The simulated experiments of woisy images using ICA algorithms aften wiener de-noising are implemented and the PSNR(Peak Signal to Noise Ratio)is used to provide quantitative evaluations of the feasibility and efficiency of the method.

应用上述算法,对有噪图像信号进行仿真实验研究,通过比较降噪前后的峰值信噪比(Peak Signal to Noise Ratio,PSNR),说明该算法的可行性和有效性.

This paper elaborates the key technologies and realization models of baseband transmitter,and carried out the design and simulation based on TMS320C6713 DSP integrated development environment CCS. The simulation results provide theoretical guidance in constructing a SR transmitter.

Gindikin that complex analytic objects related to these domains will provide explicit realizations of unitary representations ofH?.

Here we provide certain conditions (more general than those in [Ka1]) which guarantee preservation of the topology under a modification.

We investigate conditions on kernel operators in order to provide prescribed orders of approximation in the Triebel-Lizorkin spaces.

We provide a direct computational proof of the known inclusion ${\cal H}({\bf R} \times {\bf R}) \subseteq {\cal H}({\bf R}^2),$ where ${\cal H}({\bf R} \times {\bf R})$ is the product Hardy space defined for example by R.

The space spanned by the translates of φv can only provide approximation order if the refinement maskP has certain particular factorization properties.

With the rapid progress of the electronic art, there has been a steadily increasing demand for still wider-bandwidth amplifiers. The recent introduction of distributed amplification concept has provided a new technique and powerful means for designing amplifiers with top cutoff frequencies far in excess of those previously obtainable with,ordinary amplifier circuits. A distributed amplifier of the low-pass type can easily be constructed to have a uniform frequency response from audio frequencies to frequencies...

With the rapid progress of the electronic art, there has been a steadily increasing demand for still wider-bandwidth amplifiers. The recent introduction of distributed amplification concept has provided a new technique and powerful means for designing amplifiers with top cutoff frequencies far in excess of those previously obtainable with,ordinary amplifier circuits. A distributed amplifier of the low-pass type can easily be constructed to have a uniform frequency response from audio frequencies to frequencies as high as several hundred me using conventional vacuum tubes. Unlike conventional circuits, distributed amplifiers have an attainable gain-bandwidth product which is not limited by shunt capacitance associated with the vacuum tubes and circuit wiring; the high-frequency limit being determined entirely by high-frequency effects within the tube proper.The purpose of this paper is to describe the basic principle of distributed amplification and to show how such an amplifier employing various types of transmission lines may be designed. Practical methods hi design and design details are given for a three-stage distributed amplifier, using fourteen 6AK5 pentodes with a frequency response of 0.1 mo to 140 mo and a gain of 33±1 db.Both the negative mutual-inductance m-derived and constant K artificial delay lines hare been used. The former offers the advantage of a more linear phase characteristic and a more uniform response both in amplitude and delay time.The experimental results corroborate the predictions based on the first-order theory described in this paper.

In this paper a group of sequencss wtih a δ-function autocorrelation are defined as the complementary-type sequences.Six kieds of sequences are presented and compared, and then it is shown that the basic form among them is the complementary series, and the others can be derived or generalized from it.The progresses of signal processing techniques have now led to a successful development of the complementary series phase reversal coded pulse compression radars, which have a low range sidelobe level within a narrow...

In this paper a group of sequencss wtih a δ-function autocorrelation are defined as the complementary-type sequences.Six kieds of sequences are presented and compared, and then it is shown that the basic form among them is the complementary series, and the others can be derived or generalized from it.The progresses of signal processing techniques have now led to a successful development of the complementary series phase reversal coded pulse compression radars, which have a low range sidelobe level within a narrow doppler spread range. A pair of the complementary series can be isolated from each other via time-sharing, frequency sharing, etc. The optimal pairs of complementary series and the optimal series should be selected according to the sidelobe level of the ambiguity function that is as low as possible.It had been demonstrated that the compleimntary series formed by concatenating, interleaving and mating, provide some radio navigation systems, such as Loran-C, with the possibilities of supressing the sky wave interference, automatic searching, recognizing and tracking.The complementary series also find its applications in the acoustic surface, -wave transducer coding and the infrared spectrometry.

The algorithms of sequence to one-one mapping and the principle of jump to one-one mapping are investigated. The mathematical models of basis structure for the binary code, left shifting code and Grey code, which provide bases for application of the mathematical models and physical models of Walsh function, is proposed. The partitioned sequential algorithms of higher space or multiinput binary rate multiplier is derived, the state matrix of the binary rate multiplier (proposed by D. J. Lancaster) is extended...

The algorithms of sequence to one-one mapping and the principle of jump to one-one mapping are investigated. The mathematical models of basis structure for the binary code, left shifting code and Grey code, which provide bases for application of the mathematical models and physical models of Walsh function, is proposed. The partitioned sequential algorithms of higher space or multiinput binary rate multiplier is derived, the state matrix of the binary rate multiplier (proposed by D. J. Lancaster) is extended to sequential matrix or multiinput matrix.