Abstract
A novel procedure for explicit construction of the entries of real symmetric matrices with assigned spectrum and the entries of the corresponding orthogonal modal matrices is presented. The inverse eigenvalue problem of symmetric matrices with some specific sign patterns (including hyperdominant one) is explicitly solved too. It has been shown to arise thereof a possibility of straightforward solving the inverse eigenvalue problem of symmetric hyperdominant matrices with assigned nonnegative spectrum. The results obtained are applied thereafter in synthesis of driving-point immittance functions of transformerless, common-ground, two-element-kind RLC networks and in generation of their equivalent realizations.
1. Introduction
During the past few decades, many papers [1–16] have studied the inverse eigenvalue problems (IEPs) of various types. The solution existence of the specific IEPs was generally considered in [1, 3–8, 10, 11, 13, 14] without explicit formulation of the corresponding procedure for solution construction, whereas in [2, 9, 12, 15, 16] this has been accomplished. The main result of [16] is the proof that IEP of symmetric hyperdominant (hd) matrices with assigned nonnegative spectrum has at least one solution which has also been constructed. This settled an old IEP opened in [17]. Hyperdominant matrices have nonnegative diagonal and nonpositive off-diagonal entries and nonnegative hd margins of rows (hd margin of a row is the sum of entries in that row). The tool used in [16] to construct the th-order hd matrix with assigned spectrum was the th-order orthogonal Hessenberg matrix constructed as a special product of plane rotations [15]. Hessenberg matrices naturally arise in study of symmetric tridiagonal matrices, skew symmetric, and orthogonal matrices [13, 14, 18]. A matrix is upper (lower) Hessenberg if its entry (, ) vanishes whenever .
In practical work, it is commonly assumed to be better not to form Hessenberg matrices explicitly, but to keep them as products of plane rotations. On the other hand, explicit construction of real symmetric matrices with nonnegative spectrum, which either have hd sign pattern or are truly hd, is proved to be an inevitable task in considering the synthesis of driving-point immittance functions of passive, transformerless, common-ground, two element-kind networks and in generation of their equivalent realizations [17–19]. networks are comprised solely of resistors (), inductors (), and capacitors (). Driving-point immittance function of a lumped, time invariant, linear electrical network is either a driving-point impedance , or a driving-point admittance ( is the complex frequency; , are real numbers; ). It is well known that a real rational function in can be driving-point immittance function of network if and only if it is positive real function in ; or similarly, a necessary condition for a stable square matrix of real rational functions in to be driving-point immittance matrix of a passive network is that be positive real matrix [20, 21]. A few tests for ascertaining positive real properties of functions and/or matrices can be found in [20, 21]. In [22] it has been pointed out the role of hd matrices in synthesis of both passive and active, transformerless, common-ground multiports. Unlike [16], this paper presents explicit construction of entries of real symmetric matrices with arbitrarily assigned spectrum and the entries of the corresponding orthogonal modal matrices. It also presents explicit construction of real symmetric matrices with assigned spectrum and with specific sign patterns (including hd one). Thereof, a solution to the IEP of symmetric, truly hd matrices with assigned nonnegative spectrum is produced. Some of the obtained results are then applied in synthesis of driving-point immitances of transformerless, common-ground, two-element-kind networks and in generation of their equivalent realizations. The two proposed realization procedures are illustrated by an example. Throughout the paper denotes direct sum, denotes transpose of , bold capital letters denote matrices and stands for the th-order unit or identity matrix.
