The elected master nodes are
nodes C and G. First, the master nodes send a message to their neighbors to
measure the round-trip times. Once the neighbors receive the message, they
self-determine if they should become di?®use leader nodes. The ones elected
to become di?®use leader nodes reply to the master nodes and start sending a
message to measure the round-trip to their neighbors. As shown in Figure 5,
nodes M, N, and D are the di?®used leader nodes of node C. Once the replies are
received by the master nodes, the round-trip time and the standard deviation
of the round-trip time are calculated. The one-way delay from the master
nodes to the neighbor nodes is half of the measured round-trip time. After
the calculation, the master nodes send a time-stamped message containing
the standard deviation to the neighbor nodes. The time in the time-stamped
message is adjusted with the one-way delay. Once the di?®use leader nodes
receive the time-stamped message, they broadcast the time-stamped message
after adjusting the time, which is in the message, with their measured oneway
delay and inserting their standard deviation of the round-trip time. This
di?®usion process continues for n times, where n is the number of hops from
the master nodes. From Figure 5, the time is di?®used three hops from the
master nodes C and G. The nodes D, E, and F are the di?®used leader nodes
that di?®use the time-stamped messages originated from the master nodes.
Pages:
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378