3.4 System Design In this section, we provide the proposed work using graphic system such as: class diagram, system architecture, activity diagram etc.
3.3 Algorithm Design and Mathematical Model
3.3.1 Existing Clustering Algorithm
The weighted variants of k-means clustering, with an objective function, but instead of this generic form propose a Block-based Weighted Clustering scheme for the objective function. Here we are considering the proposed algorithm is based on WLAN (wireless local area network) by designing temporary Mobile Ad Hoc Networks using 802.11 MAC protocol and AODV routing protocol.
Algorithm 1: Block Based Weighted Clustering (BWC)
Input: {MANET Access point 1, MANET Access point 1, MANET Access point 1}
Concept and Notations:…show more content… A record is a vector r =[x, s], where x ϵ IRDG represents a geographical position and s ϵ IRDR is a measurement vector in radio feature space, describing the radio signal at location x. The included positions in the database may be numbered from 1 toM and given by the set:
X= [x1. . . xm, . . . , xM]
In this weighting scheme, we impose equal weight factors to all the components that belong to the same feature type; thus, the weight vector ω consists of Nf blocks of equal values. The corresponding objective function may be represented as follows:
J4 (U,R,ω)= ∑_(n=1)^N▒∑_(m=1)^M▒∑_(h=1)^(N_f)▒〖u_mn ω_h^β d^2 (ρ_nh^0,ρ_mh)〗
