This process continues until each node in the sensor field has a
level number.
The root node sends a time sync packet to initialize the time synchronization
process. Afterwards, the nodes in level 1 synchronize to level 0 by
performing the two-way handshake as shown in Figure 3. This type of handshake
is used by the NTP to synchronize the clocks of distributed computer
systems. At the end of the handshake at time g4, node A obtains the times
g1, g2, and g3 from the acknowledgement packet. The times g2 and g3 are obtained
from the clock of sensor node B while g1 and g4 are from the node A.
After processing the acknowledgment packet, the node A readjusts its clock
by the clock drift value ??, where ?? = (g2??’g1)??’(g4??’g3)
2 . At the same time, the
level 2 nodes overhear this message handshake and wait for a random time
before synchronizing with level 1 nodes. This synchronization process continues
until all the nodes in the network are synchronized. Since TPSN enables
time synchronization from one root node, it is the timing technique of type
(1).
The TPSN is based on a sender-receiver synchronization model, where the
receiver synchronizes with the time of the sender according to the two-way
message handshake as shown in Figure 3. It is trying to provide a light-weight
and tunable time synchronization service. On the other hand, it requires a time
server and does not address the robust and energy aware design goal.
Pages:
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375