WiFi Signal Strength Calculator

Calculating WiFi signal strength is the scientific process of predicting how invisible radio waves degrade as they travel through physical space from a transmitter to a receiver. By utilizing mathematical models like Free-Space Path Loss (FSPL) and comprehensive link budgets, network engineers can accurately estimate wireless range, optimize antenna placements, and ensure reliable connectivity before a single piece of hardware is installed. Understanding these calculations empowers you to transform wireless networking from a frustrating guessing game into a precise, predictable, and highly measurable engineering discipline.

What It Is and Why It Matters

At its absolute core, WiFi is simply two-way radio communication operating at specific, incredibly high frequencies—most commonly 2.4 gigahertz (GHz), 5 GHz, and increasingly 6 GHz. When a wireless router transmits a signal, it pushes electrical energy into an antenna, which converts that energy into electromagnetic waves that radiate outward at the speed of light. However, as these waves expand outward in a spherical pattern, the energy is distributed over an increasingly larger area, causing the signal strength to diminish rapidly over distance. A WiFi signal strength calculation is the mathematical method used to quantify this exact degradation, allowing us to predict exactly how much of the original transmitted energy will successfully reach the receiving device, such as a smartphone or a laptop. This calculation takes into account the initial transmit power, the amplifying effects of the antennas on both ends, and the inevitable energy loss that occurs as the signal traverses the physical environment.

Understanding and applying these calculations matters profoundly because wireless communication is inherently fragile and highly susceptible to the laws of physics. Without a mathematical understanding of signal strength, deploying a wireless network relies entirely on trial and error, which is both expensive and highly inefficient. For a homeowner, a lack of understanding leads to frustrating "dead zones," buffering videos, and dropped video calls, usually resulting in the misguided purchase of unnecessary network extenders. For enterprise network engineers, the stakes are exponentially higher; deploying wireless coverage for a 500,000-square-foot automated warehouse or a 50,000-seat sports stadium requires exact calculations to ensure every single square inch has adequate coverage without overlapping signals causing destructive interference. By calculating the link budget and free-space path loss beforehand, you can determine exactly what type of equipment to buy, what antennas to use, and where to mount them to achieve flawless connectivity. Ultimately, this mathematical approach bridges the gap between the invisible physics of radio frequency and the tangible, everyday necessity of reliable internet access.

History and Origin

The foundation of calculating wireless signal strength begins long before the invention of WiFi, tracing its roots back to the late 19th century. In 1865, Scottish physicist James Clerk Maxwell published his groundbreaking theory of electromagnetism, mathematically predicting the existence of radio waves. Two decades later, in 1887, German physicist Heinrich Hertz successfully proved Maxwell's theories by building the first apparatus to transmit and receive radio waves in a laboratory. However, the specific mathematical framework we use today to calculate signal loss over distance was formalized by Harald T. Friis, a Danish-American radio engineer working at Bell Labs. In 1946, Friis published a seminal paper detailing what is now universally known as the Friis Transmission Equation. This elegant formula provided the world's first standardized method for calculating the power received by one antenna from another under idealized, unobstructed conditions, fundamentally birthing the concept of Free-Space Path Loss (FSPL).

The Friis equation remained a cornerstone of radio and television broadcasting for decades, but it found a completely new application in the late 1990s with the dawn of consumer wireless networking. In 1997, the Institute of Electrical and Electronics Engineers (IEEE) released the original 802.11 standard, championed by a Dutch engineer named Vic Hayes, who is now widely recognized as the "Father of WiFi." This early standard allowed computers to communicate wirelessly at a mere 2 megabits per second using the 2.4 GHz industrial, scientific, and medical (ISM) radio band. Because these early wireless cards operated on strict power limitations imposed by regulatory bodies like the Federal Communications Commission (FCC) in the United States, engineers had to rely heavily on Friis's mathematical principles to maximize range and reliability. As WiFi evolved through subsequent generations—802.11b, 802.11g, 802.11n (WiFi 4), 802.11ac (WiFi 5), 802.11ax (WiFi 6), and 802.11be (WiFi 7)—the frequencies grew higher and the modulations grew more complex, making the foundational calculations of signal strength and path loss more critical than ever before. Today, every enterprise wireless deployment tool and predictive modeling software relies on the exact same physical principles established by Harald Friis over 75 years ago.

