Wi-Fi

• 1. Introduction

  ›

  Wi-Fi is the invisible infrastructure that carries more than 3.05 billion devices into the world's networks every single year. It fills coffee shops, hospital wards, airport terminals, and remote mountain villages in Peru with a signal most people never think about until it disappears. How did a wireless standard developed for supermarket cashier systems become the connective tissue of modern life? What makes that familiar yin-yang logo a legal promise rather than just a decorative sticker? And why do Australia, the United States, and the Netherlands each claim they invented it? The answers involve a Dutch engineer who chaired a standards committee for a decade, a team of Australian scientists racing to build a prototype in a radiophysics laboratory, and a brand consultancy hired to make a string of letters catchier than "IEEE 802.11b Direct Sequence."

• 2. Origins and the Patent Wars

  ›

  In 1991, inside a cashier-systems project in the Dutch city of Nieuwegein, engineers at NCR Corporation and AT&T built something called WaveLAN. It was not yet Wi-Fi by name or standard, but it was the direct ancestor of 802.11. NCR's Vic Hayes, who would hold the chair of the IEEE 802.11 working group for ten years, partnered with Bell Labs engineer Bruce Tuch to bring the technology to the Institute of Electrical and Electronics Engineers. Together they shaped the initial 802.11b and 802.11a specifications. Both have since been inducted into the Wi-Fi NOW Hall of Fame.

  A parallel effort was underway on the other side of the planet. A team of scientists in Australia had begun working on wireless LAN technology in 1989, and by 1992 researchers from the Radiophysics Division of the CSIRO had built a prototype test bed. That team was led by John O'Sullivan. The CSIRO lodged a patent for Wi-Fi in 1992, the same year the prototype appeared. Years later that patent would become the center of a protracted legal battle: in 2009 the CSIRO was awarded $200 million after a settlement with 14 technology companies, and a further $220 million followed in 2012 after proceedings against 23 more. The 1985 ruling by the U.S. Federal Communications Commission, which opened parts of the ISM bands for unlicensed use, had created the regulatory soil in which all of this growth became possible.

  The first version of the 802.11 protocol was released in 1997, offering link speeds of up to 2 Mbit/s. Two years later, in 1999, the standard was updated as 802.11b and pushed speeds to 11 Mbit/s. Also in 1999, the Wi-Fi Alliance formed as a trade association to hold the Wi-Fi trademark and enforce it across the industry. In 2016, the CSIRO's WLAN prototype test bed was selected as Australia's contribution to the exhibition A History of the World in 100 Objects, held at the National Museum of Australia.

• 3. The Name Behind the Signal

  ›

  The phrase "IEEE 802.11b Direct Sequence" was never going to sell routers. The Wi-Fi Alliance recognized this and hired Interbrand, a brand-consulting firm, to invent a name that people would actually want to say. The result, commercially in use as early as August 1999, came from a list of ten candidate names Interbrand proposed. Phil Belanger, a founding member of the Wi-Fi Alliance, confirmed that Wi-Fi was chosen from that list. Interbrand also designed the yin-yang logo that signals product interoperability.

  The name has attracted a persistent myth: that "Wi-Fi" stands for "Wireless Fidelity." It does not. The Wi-Fi Alliance did use the advertising slogan "The Standard for Wireless Fidelity" briefly after the brand launched, and some publications even called the organization the "Wireless Fidelity Alliance Inc.," which helped the misconception spread. The IEEE compounded the confusion by stating on its own website that "WiFi is a short name for Wireless Fidelity." In reality, the name was partly chosen because it sounds similar to Hi-Fi, and Interbrand hoped consumers would assume this wireless protocol delivered high quality because of that sonic resemblance. Spellings such as WiFi, Wifi, and wifi are common but none is approved by the Wi-Fi Alliance. Only Wi-Fi is official.

• 4. Apple and the Consumer Breakthrough

  ›

  By 1999, the 802.11 standard existed and the Alliance held the trademark, but Wi-Fi had not yet reached ordinary households. The major commercial breakthrough came when Apple Inc. adopted Wi-Fi for its iBook series of laptops. The iBook became the first mass consumer product to offer Wi-Fi network connectivity, which Apple marketed under its own brand name: AirPort. The collaboration behind that launch included Vic Hayes and Bruce Tuch, the engineers who had shaped the original IEEE specifications, along with Cees Links, Rich McGinn, and others from Lucent.

  Carnegie Mellon University had built the first campus-wide wireless Internet network, called Wireless Andrew, at its Pittsburgh campus in 1993, years before the Wi-Fi brand even existed. The iBook launch in 1999 pushed the technology into retail stores and living rooms. Separately, in 2000, a group of Australian scientists called Radiata, connected to the CSIRO, became the first to use the 802.11a standard on chips connected to a Wi-Fi network. From that point forward, chipset prices fell steadily, and manufacturers began building wireless network adapters into most laptops from the early 2000s onward.

