1: \documentclass[12pt,fleqn]{article}
2: \usepackage{epsfig,times,latexsym,enumerate,lscape}
3: \mathindent = 0.0cm
4: \newcommand{\prd}{Phys. Rev. D }
5: \newcommand{\pp}{Preprint gr-qc/}
6: \newcommand{\mnras}{Montly Not. RAS }
7: \newcommand{\prl}{Phys. Rev. Lett. }
8: \hoffset = 0.71in
9: \textwidth=6.0in
10: \textheight=8.50in
11: \oddsidemargin=0.0in
12: \evensidemargin-0.0in
13:
14:
15: \begin{document}
16:
17: \title{Data analysis of continuous gravitational wave: Fourier transform-II}
18: \author{D.C. Srivastava$^{1,2}$\thanks{e-mail: dcsrivastava@now-india.com}
19: $\,$ and S.K. Sahay$^1$\thanks{e-mail: ssahay@iucaa.ernet.in}\thanks{Present address: Inter University Centre for Astronomy and Astrophysics, Post Bag 4, Ganeshkhind, Pune - 411007, India}\\ \\
20: \normalsize $^1$Department of Physics, DDU Gorakhpur University, Gorakhpur-273009, U.P., India.\\
21: \normalsize $^2$Visiting Associate, Inter University Centre for Astronomy and Astrophysics, Post \\
22: \normalsize Bag 4, Ganeshkhind, Pune-411007, India.}
23: \date{}
24: \maketitle
25:
26:
27: \begin{abstract}
28: In this paper we obtain the Fourier Transform of a continuous gravitational
29: wave. We have analysed the data set for (i) one year observation time and (ii)
30: arbitrary observation time, for arbitrary location of detector and source
31: taking into account the effects arising due to rotational as well as orbital
32: motion of the earth. As an application of the transform we considered spin down and
33: N-component signal analysis.
34: \end{abstract}
35: \section{Introduction}
36: Detection of gravitational waves (GW) mainly depends on the efficient Fourier analysis
37: of the data of the output of the detector such as LIGO/VIRGO. In an earlier
38: paper we presented Fourier analysis of one day observation data set of the
39: response of a laser interferometer (Srivastava and Sahay, 2002 a). Hereafter
40: the reference to this paper and its results are noted by the prefix I.
41: We have seen that the
42: amplitude and the frequency modulations result into a large number of side bands
43: about the signal frequency $f_o$. Consequently, the maximum power lies in the
44: frequency $f_o + 2f_{rot}$ with amplitude reduction by 74\% to what one would
45: have expected due to increased data interval. Here $f_{rot}$ is the rotational
46: frequency of the earth. Hence, for GW detection it is
47: desirable to obtain FT for longer observation data. In the next
48: section we investigate one year observation data. Incidently there exists correspondences
49: and identifications of the present analysis to the analysis of paper-I. To facilitate
50: analogous modifications we introduce corresponding quantities with tilde viz.,
51: $\tilde{\cal C}$ and $\tilde{\cal D}$ in place of ${\cal C}$ and ${\cal D}$.
52: We obtain in section 3 generalisations of the results for arbitrary observation time.
53: As an application of the results obtained we consider spin
54: down and N-component signal analysis in sections 4 and 5. Last section contains
55: the discussion and conclusions of the paper.
56:
57: \section{Fourier transform for one year integration}
58: \label{sec:fty}
59: The phase of the GW signal of frequency $f_o$ at time $t$ is given via (I-25) i.e.