Key Concepts and Terminology

To accurately calculate and understand wireless signal strength, you must first master the specific vocabulary and units of measurement used in radio frequency (RF) engineering. The most fundamental unit is the decibel (dB), which is a logarithmic unit used to express the ratio of two values of a physical quantity, often power or intensity. Because wireless signals degrade exponentially rather than linearly, using a linear scale like milliwatts (mW) becomes incredibly cumbersome; a signal might drop from 100 mW to 0.000000001 mW over a short distance. The decibel scale compresses this massive range into manageable, easy-to-read numbers. Furthermore, RF engineers use a specific variation called dBm (decibel-milliwatts), which is an absolute measurement of power referenced to exactly one milliwatt. On this scale, 0 dBm equals precisely 1 milliwatt of power. Because it is a logarithmic scale, the "Rule of 3 and 10" applies: every increase of 3 dB doubles the power, and every increase of 10 dB multiplies the power by ten. Therefore, 20 dBm is 100 milliwatts, and 30 dBm is 1,000 milliwatts (or 1 Watt). Conversely, negative dBm values represent power levels smaller than one milliwatt, which is how we measure the incredibly faint signals received by your smartphone.

Beyond power measurements, you must understand the components of the physical link, starting with Transmit Power (Tx), which is the raw electrical energy pushed by the router's internal radio. This energy is then shaped and directed by the antenna, which introduces Antenna Gain, measured in dBi (decibels relative to an isotropic radiator). An isotropic radiator is a theoretical, perfect antenna that radiates energy equally in all directions, forming a perfect sphere. Real-world antennas, like the "rubber duck" antennas on a router, squeeze this spherical energy into a flatter, donut-like shape, pushing the signal further horizontally at the expense of vertical coverage; this "squeezing" is the gain. As the signal travels through the air, it experiences Free-Space Path Loss (FSPL), which is the natural spreading and weakening of the electromagnetic wave. Finally, the signal arrives at the client device, which has a specific Receiver Sensitivity (Rx Sensitivity). This is the absolute minimum signal strength, usually expressed as a highly negative number like -82 dBm, that the device's radio requires to successfully decode the data. The mathematical sum of all these gains and losses is known as the Link Budget, which tells you exactly how much signal margin you have to ensure a stable connection.

How It Works — Step by Step

Calculating WiFi signal strength requires building a complete Link Budget, which is a simple accounting equation that adds up all the power gains and subtracts all the power losses from the transmitter to the receiver. The mathematical formula for a link budget is: Received Power ($P_{rx}$) = Transmit Power ($P_{tx}$) + Transmitter Antenna Gain ($G_{tx}$) - Transmitter Cable Loss ($L_{tx}$) - Free-Space Path Loss ($FSPL$) + Receiver Antenna Gain ($G_{rx}$) - Receiver Cable Loss ($L_{rx}$). To solve this, the most complex variable we must calculate first is the Free-Space Path Loss. The standard formula for FSPL in decibels is: $FSPL = 20 \log_{10}(d) + 20 \log_{10}(f) + 32.44$, where $d$ is the distance between the two antennas in kilometers, $f$ is the frequency of the wireless signal in megahertz (MHz), and 32.44 is a constant that aligns the mathematical units of kilometers and megahertz with the speed of light. Let us walk through a complete, realistic calculation to see how this works in practice.

Imagine you are setting up an outdoor WiFi link between two buildings using the 2.4 GHz frequency band, which operates specifically at 2400 MHz. The distance between the two buildings is exactly 0.5 kilometers (500 meters). Your transmitting router outputs a power of 20 dBm (100 milliwatts). You attach a directional antenna that provides a gain of 8 dBi, and the short cable connecting the router to the antenna loses 1 dB of signal. On the receiving building, you have an identical setup: an 8 dBi antenna and a 1 dB cable loss. First, we calculate the FSPL. We plug in our variables: $FSPL = 20 \log_{10}(0.5) + 20 \log_{10}(2400) + 32.44$. Step one: the base-10 logarithm of 0.5 is -0.301. Multiply that by 20 to get -6.02. Step two: the base-10 logarithm of 2400 is 3.380. Multiply that by 20 to get 67.60. Step three: add them all together. $FSPL = -6.02 + 67.60 + 32.44 = 94.02$ dB. This means the signal will lose just over 94 decibels of strength simply by traveling 500 meters through empty air.

