• Korean Journal of Plant Resources
• /
• v.23 no.4
• /
• pp.360-367
• /
• 2010

The genetic diversity and structure of P. tongkangensis in 5 populations from 3 regions was investigated using 56 markers derived from 6 ISSR primers. Genetic diversity at the species level (P=94.6, SI=0.377, h=0.240) was substantial considering the limited distribution and small size of populations. Genetic differentiation among regions (12%) and among populations (13%) in the region was not clearly evident, which suggested a moderate level of gene flow among adjacent populations. The Mantel test revealed a significant correlation between genetic differentiation (${\Phi}_{ST}$) and geographic distance among populations. This was supported by cluster analysis and principal coordinate analysis (PCoA). The significant difference in marker band frequency at many loci and their fixation in opposite directions in the smallest and most isolated population SC were considered due to genetic drift. Therefore, the genetic diversity of P. tongkangensis could be compromised if the distribution area or the size of the population was further reduced. In particular, small and isolated populations could be at great risk of extinction. Considering this, the unique habitats of P. tongkangensis should be protected and the reduction of population size should be closely monitored. Conservation efforts including the seeding and planting of seedlings should be done carefully based on their genetic and ecological traits. Our data support the argument that establishing an integrated management system for the efficient conservation of P. tongkangensis is critical.

• Journal of the Institute of Electronics Engineers of Korea TC
• /
• v.46 no.2
• /
• pp.27-37
• /
• 2009

Sensor nodes forming a sensor network have limited energy capacity such as small batteries and when these nodes are placed in a specific field, it is important to research minimizing sensor nodes' energy consumption because of difficulty in supplying additional energy for the sensor nodes. Clustering has been in the limelight as one of efficient techniques to reduce sensor nodes' energy consumption in sensor networks. However, energy saving results can vary greatly depending on election of cluster heads, the number and size of clusters and the distance among the sensor nodes. /This research has an aim to find the optimal set of clusters which can reduce sensor nodes' energy consumption. We use a Genetic Algorithm(GA), a stochastic search technique used in computing, to find optimal solutions. GA performs searching through evolution processes to find optimal clusters in terms of energy efficiency. Our results show that GA is more efficient than LEACH which is a clustering algorithm without evolution processes. The two-dimensional GA (2D-GA) proposed in this research can perform more efficient gene evolution than one-dimensional GA(1D-GA)by giving unique location information to each node existing in chromosomes. As a result, the 2D-GA can find rapidly and effectively optimal clusters to maximize lifetime of the sensor networks.