than LTE?

This is related to the Friis transmission equation but, more simply, you can think about the same effect with light bulbs; consider a 1000 watt lamp and the distance you could see this at night time - you would probably see it fairly clearly from 10 km away and, if you walked a further 100 metres, it wouldn't look significantly dimmer.

Compared with a small 1 watt lamp, you might see it glowing at 100 metres but, if you walked away a further 100 metres, it would be noticeably dimmer.

There are a bunch of other factors too such as operating frequency - WiFi can operate at a higher carrier frequency and the Friis transmission equation informs you that as frequency rises, the path-loss increases: -

Path loss (dB) = 32.45 + \$20log_{10}\$(F in MHz) + \$20log_{10}\$(D in kilometres).

In other words at ten times the frequency, the path loss increases by 20 dB.

Andy aka
  • 434,556
  • 28
  • 351
  • 777
  • 3
    My phone carrier is at LTE B40 - 2300-2400Mhz . So I guess there is not much difference in frequencies. I think the reciever of WiFi routers is less sensitive than receiver of BTS of LTE – ObsessionWithElectricity Sep 06 '18 at 09:58
  • Or maybe my phone or the wifi router emitts less power on WiFi than LTE – ObsessionWithElectricity Sep 06 '18 at 10:02
  • That and, depending on location, it may be fighting for the same frequencies as 15 neighbors. – rdtsc Sep 06 '18 at 10:11
  • @ObsessionWithElectricity as I said in my answer, WiFi has a naturally larger data rate hence it has a naturally larger BW hence it attracts a naturally larger level of background noise. Plus it may be working at 5 GHz and this means a 6 dB decrease in received signal level compared to 2.5 GHz. – Andy aka Sep 06 '18 at 10:28
  • My area has no WiFi other than mine. WiFi frequency : 2.4 GHz channel 12. – ObsessionWithElectricity Sep 06 '18 at 10:34
  • 3
    Also, you don't tend to go and mount your WiFI access point at optimum height on a huge honking pole, unlike commercial mobile carrier equipment :) – rackandboneman Sep 06 '18 at 12:11
  • 2
    The wifi frequency is irrelevant to a discussion about _band width_. – pjc50 Sep 06 '18 at 12:19
  • The base station reciever will be both much larger and much more sensitive than the average wifi reciever. – pjc50 Sep 06 '18 at 12:24
  • 2
    @rackandboneman: And if you _do_ mount a Wi-Fi access point like that, decent equipment (with a directional antenna like LTE, not omnidirectional like cheap home "routers") can easily reach several km. Long range point-to-point Wi-Fi links are rather common. – user1686 Sep 06 '18 at 12:26
  • Also, don't forget that phone antenas are directional: they are, basically, a beam that goes through a certain distance. Also, won't the size of the antenna matter? – Ismael Miguel Sep 06 '18 at 16:16
  • 1
    Phone antenna's are generally *not* heavily directional, or you would expect your phone to lose signal as you changed its orientation. The base stations can be more directional, but they generally still a broad region, potentially doing software based beamforming. – mbrig Sep 06 '18 at 20:47
  • @pjc50 Frequency is hardly irrelevant. The higher the base frequency, the worse the reception, generally speaking. But you need a chunkier antenna. For what it's worth, LTE Cat M1 narrowband IoT has better reception than the regular LTE and the very narrow band LTE Cat NB1 has *much* better reception. LTE Cat NB1 is not based on LTE, whatever the name suggests. – Barleyman Sep 07 '18 at 12:13
9

in addition to Andy's answers, WIFI is usually limited in power, 30dBm in north America, lower levels in most of the world. LTE usually can transmit up to 4W (36dBm) And the towers transmit at a much higher power.

Also, LTE has much better network management capabilities (automatically finding the best channel and data rate), the towers have much better clock sources than WIFI (this affects receiver sensitivity) and the towers have higher placed antenna (10-30m) than the common wifi router.

Lior Bilia
  • 7,282
  • 1
  • 20
  • 30
  • But I heard that LTE transmit upto 200mW in phone's. For towers I don't know. Which is correct 4 W or 200mW? – ObsessionWithElectricity Sep 06 '18 at 11:43
  • 5
    I'm fairly certain it's 200mW. GSM (2G) used to be 2W, but that was dropped first to 1W and then 500mW. Battery life at 4W would be atrocious. Note that even GSM would use these figures as absolute maximums, and dynamically lower it where possible. – MSalters Sep 06 '18 at 12:18
  • 6
    Yes, it's 200mW. This paper is interesting: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=7879218 - the *average* power of the 95th percentile of transmissions was 4mW! – pjc50 Sep 06 '18 at 12:23
  • It is up to. The limit is regulatory, not technical. Obviously, you want to use as little power as possible. If you are in an urban setting, with lots of towers, sure your phone will transmit few mW. If you are out in the wilderness, with the nearest tower 5km away from you, your phone will transmit at maximum power. – Lior Bilia Sep 06 '18 at 13:31
  • How much is power emitted by tower of LTE? – ObsessionWithElectricity Sep 06 '18 at 15:52
  • @ObsessionWithElectricity as much as necessary to be heard by the phone. I think a typical max is 20W, but most of the time, much less :) – hobbs Sep 06 '18 at 16:06
  • I know that maximum permissible limit for public exposure is set at 1 mW/cm2 for towers emitting above 1.5GHz . But how can it relate to the 20W you talked about. – ObsessionWithElectricity Sep 06 '18 at 16:16
  • @pjc50 This obviously depends on what you're doing. Maximum transmit power would be achieved when transferring large amounts of data at poor reception. That might be a realistic use scenario if you're using mobile data for Netflix or something but not otherwise. LTE user end 200mW power limit translates to 23dBm. – Barleyman Sep 07 '18 at 12:09
0

For comparison, these are the transceiver part of a LTE network(from two different vendors)(antennas are mounted on towers/poles and connected via cables to the eNodeB), equivalent to the antenna part plus the transceiver circuit of a wifi AP.

https://www.motorolasolutions.com/en_xl/products/lte-broadband-systems/broadband-systems-equipment/enhanced-node-b/rbs6101.html#tabproductinfo

https://www.scribd.com/document/204866576/RBS-6000-Spec-Sheet

A LTE eNodeB can handle hundred of users(depending on configuration) in a controlled manner. How many can a wifi AP handle??? You cannot really compare Wifi to LTE. It is two completely different kind of system meant to be used in quite different scenarios.

Strider H
  • 1