60: \begin{eqnarray}
61: \Phi(t) & = & 2\pi f_o t + {\cal Z}\cos (w_{orb}t - \phi ) +
62: {\cal N}\cos (w_{rot}t - \delta ) - {\cal R} - {\cal Q}
63: \end{eqnarray}
64:
65: \noindent where \\
66:
67: \begin{equation}
68: \label{eq:daypq}
69: \left.\begin{array}{lcl}
70: \vspace{0.2cm}
71: {\cal P}& = & 2\pi f_o {R_e\over c} \sin\alpha (\cos\beta _o(\sin
72: \theta \cos\epsilon \sin\phi + \cos\theta \sin\epsilon )
73: - \sin\beta _o \sin\theta \cos\phi )\, ,\\
74: \vspace{0.2cm}
75: {\cal Q}& = & 2\pi f_o {R_e\over c}\sin\alpha (\sin\beta _o (\sin\theta \cos
76: \epsilon \sin\phi + \cos\theta \sin\epsilon )
77: + \cos\beta _o \sin\theta\cos\phi ) \, ,\\
78: \vspace{0.2cm}
79: {\cal N}& = & \sqrt{ {\cal P}^2 + {\cal Q}^2 }\, ,\\
80: \vspace{0.2cm}
81: {\cal Z}& = & 2\pi f_o {R_{se}\over c}\sin\theta \, , \\
82: {\cal R}& = & {\cal Z}\cos\phi\, ,\\
83: \end{array} \right\}
84: \end{equation}
85:
86: \vspace{0.2in}
87:
88: \begin{equation}
89: \label{eq:daydelta}
90: \left.\begin{array}{lcl}
91: \vspace{0.2cm}
92: \delta & = & \tan^{- 1}\frac{{\cal P}}{{\cal Q}}\, ,\\
93: \vspace{0.2cm}
94: \phi' & = & w_{orb}t - \phi\, ,\\
95: \vspace{0.2cm}
96: \xi_{orb} & = & w_{orb}t\; = \; a\xi_{rot};\quad a \;= \; w_{orb}/w_{rot}\; \approx \; 1/365.26\, , \\
97: \xi_{rot} & = & w_{rot}t
98: \end{array} \right\}
99: \end{equation}
100:
101: \noindent where $R_{e}$, $R_{se}$, $w_{rot}$ and $w_{orb}$ represent respectively the
102: earth's radius, average distance of earth's centre from the origin of
103: Solar System Barycentre (SSB) frame, the rotational and the orbital angular velocity of the
104: earth. $\theta$ and $\phi$ denote the celestial colatitude and celestial longitude
105: of the source. $\epsilon$ and $c$ represent the obliquity of the ecliptic and the velocity of
106: light. $\alpha$ is the colatitude of the detector. Here $t$ represents the
107: time in seconds elapsed from the instant the sun is at the Vernal Equinox and $\beta_o$
108: is the local sidereal time at that instant, expressed in radians. \\
109:
110: \noindent The Fourier transform (FT) for one year observation time, $T_{obs}$ is given as
111:
112: \begin{equation}
113: \left[\tilde{h}(f)\right]_y = \int_0^{\bar{a}T} \cos[\Phi (t)]e^{-i2\pi ft}dt\; ;
114: \label{eq:hfy1}
115: \end{equation}
116: \begin{equation}
117: \bar{a} = a^{-1} = w_{rot}/w_{orb}\; ;\quad T =\; one\; sidereal\; day \; ;
118: \end{equation}
119: \begin{equation}
120: T_{obs} = \bar{a}T \simeq 3.14 \times 10^{7}\; sec.
121: \end{equation}
122:
123: \noindent This splits, similar to Eqs. (I-33-35), into two terms as
124: \begin{equation}
125: \left[\tilde{h}(f)\right]_y = I_{\bar{\nu}_-} + I_{\bar{\nu}_+} \; ;
126: \label{eq:inuy}
127: \end{equation}
128: \begin{eqnarray}
129: \label{eq:inubar}
130: I_{\bar{\nu}_-}& =& {1\over 2 w_{orb}}\int_0^{2\pi} e^{i \left[
131: \bar{\xi}\bar{\nu}_- +
132: {\cal Z}\cos (\bar{\xi} - \phi ) + {\cal N}\cos (\bar{a}\bar{\xi} - \delta ) -{\cal R} - {\cal
133: Q} \right] } d\bar{\xi}\, ,\\
134: I_{\bar{\nu}_+}& = & {1\over 2 w_{orb}}\int_0^{2\pi} e^{- i \left[
135: \bar{\xi}\bar{\nu}_+
136: + {\cal Z} \cos (\bar{\xi} - \phi ) + {\cal N}\cos (\bar{a}\bar{\xi} - \delta ) - {\cal R} - {
137: \cal Q} \right] } d\bar{\xi}\, , \\
138: \bar{\nu}_{\mp}& = & \frac{f_o \mp f}{f_{orb}} ; \quad \bar{\xi} \; = \; \xi_{orb} \; = \;
139: w_{orb}t
140: \end{eqnarray}
141:
142: \noindent where $f_{orb}$ is the orbital frequency of the earth. Hereafter, we neglect
143: $I_{\bar{\nu}_+}$ as it oscillates rapidly and contributes very
144: little to $\left[\tilde{h}(f)\right]_y$ and write $\bar{\nu}$ in place of $\bar{\nu}_-$. A
145: careful comparison of
146: Eq.~(\ref{eq:inubar}) with (I-33) i.e.