2. Explicit Solution to the IEP of Real Symmetric Matrices by Using Canonic Orthogonal Transformations
Let be assigned spectrum of the sought real symmetric matrices and let be spectral matrix. Consider a set of orthogonal matrices : which are either rotators ( and ) or reflectors ( and ). A useful set of orthogonal matrices is From the following two matrix recurrent relations we readily obtain real symmetric matrices and , which are both congruent and similar to Columns of the orthogonal modal matrix correspond to eigenvectors of . Out of different possibilities of using (2.3) in generation of and , only the two selected by (2.4) produce explicit expressions of entries of and in terms of and the entries of . from (2.4) will be shown later to take on lower (upper) Hessenberg form with the entries explicitly expressed too. For the sake of brevity, we will restrict our consideration only to the first of relations (2.3), bearing on mind the possibility of treating the second one similarly. For and we readily obtain and , by using (2.1) and (2.3): Let , , and , and let us firstly introduce in (2.5) the following notation: Thereafter, observing the partition of and obtained in (2.5) it can readily be anticipated the partition of subsequent matrices as follows: where is the symmetric matrix, is row vector, is matrix, is modified eigenvalue , and For from (2.1)–(2.3), (2.8) it follows that For , let us define: , , and thereafter and Then, from (2.8)-(2.9) it follows the identification which enables the partition of (2.10) in () to be like that of (2.10) in (), and that partition of (2.10) be rather simple Let . Having uncovered the partition pattern of , we can pursue partitioning of backwardly from to , by using (2.10)). Afterwards, we can produce , by using (2.10)-(2.11). The results are Since and then after defining and it follows from (2.12)-(2.13) Since and then on introducing , and we obtain from (2.14)-(2.15) the partition of which is amenable to the production of its entries in explicit form and is suitable for further discussion about solving some specific IEPs For , we consecutively obtain from and that generally it holds Since (2.17) and (2.17), then from () it follows that Observe that it is not necessary to calculate “”s from (2.18), but only the modified eigenvalues from (2.17) since it holds and . As it is , then for from (2.10)) it follows that The real symmetric matrix with assigned spectrum and the explicitly expressed entries can be derived from (2.16) and (2.20), bearing on mind that “”s and “”s are calculated by using , , modified eigenvalues (2.17) and : where denotes , denotes , denotes , and denotes . Entries of are , , and . They are calculated according to the following steps:
(a)Select arbitrarily the entries of orthogonal matrices , given by (2.1);(b)with , calculate the modified eigenvalues by using (2.17);(c)calculate and ;(d)calculate the entries of , by using (2.21).
Matrix (2.4) is orthogonal modal matrix established from eigenvectors of . We will now prove that is not only orthogonal, but also lower Hessenberg with explicitly expressed entries. Let us firstly produce and whose partition will enable us to anticipate the partition of If we now suppose that where is orthogonal upper Hessenberg matrix then since according to (2.2), it holds we may write further for By using (2.2), (2.23)-(2.24), it follows that and thereby it is proved our previous assumption that , where (2.23) is the orthogonal upper Hessenberg matrix with entries expressed explicitly. And finally, for from and (2.4), (2.23), we obtain and The entries of the orthogonal lower Hessenberg matrix are defined as follows: By using the similar arguments as in derivation of entries of matrix , the orthogonal matrix which is to be produced by using (2.4) can be shown to take on upper Hessenberg form. Proving of this fact goes with similar paces that were used for obtaining and it is left to the reader.
3. The Explicit Solution of the IEP of Real Symmetric Matrices with Some Specific Sign Patterns
Let the real eigenvalues from the spectrum be arbitrarily enumerated, thereby establishing the sequence . The nonnegative sequence will be denoted by , and the nonpositive one by Firstly, we will prove two lemmas.
Lemma 3.1. If the sequence is increasing [decreasing], then in (2.21) and the sequence .
Proof.Since , then it is trivial to see from (2.17) and (2.18) that all diagonal entries of are nonnegative, that is, and no matter whether the sequence is increasing or decreasing. By virtue of orthogonality of , we have . If is increasing sequence, then for we have and for we obtain From (2.17) and the last of inequalities (3.1) it follows and . If is decreasing sequence, then for we have and for we obtain From (2.17) and the last of inequalities (3.2) it follows and . This completes the proof of lemma. For a nonpositive sequence, an analogous lemma can be formulated.
Lemma 3.2. If the sequence is increasing [decreasing], then in (2.21) and the sequence .
Proof. It is similar to that of Lemma 3.1, but in this case the diagonal entries of are nonpositive, that is, and , no matter whether the sequence is increasing or decreasing (see (2.18)).
Now, we shall formulate a new theorem related to explicit solving of IEP of real symmetric matrices with some specific sign patterns.
