Cognitive radio network is enormously used to improve the spectrum utilization by enabling the dynamic spectrum access. The performance of the cognitive radio network is affected due to the denial of service caused by the primary user emulation attack. This primary user emulation attack restricts the spectrum access opportunity of secondary users. In this research, the integration of Secure Hash Algorithm-1 (SHA-1) and neural network (i.e., soft computing method) is proposed to avoid the primary user emulation attack. The Received Signal Strength (RSS) and Direction of Arrival (DoA) are used to achieve the localization between the primary and secondary users. The main objective of the proposed method is to minimize the loss ratio during the communication process. The performance of the RSS-DoA-SHA method is analyzed in terms of routing overhead, End-to-End Delay (EED) and loss ratio. The proposed RSS-DoA-SHA method is compared with two existing methods such as RSS-Angle of Arrival (AOA) method and Hybrid Routing Algorithm-Symmetric Key Cryptography (HRA-SKC) method to evaluate the performance measures. The end to end delay of the proposed method for 500 PU is 1ms, it is less when compared to the RSS-AOA method.