147: \begin{equation}
148: \label{eq:dayinu}
149: I_{\nu_-} = {1\over 2 w_{rot}}\int_0^{2\pi} e^{i \left[ \xi\nu_- +
150: {\cal Z}\cos (a\xi - \phi ) + {\cal N}\cos (\xi - \delta ) -{\cal R} - {\cal
151: Q} \right] } d\xi
152: \end{equation}
153:
154: \noindent obtained for one day observation data reveals that the integrand of the
155: equations are identical with following identifications and correspondences.
156:
157: \begin{equation}
158: \left.\begin{array}{ccc}
159: \delta &\leftrightarrow &\phi\; ,\\
160: {\cal Z}&\leftrightarrow &{\cal N}\; , \\
161: a &\leftrightarrow & \bar{a}.
162: \end{array} \right\}
163: \end{equation}
164:
165: \noindent We may, therefore, employ the results obtained there by introducing obvious
166: corresponding quantities i.e. $\bar{\cal B}$, $\bar{\cal C}$, $\bar{\cal D}$
167: in place of ${\cal B}$, ${\cal C}$, ${\cal D}$ leaving ${\cal A}$ unchanged. Hence
168:
169: \begin{equation}
170: \label{eq:hfy}
171: \left[\tilde {h}(f)\right]_y \simeq \frac{\bar{\nu}}{2 w_{orb}} \sum_{k = - \infty}^{k =
172: \infty} \sum_{m = - \infty}^{m = \infty} e^{ i {\cal A}}\bar{{\cal B}}[
173: \bar{{\cal C}} - i\bar{{\cal D}} ]\; ;
174: \end{equation}
175:
176: \begin{equation}
177: \left.\begin{array}{lcl}
178: \vspace{0.2cm}
179: {\cal A}& = &{(k + m)\pi\over 2} - {\cal R} - {\cal Q} \; ,\\
180: \vspace{0.2cm}
181: \bar{{\cal B}} & = & {J_k({\cal N}) J_m({\cal Z})\over {\bar{\nu}^2 - (\bar{a} k + m)^2}} \; , \\
182: \vspace{0.2cm}
183: \bar{{\cal C}} &= & \sin 2\bar{\nu}\pi \cos ( 2 \bar{a} k \pi - k \delta - m \phi ) -
184: { \bar{a} k + m \over \bar{\nu}}\{\cos 2 \bar{\nu} \pi \sin ( 2 \bar{a} k \pi - k \delta - m \phi )
185: + \sin ( k \delta + m \phi )\}\; , \\
186: \vspace{0.2cm}
187: \bar{{\cal D}} & = & \cos 2\bar{\nu}\pi \cos ( 2 \bar{a} k \pi - k \delta - m \phi ) +
188: {k \bar{a} +m \over \bar{\nu}}\sin 2 \bar{\nu} \pi \sin ( 2 \bar{a} k \pi - k \delta - m \phi )
189: - \cos ( k \delta + m \phi )
190: \end{array}\right\}
191: \end{equation}
192:
193: \noindent where $J$ stands for the Bessel function of first kind. Now the FT of the two
194: polarisation states can be written as
195:
196: \begin{eqnarray}
197: \label{eq:hfpy}
198: \left[\tilde{h}_+(f)\right]_y&=&h_{o_+}\left[\tilde{h}(f)\right]_y \nonumber \\
