Design of Orthogonal Matching Pursuit Architecture based on Compressed Sampling Using FPGA

In our work, a Field-programmable gate array (FPGA) based architecture for compressive sensing using Orthogonal Matching Pursuit (OMP) algorithm is suggested. The storing and transmission of data is done and it will be made efficient when the sampled signal is compressed. In this technique of compressive sensing, the demerits of directly achieving compressed signal without coming across any transitional stage by considering all samples is reduced. Reconstruction of sampled data will be done using numerical computing intensive theorems which has many flaws. Implementation of Reconstruction stage is complex software implementation of these algorithms is awfully slow and consumption of the power will be more as the system has large number of layers due to most of the resource sharing at multiple levels. Contrarily implementation in hardware has much advantage in hardware parallelism, ability of creating custom data path and hardware dedication for each task. In proposed method hardware implementation OMP algorithm is utilized because of its shorter computation time and low complex nature. The main objective of the proposed method is to implement an OMP algorithm to optimize essential reconstruction time without negotiating intended accuracy.