A Novel Hierarchical Crayfish Auto-Associative Polynomial Network-based Signal Degradation Prediction for Reliable Communication in Urban VANETs

Buildings are important in an urban Vehicular Ad Hoc Network (VANET) environment because they might serve as barriers that weaken the signal. These barriers result in multipath effects, which can significantly affect data transmission because of fading. As a result, measuring how buildings affect transmission quality is an important propagation model component, particularly in rescue circumstances where dependable communication is crucial. Based on the obstacles that moving emergency vehicles perceive in their environments, this study proposed a Hierarchical Crayfish auto-associative Polynomial Network (HCAPN), a hybrid of Hierarchical Auto associative Polynomial Network architecture and crayfish optimization, to create a predictive model that can estimate the degree of signal degradation, represented by the Bit Error Rate (BER). For the improved prediction, the input data was generated using correlation coefficients and min-max normalization. The model's higher prediction efficiency was demonstrated by its extremely low 1.35×10<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">−6</sup>RMSE, 1.82×10<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">−6</sup> MSE,1.12×10<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">−6</sup>MAE, and 5.67×10<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">−6</sup> MaxPE values when compared to the current models. Its ability to learn, generalize, and accurately forecast the BER for a new data instance was demonstrated by the results of the HCAPN simulation. All things considered, the study helps to improve the effectiveness and dependability of communication in urban VANET.