We present a new color computational ghost imaging strategy using sole single-pixel detector and training by simulated dataset, which can eliminate the actual workload of acquiring experimental datasets reduce sampling times for experiments. First, relative responsibility device to different channels is experimentally detected, then enough data sets are neural network based on response value. Because simulation process much simpler than experiment, set be almost unlimited, trained model has...

A full-duplex radio over fiber link scheme with 16 quadrature amplitude modulation (16-QAM) 10 Gb/s downstream and 2.5 upstream signals is proposed. In the downlink, 60 GHz dual-tone optical millimeter- (mm-) wave signal carrier generated by 15 local oscillator via a Mach–Zehnder modulator frequency quadrupling. The 16-QAM 5 IF up-converted to 65 modulating it on one of two tones single-sideband (SSB) at central station. At remote base station, SSB mm-wave detected high-speed photodiode....

Abstract An encryption method based on computational ghost imaging (CGI) with chaotic mapping and DNA encoding is proposed. To reduce the amount of keys in CGI-based system, algorithm used to generate random sequence as speckle measurement matrix CGI system. The data bucket detector subjected block operations, which introduce nonlinear characteristics process. problem linear vulnerability system has been greatly improved. Numerical simulation results show that, compared traditional method,...

Abstract An imaging method through a dynamic scattering medium is presented based on computational ghost (CGI) and convolutional neural network (CNN). The CNN adopted to improve CGI quality, its training set obtained from numerical simulation rather than actual experiments, which greatly reduces the workload. A concise mathematical model given reflect absorption effects of medium. By adding Gaussian white noise with different intensities detected light intensity sequence, undulation caused...

An identity-based matchmaking encryption (IB-ME) scheme proposed at JOC 2021 supports anonymous but authenticated communications in a way that communication parties can both specify the senders or receivers on fly. IB-ME is easy to be used several network applications requiring privacy-preserving for its efficient implementation and special syntax. In literature, schemes are built from variants of Diffie-Hellman assumption all fail retain security quantum attackers. Despite rigorous proofs...

A very simple reorganization of a Booth-encoded carry-save array multiplier is shown to reduce power dissipation in DSP applications. The proposed architecture, rather straightforward rearrangement the standard array, particularly suitable for applications where one multiplier's input signals has large dynamic range. We take advantage sign-extension zero-encoding property Booth encoder total switching activity adder hence lowering dissipation. structure been synthesized using Synopsys Design...

We derive exact expressions of the symbol error rate (SER) performance orthogonal, frequency-hopped, noncoherent M-ary frequency-shift-keying (FH/MFSK) receivers in presence independent multitone jamming (MTJ), as well thermal and other wideband Gaussian noises. Both signal tones multiple are affected independently by Rician fading. slow-hopping fast-hopping systems considered. In our derivation, we exploit fact that multipath components can be combined with noises to form a single process....

This paper shows that a simple modification to the Booth-encoding algorithm can be used increase probability of zero coded digit. increases in partial product bits Booth-encoded multiplier and reduces average number transitions by 3.75% over traditional for random input sequence. In addition, we show transition carry-bits adders is directly related bits, reduced approximately 7%. HSPICE simulations proposed encoding reduce power dissipation more than 4% 16/spl times/16 two's complement...

Compressive ghost imaging (CGI) can effectively reduce the number of measurements required for reconstruction. In most cases, however, when using illumination patterns as measurement matrices, CGI has not demonstrated ability to reconstruct high-quality images at an ultra-low sampling rate perfect claimed by compressive sensing theory. According our analysis, reason is that non-negative nature light intensity causes matrix in be inconsistent with essential requirements good theory, leading...

While arithmetic circuits using 2's complement representation are easy to implement, it is well-known that the sign-extension bits of a number cause high switching activity in digital circuits. Such undesirable low power applications. In this work, we exploit property and propose reduced numbers avoid sign-extension. Instead having at MSB side datapath, proposed avoids MSBs altogether, therefore reduces dissipation With technique, maximum magnitude detected dynamically generated represent...

Joint module selection and retiming is a powerful technique to optimize the implementation cost speed of circuit specified using synchronous data-flow graph (DFG). In implementing high-speed circuits, use carry-save signal representation also arithmetic circuits. This paper first combine these two techniques solve joint problem while allowing representation. To this efficiently, we propose mixed-representation (MFG) that allows signals be expressed in We accurately model costs associated...

This paper presents an efficient architecture for the first carry-save integrator stage in a high-speed cascaded integrator-comb (CIC) decimation filter based on exploiting carry propagation properties accumulator. The can reduce number of registers (by 6.3% to 13.5% our examples) and replace large full-adders by half-adders (18% 42% examples), thus saving area power. Significant savings are achieved when rate is high stages small.

An analytical model that describes small-scale local area fading experienced by narrowband radio receivers consisting of N dominant specular waves and an arbitrary number additional weaker lumped into a Gaussian distributed diffuse component was recently proposed. The exact envelope pdf this is in definite integral form involving product zeroth-order Bessel functions. A closed formula for known only = 2. For analytic studies, closed-form expression 3 desirable. In paper, we derived new terms...

Instance search is an interesting task as well a challenging issue due to the lack of effective feature representation. In this paper, instance level representation built upon fully convolutional instance-aware segmentation proposed. The ROI-pooled from segmented region. So that instances in various sizes and layouts are represented by deep features uniform length. This further enhanced use deformable ResNeXt blocks. Superior performance observed terms its distinctiveness scalability on...

Abstract A real single-channel color image encryption method based on computational ghost imaging is proposed. Compared with the currently proposed scheme, does not decompose into three gray-scale images or encode it a grayscale for encryption. Instead, speckle patterns are used as key to directly encrypt. The ciphertext and thus has camouflage properties certain extent. Furthermore, optical setup easy be implemented generated by chaotic system, which only guarantees security but also...

Abstract Compressed sensing algorithm can be used in color ghost imaging to restore the image of object at less demand measurements times. However, quality reconstructed is usually not satisfactory. In this paper, we propose a sequential method that complete simple architecture and improve image, which optimize compressed get better result by post-processing measurement matrix establishing new process. Under condition same times reconstruction algorithm, compared with unpost-processing one,...

Ghost imaging technology has a great application potential in optical security because of its non-local characteristics. In this paper, on the basis computational ghost imaging, an authentication scheme is proposed that utilizes correspondence technique for preliminary reconstruction object image, and then authenticates image by nonlinear correlation algorithm. Different from previous schemes usually adopted random selection measurements, method consciously selects bucket detector...

Abstract Optical encryption based on ghost imaging has the advantages of high-speed parallel processing and multi-dimensional information. However, in practical application, a large number speckle patterns are required to achieve encryption, resulting significant increase transmission costs low efficiency. In this paper, compressive sensing-based method using sparse speckles is proposed. The used characteristics easy compression greatly reducing amount keys. According calculation, memory...