This book is great - I've owned (and used) both editions (lent the first edition out and ended up buying the second as a replacement). I find it particularly useful in that it answers design questions in a straight forward manner and shows you how to solve whatever problem you might be working with while at the same time giving solid theoretical explanations (the water pressure motif used throughout the book is quite good).
I also own The Art of Electronics but usually turn to Practical Electronics for Inventors first.
I'm currently working on implementing mesh on the OpenWRT platform and have made some significant progress...
Doing AODV mesh stuff (what's currently available on most existing mesh routers) is easy. But it's not the best approach. It's opportunistic routing and it has too much overhead on busy networks (but it is good in situations with moving clients/hubs - which is why I'm still working on it).
So as an alternative I'm looking at implementing MIT's Roofnet platform on OpenWRT. This is much more complicated but it would be very nice to have running! MIT's protocol is fabulous...
If folks are interested please let me know - I hope to have my own boxes running soon (any day now) and would love to have a large install to test things on. (email me: info at jump9.com if you're interested!)
Slashdot Mirror
Juice Analytics did this nifty research project on Colbert Bump as well: http://www.juiceanalytics.com/writing/colbert-bump/
This book is great - I've owned (and used) both editions (lent the first edition out and ended up buying the second as a replacement). I find it particularly useful in that it answers design questions in a straight forward manner and shows you how to solve whatever problem you might be working with while at the same time giving solid theoretical explanations (the water pressure motif used throughout the book is quite good).
I also own The Art of Electronics but usually turn to Practical Electronics for Inventors first.
I'm currently working on implementing mesh on the OpenWRT platform and have made some significant progress... Doing AODV mesh stuff (what's currently available on most existing mesh routers) is easy. But it's not the best approach. It's opportunistic routing and it has too much overhead on busy networks (but it is good in situations with moving clients/hubs - which is why I'm still working on it). So as an alternative I'm looking at implementing MIT's Roofnet platform on OpenWRT. This is much more complicated but it would be very nice to have running! MIT's protocol is fabulous... If folks are interested please let me know - I hope to have my own boxes running soon (any day now) and would love to have a large install to test things on. (email me: info at jump9.com if you're interested!)