example, the time representation in sensor networks may be just an aggregate
sum of the time di?®erence between sensor nodes. Thus, it has to be
translated into the universal coordinated time. In the following, examples of
these types of timing techniques are described, namely the Network Time
Protocol (NTP) [14], Timing-sync Protocol for Sensor Networks (TPSN) [7],
Reference-Broadcast Synchronization (RBS) [6], Time-Di?®usion Synchronization
Protocol (TDP) [18] and asynchronous rate-based di?®usion algorithm [11].
In Internet, the NTP is used to discipline the frequency of each node??™s
oscillator. The accuracy of the NTP synchronization is in the order of milliseconds
[8]. It may be useful to use NTP to discipline the oscillators of the
sensor nodes, but the connection to the time servers may not be possible because
of frequent sensor node failures. In addition, disciplining all the sensor
nodes in the sensor field may be a problem due to interference from the environment
and large variation of delay between di?®erent parts of the sensor
field. The interference can temporarily disjoint the sensor field into multiple
smaller fields causing undisciplined clocks among these smaller fields. The
NTP protocol may be considered as the timing technique of type (1). In addition,
it has to be refined in order to address the design challenges presented
by the sensor networks.
As of now, the NTP is very computational intensive and requires a precise
time server to synchronize the nodes in the network.
Pages:
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373