Mobile Intelligence

24.2 BASIC DATA-CENTRIC MECHANISMS

The basic data-centric mechanism simply applies hashing function that maps the event type to a location of the monitoring region. The geographic hash table (GHT)[2] mechanism is the most famous approach that stores the detected events at the sensor closest to the mapped location. The GHT operates on the network environment that consists of a large number of connected sensors and a small number of static sinks that connect to the outside world. To achieve this, the GPSR routing protocol is applied to find a route from the sensor that detects the event to the target location for storing or retrieving data. By using a uniform geographical hashing function, the GHT spreads storage and communication loads over the entire network.

Two operations, Put() and Get(), are supported by GHT. The Put(key, value) operation stores value at the sensor closest to the location that is generated by hashing on the key. On the other hand, the Get(key) operation retrieves all values associated with the key. In general, a key corresponds to a special event. For example, as shown in Figure 24.1, when the sensor node detects a lion, it stores the event using Put(“lion,” data) operation. Then the sensor node hashes the key, “lion,” to the corresponding locations, uses GPSR to reach the storage location, and puts the sensing data to the target sensor node. Since it is possible that no sensor node exists in the storage location, a node that is geographically closest to the ...