MIT, Harvard and Others Add Math to the Bandwidth Tool Chest

Carl Weinschenk

The good news about increasing usable bandwidth is that there are many ways to attack the problem and many areas in which improvements — some big and some small — can be made.

Not all data gets through networks. This problem is worse in wireless, where there is little control over the conditions through which the signals travel. In cases in which a replacement must be sent — which is not all cases — bandwidth is spent in a number of ways: the original packet — which may not be lost until it is near its destination and thus has used capacity — the request for the retransmission and the new packet itself. How this is done has impact on the efficiency of the system in use and the amount of content that can be accommodated.

A new approach, coded TCP, has been developed. The idea is to drastically reduce the amount of data that needs to be transported in real time by setting up easy mathematical constructs to allow missing data to be resurrected without it actually being transmitted. MIT Technology Review quoted Muriel Medard, a professor at MIT's Research Laboratory of Electronics, about the idea:

The technology transforms the way packets of data are sent. Instead of sending packets, it sends algebraic equations that describe series of packets. So if a packet goes missing, instead of asking the network to resend it, the receiving device can solve for the missing one itself. Since the equations involved are simple and linear, the processing load on a phone, router, or base station is negligible, Medard says.

ExtremeTech also explained and added the detail that the packets are bunched together. That is why it is easy to solve for the one that is missing:

With coded TCP, blocks of packets are clumped together and then transformed into algebraic equations that describe the packets. 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.

This conceptually is reminiscent of video compression transmission schemes in which ways around actually sending the content are utilized. The idea, which has been licensed by several companies, is backed by some heavy hitters (or thinkers). In addition to MIT, the project is backed by the University of Porto in Portugal, Harvard, CalTech and the Technical University of Munich. An MIT/CalTech startup, Code-On Technologies, is licensing the approach.

The link is given a bit of extra interest by a comment by an individual who clearly isn’t impressed. The information in the article and the commenter's reservations get into some pretty heavy technology. The problem most likely is in the reporting: Harvard and CalTech aren’t likely to have gotten involved in a project that can be disqualified in a 300-word or so comment. In any case, it’s a very interesting back and forth.

A closely related topic, forward error correction, focuses on catching and handling problems before transmission. FEC had a brief cameo in the news last week as the cable industry introduced plans for the next version of its Data Over Cable Service Interface Specification (DOCSIS). DOCSIS 3.1, according to reports at Broadband Technology Report and elsewhere, will use advanced FEC, new modulation techniques and other approaches to expanding capacity on the industry’s hybrid fiber/coaxial cable infrastructure.

Whether or not coded TCP is the answer remains to be seen. Clearly, however, many groups of very smart researchers see it as a tool in the fight to alleviate the capacity crunch.

Add Comment      Leave a comment on this blog post

Post a comment





(Maximum characters: 1200). You have 1200 characters left.



Subscribe to our Newsletters

Sign up now and get the best business technology insights direct to your inbox.


Thanks for your registration, follow us on our social networks to keep up-to-date