Mutation-based receiver search in a network for temporary information sharing

This paper examines our previously proposed mutation-based search method for temporary information sharing in a wireless mesh network. The temporary information sharing indicates that for the case in which information senders and receivers freely appear in the network, information is forwarded from an information sender to a receiver. The mutation-based search method determines a unique route to find receivers who want the forwarded information by operations among numerical values assigned to each node and probabilistically changes the assigned numerical values according to the search results, which can be regarded as a sort of mutation in evolutionary computation. A variety of types of operations and numerical values can be used for this search method. In simulations, the method is evaluated in terms of search reliability under two appearance patterns of an information sender-receiver pair. One is the appearance pattern that locations of the receiver and the receiver are randomly determined. The other is that only a location of the receiver is fixed. Simulation results reveal that search reliability depends on the types of operations and numerical values and is related to how frequently a closed path is generated in the search. The results also suggest that the degree of randomness in search is important for reliable search and varies with the types of operations and numerical values.