Orthogonal Frequency Division Multiple Access is like carpooling for the air. Instead of one transmitter using all subcarriers for a whole frame, the access point assigns subsets of subcarriers called resource units to different clients at the same time. Uplink and downlink versions let multiple devices send or receive simultaneously. This reduces contention and cuts latency, especially for many small packets from Internet of Things devices or phones.
Target wake time coordinates when battery powered devices wake up to send or receive. The access point and client agree on a schedule. The client sleeps for longer stretches, then wakes just in time to grab its resource unit. This saves power and reduces channel contention because many devices avoid trying to talk at random moments.
Multi link operation in Wi‑Fi seven lets a device use more than one link at the same time. A phone could transmit on five gigahertz and six gigahertz concurrently, splitting traffic for redundancy or speed. If one band becomes noisy, the system can steer packets to the cleaner link in real time. Combined with puncturing, which lets devices avoid small slices of interference inside a wide channel, the result is high reliability in busy apartments and offices.
Security evolved alongside speed. The original Wired Equivalent Privacy used a stream cipher with a weak key schedule and short initialization vectors. It failed under practical attacks. The fix was Wi‑Fi Protected Access. The first version improved key mixing and added message integrity, but it still relied on the old cipher for privacy. WPA two introduced the Advanced Encryption Standard in Counter Mode with Cipher Block Chaining Message Authentication Code Protocol. WPA three brought Simultaneous Authentication of Equals, which resists offline password guessing, and forward secrecy. Enterprise modes integrate with authentication servers to assign unique keys per user. Modern Wi‑Fi also uses protected management frames to keep attackers from spoofing disassociations or deauthentications.
Now that we have a sense of how it works, who made it work? The answer involves decades of physics, engineering, and standardization. The physical foundation traces back to James Clerk Maxwell in the nineteenth century, who unified electricity and magnetism to predict electromagnetic waves. Heinrich Hertz soon produced and detected those waves in the lab. Guglielmo Marconi and others built radio systems for communication. Those early radios were analog and narrowband.
The specific idea of using many subcarriers to resist echoes appeared in the nineteen sixties, with work on multitone modulation and Orthogonal Frequency Division Multiplexing by researchers in the United States and Europe. Later, better digital signal processors made the math practical. The spread spectrum techniques used in early Wi‑Fi at two point four gigahertz had an unusual popular history. Hedy Lamarr, an actor, and George Antheil, a composer, patented frequency hopping in nineteen forty two to prevent jamming of torpedoes. Their idea used synchronized hopping across many frequencies, which foreshadowed later spread spectrum systems. While their patent expired before direct use in modern Wi‑Fi, spread spectrum principles influenced the unlicensed band rules and early Wi‑Fi modes, and their story highlights how unexpected contributors can shape foundational ideas.
By the early nineteen nineties, consumer wireless networking needed a common language. The Institute of Electrical and Electronics Engineers formed a working group named eight zero two point eleven. Engineers from companies and universities argued over modulation, coding, and medium access. In nineteen ninety seven, the first standard was published. It defined rates of one and two megabits per second using a simple spread spectrum method at two point four gigahertz. It worked, but it was limited.
Two paths emerged to boost speeds. One was high rate direct sequence spread spectrum, which pushed to eleven megabits per second. The other was Orthogonal Frequency Division Multiplexing at five gigahertz, which promised much higher speeds and better performance in multipath environments. The Wi‑Fi Alliance, an industry consortium, formed to ensure that devices from different companies could interoperate and to certify products. The Alliance also popularized the term Wi‑Fi as a consumer friendly brand.
A breakthrough came with a team at the Commonwealth Scientific and Industrial Research Organisation in Australia. In the nineteen nineties, John O’Sullivan and colleagues, including Terence Percival, Graham Daniels, Diet Ostry, and John Deane, worked on methods to handle reflections in indoor environments using Orthogonal Frequency Division Multiplexing. Their techniques improved error correction and symbol timing, making indoor multicarrier communication robust. CSIRO later enforced patents related to these inventions and secured licensing deals with major companies. Their work is one pillar of practical Wi‑Fi.
Another important contributor was the group that developed Multiple Input Multiple Output theory. In the mid nineteen nineties, Arogyaswami Paulraj and Thomas Kailath, along with researchers like Gerard Foschini and Ilya Telatar, showed how multiple antennas could increase capacity without more bandwidth by creating parallel spatial channels. These ideas moved from theory to hardware as chipmakers integrated more radio chains and as standards bodies defined how to signal and calibrate beamforming.
No single person invented Wi‑Fi. It is a federation of inventions. The medium access rules draw from older Ethernet collision avoidance concepts. The radio pieces borrow from the broader digital communications field. The standardization and negotiation come from dozens of working group meetings, ballots, and drafts. The Wi‑Fi Alliance drove interoperability and branding. Universities, companies, and government labs contributed ideas and patents. If you want names to remember, note John O’Sullivan and the CSIRO team for indoor Orthogonal Frequency Division Multiplexing, Hedy Lamarr and George Antheil for early spread spectrum inspiration, Arogyaswami Paulraj and colleagues for Multiple Input Multiple Output, and the Wi‑Fi Alliance for making sure your laptop can talk to a router from any vendor.
Now, return to the message you sent. Your laptop built the frame, won time on the air, and transmitted. The access point received it, verified the checksum, and acknowledged. Then it passed the packet from the wireless interface into its network processor. If the destination is on the internet, the access point performs network address translation, rewrites headers, and forwards the packet over Ethernet or a fiber uplink to your internet provider. Many more hops and routers move it to a server. The reply travels back. Inside your home, the access point transmits the downlink frame to your laptop. If the access point supports downlink Orthogonal Frequency Division Multiple Access, it may schedule your reply alongside small frames for a smart plug and a thermostat, each using a different resource unit. The laptop’s radio watches the preamble, decodes the header to learn the modulation and coding rate, then demodulates the payload. The network stack reassembles things, checks the Transmission Control Protocol sequence number, and flags the application that the reply arrives. All of this happens in a few tens of milliseconds.