199: &\simeq & \frac{\bar{\nu} h_{o_+}}{2 w_{orb}} \sum_{k = - \infty}^{k =
200: \infty} \sum_{m = - \infty}^{m = \infty} e^{ i {\cal A}}\bar{{\cal B}}[
201: \bar{{\cal C}} - i\bar{{\cal D}} ]\,\;
202: \end{eqnarray}
203: and
204: \begin{eqnarray}
205: \label{eq:hfcy}
206: \left[\tilde{h}_\times(f)\right]_y&=& - i h_{o_\times}\left[\tilde{h}(f)\right]_y \nonumber \\
207: &\simeq & \frac{\bar{\nu} h_{o_\times}}{2 w_{orb}} \sum_{k = - \infty}^{k =
208: \infty} \sum_{m = - \infty}^{m = \infty} e^{ i {\cal A}}\bar{{\cal B}}[
209: \bar{{\cal D}} - i\bar{{\cal C}} ]
210: \end{eqnarray}
211:
212: \par The FT obtained contains double series of the Bessel
213: functions of the orders $k$ and $m$ ranging from $-\infty$ to $\infty$. It is
214: well known that the Bessel functions decrease rapidly as the order exceeds the
215: argument. Hence possible range of $k$ and $m$ over which the summation of the series is
216: to be considered depends on the arguments of Bessel functions i.e ${\cal Z}$ and
217: ${\cal N}$. Referring to Eq.~(\ref{eq:daypq}) it is found
218: that
219:
220: \begin{equation}
221: \left.\begin{array}{ccl}
222: \vspace{0.2cm}
223: {\cal Z}_{max} & = & 3133215\left(\frac{f}{1\, KHz} \right)\\
224: {\cal N}_{max} & = &134\left(\frac{f}{1\, KHz} \right)
225: \end{array}\right\}
226: \end{equation}
227:
228: \noindent The FT of a FM signal for
229:
230: \begin{equation}
231: \left.\begin{array}{lll}
232: \vspace{0.2cm}
233: f_o = 50\; Hz\, , & h_{o_+} = h_{o_\times} = 1 & \\
234: \vspace{0.2cm}
235: \alpha = \pi /4\, , & \beta_o = 0\, , & \gamma = \pi \, , \\
236: \theta = \pi /18\, , & \phi = 0\, , & \psi = \pi /4.
237: \end{array} \right\}
238: \label{eq:yearloc}
239: \end{equation}
240:
241: \noindent is shown in Fig.~(\ref{fig:yrealfm}). The spectrum
242: resolution is kept equal to $1/T_o$ i.e. $ 3.17 \times 10^{-8}$ Hz. We have
243: convinced ourselves by plotting the FT at higher resolutions that the
244: resolution of $1/T_o$ is sufficient to represent the relevant
245: peaks. We notice that the drop in amplitude at the source frequency is about
246: 98\% which may be
247: attributed to the presence of a large number of side bands.\\
248:
249: \par The complete response $R(t)$ of the detector and its FT
250: can be obtained similar to the one presented in section 4 of the paper-I.
