The History of Wi-Fi

Abstract

Wi-Fi (wireless local area networking using the IEEE 802.11 family of standards) went from a niche industrial technology to a global utility in under a decade. The standard that nobody expected to dominate, its early versions were slow and short-range; its later versions provide multi-gigabit throughput across entire buildings. Wi-Fi eliminated the cable from computing’s last mile, created the laptop culture that defined twenty-first century knowledge work, and established the wireless access point as a mandatory piece of consumer infrastructure. It is also, legally, partly the invention of an Australian research organization that holds the fundamental patent on key aspects of 802.11 and collected over $430 million in royalties.

Before 802.11: Industrial Wireless and the FCC

Wireless data networking existed before Wi-Fi, but not as a consumer product. ALOHAnet (University of Hawaii, 1971) transmitted data by radio across the Hawaiian islands and established the packet-collision-avoidance techniques that Ethernet later adapted for wired networks. Industrial wireless systems used licensed frequencies and proprietary protocols; they were expensive, slow, and specialized.

The regulatory foundation for consumer wireless networking was the FCC’s 1985 decision to allow unlicensed use of the ISM bands, Industrial, Scientific, and Medical radio frequencies (900 MHz, 2.4 GHz, and 5.8 GHz), for low-power devices. The ISM bands were already used by microwave ovens, garage door openers, and cordless phones; the FCC’s decision allowed spread-spectrum wireless data devices without a license, provided they used spread-spectrum modulation (making them more resistant to interference from and with other devices).

The ISM band opening created the regulatory space for wireless networking. The technical challenge was building systems that worked reliably in the interference-prone shared spectrum.

The 802.11 Committee and the 1997 Standard

The IEEE 802.11 committee formed in 1990 to develop wireless LAN standards. Vic Hayes, a Dutch engineer at NCR who would later be called the “father of Wi-Fi,” chaired the committee for its most formative decade (1990–2000). Hayes’s leadership of the standardization process (navigating competing technical proposals, vendor interests, and the inherent difficulty of wireless medium access) produced a workable standard.

The original IEEE 802.11 standard was published in 1997. It supported data rates of 1 and 2 Mbps using either frequency-hopping spread spectrum or direct-sequence spread spectrum in the 2.4 GHz band. The range was limited (typically 30 meters indoors) and the speed was slower than 10 Mbps Ethernet. Adoption was minimal.

802.11b: Mass Market Arrival (1999)

The market-making version was IEEE 802.11b, published in September 1999. The “b” amendment increased maximum data rates to 11 Mbps using Direct Sequence Spread Spectrum (DSSS) with Complementary Code Keying (CCK) modulation. Range improved to approximately 100 meters outdoors, 30 meters indoors through walls.

The timing coincided with several enabling factors: laptop computers were becoming widespread; broadband internet connections were arriving in homes and offices; and the 802.11b chip-set cost had fallen enough for consumer products.

The Wi-Fi Alliance, formed in 1999 (originally as the Wireless Ethernet Compatibility Alliance, WECA), provided certification that devices from different manufacturers would interoperate. The “Wi-Fi” brand name (coined by the marketing firm Interbrand, not by any technical consideration) was applied to certified 802.11b products. The name suggested a parallel to “Hi-Fi” (high fidelity audio) that was loose enough to generate some controversy but effective enough as a brand that it stuck.

Apple’s AirPort system (1999) was among the first mass-market consumer Wi-Fi products. Introduced at Macworld in July 1999, it debuted with the iBook (the first mainstream laptop designed for integrated wireless networking) paired with the AirPort Base Station router. Apple’s decision to offer Wi-Fi as a built-in feature of consumer laptops drove mass adoption.

The CSIRO Patent

The most legally consequential aspect of Wi-Fi’s history is a patent held by CSIRO (Commonwealth Scientific and Industrial Research Organisation), the Australian federal research agency.

CSIRO researchers (principally John O’Sullivan, Terry Percival, Diet Ostry, Graham Daniels, and John Deane) had been working on a technique for reducing multipath interference in wireless signals in the late 1980s. Multipath interference occurs when a transmitted signal arrives at the receiver via multiple paths (direct and reflected off walls and furniture), causing interference between the copies. The CSIRO technique used OFDM (Orthogonal Frequency Division Multiplexing) (splitting the data stream across multiple narrowband subcarriers) combined with a specific algorithm for handling the multipath copies.

CSIRO filed a patent application in 1992 and received US Patent 5,487,069 in 1996. The patent covered key aspects of the radio transmission technique used in 802.11a, 802.11g, and all subsequent Wi-Fi standards that use OFDM.

When 802.11a and 802.11g adopted OFDM (a superior modulation technique to the CCK used in 802.11b), CSIRO’s patent applied. CSIRO initially licensed the patent on reasonable terms; when manufacturers refused licensing, CSIRO sued. The resulting litigation and settlements (with Hewlett-Packard, Intel, Dell, Microsoft, T-Mobile, and others) generated $430 million in royalty revenue for CSIRO by 2012, making the OFDM patent one of the most valuable in technology history.

