Editing ALOHAnet (section)

== History ==

One of the early computer networking designs, development of the ALOHA network was begun in September 1968 at the University of Hawaii under the leadership of [[Norman Abramson]] and [[Franklin Kuo]], along with Thomas Gaarder, Shu Lin, [[Wesley Peterson]] and Edward ("Ned") Weldon. The goal was to use low-cost commercial radio equipment to connect users on [[Oahu]] and the other Hawaiian islands with a central time-sharing computer on the main Oahu campus. The first packet broadcasting unit went into operation in June 1971. Terminals were connected to a special purpose ''terminal connection unit'' using [[RS-232]] at {{nowrap|9600 bit/s}}.<ref>{{Cite journal |last=Abramson |first=Norman |author-link=Norman Abramson |date=Mar 1985 |title=Development of the ALOHANET |journal=IEEE Transactions on Information Theory |volume=31 |issue=2 |pages=119–123 |bibcode=1985ITIT...31..119A |doi=10.1109/TIT.1985.1057021}}</ref>

ALOHA was originally a [[contrived acronym]] standing for Additive Links On-line Hawaii Area.<ref>{{Cite book |last1=Kamins |first1=Robert M. |url={{google books|plainurl=yes|id=Q8jyzfUmFvgC}} |title=Måalamalama: A History of the University of Hawai'i |last2=Potter |first2=Robert E. |publisher=University of Hawaii Press |year=1998 |isbn=9780824820060 |page=159 |access-date=August 2, 2015}}</ref>

The original version of ALOHA used two distinct frequencies in a hub configuration, with the hub machine broadcasting packets to everyone on the ''outbound'' channel, and the various client machines sending data packets to the hub on the ''inbound'' channel. If data was received correctly at the hub, a short acknowledgment packet was sent to the client; if an acknowledgment was not received by a client machine after a short wait time, it would automatically retransmit the data packet after waiting a randomly selected time interval. This acknowledgment mechanism was used to detect and correct for [[Collision (telecommunications)|collisions]] created when two client machines both attempted to send a packet at the same time.

ALOHAnet's primary importance was its use of a shared medium for client transmissions. Unlike the [[ARPANET]] where each node could only talk to a single node at the other end of a wire or satellite circuit, in ALOHAnet all client nodes communicated with the hub on the same frequency. This meant that some sort of mechanism was needed to control who could talk at what time. The ALOHAnet solution was to allow each client to send its data without controlling when it was sent, and implementing an acknowledgment/retransmission scheme to deal with collisions. This approach radically reduced the complexity of the protocol and the networking hardware, since nodes do not need to negotiate ''who'' is allowed to speak.

This solution became known as a pure ALOHA, or random-access channel, and was the basis for subsequent [[Ethernet]] development and later [[Wi-Fi]] networks.<ref name=abramson2 /> Various versions of the ALOHA protocol (such as Slotted ALOHA) also appeared later in [[satellite communications]], and were used in wireless data networks such as [[DataTac|ARDIS]], [[Mobitex]], [[CDPD]], and [[GSM]].

The Aloha network introduced the mechanism of randomized multiple access, which resolved device transmission collisions by transmitting a packet immediately if no acknowledgement is present, and if no acknowledgment was received, the transmission was repeated after a random waiting time.<ref>{{Cite book|title=Communication Networks: A Concise Introduction|last1=Walrand|first1=Jean|last2=Parekh|first2=Shyam|publisher=Morgan & Claypool Publishers series|year=2010|isbn=9781608450947|location=University of California, Berkeley|pages=28–29}}</ref> The probability distribution of this random waiting time for retransmission of a packet that has not been acknowledged as received is critically important for the stability of Aloha-type communication systems. The average waiting time for retransmission is typically shorter than the average time for generation of a new packet from the same client node, but it should not be allowed to be so short as to compromise the stability of the network, causing a collapse in its overall throughput.<ref>Carleial, A. B. and Hellman, M. E. (1975) "Bistable Behavior of Aloha-Type Systems", IEEE Transactions on Communications, vol. COM-23, pages 401-410, April 1975.</ref>

Also important was ALOHAnet's use of the outgoing hub channel to broadcast packets directly to all clients on a second shared frequency and using an address in each packet to allow selective receipt at each client node.<ref name=binder1 /> Separate frequencies were used for incoming and outgoing communications to the hub so that devices could receive acknowledgments regardless of transmissions.