Analog-to-Information　Converter　(AIC)　based　on　Compressive　Sensing　(CS)　broke　the　bottleneck　of　traditional　sampling　theorem.　The　required　data　is　much　less　than　the　amount　of　data　acquired　by　the　Nyquist　sampling.　However,　algorithms　implemented　in　software　inhibited　timely　decisions　and　prevents　the　use　of　adaptive　sensing　strategies.　Therefore,　it　is　necessary　to　achieve　hardware　acceleration.　In　this　paper,　we　mainly　proposed　the　complex-valued　system　of　the　signal　reconstruction　for　the　CS-based　AIC　structure.　In　order　to　achieve　high　performance　and　high　speed　hardware,　we　propose　a　superior　design　for　inversion　based　on　Goldschmidt　algorithm.　This　design　can　double　the　speed　of　division　than　previous　works.　In　addition,　we　have　completed　the　functionalities　on　the　Xilinx　Virtex5　FPGA.　The　implementation　results　showed　that　the　Orthogonal　Matching　Pursuit　(OMP)　algorithm　achieved　a　recovery　signal-to-noise-ratio　(RSNR)　of　21.15　dB　with　the　clock　frequency　of　130.4　MHZ.　©　2017　IEEE.