AGL 38.15 Decreased By ▼ -1.43 (-3.61%)
AIRLINK 125.07 Decreased By ▼ -6.15 (-4.69%)
BOP 6.85 Increased By ▲ 0.04 (0.59%)
CNERGY 4.45 Decreased By ▼ -0.26 (-5.52%)
DCL 7.91 Decreased By ▼ -0.53 (-6.28%)
DFML 37.34 Decreased By ▼ -4.13 (-9.96%)
DGKC 77.77 Decreased By ▼ -4.32 (-5.26%)
FCCL 30.58 Decreased By ▼ -2.52 (-7.61%)
FFBL 68.86 Decreased By ▼ -4.01 (-5.5%)
FFL 11.86 Decreased By ▼ -0.40 (-3.26%)
HUBC 104.50 Decreased By ▼ -6.24 (-5.63%)
HUMNL 13.49 Decreased By ▼ -1.02 (-7.03%)
KEL 4.65 Decreased By ▼ -0.54 (-10.4%)
KOSM 7.17 Decreased By ▼ -0.44 (-5.78%)
MLCF 36.44 Decreased By ▼ -2.46 (-6.32%)
NBP 65.92 Increased By ▲ 1.91 (2.98%)
OGDC 179.53 Decreased By ▼ -13.29 (-6.89%)
PAEL 24.43 Decreased By ▼ -1.25 (-4.87%)
PIBTL 7.15 Decreased By ▼ -0.19 (-2.59%)
PPL 143.70 Decreased By ▼ -10.37 (-6.73%)
PRL 24.32 Decreased By ▼ -1.51 (-5.85%)
PTC 16.40 Decreased By ▼ -1.41 (-7.92%)
SEARL 78.57 Decreased By ▼ -3.73 (-4.53%)
TELE 7.22 Decreased By ▼ -0.54 (-6.96%)
TOMCL 31.97 Decreased By ▼ -1.49 (-4.45%)
TPLP 8.13 Decreased By ▼ -0.36 (-4.24%)
TREET 16.13 Decreased By ▼ -0.49 (-2.95%)
TRG 54.66 Decreased By ▼ -2.74 (-4.77%)
UNITY 27.50 Decreased By ▼ -0.01 (-0.04%)
WTL 1.29 Decreased By ▼ -0.08 (-5.84%)
BR100 10,089 Decreased By -415.2 (-3.95%)
BR30 29,509 Decreased By -1717.6 (-5.5%)
KSE100 94,574 Decreased By -3505.6 (-3.57%)
KSE30 29,445 Decreased By -1113.9 (-3.65%)

Forget about watching, Big Brother may be listening. Sounds from typing on computer keyboards are distinctive enough to be decoded, allowing security breaches caused by "acoustic snooping," University of California, Berkeley researchers said on Wednesday.
The researchers said they were able to feed sound recordings of typing on keyboards into a computer and use an algorithm to recover up to 96 percent of the keyboard characters entered by typists.
"It's a form of acoustical spying that should raise red flags among computer security and privacy experts," said Doug Tygar, a Berkeley professor of computer science and information management.
"If we were able to figure this out, it's likely that people with less honourable intentions can - or have - as well," Tygar said.
The research builds on earlier work by International Business Machines Corp researchers who were able to recover 80 percent of text from keyboard recordings.
That research relied on the same typist using the same keyboard and an algorithm trained with known text and corresponding sound samples.
By contrast, the algorithm in the Berkeley study adapts to typing patterns of multiple typists and overcomes background noise such as music or ringing cell-phones. Also, no special recording equipment was required; keystroke noise could be recorded using off-the-shelf gear.

Copyright Reuters, 2005

Comments

Comments are closed.