Controller Area Network (CAN) bus forms the major medium of communication between the Electronic Control Units (ECUs) in cars. As of today, it is impossible to imagine automobiles without CAN. Some of the recent studies reveal that it is possible to access the in-vehicle network and inject malicious messages to alter the behavior of the vehicle. Researchers have shown that, it is possible to hack a car and remotely take control of brake, steering, power window systems, etc. Hence, it becomes inevitable to implement schemes that detect anomalies and prevent attacks on CAN network. Our work explores the complete anomaly detection process for CAN network that includes the techniques followed, available tools and challenges at every stage. Various existing techniques to eliminate malicious CAN messages are also mentioned. Starting from what makes CAN protocol vulnerable, we discuss case studies about attacks on CAN network with major focus on Denial of Service (DoS) attacks. We analyze the pattern of normal CAN messages from real vehicle along with patterns of simulated attack data using different methods/tools. Based on these patterns, we observe that the periodic nature of CAN messages are affected when DoS attacks occur. The work in this paper presents two algorithms “time-based” and “message-based” to detect DoS attacks in CAN network. Comparative analysis of observations and accuracy results are highlighted. The average accuracy obtained for time-based algorithm is 81% while that for message-based is 80%.