Research On Improved Compressed Sensing Algorithm And Its FPGA Implementation

With the rapid development of things in the mine, more and more problems come out, in the coal mine as well as surface wireless network nodes distributed more, especially underground. At present, enterprises tend operating unattended operations smartly, increasing safety and efficiency of coal production is inevitable. Whereby the amount of data to be collected and processed by each node will show a geometric growth which will be the normalization of the phenomenon. Brought about by a series of problems will make things serious resource consumption and shorten the life cycle.And a new type of sampling theory that is compressed sensing theory, the sampling algorithm sampling rate can be much lower than the Nyquist frequency of data collection, which can greatly reduce the amount of data sampling, improve resource utilization of Things. It have a great advantage in the data transmission and storage.In this paper, through theoretical research on compressive sensing, than to further improve compressive sensing algorithms.Came up with the compressive sensing algorithm based on wavelet transform.In-depth research on data acquisition and compression of mine IOT. First, analysis of the compressed sensing three important aspects of the basic theory that is sparse transform data observation reconstruction restore. And then combined with discrete wavelet transform algorithm in compressive sensing theory. In this paper, as the mother wavelet transform base sym3. The main aspects is improving data sparse algorithm. Sparse data and get it factor, then experiments and simulations are to be made on simulation platform Matlab7.1 version, by writing m file for simulation. Finally, designing and researching a FPGA hardware for the improved compressed sensing algorithm on platform which is mainly Cyclone IV EP4CE22F17C8 chip devices. Functional analysis of each hardware module using Verilog hardware description language to write run. The experimental data and theoretical simulation data were compared to verify the feasibility of the hardware.