Now that we have the FSPL, we can complete the Link Budget equation to find our exact Received Power ($P_{rx}$). We start with our Transmit Power (20 dBm). We add the Transmitter Antenna Gain (+8 dBi) and subtract the Transmitter Cable Loss (-1 dB). This gives us an Effective Isotropic Radiated Power (EIRP) of 27 dBm leaving the first building. Next, we subtract our calculated Free-Space Path Loss (-94.02 dB). The signal arriving at the second building's antenna is now extremely weak: 27 - 94.02 = -67.02 dBm. Finally, we add the Receiver Antenna Gain (+8 dBi) and subtract the Receiver Cable Loss (-1 dB). Our final equation is: -67.02 + 8 - 1 = -60.02 dBm. Therefore, the signal strength reaching the internal radio of the receiving router is exactly -60.02 dBm. Because most standard WiFi receivers require a signal of at least -75 dBm to maintain a fast, stable connection, our calculated result of -60 dBm gives us a highly reliable link with a comfortable 15 dB margin of safety.

Types, Variations, and Methods

While the Free-Space Path Loss (FSPL) model is the foundational bedrock of wireless calculations, it represents a mathematically perfect, unobstructed vacuum that rarely exists in real-world scenarios. Consequently, RF engineers have developed several variations and alternative mathematical models to account for different physical environments. The most common variation is the Log-Distance Path Loss Model, which introduces a variable known as the "path loss exponent" (often denoted as $n$). In a perfect vacuum, the path loss exponent is exactly 2.0, meaning the signal degrades purely based on the inverse-square law. However, in an indoor office environment with cubicles and drywall, the exponent might be increased to 3.0 or 3.5. In a dense industrial factory with heavy metal machinery, the exponent could rise to 4.5 or 5.0. By tweaking this single variable, engineers can use the base FSPL formula to generate highly accurate predictions for complex indoor environments without having to calculate the exact density of every single wall.

Another prominent method is the ITU Indoor Path Loss Model, developed by the International Telecommunication Union. This model specifically accounts for the extreme attenuation caused by signals passing through multiple floors of a multi-story building. The ITU formula adds a "floor penetration loss factor" which assigns a specific decibel penalty based on the frequency being used and the number of concrete or wood floors the signal must traverse. For example, a 2.4 GHz signal might lose 10 dB passing through the first floor, but the model mathematically accounts for the fact that subsequent floors might only add 4 or 5 dB of loss due to complex multipath reflections. Additionally, for long-distance outdoor links, engineers often use the Two-Ray Ground Reflection Model. Unlike pure FSPL which assumes the signal travels in a single straight line, the Two-Ray model calculates the primary line-of-sight signal and a secondary signal that bounces off the surface of the earth. Because the bounced signal travels a slightly longer distance, it arrives slightly out of phase with the primary signal, causing destructive interference. Understanding and choosing the correct variation of these models is what separates a basic estimate from a professional-grade wireless deployment.

Real-World Examples and Applications

To truly grasp the power of WiFi signal strength calculations, it is helpful to examine how they dictate the design of everyday technology. Consider a smart home deployment where a homeowner is upgrading from an older 2.4 GHz router to a modern 5 GHz system to achieve faster download speeds. The homeowner places the new router in the exact same location in the living room and is shocked to find that their smart security camera, located 25 meters away in the garage, suddenly drops offline. By applying the FSPL formula, the reason becomes mathematically obvious. At 25 meters, a 2.4 GHz signal (2400 MHz) experiences a free-space path loss of approximately 68.0 dB. However, a 5 GHz signal (specifically 5180 MHz) traveling that exact same 25-meter distance experiences a free-space path loss of 74.7 dB. Because the decibel scale is logarithmic, that seemingly small difference of 6.7 dB means the 5 GHz signal arrives at the garage with less than one-quarter of the raw power of the old 2.4 GHz signal. The math perfectly predicts the real-world failure, proving that higher frequencies simply cannot travel as far without specialized, high-gain antennas.

