In this article, low-cost data transmission system is designed for smart building. The proposed work uses deep learning models to compress and decompress the sensory data. The Long-Range communication protocol is used for communicating the data. A communication channel is implemented in a novel way by designing a deep learning model to reduce the transmission error. The experimental results show a fixed number of different types of Edge devices to design the low-cost data transmission system that is constrained by energy and latency restrictions.