10

I have observed that WiFi routers with 2 antennas have a better coverage than a router with a single antenna. How does that work out?

I can clearly make out that the second antenna is definitely not a repeater kind of thing, or is it?

Absolutely related: How does having more antenna help in a networking router

The answer in the question sited above speaks of Noise cancellation. I can correlate that to Efficiency in terms of reception quality, packet loss, etc, but not to the signal strength.

Community
  • 1
WedaPashi
  • 1,670
  • 17
  • 29
  • Pro tip: If you have removable antennas and two cables to spare, try attaching the antennas via cables and run a continous download or network SNR test while variying the antenna positions in relation to each other. If, like @Neil_UK says below you already have very good conditions you won't see much of a difference. In my experiments, the worse conditions you have to start with, the better the improvement of separating the antennas (by several meters). – winny Oct 20 '16 at 14:50
  • 1
    Does this answer your question? Will having two antennae at the receiver give better reception? – feetwet Jun 07 '23 at 20:10

3 Answers3

22

In free space with a nice direct path, it would make little difference.

Unfortunately, WiFi doesn't operate in free space: it operates in a complicated sea of reflections from walls, people, furniture, wiring etc. The reflections add together in more or less random phases, and from time to time the signal level can drop quite significantly. This is called Rayleigh Fading.

As the fade depends on the relative phases of the signal paths, being only half a wavelength away is enough to make the random fades at these two positions completely independent. If there is a 10% chance of each antenna having a signal strength too low for operation, then the chance of both antennas suffering is just 1%. This is called Diversity Reception. The receiver decodes both channels, and uses the one with the fewest errors.

There are other, cleverer, more complicated ways to use two antennas in reception and transmission to improve the signal strength or data rate, but this one is the easiest to understand.

Lightness Races in Orbit
  • 168
  • 8
Neil_UK
  • 166,079
  • 3
  • 185
  • 408
  • 3
    Actually, modern systems don't just select the strongest the strongest signal. They can combine all the signals in order to get better performance than just selecting the strongest. A traditional example is rake receiver. – AndrejaKo Oct 20 '16 at 15:44
  • 6
    @AndrejaKo yes, that's included in the 'other, cleverer, more complicated ways' I referred to. Also included in those are the MIMO spatial diversity ways of increasing the data rate, or MIMO beamforming to reject interference. 'Combine' I felt was a bit waffly to include in the answer, and I didn't want to go into the details of soft decisions and overload niave readers. Do put your own reply up with the full list of options if you feel it would add to the OP's or others' understanding. – Neil_UK Oct 20 '16 at 17:49
  • @AndrejaKo Expensive ones - yes. Cheap ones just rapidly switch between antennae, never using more than 1 at a time. They pick the antenna that works best at the moment, and it still gives big advantage as Neil described (could be described as 1x2 "MIMO"). In fact, every time there is an "uneven" MIMO it's just that: picking best antennae for the moment from a bigger pool. – Agent_L Oct 21 '16 at 13:29
  • @– Agent_L Not always, for example Alamouti scheme is used with asymmetric MIMO (1 Rx antenna) and is very interesting for highly mobile devices or situations where receive antennas can be a bit bulky, but transmitter has good infrastructure. – AndrejaKo Oct 21 '16 at 14:54
9

With two or more antennas it is possible to use MIMO (Multiple Input, Multiple Output) but also to use Diversity, see this article.

I do not think MIMO by itself enhances the range or signal strength. However Diversity does as it uses the two antennas in such a way that the signal is aimed in a certain direction.

This works by sending a slightly different signal to each antenna and through phase cancellation and adding, at some locations a stronger signal can be received.

It works in the other way as well, with two receiving antennas the phase differences between the signals are used to cancel out or suppress unwanted signals.

Do note that you're comparing a one antenna router versus a two antenna router. It is very well possible that the two antenna router is also more modern and is simply more sensitive than the older router. It cannot send a stronger signal though, there are regulations which dictate how much signal can be send by a router, no matter how many antennas it has.

Regarding noise cancellation: as noise is random and cannot be cancelled it is better to speak about interfering and disturbing signals. Lowering the levels of these signals does help as that increases the signal-to-disturbance ratio (and thus the signal-to-noise ratio, if we call anything we do not want "noise") of the signal you want to receive. A higher signal to noise ratio allows for a more advanced modulation scheme to be used (for example QAM 64 instead of QAM 8) so the bitrate can be higher.