In an enterprise context, consider a Wireless Internet Service Provider (WISP) attempting to beam high-speed internet across a rural valley to a remote farmhouse located 10 kilometers away. They cannot run fiber optic cables, so they must use a wireless point-to-point link. The distance is massive, and using standard indoor WiFi equipment would fail instantly. By running a link budget calculation, the WISP engineers know that a 5 GHz signal over 10 kilometers will suffer a devastating 126.4 dB of Free-Space Path Loss. If they use a standard router transmitting at 20 dBm, the received signal would be -106 dBm—far below the noise floor and completely undetectable. To solve this, the math dictates their equipment purchases. They choose specialized outdoor radios that transmit at a much higher 27 dBm (500 mW). More importantly, they utilize massive, parabolic dish antennas on both the transmitting and receiving ends, each providing an enormous 30 dBi of gain. Re-running the link budget: 27 dBm (Tx) + 30 dBi (Tx Gain) - 126.4 dB (FSPL) + 30 dBi (Rx Gain) results in a received signal strength of -39.4 dBm. Thanks to the predictive power of the math, the engineers successfully design a link that provides incredibly fast, rock-solid internet across a 10-kilometer physical gap.

Common Mistakes and Misconceptions

One of the most pervasive misconceptions among beginners is the belief that "more transmit power is always better." When faced with a weak signal, a novice will often log into their router's administrative settings and crank the transmit power to the absolute maximum, or buy a massive aftermarket antenna. While this will indeed increase the link budget in the downstream direction (from the router to the smartphone), it completely ignores the upstream direction (from the smartphone back to the router). WiFi is a two-way conversation. A smartphone has a tiny battery, a microscopic internal antenna, and strict thermal limits; it typically transmits at a mere 10 to 14 dBm (10 to 25 milliwatts). If you boost your router to blast a signal 100 meters down the street, your phone will show "full bars" because it can "hear" the router perfectly. However, when the phone tries to reply, its weak 14 dBm signal dies long before it reaches the router. This creates a highly frustrating scenario known as an "asymmetric link," where the device shows a strong connection but web pages simply refuse to load.

Another profound mistake is misunderstanding the logarithmic nature of the decibel scale, specifically regarding negative dBm values. Because signal strength is measured in negative numbers, beginners frequently misinterpret the data, assuming that -80 dBm is a "bigger" and therefore better signal than -60 dBm. In reality, -60 dBm is significantly closer to zero, meaning it is a much stronger signal. Furthermore, because it is a logarithmic scale, people vastly underestimate the difference between these numbers. A user might look at a drop from -60 dBm to -70 dBm and assume they have only lost a small fraction of their signal. In truth, a 10 dB drop means the signal power has been divided by ten; 90% of the raw electrical energy has been lost. Failing to respect the 3 dB rule (doubling/halving of power) and the 10 dB rule (tenfold increase/decrease) leads to catastrophic miscalculations when estimating how far a signal will penetrate through walls or how much an upgraded antenna will actually improve performance.

Best Practices and Expert Strategies

Professional RF engineers never design a wireless network based solely on the raw mathematical output of a Free-Space Path Loss calculator; they always build in a strategic buffer known as a "Fade Margin." Even if your link budget calculation indicates that a signal will arrive at a perfectly acceptable -65 dBm, real-world conditions fluctuate constantly. A heavy rainstorm, a passing delivery truck, or a neighbor turning on a poorly shielded microwave oven can temporarily degrade the signal. Best practice dictates designing the network with a minimum Fade Margin of 10 to 15 decibels above the absolute minimum receiver sensitivity. For example, if a manufacturer states that a wireless bridge requires a -75 dBm signal to maintain its highest data rate, an expert will design the physical link, select the antennas, and set the transmit power to achieve a target received signal of -60 dBm. This 15 dB safety net ensures that even when environmental variables inevitably degrade the signal, the link remains stable and the end-user experiences zero interruption.

Another critical expert strategy involves accounting for the Fresnel Zone (pronounced freh-NEL), which is a three-dimensional, football-shaped elliptical area surrounding the visual line-of-sight between two antennas. A common beginner mistake is assuming that as long as you can see the receiving antenna with a pair of binoculars, the WiFi signal will travel perfectly. However, radio waves spread out as they travel. If the bottom of this invisible RF "football" grazes the ground, a building roof, or the canopy of a forest, the signal will reflect off those surfaces. These reflections arrive at the receiving antenna slightly out of phase with the main signal, causing destructive interference that can ruin a connection even if the math says the signal should be strong. Experts calculate the radius of the first Fresnel zone at the midpoint of the link and ensure that at least 60% of that radius is completely free of physical obstructions. This often requires mounting antennas on significantly taller poles or towers than a pure line-of-sight visual check would suggest, marrying the theoretical math of the signal calculator with the physical realities of the deployment site.

