This study presents a novel approach to design an optimal and energy-efficient communication system tailored for wireless monitoring system in nuclear power facilities. It addresses the unique challenges of such environments, including high throughput demands for the size expansion of wireless sensor networks (WSN), limited power availability with long service time requirements, and severe signal attenuation, error rates, and loss packets in harsh through-wall scenarios. The proposed system utilizes a low-power single input multiple output (SIMO) Ultra-Wideband (UWB) system with orthogonal frequency division multiplexing (OFDM), enhancing spectrum efficiency through frequency and spatial diversities. We introduce a modified water-filling algorithm, designed to optimally allocate power across subchannels based on varying channel conditions when total power budget is undefined. This algorithm specifically targets on achieving necessary system throughput, which is a critical parameter in communication designs. Our simulation results demonstrate significant energy savings and reductions in bit error rate and outage probability, offering a robust solution for nuclear safety. Furthermore, we emphasize the gap in the existing literature regarding channel models for harsh though-wall environments by developing a straightforward and comprehensive channel model using ray-tracing techniques and Friis’ transmission equations. This model’s accuracy is validated through a comparison of calculated results against experimentally measured results, verifying its effectiveness and applicability in different through-wall communication scenarios.