A comparison of nondeterministic and deterministic switchingmethods for DC-DC power converters

An analysis and experimental results of the random pulsewidth modulation (RPWM) and random pulse-position modulation (RPPM) methods for DC-DC converters are presented. The characteristics and performance of power converters under two randomized modulations are evaluated. The theoretical relationships of discrete harmonics, continuous noise, and output-voltage ripple of the RPWM and RPPM schemes are established and compared with those of a standard deterministic pulsewidth modulation (PWM) scheme in a buck converter. Limitations in the randomly switched DC-DC converters, which have not been addressed before, are highlighted. Randomized schemes generally have inherent problems in achieving low-output-voltage ripple due to the low-frequency continuous noise within the passband of the converter filter. The inherent low-frequency noise-induced voltage ripple problems of nondeterministic switching methods for DC converters are confirmed experimentally. The performance of the RPPM method is found to be closer to the standard PWM method than that of the RPWM method. For DC-DC power conversion, the RPPM method offers much better output-voltage performance than the RPWM method     

Novel random PWM schemes with weighted switching decision

A new random pulse-width modulation (RPWM) scheme, which includes a weighted decision switching process, has been developed and tested. The switching strategy can be applied to the entire range of the modulation index. The scheme combines the advantages of the deterministic nature of the standard pulse-width modulation (PWM) schemes, and the nondeterministic spread-spectral characteristic of RPWM schemes. Test results have confirmed the improved spectral performance of the proposed schemes over a standard RPWM scheme at both low- and high-modulation indexes     

Experimental investigation of a naval propulsion drive model with the PWM-based attenuation of the acoustic and electromagnetic noise

An extensive experimental investigation of a 40-hp ac drive was conducted with the focus on mitigation of the acoustic and electromagnetic noise, and vibration, by means of random pulsewidth modulation (RPWM) employed in the drive's inverter. The drive was a laboratory model of an electric propulsion system for naval vessels, particularly electric submarines, in which the noise mitigation is crucial for survivability. Three PWM methods were compared: 1) the classic deterministic PWM, characterized by a constant switching period equal to the sampling period of the digital modulator; 2) the known RPWM technique, referred to as RPWM I, in which the switching and sampling periods are varied simultaneously in a random manner; and 3) a novel RPWM method, referred to as RPWM II, with a constant sampling period and the switching periods randomly varied around an average value equal to the sampling period. The experimental results have confirmed the mitigating properties of RPWM with respect to the acoustic and electromagnetic noise, and vibration. Because of the fixed sampling frequency, the RPWM II technique is technically more convenient than the classic RPWM I method and only marginally less effective in flattening the peaks of noise spectra. Importantly, conclusions drawn from the described study are valid for ac drives in general.     

Analysis of power and power spectral density in PWM inverters withrandomized switching frequency

A novel method for accurate calculation of power spectra of the pulsewidth modulation (PWM) voltage-source inverter with randomized switching frequency is presented. The case of a limited pool of switching frequencies, convenient and adequate for technical purposes, is considered, and a mathematical background for this type of random PWM (RPWM) is provided. It is shown that the limited-pool RPWM gives rise to not only a continuous spectrum, but, also, under certain circumstances, pure power spectral components (harmonics), typical for the deterministic PWM. Criteria for the existence of harmonics are given, and formulae for both the power and the power spectral density are derived and verified experimentally     

Noise analysis of DC-AC random PWM schemes

Random pulsewidth modulation (RPWM) techniques for DC-AC power converters usually work well with high-sampling frequency. When RPWM schemes are implemented in digital signal processors (DSPs), the sampling frequency is limited by the speed of the processors. This paper analyzes the noise spectrum of various DSP-implemented RPWM techniques using a statistical approach. The relationship of the noise components and the sampling frequency for both the standard RPWM and the weighted RPWM methods is generalized. The dependency of the noise characteristics on various factors is calculated theoretically and verified experimentally. The generalized noise theory for randomized PWM switching provides useful practical guidelines to the choice of sampling frequency     

Microprocessor-based random PWM schemes for DC-AC power conversion

Two classes of microprocessor-based random PWM (RPWM) real-time schemes for DC-AC power conversion are compared and evaluated. Performance of the RPWM schemes based on mathematical and logical approaches is examined. The proposed schemes exhibit excellent harmonic content with all low and high-order harmonics suppressed, and are suitable for both MOSFET and IGBT inverters     

