This paper proposes a privacy-preserving spatial data query scheme for cloud computing. The authors address the challenges of secure spatial data queries in the cloud, where data is outsourced to reduce local storage and computational burdens but raises privacy concerns. Existing schemes use Asymmetric Scalar-Product-Preserving Encryption (ASPE), which is insecure against known-plaintext attacks and generates large ciphertexts, leading to high storage and computational overheads. To solve these issues, the authors propose a new scheme based on Enhanced ASPE (EASPE), which provides better security and efficiency. The authors first propose a basic Privacy-Preserving Spatial Data Query (PSDQ) scheme using a new unified index structure that requires users to provide less information about the query range. They then propose an enhanced PSDQ scheme (PSDQ⁺) using a Geohash-based R-tree (GR-tree) and an efficient pruning strategy, which significantly reduces query time. Formal security analysis shows that their schemes achieve Indistinguishability under Chosen Plaintext Attack (IND-CPA), and extensive experiments demonstrate their efficiency in practice. The proposed schemes use Geohash algorithm to achieve spatial range queries, which only require users to provide less information about the query range. The GR-tree structure is designed to avoid the range intersection problem between non-leaf nodes in R-tree. Each node of GR-tree is encoded as a unified index vector, and a pruning strategy based on both spatial information and keywords is used to achieve spatial keyword queries with sub-linear search complexity. The authors also compare their schemes with previous ones, showing that their schemes are more efficient and secure. The proposed schemes achieve linear query complexity for PSDQ and sub-linear complexity for PSDQ⁺, significantly improving query efficiency. The schemes are secure against IND-CPA and have been tested to be efficient in practice, with PSDQ⁺ being about 10 times faster than the conference version.This paper proposes a privacy-preserving spatial data query scheme for cloud computing. The authors address the challenges of secure spatial data queries in the cloud, where data is outsourced to reduce local storage and computational burdens but raises privacy concerns. Existing schemes use Asymmetric Scalar-Product-Preserving Encryption (ASPE), which is insecure against known-plaintext attacks and generates large ciphertexts, leading to high storage and computational overheads. To solve these issues, the authors propose a new scheme based on Enhanced ASPE (EASPE), which provides better security and efficiency. The authors first propose a basic Privacy-Preserving Spatial Data Query (PSDQ) scheme using a new unified index structure that requires users to provide less information about the query range. They then propose an enhanced PSDQ scheme (PSDQ⁺) using a Geohash-based R-tree (GR-tree) and an efficient pruning strategy, which significantly reduces query time. Formal security analysis shows that their schemes achieve Indistinguishability under Chosen Plaintext Attack (IND-CPA), and extensive experiments demonstrate their efficiency in practice. The proposed schemes use Geohash algorithm to achieve spatial range queries, which only require users to provide less information about the query range. The GR-tree structure is designed to avoid the range intersection problem between non-leaf nodes in R-tree. Each node of GR-tree is encoded as a unified index vector, and a pruning strategy based on both spatial information and keywords is used to achieve spatial keyword queries with sub-linear search complexity. The authors also compare their schemes with previous ones, showing that their schemes are more efficient and secure. The proposed schemes achieve linear query complexity for PSDQ and sub-linear complexity for PSDQ⁺, significantly improving query efficiency. The schemes are secure against IND-CPA and have been tested to be efficient in practice, with PSDQ⁺ being about 10 times faster than the conference version.