The Research On Automatic Identification Of Custom Instructions For Compression Algorithms And Encryption Algorithms

With the development of satellite remote sensing technology,remote sensing image data develops in the direction of multi-dimensional,multi-temporal and high resolution,and the data size is also increasing geometrically.Under the existing technical conditions,the pressure of data transmission and real-time processing will be further increased,and the data transmission difficulties will seriously affect the quality of information processing.Data compression is a key technology of information acquisition and information transmission.On the other hand,Shared services and application of remote sensing image data is becoming more common,in an open network environment,the military or civilian security and credibility of the geospatial information also faces problems.Data encryption technology is an important way to ensure the security of data transmission.Effective encryption method for remote sensing image data will ensure the security and reliability of data.This paper studies the efficiency of the compression algorithms and the encryption algorithms involved in the process of data transmission of remote sensing image data.By automatically identifying the custom instructions and using the identified extended instructions set to accelerate the execution of the compression algorithms and the encryption algorithms in the process of data transmission.The main algorithms involved in this paper are JPEG2000,JPEG,AES,RSA,etc.Extensible processors provide a good balance among design cycles,flexibility,performance,and power consumption.The automatic generation of instruction set extensions(the set of custom instructions)is the key to the design of extensible processors.This paper focuses on a series of research work on the key techniques and methods involved in the automatic identification of instruction set extensions.The subgraph enumeration problem and the subgraph selection problem are two key issues involved in the process of automatic identification of instruction set extensions.Aiming at the efficiency of subgraph enumeration,the dissertation first proposes an efficient algorithm to solve the enumeration of connected convex subgraphs.Starting from the topological structure of directed acyclic graph,a BS algorithm that uses the topological properties of directed acyclic graph and the monotonicity of convex graph is proposed to generate all connected convex graph recursively.The algorithm can also be flexibly tuned to enumerate connected convex graph or enumerate connected convex graph under maximum size constraint.In order to maximize the performance gain,a maximal convex graph enumeration algorithm is proposed,which uses a sandwich method to enumerate the largest convex graph,by combining the advantages of the traditional bottom-up algorithm and the top-down algorithm.The search space can be significantly reduced.In order to further improve the efficiency of subgraph enumeration,a graph segmentation method based on runtime estimation model is introduced to implement an efficient parallel subgraph enumeration method.In order to make a good trade-off between the quality of the solution and the running time of the algorithm,the simulated annealing algorithm,tabu search algorithm,genetic algorithm,particle swarm optimization and ant colony algorithm are adopted to solve the subgraph selection problem.Then,a detail comparison and analysis of the quality of the solution and the running time of algorithms is performed.At the same time,in order to speed up the algorithm,the subgraph selection problem is transformed into the maximal clique problem,and the compatible graph is segmented.A parallel ant colony algorithm is constructed and implemented to select subgraphs.The above works provide a solid foundation for the realization of automatic identification of instruction set extensions.Finally,the diversity of constraints involved in the subgraph enumeration and the multiobjectives optimization in the subgraph selection are effectively solved by constructing the constraint satisfaction problem models of subgraph enumeration and subgraph selection.The automatically identified instruction set extensions are applied to data encryption algorithms and data compression algorithms to significantly improve the performance of these algorithms.The research on improving the performance of compression algorithm and encryption algorithm through automatic identification of custom instructions will promote the extensive application of extensible processors in the fields of remote sensing image data compression and encryption,which has a good practical value.The paper has 46 pictures,14 tables and 125 references.