Abstract
This paper explores the application of the discrete wavelet transform (DWT) using the Haar method to improve image reconstruction quality after transmission through multipath communication channels. The Rician fading channel model is employed, taking into account the influence of the direct path, reflections with different delays and gain coefficients, as well as interference and additive white Gaussian noise (AWGN). To process the frequency characteristics of the image, the Fast Fourier Transform (FFT) was applied, enabling efficient channel modeling and enhanced reconstruction. The study demonstrates the effectiveness of the proposed approach for sensor networks transmitting images under limited bandwidth and significant multipath distortions. The Haar method allows for localized image modifications and compensates for channel effects, thereby improving reconstruction accuracy. Tests conducted with predefined Rician channel parameters (K-factor = 4, reflection amplitudes [0, 0.2, 0.1], delays [0, 5, 10]) achieved an SSIM value of up to 98% while maintaining minimal computational complexity. This research is relevant for the development of sensor networks operating in multipath environments, such as environmental monitoring, medical, and industrial systems, where sensor networks are utilized for visual data transmission in interference-prone conditions. The applied algorithms ensure reliable image reconstruction even in the presence of interference and under the computational constraints of network nodes and devices.