The study of convenient methods of information dissemination has been a vital research area for years. Mobile ad hoc networks (MANET) have revolutionized our society due to their self-configuring, infrastructure-less decentralized modes of communication and thus researchers have focused on finding better and better ways to fully utilize the potential of MANETs. The recent advent of modern machine learning techniques has made it possible to apply artificial intelligence to develop better protocols for this purpose. In this paper, we expand our previous work which developed a clustering algorithm that used weight-based parameters to select cluster heads and use Artificial Neural Network to train a model to accurately predict the scale of the weights required for different network topologies.

Feature selection plays a very important role in all pattern recognition tasks. It has several benefits in terms of reduced data collection effort, better interpretability of the models and reduced model building and execution time. A lot of problems in feature selection have been shown to be NP – Hard. There has been significant research in feature selection in last three decades. However, the problem of feature selection for clustering is still quite an open area. The main reason is unavailability of target variable as compared to supervised tasks. In this paper, five properties or metafeatures like entropy, skewness, kurtosis, coefficient of variation and average correlation of the features have been studied and analysed. An extensive study has been conducted over 21 publicly available datasets, to evaluate viability of feature elimination strategy based on the values of the metafeatures for feature selection in clustering. A strategy to select the most appropriate metafeatures for a particular dataset has also been outlined. The results indicate that the performance decrease is not statistically significant.

Among the common wireless communication modules like Bluetooth and Wi-Fi, XBee modules are embedded solutions providing wireless communication standard with self-healing mesh networks, which has longer range than Bluetooth and lower power consumption than Wi-Fi. An alternative to the XBee radio modules is nRF24L01+ radio modules which are cheap and powerful, highly integrated, ultra-low power (ULP) 2Mbps RF transceiver ICs for the 2.4GHz ISM (Industrial, Scientific, and Medical) band. In this paper, performances of nRF24L01+ modules have been analyzed and compared with that of XBee ZB modules in wireless ad-hoc networks. The performance metrics for the analytical study are - 1) Throughput measurement, 2) Mesh routing recovery time and 3) Power consumption. This work has revolved around an open source library released by the developer, tmrh20 which builds a complete TCP/IP suite on top of the nRF24L01+ modules.