Edge Cases, Limitations, and Pitfalls

While Free-Space Path Loss calculators are indispensable tools, they possess distinct limitations that can lead to disastrous deployments if ignored. The most significant limitation is that FSPL assumes a perfect, obstacle-free vacuum. It does not account for the extreme attenuation caused by physical materials. For instance, calculating the signal strength of a router placed inside a modern environmentally-friendly building can be incredibly deceptive. Modern LEED-certified buildings often use specialized "Low-E" (low emissivity) glass windows. These windows are coated with a microscopic layer of metal designed to reflect solar radiation and reduce air conditioning costs. Unfortunately, this metal film is highly opaque to radio frequency waves. A mathematically perfect outdoor WiFi signal that should easily penetrate a standard glass window will be almost entirely reflected by Low-E glass, acting as a functional Faraday cage. If an engineer relies purely on an FSPL calculator without auditing the physical building materials, their deployment will fail completely.

Another major pitfall involves the concept of the "Noise Floor" in high-density environments. A signal strength calculator only tells you the absolute power of your specific signal arriving at the receiver; it tells you absolutely nothing about the other signals in the air. Imagine trying to have a conversation with a friend from 20 feet away. In a quiet library, your friend can hear you perfectly. In the middle of a packed, screaming rock concert, your friend cannot hear a word you say, even though you are speaking at the exact same volume. This is the concept of Signal-to-Noise Ratio (SNR). In dense urban environments, apartment complexes, or trade show floors, the 2.4 GHz and 5 GHz spectrums are flooded with interference from hundreds of other routers, Bluetooth devices, and wireless peripherals. This raises the background "noise floor" from a typical -95 dBm up to a highly disruptive -75 dBm. Even if your calculator proves you are delivering a respectable -80 dBm signal to your client, the connection will fail because your signal is buried beneath the ambient noise. Mathematical calculators must always be paired with an understanding of the local RF spectrum.

Industry Standards and Benchmarks

To make sense of the numbers generated by a signal strength calculator, you must compare them against established industry benchmarks. The Institute of Electrical and Electronics Engineers (IEEE) and leading enterprise networking companies like Cisco and Aruba have established rigorous standards for what constitutes "good" and "bad" Received Signal Strength Indicator (RSSI) values. At the very top end, a signal of -30 dBm is considered the absolute maximum; anything stronger than this (closer to zero) is actually detrimental. Placing a device inches away from a high-powered router can "deafen" the receiver, causing packet loss much like screaming directly into someone's ear. A signal between -40 dBm and -60 dBm is considered excellent, providing the maximum possible data rates and supporting complex, high-bandwidth modulations like 1024-QAM used in WiFi 6. In this range, users can seamlessly stream 4K video, transfer massive files, and experience zero latency.

The critical thresholds begin as the signal drops further. The universal industry benchmark for a reliable, enterprise-grade wireless network is -67 dBm. At -67 dBm, the connection is strong enough to support highly sensitive, real-time applications like Voice over IP (VoIP) phone calls and seamless roaming between different access points without dropping the connection. Once the signal drops to -70 dBm, the network is still functional for basic web browsing and email, but real-time voice and video will begin to stutter and experience jitter. At -80 dBm, the connection becomes highly unreliable; standard packets will frequently drop, causing the device to constantly retransmit data, which severely degrades battery life and slows down the entire network for other users. By -90 dBm, the signal has effectively merged with the background noise floor, and the device will completely disconnect. Professional network architects always calibrate their predictive calculators to ensure every square foot of a target coverage area meets the strict -67 dBm standard.

Comparisons with Alternatives

When planning a wireless network, relying on a mathematical WiFi signal strength calculator is just one of several available methodologies. The most common alternative used by consumers is the "Heuristic Method," which is a fancy term for relying on rules of thumb and manufacturer marketing claims. A consumer might buy a router because the box claims it "covers up to 2,500 square feet." This approach requires zero math and zero technical knowledge. However, the obvious con is that it is highly inaccurate. The manufacturer's claim is based on a completely empty, open-plan test facility. If the consumer places that router in a brick house with metal lath-and-plaster walls, the actual coverage might be less than 800 square feet. The heuristic approach is fast and easy, but it relies on blind faith and frequently results in the need to purchase additional, unplanned equipment to fix unexpected dead zones.