Bimpelrekkie
  • 80,812
  • 2
  • 94
  • 185
  • Thanks for the detailed answer. I believe by your answer then is it valid to state that with a better bit rate (through a sophisticated modulation technique) and better SNR the signals travel a longer distance? – WedaPashi Oct 20 '16 at 14:46
  • No it is the other way round, the circumstances (distance, interferers, reflections, noise) dictate the SNR. Given this SNR the modulation technique is selected. A large SNR allows for a more complex modulation scheme resulting in a higher bitrate. When SNR is bad Wifi scales back the modulation scheme resulting in a lower bitrate. So you get the largest range at the lowest bitrate because for a low bitrate only a limited SNR is needed. – Bimpelrekkie Oct 20 '16 at 14:52
  • The signals travel the same distance; the "diversity" technique lets the reciever improve its SNR. – pjc50 Oct 20 '16 at 15:21
  • MIMO systems can get higher output power due to beamforming. – AndrejaKo Oct 20 '16 at 15:41
  • @AndrejaKo MIMO by itself does not give higher output power. It is the beamforming function that does this. To have directional power and MIMO at the same time you would need more than 2 Antennas. That higher power is also directional. You get more power in one direction at the cost of power in another direction. In the higher power direction the usable range will be larger. – Bimpelrekkie Oct 20 '16 at 17:44
  • 1
    With only two antennas, the "beam" would be rather weak. One should better think of having a slight ellipse instead of a circle around the antenna. Oh, and the "beam" always goes both forward and back when you only have one line of antennas. To have a real "directional" power, you need at least three antennas arranged in a triangle with the needed phase shifts. – Janka Oct 20 '16 at 18:16
  • MIMO is all fine... yet in my practical experience in the 5 GHz band, 802.11a (single antenna) often achieves higher channel bitrates than 802.11n MIMO with two antennas, other conditions equal. This on fairly decent WiFi hardware (I'll keep the brand to myself) and in fairly good transfer conditions: direct visibility, distance not too long and not too short... I don't have an explanation. – frr Oct 20 '16 at 18:19
  • I think you have confused two terms. I may be mistaken but I believe that MiMo leverages antenna diversity to do its thing. Antenna diversity is the idea of having multiple antennas at a Tx or Rx in the hope one of them gets a better signal. Signal processing can give you a better result from the combined signal than just adding or picking one. MiMo is a set of protocols (inc signal processing ones) for handling diverse 802.11 signals within a typical environment. In a way MiMo is a specialised sub-set of antenna diversity use. – TafT Oct 21 '16 at 10:08
-1

If you use two identical antennas, Antenna diversity will help the receiver circuit fighting signal reflections, as others already wrote.

But you can also setup the router to have one antenna exclusively for sending, the other for receiving. When the regular firmware on your router doesn't have this option, there may be an OpenWrt Firmware, which usually does.

Having separated send and receive antennas is useful when the receiving antenna is a directional one. That gives you a greater range in that single direction.

Ok, you could ask: why don't have two directional antennas? Because you aren't allowed to go beyond the field power limit while sending, so if you use a directional sending or send/receiving antenna, it doesn't help you as you have to weaken the power output to it.

So: having a dedicated directional receiving antenna helps you saving some money by only having to buy one of those instead of two. (Of course, the other station needs the same setup.)

Janka
  • 14,276
  • 1
  • 21
  • 35
  • 3
    But, I can think of a reason why someone or the two of them have down-voted your answer. I agree with the first statement, You can setup the router to have one antenna exclusively for sending, the other for receiving. But it doesn't explain even slightly how does that increase signal strength. – WedaPashi Oct 20 '16 at 14:43
  • I did one down-vote, but was unable to comment at the time. My -1 comes from the first sentence:"You can setup the router to have one antenna exclusively for sending, the other for receiving." This is NOT how the modern multi-antenna systems are working. Second sentence is basically a link-only answer. – AndrejaKo Oct 20 '16 at 15:31
  • 1
    @AndrejaKo: You can still do this on a lot of popular Wifi hardware (maybe need to install OpenWRT on it). It's useful when you have two different antennas. One standard one for sending, and a directional antenna for receiving. That gives you a greater range in that single direction. (And if you ask: why don't have a directional antenna on sending also. Because you aren't allowed to go beyond the field power limit, so if you use a directional sending antenna, it doesn't help you as you have to weaken the power output to it.) – Janka Oct 20 '16 at 18:06
  • 1
    This statement is wrong: "If you use two identical antennas, Antenna diversity will help the receiver circuit fighting signal reflections, as others already wrote." Antenna diversity can work with identical and non-identical antennas. It is the given example in this answer which only benefits from non-identical antennas. 802.11(bna etc) works poorly with non-symmetric Tx Rx pathways as clients and APs must both send and receive constantly as part of the standard. When you make inter- building wireless links your suggest can work well but I do not think that uses standard WiFi signalling. – TafT Oct 21 '16 at 10:15
  • That's why I wrote the other station needs the same setup, with a directional receiving antenna. It doesn't help when only one side has directional antennas (one or two, it doesn't matter.) – Janka Oct 21 '16 at 10:52
  • I really don't like this answer. It skims several topics, but doesn't go into any deep enough to explain how are they relevant. Plus, wifi is a 2-directional communication, having only send or only receive is useless. If one antenna has better visibility to the other station, it should be used for both, relegating one side to worse visibility antenna has no good effects. To me this answer looks more like mumbo-jumbo. Even if that's because I failed to understand it, it's still poor answer, because it failed at making me understand. – Agent_L Oct 21 '16 at 13:22
  • @Agent_L Having receive only antenna helps you maintain link budget when you're EIRP-limited. On the other hand, it breaks spatial multiplexing and might break spatial diversity. For example, let's say you have a 20 dBm EIRP limit that's achieved by your router with stock antenna. If you replace the stock antenna with a directional TX antenna that has 10 or 20 dB gain over the stock antenna, you're outside of EIRP, but if you just replace the RX antenna, you're getting extra 20 dB of link budget "for free". I, personally, would just same model for all antennas and decrease TX power to fit EIRP – AndrejaKo Oct 21 '16 at 14:47