Theorem 3.3. If are arbitrarily selected angles from the range , then the entries of real symmetric matrices with assigned spectrum , produced by (2.21), can attain the following twelve sign patterns (zero entries are permitted), depending on selection of matrices (see (2.1)).
Case 1.
Case 2.
Case 3.
Proof. If , then the signs of and depend solely on selection of canonic orthogonal matrices . For any sign of sequence and its monotonicy realized through enumeration of its members, one can readily check the sign patterns stated above: by using (2.18) to determine signs of the diagonal entries in and by using Lemma 3.1 or Lemma 3.2 to determine signs of . Observe that only in Case 1 when that is, when the sequence is nonnegative and increasing (but not strictly), matrix is produced with hd sign pattern, including the possible presence of zero entries. may attain a sparse structure if, for example, some eigenvalues are equal. To see that, let us firstly suppose . Then from (2.17)-(2.18) it follows that and thus obviously making the matrix (2.21) with sparse structure. By using (2.17)-(2.18), (2.21) and both Lemmas, we can readily infer that if and the sequence is strictly monotone, then matrix (2.21) is produced with no zero entries in all three considered cases.
Remark 3.4. Let . Then, since and (recall that U is orthogonal), it follows that . Also, when the sequence is increasing (decreasing), then the sequence is decreasing (increasing). These facts and Theorem 3.3 offer a possibility of determining the sign pattern of without really inverting . Furthermore, by using (2.17)-(2.18), (2.21), can be calculated explicitly, also without really inverting .
Theorem 3.5. Let the positive increasing sequence be the spectrum of produced by using Case 1 of Theorem 3.3. Then there always exists such a diagonal matrix with positive diagonal entries which makes truly hyperdominant.
Proof. If , then by Case 1 of Theorem 3.3, the nonsingular matrix will have hd sign pattern and by Remark 3.4 will be nonnegative matrix. Since , then the nonsingular symmetric matrix is produced with hd sign pattern, but it may not be truly hd, unless hd margin of each of its rows (or columns) is nonnegative (recall that hd margin of a row or a column is sum of all entries in that row or column). If , then hd margin of the th row (or the th column) in is given by Let we arbitrarily select and let and . Then, from (3.6) it follows that , that is, and . This not only means that has hd sign pattern, but that it is truly hd furthermore. Obviously, as much as “”s are assumed greater, the greater will be row (column) hd margins of . This completes the proof of theorem.
4. Explicit Solution of IEP of Hd Matrices with Uncommitted and with Assigned Nonnegative Spectrum
Theorem 4.1. Let be a set of angles selected from the range and let be nonnegative spectrum of the real symmetric matrix which is to be produced as truly hd. Suppose that through enumeration of eigenvalues the sequence is made increasing. Then, matrix given by (2.21) will be truly hyperdominant if is sufficiently great.
Proof. Since by assumption the conditions of Theorem 3.3 (Case 1) are satisfied, then produced by using (2.21) has hd sign pattern. As it is , then from (2.17)-(2.18), (2.21) it follows that hd margin of the th row (or column) from can be in general represented as where “” coefficients are defined as follows: According to Case 1 of Theorem 3.3, both and are nonnegative when . Then, from (4.2) we see that , whereas other “”s may be nonpositive. Since “”s depend only on selection of “”s, then by presuming , we obtain from (2.21) and and from (4.1) we conclude that in general it holds: Although is produced with hd sign pattern, it will not be truly hd unless each of its row (column) hd margins is nonnegative, that is, . The column vector p with entries (4.1) can be written as From (4.4) we finally obtain Since “”s and “”s in (4.5) are not certainly nonnegative and since the sequence is increasing, then firstly by arbitrary selection of differences and and thereafter a sufficiently great , all hd margins can be made nonnegative, that is, the matrix can be always produced as truly hyperdominant. This completes the proof of this theorem.
Presentation of explicit solution to the IEP of truly hd matrices with assigned nonnegative spectrum is now in order. It has been proved in [16] that this IEP always has at least one solution and that infinitely many others can be produced thereof by using Givens rotations. Solution of that IEP is important in electrical network synthesis of driving-point immittance functions and matrices of both passive and active, common-ground, transformerless, two-element-kind networks and in generation of various classes of canonic and noncanonic equivalent realizations [19, 22]. In [16] we have proved the existence of solution to the IEP of hd matrices with assigned nonnegative spectrum, but here we shall present the explicit construction of solution matrix entries by using other arguments. This represents the explicit solution of the problem opened in [17].