The case highlighted the structural problem of standards-essential patents: companies that have intellectual property incorporated into a mandatory standard can demand royalties from every device that complies with the standard, regardless of whether they participated in the standardization process.

Evolution: 802.11a through Wi-Fi 6

The 802.11 standard has evolved continuously, with each amendment increasing throughput, range, and efficiency:

Generation	Standard	Year	Max Speed	Notes
–	802.11b	1999	11 Mbps	2.4 GHz; first mass market
Wi-Fi 3	802.11g	2003	54 Mbps	2.4 GHz + OFDM
Wi-Fi 4	802.11n	2009	600 Mbps	2.4/5 GHz + MIMO
Wi-Fi 5	802.11ac	2013	6.9 Gbps	5 GHz; multi-user MIMO
Wi-Fi 6	802.11ax	2021	9.6 Gbps	2.4/5/6 GHz; dense environments
Wi-Fi 7	802.11be	2024	46 Gbps	Multi-link operation

MIMO (Multiple Input Multiple Output): Using multiple antennas at both transmitter and receiver to transmit multiple data streams simultaneously over the same channel. 802.11n introduced MIMO and dramatically increased throughput; 802.11ac introduced multi-user MIMO, allowing simultaneous transmission to multiple devices.

OFDMA (Orthogonal Frequency Division Multiple Access): 802.11ax’s introduction of OFDMA allows the access point to simultaneously serve multiple devices on different subcarriers of the same channel, dramatically improving performance in dense environments (airports, stadiums, offices).

WEP: The Security That Lasted Two Years

The original 1997 standard shipped with an encryption scheme called WEP (Wired Equivalent Privacy). The name stated the modest goal: make wireless as private as plugging into a wired LAN, no more. It fell short of even that.

WEP used the RC4 stream cipher (designed by Ron Rivest) with a 40-bit key, later extended to 104-bit and marketed as “128-bit” by counting a 24-bit initialization vector (IV) prepended in the clear. That IV was the flaw. At 24 bits it repeats after roughly 16 million frames, and a busy access point cycles through them in hours. RC4 combined with a repeating, visible IV leaks information about the key with every reused value.

In August 2001, Scott Fluhrer, Itsik Mantin, and Adi Shamir (the S in RSA) published “Weaknesses in the Key Scheduling Algorithm of RC4,” showing that certain IVs reveal key bytes directly. The attack was passive: capture enough traffic and the key falls out. Within weeks Adam Stubblefield, John Ioannidis, and Aviel Rubin at AT&T Labs implemented it against a real network. Tools followed fast (AirSnort and WEPCrack in 2001, then the aircrack family), turning a research paper into a point-and-click operation. A key that was supposed to protect a network for its lifetime could be recovered from a laptop in a parking lot.

The attacks only got faster. The 2007 PTW method (Erik Tews, Ralf-Philipp Weinmann, Andrei Pyshkin) recovered a 104-bit key from about 40,000 captured packets, achievable in under a minute once the attacker injected traffic to generate them. The same year the FBI demonstrated breaking a WEP network in roughly three minutes with public tools.

The industry response came in two steps. The Wi-Fi Alliance shipped WPA in 2003 as an emergency patch, keeping RC4 for hardware compatibility but wrapping it in TKIP with per-packet keys. The real fix was WPA2, based on the IEEE 802.11i amendment ratified in 2004, which replaced RC4 with AES in CCMP mode. The same 802.11i amendment formally deprecated WEP-40 and WEP-104. WEP had been part of the standard for seven years and cryptographically dead for the last three of them.

Misconception: WEP was broken because 40-bit keys were too short

Longer keys did not save WEP. The 104-bit “128-bit WEP” was broken by the same key-scheduling attacks as the 40-bit version, because the weakness was in the RC4 IV construction, not the key length. Adding key bits to a structurally broken protocol bought nothing. See Myths and Misconceptions.

Wi-Fi as Infrastructure

Wi-Fi has transitioned from a convenience to a utility. The absence of Wi-Fi in a hotel, café, airport, or office is now treated as an infrastructure failure, not a feature absence. The global installed base of Wi-Fi access points is measured in the billions; consumer routers provide Wi-Fi as a matter of course, and the expectation of universal wireless connectivity shapes architectural design, urban planning, and business operations.

The laptop (which became the dominant form of personal computing in the 2000s and 2010s) is substantially a product of Wi-Fi: a laptop without wireless networking is a less mobile desktop. The coffee shop culture of knowledge workers, the wireless hospital that doesn’t trip over cables, the factory floor with tablet-based inventory systems, all depend on the reliable short-range wireless connectivity that 802.11 standardized.