Networking – Can mirrors block the wifi signal

Short Answer

In many — if not most — cases, it is not possible to compare vendors' published spec sheets between consumer Wi-Fi products and determine which will give you the greatest raw signal strength. It is never possible to tell which will give you the best range or performance-at-range just by reading the spec sheet. Instead you should look for professional reviews where rate-vs.-range (performance at various ranges) was carefully measured by an expert.

Signal Strength

The difficulty in comparing spec sheets even for raw signal strength is that many vendors don't publish any signal strength related measurements at all. I just went to the websites for D-Link, Netgear, and TRENDnet, and couldn't find any radio power information at all on the spec sheets I checked for some of their latest-and-greatest Wi-Fi routers. Among the vendors that do publish some information, different vendors often publish different kinds of information related to signal strength, and the information they give is not always directly comparable.

For example, all Apple says about their current AirPort Extreme 802.11n (Simultaneous Dual-Band II) Wi-Fi router is "Radio output power: 20 dBm (nominal)". 20 dBm means 100 milliwatts (mW). But a "radio output power" is probably a measure of the signal strength at the antenna connector inside the unit. This doesn't tell you how the antennas and antenna leads affect the power output, or how that 100mW of power gets split up amongst the multiple radio chains inherent in 802.11n, or how the power varies per band or channel (and the power usually does vary per band and channel due to regulatory requirements and antenna characteristics).

So without any antenna information from Apple, you can't quite compare it to the example you gave of the Cisco-Linksys E1000, which only gives you these numbers:

RF Pwr (EIRP) in dBm: 17.5 dBm  
Antenna Gain in dBi:   1.5 dBi  

"RF Pwr" is "Radio-Frequency Power", which might make you think it's the same kind of measurement as Apple's "radio output power", but then Linksys qualifies it with "(EIRP)". EIRP is a measurement that takes antennas and antenna leads into account.

Aside: EIRP and other RF basics

EIRP stands for Equivalent Isotropically Radiated Power. To understand EIRP you first have to understand some things about antennas and directional gain:

• An isotropic antenna is a hypothetical antenna that radiates power equally in all directions; an antenna whose three-dimensional "coverage pattern" is a perfect sphere.
• It's probably impossible to create a truly isotropic antenna, and most of the time you don't want one anyway. You usually want an antenna to focus its power in one direction.
• An antenna that focuses its power in certain directions is called a directional antenna.
• The measurement of how well a directional antenna boosts the signal strength in the intended direction(s), as compared to the hypothetical isotropic antenna, is called its directional gain.
• Directional gains are measured in decibels relative to an Isotropic antenna, or dBi.
• "Omnidirectional" is not a synonym for isotropic. Omnidirectional antennas try to focus the power in a 2D disc rather than a 3D isotropic sphere. Think of squishing a water balloon between two books. An unsquished water balloon is roughly spherical, like an isotropic antenna's coverage pattern. But when you squish it between two books, it stretches out in every (2D) direction and thus becomes less compact overall. The amount it stretches out along that 2D plane is an "omni"-directional antenna's directional gain.
• Decibels are a logarithmic scale for measuring ratios. Every 3 dB is a doubling. The ratios are usually relative to a reference value, and the reference value it denoted in the letter after the "dB". So "dBi" is decibels relative to an isotropic antenna, and dBm is decibels relative to 1 milliwatt. 0 dBm is 1mW. Positive dBm's are multiples of a milliwatt, and negative dBm's are fractions of a milliwatt.

When you want to measure how much power a radio system is giving off in the antenna's best direction, you compare it to how much power you would have to feed into an isotropic antenna to get it to radiate the same amount of power in that direction. This is the Equivalent Isotropically Radiated Power, or EIRP.

...now back to the comparison

The Linksys E1000's antenna has just 1.5 dBi directional gain, which means it's pretty close to isotropic, which is probably fine for a home wireless router. A good omnidirectional antenna can have 6 dBi gain in one plane, which might be great for a sprawling single-story house, but wouldn't cover the basement or the 2nd floor very well when oriented horizontally on the first floor in a multistory home.