Theorem 4.2. For any set of real nonnegative numbers there always exists at least one (and infinitely many) real symmetric hyperdominant matrices having these numbers as eigenvalues. In other words, IEP of symmetric hd matrices with assigned nonnegative spectrum always has at least one solution.
Proof. We will take the same assumptions as in Theorem 4.1, except for . Through enumeration of eigenvalues, the nonnegative sequence is made increasing. Then, according to Theorem 3.3 (Case 1), the symmetric matrix with spectrum and the entries determined by (2.21), is produced with hd sign pattern, no matter what selection of has been made. Observe that in Case 1 and . To make truly hd, we will prove the existence of such “”s that make all “”s (and hence all “”s) in (4.1) nonnegative. Let we introduce the following positive sequence Then, by using (4.2) we obtain a consistent set of inequalities that ensure nonnegativity of all “”s in (4.1) For from (4.8) we obtain and from (4.7) . Then, and . From (4.11)-(4.12) it follows that and [inequalty (4.12) is the same as (4.9) if . For , we obtain from (4.9) and for , we obtain from (4.10) . To summarize, we have proved that: (a) , for and (b) and , for . And finally, from (4.10) we obtain for and . Since the matrix has hd sign pattern and each of its row (column) hd margins is equal to , then is truly hd matrix. This completes the proof of the theorem.
Remark 4.3. It relates to calculation of entries of . In Theorem 4.2 it is proved that and . It is assumed . Since , then . For it follows from (4.6) By using (2.17)-(2.18), (2.21), (4.13) we can easily calculate all entries of the (initial) hd matrix . Other hd matrices having the same spectrum can be produced thereof by application of Givens rotations, one at a time.
5. Application of the Obtained Results in Electrical Network Synthesis
It is well known that synthesis methods of passive, common-ground, transformerless, two-element-kind networks yield topological configurations which are severely restricted by the method chosen [19]. By using of the results above, a new class of non-canonic, driving-point immittance realizations of passive, common-ground, transformerless, two-element-kind networks with minimum number of both nodes and elements of one kind can be generated with possibility of reduction in number of elements of other kind. The network synthesis is always performed by using normalization of both the complex frequency and the impedance . If Ω is a selected normalization frequency, then the normalized frequency is . Similarly, if is a selected normalization resistance, then the normalized impedance is . Thereby we achieve [20]: (a) lesser dispersion of coefficients in normalized functions and (b) dimensionless manipulation of quantities. The normalized resistance of resistor is . The normalized impedance of an inductor is (-normalized inductance). The normalized impedance of a capacitor is (-normalized capacitance). To physically realize a network after synthesis, a denormalization process must be performed. The actual parameter values of elements are calculated as follows: . From now on it will be assumed that normalized synthesis is being carried out, but the lower index “’’ we be dropped from component labels for brevity.
It is well known that if a real rational function in can be realized as driving-point impedance , then it can be also realized as driving-point admittance [20]. And similarly, if it can be realized as driving-point admittance , then it can also be realized as driving-point impedance . The transformation turns the synthesis of driving-point impedance to synthesis of driving-point impedance [20]. It also turns the synthesis of driving-point admittance to synthesis of driving-point admittance These driving-point imittances are at first realized by prototype networks and thereof are produced the desired networks in the following way: capacitors in and networks remain the same, but the resistor from network turns to inductor in network with the same parameter value. Also, transformation turns the synthesis of driving-point impedance to synthesis of driving-point impedance It also turns synthesis of driving-point admittance to synthesis of driving-point admittance These imittances are realized by prototype networks and the desired networks are produced thereof in the following way: the resistor from network turns to inductor in network with the same parameter value, and the capacitor from network turns to resistor in network with reciprocal parameter value. Bearing all the aforementioned on mind, we can obviously restrict our consideration only to synthesis of driving-point impedance functions of networks, which satisfy the following well known analytic necessary and sufficient conditions [20]: (a) is real rational function in , (b) It has only simple poles on negative real axis, or at the origin. At infinity it cannot have pole and (c) Residues of these poles are real and positive and
In general, the first canonic Foster's expansion (form) of [20] reads where is residue of the pole . The network which realizes driving-point impedance (5.1) with minimum number of nodes , minimum number of resistors and minimum number of capacitors is depicted in Figure 1. Observe that neither the resistors, nor the capacitors share common-node and hence the overall network realization is said to be non common-grounded.
