Compressive Sensing Based On Chaos Of Silicon Optomechanical Microcavity

Being in the torrent of big data,hundreds of millions of data will be generated every day,which will undoubtedly put great pressure on storage hardware equipment.The birth of compressive sensing theory has brought dawn to save a lot of storage space.As a newly emerging research direction,compressive sensing technology has attracted much attention in many application fields.It breaks the limitation of the traditional Nyquist-Shannon theorem(in order to obtain the complete information sampling frequency,it must reach more than twice the signal bandwidth).It can sample the signal at a lower sampling frequency and can accurately reconstruct it.Out of the original signal.Nowadays,compressed sensing technology has been widely used in all aspects of life,such as medical imaging,seismology,channel coding,image encryption and video compression.Sensing matrix is an important part of compressed sensing technology.In recent years,scientists have been struggling to produce better perceptual matrices.Thanks to the many excellent properties of chaos,chaos is a good carrier for constructing a sensing matrix.In 2010,a chaotic sequence extracted from a logic diagram by Yu et al.proposed a usable sensing matrix.In 2011,Madalin et al.obtained a sensing matrix from tent mapping.In 2017,Rontani and his collaborators used optical chaos generated by external cavity semiconductor lasers to generate a sensing matrix.In this paper,the chaos generated by the newly-designed silicon-based photonic crystal microcavity is used to build a sensing matrix,and experiments have proved that the properties of this matrix can meet the requirements of compressed sensing technology and realize the reconstruction of the original signal.The main work of this article is as follows:(1)A scheme of a compressed sensing system based on silicon photonic crystal microcavities is proposed.The optical chaos source includes a monolithic silicon-based microcavity.A tunable laser and an erbium-doped fiber amplifier are used to drive the microcavity.The fiber polarization controller selects the chip's lateral electrical polarization and optical isolator to prevent unnecessary reflected light.The generated chaos is collected into the optical fiber,and the electronic spectrum analyzer and digital oscilloscope are used to collect and analyze the chaotic data.The final compressed sensing process is done by computer.And because most of the large-scale integrated circuits at this stage are made of CMOS technology based on silicon materials,the core of the system solution we designed is silicon photonic chips,which can be better combined with the current popular CMOS technology,making it easier for the system to graft.Integrated circuit.(2)Experiments prove that the measurement matrix constructed by silicon-based optical chaos can achieve compressed sensing.The obtained optical chaos is used to form a sensing matrix,and a base tracking algorithm is used to realize the restoration and reconstruction of one-dimensional sparse signals.And compared with the experimental results of several classic perception matrices proposed by previous scholars.It is confirmed that our sensing matrix generated by using silicon-based optical chaos has good performance.In addition,we also performed experiments on two-dimensional images,and the results showed that the two-dimensional sparse images were successfully restored.