Wireless LANs

CSIRO’s wireless invention lies at the heart of what is now the most popular way to connect computers without wires. It is used in offices, public buildings, homes and coffee shops ‘ often called ‘WiFi Hotspots’. The invention came out of CSIRO’s pioneering work in radioastronomy. That work involved complex mathematics known as ‘fast Fourier transforms’ as well as detailed knowledge about radio waves and their behaviour in different environments. Indoor environments are particularly difficult for the rapid exchange of large amounts of data using radio waves. CSIRO solved these problems in a unique way at a time when many of the major communications companies around the world were trying, but with less success, to solve the same problem.

CSIRO’s invention was granted a US patent in 1996. There are corresponding patents in 18 other countries. The technology was first embodied in an industry standard in 1999 (called IEEE 802.11a) and later in other standards (IEEE 802.11g and IEEE 802.11 draft n). By 2012 CSIRO had licence agreements with 23 companies representing around 90% of the industry, with total revenue earned from this technology in excess of $430 million. As well as computers, it is now used to connect wirelessly many electrical and electronic devices including printers, game consoles, TV sets and phones. Forecasters predict that there are likely to be more than five billion devices sold worldwide by the end of 2013 using the technology invented by CSIRO scientists.

This outstanding research has been recognised as follows: John O’Sullivan was elected a Fellow of the Australian Academy of Science in 2009 and was awarded the 2009 Prime Minister’s Prize for Science. He and the Wireless LAN team were awarded the 2009 CSIRO Chairman’s Medal, a Clunies Ross award in 2010 and a European Inventor of the Year Award in 2012.

Technical background

In August 2008, the United States Court for the Eastern District of Texas issued a ruling in which the court construed various technical terms used in CSIRO’s US Patent 5,487,069, which was granted on 23 January 1996 (the ‘069 patent). In the following extract, the court described certain aspects of CSIRO’s invention.

Extract

The ‘069 Patent relates to a wireless Local Area Network (WLAN) wherein a plurality of wireless transceivers communicate with a plurality of wireless hub transceivers. Radio frequency wave propagation characteristics must be considered in implementing a WLAN. Radio waves can be reflected by some materials such as walls, furniture, and other indoor items, creating ‘multipath’ where a radio signal is dispersed and arrives at the receiver from different paths. As a result, there can be multiple copies of the signal with different signal strengths. The problem that can result is called ‘intersymbol interference’ (ISI), which is an overlap in arrival of the same symbol from different paths. ISI is the result of time differences between the arrivals of reflected copies of the same signal. This time difference is referred to as ‘delay spread’. As the data transmission speed gets faster, the time duration of the transmitted symbols (symbol period) gets smaller and more susceptible to ISI. In conventional radio transmission, the symbol period is set to be longer than the delay spread. Thus, multipath places an upper limit on data transmission rate. That is, as the delay spread increases, the symbol period must get longer, which in turn means that the data transmission rate necessarily decreases.

The ‘069 Patent provides high data transfer rates and high reliability in wireless environments with significant multipath interference. The patent teaches a combination of three key techniques: parallel sub-channels (ensemble modulation) wherein the period of a sub-channel symbol is longer than a predetermined time delay of the non-direct transmission paths, data reliability enhancement through Forward Error Correction (FEC), and data reliability enhancement through bit interleaving.

Radio transmission of information relies upon the concept of superimposing information on, or ‘modulating’, a carrier wave. In conventional radio transmission, the carrier is at a specific ‘narrowband’ frequency. The receiver must be tuned to that same narrowband frequency to receive the transmission. If there are many transmissions occurring at the same time at the narrowband frequency, interference will result. In order to minimise interference, various techniques have been developed.

One technique to avoid interference from other transmission sources is to spread the signal over a wider range of frequencies. This is referred to as ‘spread spectrum’. A particular approach to the reduction of interference is Frequency Hopping Spread Spectrum (FHSS), where the signal carrier is transmitted for a short period of time ‘ ‘dwell time’ ‘ on one narrowband frequency and is then hopped to another narrowband frequency. A WLAN that uses FHSS to reduce interference with other devices operates on a predetermined hopping sequence that is known to the receiver and can be followed by it. The dwell time, however, must be consistent with the delay spread to avoid ISI. Thus, FHSS is a wideband modulation scheme that uses multiple carriers one at a time and avoids interference with other transmission signals in the same band by hopping over many different frequencies. During any one hop, the FHSS signal appears to be a narrowband signal.

Another technique is to use multiple carriers simultaneously rather than one at a time. This is technically not a spread spectrum because the carriers remain stationary and are not moved, but it serves the same purpose of spreading the signal power over a large band. This is known as Orthogonal Frequency Division Multiplexing (OFDM) or Multicarrier Modulation (MCM). The data is broken into subparts and each subpart is simultaneously transmitted on a different carrier frequency. Again, the transmission period of each part (the sub-channel symbol period) must be consistent with the delay spread to avoid ISI. As there is simultaneous transmission of all the signal parts, the data transmission rate is higher than with FHSS.

In addition to various modulation schemes for radio frequency transmission of data, an important aspect of WLAN data transmission is the addition of data reliability enhancement afforded by using coding of the actual data prior to its conversion to a modulated transmission signal. Forward Error Correction (FEC) coding is one type of digital signal processing that improves data reliability by introducing a known structure into a data sequence prior to transmission. This structure allows a receiver to detect and possibly correct errors caused by corruption from the channel without requesting re-transmission of the original information. In a system that employs FEC, a digital information source sends a data sequence to an encoder. The encoder inserts redundant bits, thereby outputting a longer sequence of code output bits as a ‘codeword’. One type of FEC is known as ‘convolutional coding’. The incoming data is in a stream of bits. A Rate ½ convolutional encoder provides two data ‘di-bits’ for every input bit.

Additional protection to data corruption due to adjacent burst errors is data ‘interleaving’, which spreads data over a variable period of time. With data interleaving, data is transmitted by spacing the content of consecutive data packets. Interleaving is used in conjunction with FEC. Burst errors are distributed over many data packets and the FEC has fewer errors to correct in each packet. Data interleaving shuffles the data to reduce the error rate.

Honours and awards

Awards