Lattice fitter is superior to its counterpart of direct structure in numerical properties and modularity etc., but it is more complex to realize. This paper suggests a simplified lattice structure for the realization of FIR digital filters, so that the hard-ware requirement is only half as many as that of the conventional lattice filter, comparable to that of the direct structure. The synthesis algorithms that transform the direct structure into the simplified lattice structure are tfiven. Furthermore, the lattice synthesis algorithms for linear phase FIR digital filters are simplified. The main results are illustrated by a practical example.

The new concepts of reducible fundamental cutset and reducible fundamental loop through an analysis of ladder network is developed. A new analytic method, named structural analysis method, is presented. It evaluates the controllability and measurablity of network based on its structure rather than on its state equation. Some problems in the conventional network analysis can be solved by this method. Results obtained by this method are in-good agreement with that obtained by state space analysis, but this method is much simpler for use. The practical examples are given.

The partial charge-simulation method is presented for calculating the capacitance of the rectangular coaxial line with offset inner conductor. The capacitance calculated using this method is in good agreement with the experimental results. The method can im-prove the accuracy by increasing the number N of the series.

The optimal magnetized field distributions of variable polarization effect of ferrite are studied by using the coupled-wave theory and variation principle. Several kinds of optimal shapes of magnetized field are given. The theoretical calculation are in good agreement with the experimental results.

The canonical piecewise-linear method is used to analyse the nonlinear DC fault circuits and determine the potential at the preseleted test ports in the circuit. The method needs less storage capacity and has higher computing efficiency. It is a good method for analysis of fault circuits.

Main reflector module sharing technique for dual reflector antennas is pre-sened in this paper, which makes main reflectors with same focus but different aperture share one set of module by means of optimizing subreflectors. To optimize the subreflector, a fast convergent optimization method is given and satisfactory results are obtained.

The lasertron is a kind of superpower microwave source, originated in1980 s. At present the lasertron is still under development. In this paper, the basic principle a introduced and one-dimensional analysis is presented. The computing model is rather simple, while the characteristics and the differences with respect to traditional microwave tubes are-figured out. From numerical results the significance of space-charge effect is emphasized. The main problems of lasertron are pointed out at the last part.

It is proposed to introduce a specially designed cavity: in the interaction region, the circular cross-section is kept, beyond which a slowly varying elliptical auxiliary transition section is used to suppress the unwanted modes. The cavity is so designed: the fields of unwanted modes can penetrate to the auxiliary section, but that of the operating mode can not. If we then load the auxiliary section, the unwanted modes can be attanuated but the operating mode (_cTE_0n and _sTE_1n) maintained.

A new active C notch filter circuit with two operational amplifiers and five capacitors is presented. The circuit can fulfil the second-order lowpass, band-pass, band-stop and symmetrical band-stop functions. Its pole parameters Q and (ω₀ can be adjusted independently. Design examples and experimental results are given. It is investigated why the attenuation is not infinite at notch frequency ω₀.

In this paper, the dispersion equations of the waveguide filled with N dielectric slabs in E-plane are presented. The characteristics of the dispersion equations of the waveguide filled with N dielectric slabs in E-plane are studied. The effective dielectric constants of the waveguide filled with two dielectric slabs are calculated.

The problem of wave propagation in moving optical waveguide is solved by using the Lorents transformation, that is, the solutions of the problem are derived first in the reference frame in which the optical waveguide is at rest, next the obtained results are transformed into laboratory frame, and then the behaviors of wave propagation in moving optical waveguide are Seduced.

According to the transverse resonance method the dispersion relation is derived based on the analysis of fringe equivalent circuit by using Babinet principle. The attenuation constant causing by radiation is obtained. The explanation of the leakage waves is-given. The theoretical results agree with the experimental ones.

Some experimental results are reported and discussed. The n+p junction leakage currents and the MOS generation lifetimes of wafers pretreated by the 1G method can be improved by 1-3 and 1-2 orders of magnetitute respectively, while the yields of diodes mode from silicon wafers pretreated by the chlorine oxidation gettring are approximately 3 times of those untreated. It is suggested that a combination of 2 or more defect control methods are necessary to meet the increasingly stringent demands of silicon substrates quality required for LSI, VLSI circuits.

A new processing method in which the antireflective coating (AR) Si₃N₄ is deposite by LPCVD, at the same time, the Al-alloyed high low junction BSF is formed, is described. In the method, the procedure is simplified for fabricating high efficiency silicon solar cells, that is the AR-BSF to be formed in one step, and because the system is in high vacuum and the cells are in the ambinet of N₂ as well as H₂, so that the surface characteristics of emitters in the cells is improved.