Wireless Sensor Networks (WSNs) increases the usage of integrated systems and areas which attracts the attention of attackers. However, WSNs are vulnerable to diferent kinds of security threats and attacks. To ensure their security, an effective Intrusion Detection System (IDS) need to be in place to detect these attacks even under these constraints. The traditional IDS are less effective as these malicious attacks are becoming more intelligent, frequent, and complex. To overcome these challenges, this paper proposes a novel Improved Deep Neural Network Integrated Intrusion Detection System in WSN (IIDS-NET) technique has been proposed, which increases the energy efficiency in the WSN network. Initially, an optimal CH is selected via Tom and Jerry optimization algorithm (TJOA) based on the Residual energy and Node Centrality. The proposed scheme makes use of Improved Elman Spike Neural Network (IESNN) technique is implemented to detect the intrusion nodes and to blocks the suspicious or malicious activity in the wireless networks. Finally, the Aquila-Sooty Tern Optimization Algorithm (AQSOA) is used to find the optimal route for sending the data between the sensor nodes and the base stations. The proposed scheme is simulated by using Cloud simulator (CloudSim) and a comparison is made between proposed IIDS-NET and existing approaches such as GWOSVM-IDS, EPK-DNN, FL-SCNN-Bi-LSTM and SG-IDS in terms of detection accuracy, energy consumption, and throughput. The proposed HOPI-NET approach outperforms the existing techniques such as GWOSVM-IDS, EPK-DNN, FL-SCNN-Bi-LSTM and SG-IDS in terms of energy consumption of 120.73%, 198.68%, 193.34%, 187.73%, and 165.88% respectively.