The invention discloses an index inquiry method suitable for a high-dimensional scattered point cloud. The index inquiry method suitable for the high-dimensional scattered point cloud comprises the following steps that (1) the coordinates of each point in scattered point cloud data are stored in a logic main node; (2) the id and the jth-dimensional coordinate information of each point are sent to a logic sub-node numbered as j through the logic main node, and the logic sub-node builds a B+ tree index after receiving the id and the jth-dimensional coordinate information; (3) the logic main node sends the projection of an inquiry area R on the jth dimension to the logic sub-node numbered as j, and the logic sub-node searches for points in the projection area after receiving the projection, and feeds the ids of the points back to the logic main node; (4) the logic main node calculates the intersection set of all the ids of the points fed back by the logic sub-node. The index inquiry method suitable for the high-dimensional scattered point cloud converts the inquiry of the range of the nth-dimensional space into the inquiry of the range of n parallel one-dimensional spaces, largely improves the inquiry efficiency of the high-dimensional scattered point cloud, and effectively resolves the problems that an existing index inquiry method is simplex in resolution ratio, low in universality, redundant in data and the like.