Time-frequency distributions

Figure Al.6.30. (a) Two pulse sequence used in the Tannor-Rice pump-dump scheme, (b) The Husuni time-frequency distribution corresponding to the two pump sequence in (a), constmcted by taking the overlap of the pulse sequence with a two-parameter family of Gaussians, characterized by different centres in time and carrier frequency, and plotting the overlap as a fiinction of these two parameters. Note that the Husimi distribution allows one to visualize both the time delay and the frequency offset of pump and dump simultaneously (after [52a]).

The late portion of the EDR can be described in terms of the frequency response envelope G(a) and the reverberation time Tr (co), both functions of frequency [Jot, 1992b], G(co) is calculated by extrapolating the exponential decay backwards to time 0 to obtain a conceptual EDR(0, co) of the late reverberation. For diffuse reverberation, which decays exponentially, G(co) = EDR(0, co). In this case, the frequency response envelope G(co) specifies the power gain of the room, and the reverberation time Tr (CO) specifies the energy decay rate. The smoothing of these functions is determined by the frequency resolution of the time-frequency distribution used. 

Fourier analysis. Among several different kinds of time-frequency distributions, we employ the Wigner distribution function [12], defined as... 

Cohen, L. (1989) Time-frequency distribution - A review. Proceedings of the IEEE, 77 941-81. 

The evolutionary power-spectral density functions proposed by the authors are displayed in Fig. 4a, b in general, due to the nonstationarity and the approximations involved in the models, the iterative scheme as seen in Eq. 52 is required. Therefore Fig. 4c, d shows the iterated power-spectral density functions markedly it should be observed that although the different joint time-frequency distribution both the models can be used according to the seismic provisions for simulating ground motion accelerograms and evidently tending to reach the spectrum-compatibility criteria as shown in Fig. 5. 

The computed CWT leads to complex coefficients. Therefore total information provided by the transform needs a double representation (modulus and phase). However, as the representation in the time-frequency plane of the phase of the CWT is generally quite difficult to interpret, we shall focus on the modulus of the CWT. Furthermore, it is known that the square modulus of the transform, CWT(s(t)) I corresponds to a distribution of the energy of s(t) in the time frequency plane [4], This property enhances the interpretability of the analysis. Indeed, each pattern formed in the representation can be understood as a part of the signal s total energy. This representation is called "scalogram". 

The check sheet shown below, which is tool number five, is a simple technique for recording data (47). A check sheet can present the data as a histogram when results are tabulated as a frequency distribution, or a mn chart when the data are plotted vs time. The advantage of this approach to data collection is the abiUty to rapidly accumulate and analy2e data for trends. A check sheet for causes of off-standard polymer production might be as follows ... 

Larsen (18-21) has developed averaging time models for use in analysis and interpretation of air quality data. For urban areas where concentrations for a given averaging time tend to be lognormally distributed, that is, where a plot of the log of concentration versus the cumulative frequency of occurrence on a normal frequency distribution scale is nearly linear,... 

The use of various statistical techniques has been discussed (46) for two situations. For standard air quality networks with an extensive period of record, analysis of residuals, visual inspection of scatter diagrams, and comparison of cumulative frequency distributions are quite useful techniques for assessing model performance. For tracer studies the spatial coverage is better, so that identification of meiximum measured concentrations during each test is more feasible. However, temporal coverage is more limited with a specific number of tests not continuous in time. 

Besides the deviation mentioned above, the main problem with the dynamical information from the MF approximation is that it contains only one positive frequency and so the resulting real-time correlations cannot be damped or describe localizations on one side of the double well due to interference effects, as one expects for real materials. Thus we expect that the frequency distribution is not singly peaked but has a broad distribution, perhaps with several maxima instead of a single peak at an average mean field frequency. In order to study the shape of the frequency distribution we analyze the imaginary-time correlations in more detail. 

When considering environment it generally becomes difficult since actual service conditions are most of the time unpredictable. As an example, there is a systematic difference in the frequency distributions of liquid water content in rain. It appears that the areas most likely to have high values of liquid water are where there is a plentiful supply of moisture and a high instability in the atmosphere. The lowest values of liquid water are obtained from the climatic areas of light continuous rains such as that found along the northwest coast of the United States. 

With the aid of the power density spectrum, we can now give a complete description of how a linear, time-invariant filter affects the frequency distribution of power of the input time function X(t). To accomplish this, we must find the relationship between the power... 

The shape of a frequency distribution curve will depend on how the size increments were chosen. With the common methods for specifying increments, the curve will usually take the general form of a skewed probability curve with a single peak. However, it may also have multiple peaks, as in Fig 2, There are various analytical relationships for representing size distribution. One or the other may give a better fit of data in a particular instance. There are times, however, when analytical convenience may justify one. The log-probability relationship is particularly useful in this respect... 

Thru a combination of sedimentation and transmission measurements, a particle size distribution can be found. Tranquil settling of a dispersion of non-uniform particles will result in a separation of particles according to size so that transmission measurements at known distances below the surface at selected time intervals, will, with Stokes law, give the concn of particles of known diameter. Thus, a size frequency distribution can be obtained... 

Cumulative Frequency Distribution of 3 D-optimal sampling times... 

Figure 3.7 Frequency and cumulative frequency distributions of 3D-optimal sampling times for the Gompertz model, given the observations for subject 4. Vertical lines split the cumulative empirical distribution into equal probability regions.

Speleothem frequency distributions have provided a useful tool for broad comparisons, but they suffer from the problem of biased sampling strategies and low resolution at times of known abrupt change. The increased precision afforded by mass-spectrometric techniques will result in fewer studies using this approach to assess of growth frequency and, more often, records of continuous deposition and growth rate studies will be graphically illustrated. 

The ACAT model is loosely based on the work of Amidon and Yu who found that seven equal transit time compartments are required to represent the observed cumulative frequency distribution for small intestine transit times [4], Their original compartmental absorption and transit (CAT) model was able to explain the oral plasma concentration profiles of atenolol [21]. 

---