Now, we will present our synthesis procedure. If for a given driving-point impedance we found that and/or , then in the preamble of the realization procedure and/or should be at first extracted from (5.1) and realized by a series connection of resistor and capacitor , thereby leaving for realization the driving-point impedance with solely poles lying on the negative real axis. In the sequel we will assume that has only such poles.
Let and be diagonal matrices with strictly positive diagonal entries corresponding to the normalized capacitances and conductances, respectively. If we arbitrarily choose a nonsingular matrix T, then the reciprocal passive networks which realize and will have the same natural frequencies. By arbitrary selection of nonsingular diagonal matrices , a broad class of nonsingular matrices can be generated with assumption , where and are orthogonal matrices. Since , thenVarious network topologies can be produced by different choices of and . But, only by selecting and , the networks with minimum number of common-ground capacitors are produced; and only by selecting and , the networks with minimum number of common-ground resitors are produced. Let us select and , and let us assume in (5.1): and Since , then from (5.2) it follows that The matrices which are effectively realized by common-ground network with nodes [th node is the common-ground] are and provided that both are truly hd. According to (5.1) and (5.3) it holds and By using Theorem 3.3 [Case 1, and ] we infer that (2.21) is produced with hd sign pattern and no zero entries and with strictly positive inverse. Matrix (2.26) is lower Hessenberg with nonnegative entries, except for negative “”s. The same conclusions relating to and also hold if we apply Case 1 of Theorem 3.3 with and , except for “”s in (2.26) are then negative and “”s are positive. To realize we must select in (5.3) either or thus obtaining either or By assuming and it follows from (2.26), (5.1), (5.3) To prove the existence of a physical realization of both and we still have to determine the positive column vector which, according to Theorem 3.5, makes truly hd with possibly zero hd margins of at most rows. Let hd margin of the th row in be and let . If ( unities), then Let we introduce a column vector () of arbitrarily assumed real nonegative numbers. Since is strictly positive, then it always can be find a diagonal matrix with positive diagonal entries, such that . Herefrom, we obtain , that is, that . Since , then it follows , bearing on mind that at most “”s can be equal to zero. These “”s indices correspond to indices of those rows (or columns) in which have zero hd margins. Then, from the overall network vanish resistors connecting common-ground to nodes with the same indices as that of rows (columns) with zero hd margins [22]. For different selections of , different algorithms and different topologically and parametrically equivalent realizations emerge. For example, if we select where it is according to (2.26), (5.4) then the column vector of row (column) hd margins of matrix with hd sign pattern reads This means that is truly hd. We will now present two algorithms for realization of driving-point impedances which rely on the results developed above.
Algorithm 1. Realization of with minimum number of common-ground capacitors and non-reduced number of resistors
(10) Commencing with calculate the entries of , by using (2.26) and (5.4).
(20) Arbitrarily select some and then calculate and .
(30) Calculate by using (5.5) and the entries of hd matrix Calculate , by using (5.6).
(40) Realize by common-ground, transformerless, conductance network. This can be done easily, almost by visual inspection of [22]. Attach to the ports of that network, enumerated by 1, and , the common-ground capacitors with normalized capacitances respectively. The th port of the overall network realizes driving-point impedance , provided that all other ports are left open-circuited.
Algorithm 2.
Realization of
with
minimum number of common-ground capacitors and the reduced number of
resistors
(10) The same as step of Algorithm 1. Let .
Recall that .
(20) Select and Thereafter, by using (2.21) and (5.4), calculate Calculate ,
where is -dimensional column vector.
(30) Calculate the entries
of and its hd margin where Set for other hd margins
(40) Realize by common-ground, transformerless, conductance
network and attach to its th port
the common-ground capacitor with normalized capacitance The th
port of the overall network realizes driving-point impedance provided that all other ports are left
open-circuited.
