Basis transformation

The basic idea of symmetry analysis is that any basis of orbitals, displacements, rotations, etc. transforms either as one of the irreducible representations or as a direct sum (reducible) representation. Symmetry tools are used to first determine how the basis transforms under action of the symmetry operations. They are then used to decompose the resultant representations into their irreducible components. 

To balance a reaction, we write swap reactions relating B to B. Returning to the previous example, we wish to compute the reaction by which Ca-clinoptilolite transforms to muscovite and quartz. Reserving the clinoptilolite for the reaction s left side, we write the swap reactions for the basis transformation in matrix form. 

The natural localized orbitals introduced in Section 1.5 provide a useful alternative to the canonical delocalized MOs (CMOs) that are usually employed to analyze chemical bonding. The NAO and NBO basis sets may be regarded as intermediates in a succession of basis transformations that lead from starting AOs /, to the final canonical MOs , ... 

As eq. (24) shows, the 4x4 matrix B has been reduced by this basis transformation into the direct sum... 

For tensors of higher rank we must ensure that the bases are properly normalized and remain so under the unitary transformations that correspond to proper or improper rotations. For a symmetric T(2) the six independent components transform like binary products. There is only one way of writing xx xx, but since xx x2 = x2 xx the factors xx and x2 may be combined in two equivalent ways. For the bases to remain normalized under unitary transformations the square of the normalization factor N for each tensor component is the number of combinations of the suffices in that particular product. F or binary products of two unlike factors this number is two (namely ij and ji) and so N2 = 2 and x, x appears as /2x,- Xj. The properly normalized orthogonal basis transforming like... 

In these expressions differentiation and one-index transformations refer to the g integrals only of the Fock matrix [Eqs. (235) and (236)], treating the t elements as densities. The Fock matrix density elements in (Dj 1 and to the contravariant representation. If first derivative integrals in the MO basis is reduced to two occupied and two unoccupied indices (Handy et al., 1986). Note that 7] + T2 [Eqs. (257) and (258)] has the same structure as the <2) part of the MRCI Hessian (129). 

This situation is of course not satisfactory as observable quantities should be invariant with respect to unitary basis transformations. " Here, we outline the adiabatic route to a basis-invariant formulation of the theory. 

For the current MO basis, transform the required one- and two-electron integrals exactly. These integrals define the exact Hamiltonian operator H. For this operator, update the current CSF expansion vector c. 

In many applications, we are not able to distinguish all the N spins in the spectrum whose polarizations appear in P(t) due to spectral overlap. For example, if we take proton spectroscopy in a typical polymer, we are often only able to distinguish aliphatic from aromatic resonances. We can take the limited resolution into account by a basis transformation T of the polarization vector P(t). If there are M distinguishable lines in the sample, we rearrange the population vector so that the M sum polarizations, S = S,e p, form a vector S. The remaining N - M components of the transformed polarization vector are assembled into a vector D. 

The creation operators and the basis transform in the same manner, so when we apply a general unitary transformation to the basis, the creation operators suffer the same transformation. We can write in matrix form... 

When the Chebychev polynomial series is developed until the order k equals n - 1, i.e. until there are as many polynomial coefficients as there are observations, the matrix X can be considered an orthogonal basis for n-di-mensional space. The coefficients Pj are the co-ordinates in this alternative system of axes, this other domain, as it is often called. We could speak of the Chebychev domain in this case. Eq. (10) describes the basis transformation, i.e. the projection of the signal onto the alternative basis. The transform has only changed our perspective on the data nothing has been changed or lost. So we could also transform back, using the model at the start of all this ... 

People working in chemometrics will be familiar with another kind of basis transformation principal component analysis (PCA). They may be puzzled by the differences between PCA and orthogonal polynomials. Therefore we will compare the two. 

The Chebychev polynomial is just one possibility to construct a fixed orthogonal basis for n-dimensional space. There are many others. Interesting members of the family are the Hermite polynomial. Fourier and Wavelets. As there are many, the question arises how to choose between them. Before we are able to answer that question, we need to deal with another, more fundamental one why do a basis transformation in the first place ... 

Together, the sixteen elements of the columns a" and "d" form an alternative representation of the signal. We could say that we have just performed a basis transformation. The basis functions are presented in Table 2. The first element of column a is the inproduct of the signal and the basis function given in the first column of Table 2. The inproduct means that we calculate the product of the first element of the signal and the first element of the basis function, the product of the second element of the signal and the second element of the basis function, etc. Then we sum the products. As all but the first two products are zero, it is easy to see that the inproduct boils down to the sum we calculated earlier. The basis of Table 2 can be regarded a short-time Fourier basis for a window width of 2 points. 

The basis transformation can be calculated as a convolution, or as a matrix multiplication. When we call the matrix of Table 2 W , and the signal x, that multiplication is simply ... 

The time-complexity is generally 0(N ) for the diagonalisation algorithm and O(N ) for the basis transformation. 

The basis is a function of the image data since it depends on the autocovariance matrix for the given image. That is, each image will have its own basis transform. 

For these reasons the KLT is seldom used in practice. To circumvent these problems, sub-optimal basis transforms are employed which effectively decorrelate the image but are image-independent and have a reduced (linear or linear-log) time-complexity. 

---