On 29 Jul 2003 at 17:03, Joel Jaeggli wrote: > The mathematicians were absolutlely correct about the physical > limitations of the materials they were dealing with. what changed were > the materials being used and the fashion in which the data was > encoded. no-one dodged any laws of physics.
The operative concept here is "encoding method", as the primary advances in data storage communications bandwidth for the past 15 years or so have been the result of improvements in encoding. OFDM offers improvements in NLOS systems because it sprays bit symbols in parallel over subcarriers from which it's easier to recover them than it would be from a serial system. While this is clearly an advantage in an environment with limited sources of RF noise, it's not clear, at all, that it's best in one with multiple signallers overlapping each other. One of the big challenges in building robust RF systems is test envronments that model real-world sources of noise, degradation, and interference. Remember what happened at Networld+Interop this year? It wasn't pretty, but it's something that was predicted for a long time. The final exam for RF systems hasn't been written yet, but before it is we need to have a better appreciation for sharing channels than present-day products typically display. RB -- Richard Bennett http://www.bennett.com -- general wireless list, a bawug thing <http://www.bawug.org/> [un]subscribe: http://lists.bawug.org/mailman/listinfo/wireless
