When the data being
transmitted is redundant in all or part of the transmitting nodes, data can
be aggregated. The intermediate nodes can sequentially fuse (aggregate) the
data as it is routed to the sink. Data-centric protocols inherently introduce a
latency into the transmissions by avoiding the optimal route to pursue data
fusion, but the tradeo?® in energy savings is more than justified.
Sensor Protocols for Information via Negotiation (SPIN) [5] is the first
data-centric protocol which considered interest negotiation before the actual
data forwarding in order to eliminate redundant data and save energy. In
SPIN, data is named with high-level descriptors or meta-data. SPIN uses
meta-data negotiation where each node that receives new data advertises it
to its neighbors (rather than sending the actual data) and waits to hear their
interest in the data before forwarding it. This procedure solves the problems
of the classical flooding mechanism in which each node forwards the data to
all of its neighbors. SPIN avoids data redundancy and resource blindness, thus
achieving significant energy savings.
Directed Di?®usion [7], Energy Aware Routing (EAR) [23], Rumor Routing
[1], Gradient Based Routing [22], Constrained Anisotropic Di?®used Routing
[2], COUGAR [28], and ACQUIRE [21] are other examples of data centric
protocols.
3.2 Hierarchy-Centric Routing
Borrowing ideas from infrastructured networks, hierarchical based routing
schemes take the large number of nodes and organize them into clusters, and a
368
Chapter 15 Information Forwarding and Tra?±c Engineering
clusterhead is elected.
Pages:
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588