In high-level synthesis, it is necessary to find the design with the highest suitability to the design directives under various quality criteria. Optimization techniques result in quality increases impossible to achie...
详细信息
In high-level synthesis, it is necessary to find the design with the highest suitability to the design directives under various quality criteria. Optimization techniques result in quality increases impossible to achieve in other steps of the synthesis process. Two optimization techniques are presented, one to raise the throughput and another to reduce the area and output latency of the design. A scheduling method which can obtain rate-optimal schedules with high processor utilization is also presented. These techniques form an approach for the high-level synthesis of digital signal processing algorithms for high performance designs.< >
The authors discuss a transient analysis of transmission lines terminated by nonlinear subnetworks. In this analysis, the circuit is partitioned into the transmission line and nonlinear subnetwork by using a substitut...
详细信息
The authors discuss a transient analysis of transmission lines terminated by nonlinear subnetworks. In this analysis, the circuit is partitioned into the transmission line and nonlinear subnetwork by using a substitution source. The transmission line is solved by the phasor technique or perturbation method, and the nonlinear subnetwork by a numerical integration technique. The substitution source giving rise to the transient response is calculated by application of the relaxation hybrid harmonic balance method.< >
An approach for the evaluation of the coefficients of the generalized McClellan transform is presented in order to design 2-D finite impulse response (FIR) filters with arbitrarily shaped frequency responses. Two clas...
详细信息
An approach for the evaluation of the coefficients of the generalized McClellan transform is presented in order to design 2-D finite impulse response (FIR) filters with arbitrarily shaped frequency responses. Two classes of transforms which fit to the design of quadrantally symmetric filters and that of general centro-symmetric filters, respectively, are investigated. An analytical solution for the transform coefficients is derived by formulating a least-square approximation problem for the designed transform. This solution permits fast calculation of the coefficients in terms of the expected contour mapping requirements for the designed 2-D filter. In this method, the transform coefficients can be approximately scaled without requiring an optimization procedure.< >
The geometric methods introduced in H. Wong, 'A geometric view of /spl Sigma/-/spl Delta/ modulation,' ieee Trans. Circ. Sys., Vol. 39, no. 6, pp. 402-405, are applied to the basic architecture of third-order,...
详细信息
ISBN:
(纸本)0780312813
The geometric methods introduced in H. Wong, 'A geometric view of /spl Sigma/-/spl Delta/ modulation,' ieee Trans. Circ. Sys., Vol. 39, no. 6, pp. 402-405, are applied to the basic architecture of third-order, single-path sigma-delta modulation. An attempt is made to answer the fundamental questions on the stability problem of the third-order system, i.e., can such a modulation be stable, and if so, under what conditions? Combined with the application of bifurcation theory, the stability and other dynamical properties of the modulation are revealed, and certain guidelines directly applicable to the design are provided.< >
The Curtice quadratic, Materka, Statz and Rodriguez nonlinear models are compared from DC, constant voltage (CV) and RF points of view in order to determine which is most suitable for nonlinear wideband circuit design...
详细信息
The Curtice quadratic, Materka, Statz and Rodriguez nonlinear models are compared from DC, constant voltage (CV) and RF points of view in order to determine which is most suitable for nonlinear wideband circuit design. For this comparison, a variety of different size GaAs MESFET devices is employed. Data for a 25-/spl mu/m and 900-/spl mu/m gate-width device with a pinch-off voltage of -2V and a gate-length of 0.5-/spl mu/m are extensively used for the model comparison work.< >
An arbitrary hyperplane approach for both codebook generation and search for vector quantization (VQ) is introduced. It is a divisive algorithm and aims at minimizing the sum-of-squared-errors. The relation to tree-st...
详细信息
An arbitrary hyperplane approach for both codebook generation and search for vector quantization (VQ) is introduced. It is a divisive algorithm and aims at minimizing the sum-of-squared-errors. The relation to tree-structured vector quantization is discussed, and the superiority of the proposed algorithm is demonstrated through a comparison with the LBG (generalized Lloyd) algorithm for a 16-dimension VQ compression of still pictures.< >
The authors propose a class of single hop multiple bus networks based on balanced incomplete block design (BIBD). After introducing the basic concepts of multiple bus networks, a theorem is presented, which justifies ...
详细信息
The authors propose a class of single hop multiple bus networks based on balanced incomplete block design (BIBD). After introducing the basic concepts of multiple bus networks, a theorem is presented, which justifies the restriction of critically single hop bus (CSHB) networks. A model for performance analysis of CSHB networks is introduced, and it is proved that a CSHB network is optimal if both bus load and processor fanout are constant. Since topologies of such a balanced CSHB based on BIBD do not exist for all n, the number of processors, two theorems are included for the construction of CSHB networks for all n, with bounded skew in the bus loading and processor fanout.< >
A BiCMOS biquadratic continuous-time filter is presented. The cell has been designed as 'noise shaper' for signal processing in the electronic read-out channel of elementary particles detectors. The network sy...
详细信息
A BiCMOS biquadratic continuous-time filter is presented. The cell has been designed as 'noise shaper' for signal processing in the electronic read-out channel of elementary particles detectors. The network synthesizes a semi-Gaussian response where the shaping time (nominally 20 ns) and the quality factor (nominally 1/2) can be externally adjusted to realize the desired transient response. A linearity error lower than 1% is achieved for step input signal amplitudes up to 240 mV. The input referred noise is about 40 nV//spl radic/Hz, and the total current consumption is 200 /spl mu/A.< >
The authors present a new boost topology which can be operated in constant switching frequency and zero-current-switching. In addition, the voltage stresses on the active switches are as low as the traditional boost p...
详细信息
The authors present a new boost topology which can be operated in constant switching frequency and zero-current-switching. In addition, the voltage stresses on the active switches are as low as the traditional boost pulse width modulated (PWM) converter. The control curves are derived through the normalized state-plane trajectory. The design procedure is given through a specific example. Simulation waveforms are provided to verify the analytical results.< >
S. C. Chan and C. W. Kok (1993) established the conditions on the modulation for the modulated filter bank to be perfect-reconstruction (PR) when the linear phase prototype filter is of length 2mM and satisfies the po...
S. C. Chan and C. W. Kok (1993) established the conditions on the modulation for the modulated filter bank to be perfect-reconstruction (PR) when the linear phase prototype filter is of length 2mM and satisfies the power complementary condition. It was found that the modulation is generated by the class of unitary matrices. It is shown that similar unitary classes of solutions also exist for the case M = 2mM + /spl beta/. This greatly increases the possible choices of modulation and filter length in the design of finite impulse response (FIR) PR quadrature mirror filter (QMF) banks. Moreover, additional regularity conditions are imposed on the filter banks to obtain a unitary class of modulated orthonormal compactly supported wavelets.< >
暂无评论