A concept known as the Internet of Things (IoT) connects a wide range of gadgets found in homes, cars, and commercial spaces. Security issues arise from the expansion of linked devices, leaving IoT networks vulnerable to replay and eavesdropping attacks. This paper proposes a lightweight key generation approach that improves quantization and information reconciliation in order to overcome this problem. For key generation, the suggested method uses a modified window in conjunction with the discrete wavelet transform (DWT). Simulations are performed for various numbers of device nodes in both indoor and outdoor environments to verify the efficacy of the suggested strategy. According to the analysis of the data, the suggested method outperforms current key generation techniques in terms of dependability and performance. Moreover, comparative studies show that our approach performs better than other pertinent approaches For the Internet of Things (IoT), we proposed a more advanced key generation technique using a modified window in combination with discrete wavelet transform (DWT).The modification comprises the addition of bit sequencing, an alternative signal processing method for shaping signal patterns. This proposed method aims at increasing the efficiency of the key generation process by decreasing DWT disparities. The modified discrete wavelet transform output is assessed during the key generation process using multi-bit quantization. The effectiveness of the proposed method is confirmed through simulations conducted for varying numbers of device nodes in indoor and outdoor settings. MATLAB tools are used to monitor standard characteristics such as Average Key Length (AKL) and Key Disagreement Rate (KDR) throughout the simulation.