This paper presents efficient and secure schemes for privacy-preserving keyword searches on remote encrypted data. The goal is to allow a user to store files on a remote server in encrypted form and later retrieve them based on specific keywords without revealing the keywords or compromising the security of the stored files. The schemes do not require public-key cryptography and are efficient in terms of bandwidth and storage. The schemes are based on a keyword index created by the user, which maps keywords to the files they appear in. The user can then perform keyword searches using this index. The schemes are incremental, meaning new files can be added securely against previous queries but still be searchable against future queries. The schemes also handle the secure update of files, ensuring that newly submitted files are secure against previous queries but remain searchable against future ones. The paper discusses two scenarios: one where the user can store a dictionary on their mobile device and another where they cannot. In both cases, the user uses pseudo-random functions and permutations to mask the keyword index, allowing the server to recover selective parts of the index while keeping the rest pseudo-random. This approach ensures that the server learns only that the encrypted files contain the keyword, not the specific files. The schemes are proven to be secure against malicious servers and provide strong theoretical guarantees. They are efficient, practical, and work with various file formats, including compressed and multimedia files. The paper also addresses open problems, such as handling Boolean operations on multiple keywords and general pattern matching, which remain challenging. The schemes can be extended to handle occurrence queries, though this increases storage overhead. Finally, the paper notes that existing schemes lack the ability to securely update and delete files, a limitation that applies to all known schemes.This paper presents efficient and secure schemes for privacy-preserving keyword searches on remote encrypted data. The goal is to allow a user to store files on a remote server in encrypted form and later retrieve them based on specific keywords without revealing the keywords or compromising the security of the stored files. The schemes do not require public-key cryptography and are efficient in terms of bandwidth and storage. The schemes are based on a keyword index created by the user, which maps keywords to the files they appear in. The user can then perform keyword searches using this index. The schemes are incremental, meaning new files can be added securely against previous queries but still be searchable against future queries. The schemes also handle the secure update of files, ensuring that newly submitted files are secure against previous queries but remain searchable against future ones. The paper discusses two scenarios: one where the user can store a dictionary on their mobile device and another where they cannot. In both cases, the user uses pseudo-random functions and permutations to mask the keyword index, allowing the server to recover selective parts of the index while keeping the rest pseudo-random. This approach ensures that the server learns only that the encrypted files contain the keyword, not the specific files. The schemes are proven to be secure against malicious servers and provide strong theoretical guarantees. They are efficient, practical, and work with various file formats, including compressed and multimedia files. The paper also addresses open problems, such as handling Boolean operations on multiple keywords and general pattern matching, which remain challenging. The schemes can be extended to handle occurrence queries, though this increases storage overhead. Finally, the paper notes that existing schemes lack the ability to securely update and delete files, a limitation that applies to all known schemes.