Internet of Things (IoT) based methods have enhanced operations on several applications in the recent year. Cyber-attacks are made more likely by inadequate security protocols in the Internet of Things network. This paper proposed a Honeypot enabled Intrusion Detection using Optimized deep Learning (HIDOL) technique to identify the attack on IoT device. Initially, a honeypot server is placed to find malevolent activity in IoT devices and the patterns are gathered from the attacker. The necessary data are extracted during the feature extraction procedure using a one-hot encoder. By using the Dingo Optimization, the essential features are selected. The Conv-BiLSTM model will classify the vulnerabilities like DDoS attacks, physical attacks and Ransomware attacks. The KDDCUP 19 dataset are used to evaluate the suggested system's efficiency and the evaluation measures like Throughput, Detection Rate and Time efficiency have been utilized to assess the efficacy of the suggested intrusion detection technique. By the comparison analysis the proposed HIDOL archives detection rate of 28.07% which is comparatively higher than the existing system respectively.