On the opposite end of the spectrum is the "Empirical Measurement" method, commonly known as an Active Site Survey. Instead of predicting the signal using math, an engineer temporarily mounts a real WiFi access point on a tripod at the deployment site. They then walk the entire facility carrying a specialized, highly calibrated measurement device (like an Ekahau Sidekick) that physically records the exact signal strength, interference, and noise floor at thousands of different data points. The software then generates a highly accurate, color-coded heat map of the actual coverage. The massive advantage of an active site survey is that it is 100% accurate; it inherently accounts for every single wall, window, and source of interference because it is measuring physical reality. The downside is that it is incredibly time-consuming, highly expensive, and requires the building to already be fully constructed. In professional environments, the two methods are used synergistically: mathematical calculators and predictive modeling are used to design the network while the building is still just a blueprint, and an empirical active site survey is performed after installation to validate that the real-world performance matches the mathematical predictions.

Frequently Asked Questions

What is a good WiFi signal strength for everyday use? For the vast majority of standard internet tasks, a received signal strength between -50 dBm and -67 dBm is considered excellent. In this range, you will experience maximum download speeds, stable connections, and flawless video streaming. If the signal drops to between -68 dBm and -75 dBm, the connection is still usable for basic web browsing and email, but you may notice occasional buffering or slower download speeds. Anything below -80 dBm is considered a weak and unreliable connection, likely resulting in frequent disconnects and high latency.

Why does my 5 GHz WiFi have significantly less range than my 2.4 GHz WiFi? This is a fundamental law of physics governed by the Free-Space Path Loss formula. Higher frequency waves, like 5 GHz, oscillate much faster than lower frequency waves like 2.4 GHz. Because they oscillate faster, they expend their energy more quickly as they travel through space and interact with the environment. Furthermore, higher frequency waves have shorter wavelengths, making them much worse at penetrating solid objects like drywall, wood, and furniture. While 5 GHz can carry significantly more data at faster speeds, the mathematical trade-off is a drastically reduced physical range.

How do walls and physical obstacles affect my signal strength calculation? The standard Free-Space Path Loss formula assumes the signal is traveling through a perfect, empty vacuum. In reality, every time a WiFi signal passes through a physical object, a portion of the radio energy is absorbed, reflected, or scattered. A standard interior drywall might reduce the signal by 3 to 4 dB, effectively halving the power. A heavy brick wall might reduce the signal by 12 to 15 dB, and solid concrete or metal can block the signal entirely. To calculate accurate indoor coverage, engineers must add these specific material attenuation penalties to their overall link budget calculation.

Can I just buy a bigger, higher-gain antenna to fix my weak WiFi? While upgrading to a higher-gain antenna will mathematically increase your link budget, it is not a magical cure-all and comes with significant trade-offs. Antenna gain does not create new electrical power; it simply reshapes the existing power. A high-gain omnidirectional antenna squashes the spherical signal into a flat pancake shape, pushing the signal further horizontally but drastically reducing vertical coverage. If you put a high-gain antenna on a router in a two-story house, you might get better range in the yard, but the bedrooms directly upstairs will likely lose their connection entirely.

What is the exact difference between dB, dBm, and dBi? These three terms are the most commonly confused metrics in wireless calculations. A simple decibel (dB) is a relative measurement used to express a change or a ratio, such as a cable losing 2 dB of signal or an amplifier adding 10 dB of power. The term dBm (decibel-milliwatt) is an absolute measurement of raw electrical power, where 0 dBm equals exactly 1 milliwatt. Finally, dBi (decibels relative to isotropic) is used exclusively to measure the directional gain of an antenna compared to a theoretical, perfect spherical antenna. You use dB to describe changes, dBm to describe power levels, and dBi to describe antennas.

How does the "noise floor" affect my calculated signal strength? Your calculated signal strength (e.g., -65 dBm) is only half of the equation; the other half is the background radio noise in your environment. The noise floor is the combined sum of all ambient radio frequency energy, including neighbor's routers, Bluetooth devices, and microwave ovens. If your noise floor is very quiet (-95 dBm), your -65 dBm signal provides a massive, 30 dB Signal-to-Noise Ratio (SNR), resulting in a perfect connection. If you are in a crowded apartment building and the noise floor rises to -70 dBm, that exact same -65 dBm signal now only has a 5 dB SNR, resulting in a terrible, constantly dropping connection despite the "strong" mathematical signal.