Analysis, design, and implementation of a random frequency PWMinverter

Random pulse width modulation (RPWM) approaches can make the harmonic spectrum of inverter output voltage be continuously distributed without affecting the fundamental frequency component, and thus the acoustic noise and mechanical vibration of an inverter-fed AC motor drive are greatly reduced. However, the analysis and design of the RPWM mechanisms are generally not so trivial for practical engineers that their applications are limited. In this paper, a random frequency PWM (RFPWM) inverter and its practical design procedure are presented. First, the effects of the attributes of a random signal on the inverter output harmonic spectrum distribution characteristics are analyzed using an intuitive concept, then based on which, the quantitative design. Simulink simulation and implementation of the proposed RFPWM inverter are introduced. The proposed RFPWM inverter is employed to power an indirect field-oriented induction motor drive. The simulated and measured results indicate that the uniform random distribution of inverter output harmonic spectrum and thus smaller acoustic noise and mechanical vibration are obtained by the proposed RFPWM scheme     

Spread spectrum in three-phase inverter by an optimised dual randomised PWM technique

Randomised pulse width modulation (RPWM) technique has become a viable alternative to deterministic pulse width modulation (DPWM). By spreading the power spectrum in a continuous noise, this new technique better complies with electromagnetic compatibility (EMC) requirements for conducted electromagnetic interferences (EMI) and allows reducing the emitted acoustic noise in variable speed drives (VSDs). The most popular RPWM schemes are randomised pulse position modulation (RPPM) and randomised carrier frequency modulation (RCFM). The combination (RCFM-RPPM) or dual RPWM (DRPWM) has also been proposed. In this article, we propose an optimised DRPWM (ODRPWM) for the three-phase inverter. First, the modulating principle is proposed, and then, a mathematical model of power spectral density (PSD) of the output voltage is developed and validated for the three schemes, namely RPPM, RCFM and RCFM-RPPM. PSD analysis shows that the proposed scheme is more effective on spreading PSD. Moreover, this analysis reveals optimal parameters of randomisation for a maximum spread of the PSD. The optimisation problem is then modelled and solved using two powerful non-linear methods.     

基于噪声调制的薄膜电阻加热温度控制系统

基于脉宽调制（PWM）技术的控制系统广泛应用于工业和日常生活中,随机脉宽调制（RP-WM）技术也已提出.利用耦合到系统的环境噪声和系统本身的噪声经加工处理后调制控制信号,由噪声信号的自相关函数和功率谱密度函数给出噪声调制脉冲的平均开关频率和平均加热功率的计算方法.利用模拟器件设计了一个简单的基于噪声调制的薄膜电阻加热温度控制系统.实验表明：控制系统能够稳定地运行,在一定范围内能抑制噪声,实现了高精度温度控制.     

Nichtlineare Signalverarbeitung für die Sprachkommunikation

This paper provides an elementary introduction to problems and methods of nonlinear signal processing and presents on overview of recent developments in this field. An important topic is the modeling and processing of one-dimensional signals as exemplified in simple voice communication systems. The interplay and integration of deterministic and nondeterministic approaches is illustrated with recent research findings.     

Analysis of random PWM switching methods for three-level powerinverters

This paper presents a generalized theory which covers both two-level and three-level random pulsewidth modulation (RPWM) schemes. Various three-level RPWM schemes with low switching frequency are presented and compared with two-level schemes. Three-level RPWM schemes have less discrete harmonics and continuous noise than two-level RPWM schemes. They have desirable spectral characteristics and can be employed in high-voltage inverter-fed motor drives. Measurements have confirmed the theory and the attractive features of three-level RPWM schemes     

Improved frequency characteristics of the randomized PWM boost rectifier

A randomized pulse width modulation (RPWM) algorithm is applied in the control unit of the boost rectifier to achieve improved frequency characteristics in the wide band. First, the introduction of the RPWM switching is reflected in a smaller increase of the total harmonic distortion (THD) factor in the input current. Nevertheless, decrease of the power factor is negligibly small. Second, the power spectrum density (PSD) of the input current is estimated and measured to evaluate the influence of randomization in the high-frequency range. This approach offers an effective and credible prediction method for reduction of conductive electromagnetic interference (EMI) by using the RPWM switching.     

Current control for induction motor drives using random PWM

Current control in voltage-source inverters with random pulsewidth modulation (RPWM) is investigated. The random modulation is introduced to alleviate the undesirable acoustic, vibration, and EMI effects in inverter-fed AC drive systems. A novel RPWM digital technique with dithering of the switching frequency and compensation of the processing time is described. Design of the current control loop is discussed. Results of investigation of an experimental drive system are presented, proving the feasibility of the proposed solutions     

A New Hybrid Random PWM Scheme

This paper proposes a new hybrid random pulsewidth modulation (PWM) scheme based on a TMS320LF2407 DSP, in order to disperse the acoustic switching noise spectra of an induction motor drive. The proposed random PWM pulses are produced through the logical comparison of a pseudorandom binary sequence (PRBS) bits with the PWM pulses corresponding to two random triangular carriers. For this reason, the PWM pulses of the proposed scheme possess the hybrid characteristics of the random pulse position PWM and the random carrier frequency PWM. In order to verify the validity of the proposed method, the simulations and experiments were conducted with a 1.5-kW three-phase induction motor under the 2.5-A load condition. The DSP generates the random numbers, the PRBS bits with a lead-lag random bit (8 bit) and the three-phase reference signals. Also, a frequency modulator MAX038 makes the randomized frequency triangular carrier (3 $pm$ 1 kHz). From the results, the proposed scheme shows good randomization effects of the voltage, current, and acoustic noise of the motor as compared with conventional scheme (3 kHz).      

