On Computing Mobile Agent Routes for Data Fusion in Distributed Sensor Networks

The problem of computing a route for a mobile agent that incrementally fuses the data as it visits the nodes in a distributed sensor network is considered. The order of nodes visited along the route has a significant impact on the quality and cost of fused data, which, in turn, impacts the main objective of the sensor network, such as target classification or tracking. We present a simplified analytical model for a distributed sensor network and formulate the route computation problem in terms of maximizing an objective function, which is directly proportional to the received signal strength and inversely proportional to the path loss and energy consumption. We show this problem to be NP-complete and propose a genetic algorithm to compute an approximate solution by suitably employing a two-level encoding scheme and genetic operators tailored to the objective function. We present simulation results for networks with different node sizes and sensor distributions, which demonstrate the superior performance of our algorithm over two existing heuristics, namely, local closest first and global closest first methods.

MULTIPLE sensor systems have been the target of active research since the early 1990s [1] with a particular emphasis on the information fusion methods for distributed sensor networks (DSNs) . Recent developments in the sensor, networking, and computing areas now make it possible to deploy a large number of inexpensive and small sensors to “achieve quality through quantity” in complex applications. In an important subclass of DSNs that are deployed for remote operations in large unstructured geographical areas, wireless networks with low bandwidth are usually the only means of communication among the sensors. These sensors are typically lightweight with limited processing power, battery capacity, and communication bandwidth. The communication tasks consume the limited power available at such sensor nodes and, therefore, in order to ensure their sustained operations, the power consumption must be kept to a minimum. Furthermore, the massively deployed sensors typically generate a huge amount of data of various modalities, which makes it critical to collect only the most important data and to collect it efficiently. In addition, despite the abundance of deployed sensors, not all sensor data is critical to ensuring the quality of fused information such as adequate detection energy for target detection or tracking. In conventional fusion architectures, all the sensor data is sent to a central location where it is fused. But, the transmission of noncritical sensor data in military DSN deployments increases the risk of being detected in addition to consuming the scarce resources such as battery power and network bandwidth. To meet these new challenges, the concept of mobile agent-based distributed sensor networks (MADSNs) has been proposed by Qi et al. [5] wherein a mobile agent selectively visits the sensors and incrementally fuses the appropriate measurement data. Mobile agents are special programs that can be dispatched from a source node to be executed at remote nodes. Upon arrival at a remote node, a mobile agent presents its credentials, obtains access to local services and data to collect needed information or perform certain actions, and then departs with the results. One of the most important aspects of mobile agents is the security, which is not addressed here, but is being actively investigated . The transfer of partial fusion results by a mobile agent may still have the risk of being spied on with hostile intent; the serial data collection process employed by the mobile agent obviously decreases the chance of exposing the individual raw data. Although there are advantages and disadvantages of using mobile agents in a particular scenario, their successful applications range from e-commerce [9] to military situation awareness [10]. They are found to be particularly useful for data fusion tasks in DSN. The motivations for using mobile agents in DSN have been extensively studied [5]. For a particular multiresolution data integration application,

Free download research paper