If you take the E1000's EIRP of 17.5 dBm and subtract the 1.5 dB of directional gain the antenna gives you, you can figure the radio's nominal output power to be around 16 or 17 dBm (assuming the internal antenna cables eat between 0-1 dBm of the radio power).

So will the E1000's 16-17 dBm of radio power make have less signal strength in your application than the AirPort Extreme's 20 dBm? Well, we don't really know, because we don't know how much signal the AirPort Extreme's antenna cables and antennas eat up, or how much of that signal gets focused in a direction that doesn't matter for you given where you're going to locate the router in your home, and where your clients are.

Even if you assume, for the sake of argument, that the AirPort Extreme's antennas will be about the same as the Linksys's, you still have to consider that signal strength is just one component in range, and performance-at-range. Receiver sensitivity and noise floor are big components too, and vendors rarely publish their specs for that. Some Wi-Fi radios are sensitive all the way down to signals as tiny as -95 dBm (we're talking femtowatts here), whereas others don't break -90 dBm. I've seen poorly-designed or poorly-assembled/poorly-tested Wi-Fi routers with RF noise sources inside the box as high as -85 dBm, so regardless of the receiver sensitivity, it wouldn't be able to receive signals lower than -85 dBm.

Myth Busting

There are a lot of myths and misinformation about Wi-Fi product signal strength and range out there, some of which got repeated in the other Answers to your Question. I'd like to take a moment to bust some myths here.

1. Not all Wi-Fi routers have the same output power or range. There really are significant differences in radio output power, and antenna design, and receiver sensitivity, and noise floor, and everything else that goes into range and performance-at-range. Of course if you have a lousy wireless client system at the other end of the link, it's lousiness can mask a lot of the greatness of a great router.

2. The FCC doesn't regulate the "default" output power, just the maximum. And the maximum is 1 Watt (1000 mW).

3. You'd think that every vendor would design products to bump right up against that limit, but that doesn't happen either. It turns out it's hard to make a good power amplifier (PA) that can reach 1000 mW without distorting the signal so much that the receiver can't decode the packets. Vendors don't want to price themselves out of the market, so they don't tend to create designs with super expensive PAs that can reach those power levels without hopelessly corrupting the signal. Also, it doesn't matter how loud your Wi-Fi router can shout and far it can be heard if your Wi-Fi client laptop or smartphone can't shout loud enough to be heard at the same distance. And most laptops and smartphones can't burn a full watt of power on networking without destroying battery life. So Wi-Fi radios in both APs and clients are rarely designed for more than 100 mW (20 dBm).

Update, late 2014: In the 3.5 years since I wrote this, high-power PAs and high quality LNAs (Low Noise Amplifiers which increase the receive gain for the AP) have caught on. If you buy a top-of-the-line consumer AP, (which at this time means an 802.11ac AP that supports 3 spatial streams or more) you're probably buying a product with good high-power PAs that allow it to bump up against the regulatory limits without distorting. This not only changes what I said above in #3, but it also changes what can be considered common Tx power ratings for APs in #4 below. With high-power PAs, 30dBm ( == 1000mW == 1 full Watt) APs are increasingly common.

4. Someone said the default for most routers is 9 mW, which is nonsense. 20 mW is about the lowest I've seen, and 32-100 mW (15-20 dBm) is common. The specs we found for the E1000 and the AirPort Extreme attest to this.

There are usually no legal ways to make the signal stronger. Most wireless access points broadcast in the 2.4GHz or 5GHz range, which is legal for everyone within certain (low) power levels.

That means that you will need to get different hardware and a license (and pass training/exams for that in order to transmit with your more powerful custom hardware. For most of us this is not an option.

More practical options comes down to 'do not use wireless to go though walls'.

E.g.

1. use regular wired (always works, but you need a cable).
2. use regular wired via existing wall sockets (powerline).
3. use either of the two above and connect to a second wireless access point in the cellar (still using wired/powerline to get the data to the cellar).
4. Be very lucky and play with the antenna's. Since a wireless signal is often broadcast in a plane this might boost signals upstairs and downstairs, but at a cost to the signal strenght on the original floor.