Analysis and spectral characteristics of a spread-spectrumtechnique for conducted EMI suppression

Frequency modulation (FM) and random switching methods have been used for reducing conducted electromagnetic interference (EMI) in power converters. Limited theoretical studies and comparisons of these schemes, however, are available. In this paper, a detailed analysis and the spectral characteristics of a random carrier-frequency (RCF) technique for suppressing conducted EMI in an offline switched-mode power supply are presented. The analysis provides a theoretical platform for studying the characteristics of this random switching scheme. The level of randomness is defined for the RCF scheme and varied in the converter example so that the effects on the power spectra can be demonstrated. Theoretical predictions of the spectral characteristics of this scheme are confirmed with measurements. The RCF scheme has been compared with the standard constant-frequency pulsewidth modulation (PWM) scheme and the FM scheme. Comparisons of their spectral performance show that the RCF scheme has better conducted EMI suppression than the FM and standard PWM schemes     

OFDM系统中的盲信道估计

本文从OFDM信号的矩阵表示出发,分析比较了OFDM系统中现有的各种盲信道估计方法。OFDM盲信道估计方法分为两类,一类是统计型方法,它利用了发送信号和接收信号的统计特性;另一类是确定型方法,它利用了发送调制信号的固有特性。一般而言,统计型方法的计算量较小,但是估计精度不高且估计的实时性不好;而确定型方法的估计精度较高,实时性较好,但是其计算量较大。计算机仿真表明,这些盲信道估计方法的性能受信道参数尤其是多普勒频率影响很大,盲信道估计的实用化有待进一步研究。     

Maximum value determination of non-linear products in cable television networks

This article presents a summary of methods for the determination of non-linear products in cases when they are nondeterministic but depend on random parameters. When applied in the design of a cable television network, this method makes clear, through its results, the impact of non-linear products on the level of output voltages of backbone amplifiers. It allows quick selection of the optimum solution in accordance with the max/min number of transferred TV signals and the output voltage of the amplifiers.     

Unity PF current-source rectifier based on dynamic trilogic PWM

One remaining step in perfecting the stand-alone, unity power factor, regulated current-source pulse-width-modulated (PWM) rectifier is to reduce cost, by bringing the twelve-valve converter (consisting of three single-phase full bridges that operate with two-level or bilogic PWM) to the six-valve bridge. However, the six-valve topology requires a three-level or trilogic PWM strategy that can handle feedback signals. A general method of translating three-phase biologic PWM signals into three-phase trilogic PWM signals is presented. The method of translation retains the characteristics of the bilogic PWM, including the frequency bandwidth. Experiments show that the trilogic PWM signals produced by the method can handle not only stabilizing feedback signals but also signals for active filtering     

Training set design for connected speech recognition

Two methods for generating training sets for a speech recognition system are studied. The first uses a nondeterministic statistical method to generate a uniform distribution of sentences from a finite state machine (FSM) represented in digraph form. The second method, a deterministic heuristic approach, takes into consideration the importance of word ordering to address the problem of coarticulation effects. The two methods are critically compared. The first algorithm, referred to as MARKOV, converts the FSM into a first-order Markov model. The digraphs are determined, transitive closure computed, transition probabilities are assigned, and stopping criteria established. An efficient algorithm for computing these parameters is described. Statistical tests are conducted to verify performance and demonstrate its utility. A second algorithm for generating training sentences, referred to as BIGRAM, uses heuristics to satisfy three requirements: adequate coverage of basic speech (subword) units; adequate coverage of words in the recognition vocabulary (intraword contextual units); and adequate coverage of word pairs bigrams (interword contextual units)     

An approach to include stochastic rough surface scattering into deterministic ray-optical wave propagation modeling

A new method to include stochastic rough surface scattering into deterministic ray-optical wave propagation modeling is derived. It can be utilized in conjunction with the concept of ray launching. Similar to the Kirchoff formulations, the approach is based on a tangential plane approximation, i.e., it is applicable to surfaces with gentle undulations, whose horizontal dimensions are large compared to the incident wavelength. However, in contrast to the Kirchoff models, which are only valid for either slightly rough or very rough surfaces, the proposed stochastic scattering approach includes both the coherent and incoherent components at the same time. The purely deterministic ray-based modeling is expanded by a "stochastic" component, allowing, for the first time, to account for nondeterministic scattering in ray-optical wave propagation modeling.