Published Paper Details:

Web applications remain the most exploited entry point in modern cyber attacks, with industry reports attributing nearly one in four data breaches to vulnerabilities in web-facing software. Web Application Firewalls (WAFs) have emerged as a critical defense layer, inspecting HTTP traffic to identify and block malicious requests before they reach the application. This survey provides a comprehensive review of WAF technologies spanning signature-based, anomaly-based, and machine learning-driven detection methods. We examine the evolution of WAFs from early rule-matching systems such as Snort and ModSecurity through modern cloud-native solutions offered by AWS, Cloudflare, and Azure, and into the emerging class of AI-powered next-generation firewalls. We present a structured taxonomy of web application attacks aligned with the OWASP Top 10 (2021) and their corresponding Common Weakness Enumerations (CWEs), and we systematically compare how different WAF approaches address each category. Our analysis covers deployment architectures (reverse-proxy, embedded, cloud-based, and hybrid), real-time monitoring and alerting mechanisms, and the trade-offs among detection accuracy, latency overhead, and operational complexity. We consolidate findings from over 40 studies published between 2003 and 2025, identify seven persistent research gaps--including the shortage of public evaluation benchmarks, the opacity of commercial solutions, and the difficulty of detecting novel zero day payloads--and outline promising future directions such as federated learning for distributed WAF deployments and explainable AI for security decision-making.