• 5. How the Signal Travels

  ›

  Wi-Fi stations communicate by exchanging data packets over radio channels, using modulation and demodulation of carrier waves. The 802.11 standard provides frequency ranges at 900 MHz, 2.4 GHz, 3.6 GHz, 4.9 GHz, 5 GHz, 6 GHz, and 60 GHz. Most everyday Wi-Fi runs on the 2.4 GHz UHF and 5 GHz SHF bands, with newer generations adding the 6 GHz band. Within a band, channels are numbered at 5 MHz spacing, though transmitters generally occupy at least 20 MHz.

  The scheme that governs how stations share channels is called carrier-sense multiple access with collision avoidance, or CSMA/CA. Stations listen before transmitting and delay if the channel seems occupied, but two stations can still sense silence simultaneously and collide. When a collision occurs, the transmitted data is corrupted and stations must re-transmit, reducing throughput.

  Range and signal strength are shaped by the physical environment. An access point has a range of roughly 20 metres indoors; some claim up to 150 metres outdoors. Metallic structures, marble, and even water in vegetation absorb or deflect signals heavily. Running non-standard equipment at extreme distances, Ermanno Pietrosemoli and EsLaRed of Venezuela transferred about 3 MB/sec between mountain tops 382 km apart in June 2007. The Swedish National Space Agency pushed data 420 km using 6 watt amplifiers to reach an overhead stratospheric balloon. Wi-Fi 5, which operates exclusively on the 5 GHz band, is capable of multi-station throughput of at least 1 gigabit per second and single-station throughput of at least 500 Mbit/s. Some versions of Wi-Fi, running on suitable hardware at close range, can reach speeds of 23 Gbit/s.

• 6. Security and Its Failures

  ›

  Wired Equivalent Privacy, the first encryption standard built for Wi-Fi, was designed to deter casual snooping. Tools such as AirSnort and Aircrack-ng can quickly recover WEP keys, and the protocol is no longer considered secure. The Wi-Fi Alliance responded by approving Wi-Fi Protected Access, or WPA, which uses TKIP and was designed to run on older hardware through firmware upgrades. WPA carries known vulnerabilities as well.

  WPA2, using Advanced Encryption Standard, arrived in 2004 and is supported by most newer Wi-Fi devices. In 2017, a flaw in the WPA2 protocol was discovered that allowed a key replay attack, publicly named KRACK. Separately, a feature added to Wi-Fi in 2007 called Wi-Fi Protected Setup introduced its own vulnerability, bypassing WPA and WPA2 security entirely; the only remedy was to disable Wi-Fi Protected Setup, which is not always an option on consumer hardware. WPA3 was announced in 2018 as WPA2's replacement, with its rollout beginning on the 26th of June that year.

  Beyond encryption failures, Wi-Fi access points typically default to an open, encryption-free mode. On such networks, any device within range can monitor and record traffic, including personal information. The practice of passively logging and mapping other people's access points became known as wardriving. A related proposal called warchalking suggested leaving graffiti to describe available open services nearby.

• 7. Wi-Fi Beyond the Router

  ›

  In 2004, Mysore, in India, became the country's first Wi-Fi-enabled city, with a company called WiFiyNet setting up hotspots across the whole city and a few surrounding villages. In 2005, St. Cloud, Florida, and Sunnyvale, California, became the first cities in the United States to offer citywide free Wi-Fi, through a provider called MetroFi. Minneapolis generated $1.2 million in annual profit for its Wi-Fi provider.

  New York City's LinkNYC project, announced in 2014, began installing digital kiosks to replace old phone booths. Installation started in late 2015, and the city planned to deploy more than seven thousand kiosks, aiming to create the largest and fastest public, government-operated Wi-Fi network in the world. Seoul was planning to provide free Internet access at more than 10,000 locations, with KT, LG Telecom, and SK Telecom set to invest $44 million.

  In rural Peru in 2007, engineers erected a 450 km network between Cabo Pantoja and Iquitos, powered entirely by solar panels, connecting the central hospital in Iquitos to 15 medical outposts for remote diagnosis. In June 2014, Texas Instruments introduced the SimpleLink CC3200, the first ARM Cortex-M4 microcontroller with an onboard dedicated Wi-Fi chip, making it practical to build Wi-Fi connectivity into single-chip, low-cost embedded devices. A portable ECG monitor that sends patient data over Wi-Fi to a remote hospital is one example of what that shift made possible.