The Narrowband Internet of Things (NB-IoT) has been presented recently to offer inexpensive, low powered, wide area cellular connectivity for Internet of Things (IoT). With the integration of NB-IoT, low power wide area network (LPWAN) becomes a familiar low rate long range radio communication technology. But resource management such as power consumption reduction in NB-IoT is yet to be improved to attain energy efficient architectures for standardization and commercialization. This paper designs an optimal DBN for power consumption reduction (ODBN-PCR) model in LPWAN using NB-IoT environment. The proposed ODBN-PCR model intends to minimize the power utilization by the use of DBN model, which helps to improve the battery lifetime of the LPWAN device. The proposed model involves data pre-processing in the initial stage in two levels namely data splitting and data scaling. Also, the hyperparameter tuning of the DBN technique take place utilizing Water Strider Optimization (OWSO) technique and thereby improves the performance of the DBN model. For examining the enhanced performance of the ODBN-PCR technique, a wide range of simulations take place. The comparative results analysis portrayed the better performance of the ODBN-OPCR technique over the other others interms of power consumption.