5.1. A Numerical Example
Consider realization of the real rational function as driving-point impedance of common-ground transformerless network with minimum number of capacitors and the reduced number of inductorsThis function satisfies the necessary and sufficient conditions for driving-point immittances of networks: (a) it is an odd real rational function in ; (b) it has only simple poles located strictly on imaginary axis; and (c) residues of those poles are real and positive. Therefore, can be realized both in two Foster's and in two Cauer's canonic forms [20]. The partial fraction expansion of reads The reactance function corresponding to is and it is depicted in Figure 2. The first canonic Foster's realization of with minimum number of nodes, noncommon-ground capacitors and inductors is depicted in Figure 3. Thereon are denoted the normalized values of parameters.
In Figure 4, it is depicted the first canonic Foster's realization of driving-point impedance from Figure 3 with selected normalization frequency and selected normalization resistance . The network is excited by a sinusoidal current generator having constant current amplitude and discretely varying frequency within the range kHz. If the complex representative of generator current is and the complex representative of the voltage across its terminals is , then the complex driving-point impedance of the overall network is The modulus of that is, (usually called impedance) obtained through PSPICE simulation within the range kHz is depicted in Figure 5.
Now, we will realize by using the proposed Algorithm 2. After transformation, we firstly produce the function where is driving-point impedance of network which should be expanded into partial fractions as follows: In step of Algorithm 2 we determine the orthogonal matrix by using and (see (2.26) and (5.4))By assuming in step , we further easily obtain and In step we firstly calculate and then hd margins of its rows (columns), In step we calculate the normalized capacitances of common-ground capacitors: and Realization of driving-point impedance by transformerless, common-ground network with minimum number of nodes reduced number of inductors and minimum number of common-ground capacitors begins by realization of conductance matrix which can be accomplished almost by inspection of that matrix [22]. Then, to the th port of the realized conductance network, it should be connected to the capacitor The third port of the overall network realizes the driving-point impedance provided that all other ports are left open-circuited. By embedding in the third port a series connection of resistor and capacitor with the normalized parameter values 1 and 2/5, respectively, and by applying transformation thereafter, we finally produce noncanonic network which realizes with minimum number of nodes and capacitors and with reduced number of inductors. That network is depicted in Figure 6 whereon are denoted the normalized (dimensionless) values of parameters.
In Figure 7, is depicted the noncanonic realization of driving-point impedance from Figure 6 with selected normalization frequency and selected normalization resistance The network is excited by a sinusoidal current generator with constant current amplitude and with discretely variable frequency within the range kHz. If the complex representative of generator current is and the complex representative of the voltage across its terminals is , then the complex driving-point impedance of the overall network is The modulus of that is, (usually called impedance) obtained through PSPICE simulation within the range kHz is depicted in Figure 8. Since networks in Figures 4 and 7 are intentionally designed to be equivalent, then their driving-point impedances must have the same variations in frequency, as can be verified from Figures 5 and 8 qualitatively and more precisely by using numerical results of simulation.
6. Conclusions
A novel procedure for explicit construction of entries of real symmetric matrices with assigned spectrum is developed by using a group of four types of canonic, second-order, orthogonal transformations. It has been also shown that the orthogonal modal matrices corresponding to the produced real symmetrix matrices, are either lower or upper Hessenberg with explicitly constructed entries too. Thereafter, the inverse eigenvalue problems of real symmetric matrices with twelve specific types of sign patterns (including hyperdominant one) are explicitly solved providing that the signs of eigenvalues are the same (zeros are permitted) and that they are enumerated such as to establish the increasing or decreasing sequence. It is proved to arise thereof a possibility of explicit solving the inverse eigenvalue problem of symmetric hyperdominant matrices having either uncommitted or assigned nonnegative spectrum. The results obtained are then applied in synthesis of driving-point immittance functions of transformerless, common-ground, two-element-kind networks and in generation of their equivalent realizations with minimum number of nodes. The synthesis procedures proposed herein turn the synthesis problem of any immittance function of the two-element-kind network to the synthesis problem of impedance function of a prototype network.