In today's time, there is a need to continually monitor the ocean shore in real time for different applications such as coastal monitoring, disaster prevention, assisted navigation etc. This requires a high data rate (order of Mbps) communication for transfer of data from an underwater environment to surface. This cannot be achieved using the traditional acoustic networks as the data rate offered by acoustic networks is in the order of kbps. Notably, for the shallow underwater environment, radio frequency (RF) based communication provides a better data rate than acoustic techniques. However, propagation loss using RF can be higher as compared to acoustic and suffers from less communication range. As RF nodes in water are battery operated, high energy-efficiency is extremely significant. Further, there is no analysis of throughput using realistic energy model for RF shallow underwater. Hence, a realistic energy model, RSU-Energy, a cluster-based communication architecture in shallow water (CRUSE) and, a mechanism for RSU-Energy based throughput analysis (E-CRUSE) were proposed in this work. The performance in terms of duration after which the first node dies and network life is then verified using RSU-energy. Moreover, the throughput of the network using RSU-energy was also analysed for CRUSE.