Study And Implementation Of Compression Algorithm For Network Transmission

As a new area of information technology, data compression is intended toreduce storage room and increase efficiency of transferring, storing andprocessing to meet the special demand. Due to the extreme complexity ofnetwork, there is no method of data compression specific to networktransferring. This thesis compares various popular methods of datacompression and makes some selections. It chooses lossless compressionalgorithm LZ as the basis, combining it with the real need for network topropose a way of data compression specific to network transferring. The thesisdeals with the following items:(1) Making an in-depth study in data compression technologies, describes the cause ofdata redundancies and basic method of compression, and presents a model design for thedata compression and a pattern of thinking for coding.(2) Analyzing the classification of data compression technologies, discusses the meritsand demerits of these technologies, and chooses a basic algorithm that is the mostsuitable for network data compression.(3) Designing a primary and secondary table dictionary structure, index of the Hashand attribute inline, the parameter adjustment of the variable dual-format outputoptimization scheme by combining the main issues of the present network datatransmission and the characteristics of the basic algorithm.(4) Testing the performance of the algorithm in the experiment, and obtains theexperienced value of two parameters reference.The algorithm proposed in this thesis can be used widely, achieving theaim of transmitting more source information in time lag, reducing the cost,developing more channels for communications and decreasing the power ofthe transmitter circuit and minimizing the cost of data compression. Moreover,the dictionary of the algorithm can support the online need for Data Mining.