second time in two weeks we have had a leap in tech increasing speeds,
this time TCP (transport control protocal) improvements using the same forward looking forward correcting (FEC) as my thread on the opus codec but
implemented in the TCP layer.
the way it works,
when transmitting pakets over wifi a single packet loss requires re transmition of packets following the lost packet,
this signals the "congestion" scale back function and bit rates are reduced accordingly,
in the new schema,
as with the opus codec the forward facing error correction can "account" for the lost packet "on the fly" without re transmition of all the following
packets,
this means that the "scale back" of bitrate is not a reaction to single packet loss,
how it does this without reducing the rate and retransmitting the packet sequence in order,
well algebra is used to encode the "clumped" packet data,
and if a packet is lost, rather than re send the clump,
the algebra is trasformed to complete the missing packet.
ON THE FLY
If part of the message is lost, the receiver can solve the equation to derive the missing data. The process of solving the equations is simple and
linear, meaning it doesn't require much processing on behalf of the router/smartphone/laptop. In testing, the coded TCP resulted in some dramatic
improvements. MIT found that campus WiFi (2% packet loss) jumped from 1Mbps to 16Mbps. On a fast-moving train (5% packet loss), the connection speed
jumped from 0.5Mbps to 13.5Mbps. Moving forward, coded TCP is expected to have huge repercussions on the performance of LTE and WiFi networks — and
the technology has already been commercially licensed to several hardware makers.
hardware.slashdot.org...
so why is this so cool?
perceived congestion "symptoms" wont lower through put (wireless always losses packets)
and is perceived as line conjestion causing slower bit rates,
and the over head to "solve" the algebra is low.
meaning that even mobiles can use this new TCP format.
explination from slashdot member rabtech (under main article)
1. A lost packet means a collision or network congestion, therefore the proper response to a lost packet notification is to slow down the transmit
rate
2. When packet #2 is lost, even though the client has perfectly cromulent copies of packets 3-1094932, they must *all* be thrown away and the entire
stream is restarted. There is no facility to go back and retransmit just the missing packet - the ACK can't say "I got packets 1,3-1094932, please
just re-send #2".
This new scheme reconstructs packet #2 in this scenario by using the FEC data in the other packets. This allows the link to tolerate a certain % of
packet loss without requiring any re-transmits, thus all those packets from 3 upwards don't have to be retransmitted. It also greatly reduces latency
as reconstructing packet #2 is faster (due to the computationally efficient FEC algorithm) than requesting a retransmit. This also prevents the TCP
link from scaling back its transmit rate, further improving performance.
this is a slightly different approach than the opus codec uses but is at a different layer,
imagine if this becomes the TCP standard?
imagine the speed increase (throughput) and latency decrease (ping)
very elegant solution to freeing up bandwidth,
while reducing mulitipule sends of the same packets
the future (if adapted) could see TCP at many times the efficiency we expect today,
this could prove a short term answer for limited bandwidth customers with lossey/noisy lines,
or very low band widths, that have very poor through put
and guess what,
IT USES CURRENTLY INSTALLED LINES and just requires adoption as a standard,
and software updates.
for all those people on slow connections, with very little chance of near future upgrades,
the internet could possibly double in speed in the near future,
without extra cost to the customers
gamers pre-pair for low ping and high download rates
on your old crusty connection
By providing new ways for mobile devices to solve for missing data, the technology not only eliminates this wasteful process but also can
seamlessly weave data streams from Wi-Fi and LTE—a leap forward from other approaches that toggle back and forth. "Any IP network will benefit from
this technology," says Sheau Ng, vice president for research and development at NBC Universal.
www.technologyreview.com...
if you use wireless your going to love this if its rolled out as net wide standard,
cable will see similar improvements as well but not to the degree of wireless,
although major improvements can be expected
xploder
edit on 23-10-2012 by XPLodER because: (no reason given)
edit on 23-10-2012 by XPLodER because: (no reason given)