251: Taking the results straight away from (I-43, 44, 45) one may get
252:
253: \begin{eqnarray}
254: \label{eq:rfy}
255: \left[\tilde {R}(f)\right]_y &= & \left[\tilde{R}_+(f)\right]_y +
256: \left[\tilde{R}_\times (f)\right]_y \nonumber \\
257: & = & e^{-i 2 \beta_o}\left[\tilde{h}( f
258: + 2 f_{rot})/2\right]_y\left[ h_{o_+}( F_{1_+} + i F_{2_+} )
259: + h_{o_\times} ( F_{2_\times} - i F_{1_\times} )\right] +\nonumber \\
260: && e^{i2\beta_o}\left[\tilde{h}( f - 2 f_{rot})/2\right]_y\left[ h_{o_+}
261: ( F_{1_+} - i F_{2_+} )
262: - h_{o_\times} ( F_{2_\times} + i F_{1_\times} )\right] +\nonumber \\
263: & & e^{-i\beta_o}\left[\tilde{h}( f + f_{rot})/2\right]_y\left[ h_{o_+}( F_{3_+} + i F_{4_+} )
264: + h_{o_\times} ( F_{4_\times} - i F_{3_\times} )\right] +\nonumber \\
265: && e^{i\beta_o}\left[\tilde{h}( f - f_{rot})/2\right]_y\left[ h_{o_+}( F_{3_+} - i F_{4_+} )
266: - h_{o_\times} ( F_{4_\times} + i F_{3_\times} )\right] +\nonumber \\
267: &&\left[\tilde{h}(f)\right]_y\left[ h_{o_+}F_{5_+} - i h_{o_\times}F_{5_\times}\right]
268: \end{eqnarray}
269:
270: \par Figure~(\ref{fig:ycr}) shows the power spectrum of the complete response of
271: the Doppler modulated signal. We have kept here all the parameters same as in FM.
272: In this caee also we observe that maximum
273: power is associated with $f_o + 2f_{rot}$ and the least lies in $f_o - f_{rot}$.
274: \section{Fourier transform for arbitrary observation time}
275: \label{sec:ftt}
276: It is important to obtain the FT for arbitrary observation
277: time. The results obtained will be
278: employed to outline how spin down of a pulsar due to the gravitational
279: radiation back reaction or due to some other mechanism can be taken into account.
280:
281: \noindent The FT for a data of observation time $T_o$ is given via
282:
283: \begin{equation}
284: \tilde{h}(f) \;= \;\int_0^{T_o}\cos [\Phi (t)]e^{-i2\pi f t} dt
285: \end{equation}
286:
287: \noindent which may be split
288:
289: \begin{equation} \tilde{h}(f)\ = I_{\nu_-} + I_{\nu_+} \; ;
290: \label{eq:inut1}
291: \end{equation}
292: \begin{eqnarray}
293: \label{eq:inut}
294: I_{\nu_-}& =& {1\over 2 w_{rot}}\int_0^{\xi_o} e^{i \left[
295: \xi\nu_- +
296: {\cal Z}\cos (a\xi - \phi ) + {\cal N}\cos (\xi - \delta ) -{\cal R} - {\cal
297: Q} \right] } d\xi\, ,\\
298: I_{\nu_+}& = & {1\over 2 w_{rot}}\int_0^{\xi_o} e^{- i \left[
299: \xi\nu_+
300: + {\cal Z} \cos (a\xi - \phi ) + {\cal N}\cos (\xi - \delta ) - {\cal R} - {
301: \cal Q} \right] } d\xi\, , \\
302: \nu_{\mp}& = & \frac{f_o \mp f}{f_{rot}} ; \quad \xi_o \; = \; w_{rot}T_o ;
303: \quad \xi \; = \; \xi_{rot} \; = \; w_{rot}t \, ,
304: \end{eqnarray}
305:
306: \noindent [refer to Eqs. I (33 - 35)].
307:
308: \noindent As $I_{\nu_+}$ contributes very little to $\tilde{h}(f)$
309: we drop $I_{\nu_+}$ and write $\nu$ in place of $\nu_-$. Using
310: the identity
311:
312: \begin{equation}
313: e^{\pm i\kappa\cos\vartheta} = J_o (\pm\kappa) + 2 \sum_{l = 1}^{l =
314: \infty} i^l J_l (\pm\kappa)\cos l\vartheta
315: \label{eq:bessel}
316: \end{equation}
317:
318: \noindent we obtain
319: \begin{eqnarray}
320: \tilde{h}(f) &\simeq & \frac{1}{2 w_{rot}} e^{i (
321: - {\cal R} - {\cal Q})} \int_0^{\xi_o} e^{i\nu\xi} \left[ J_o( {\cal Z} ) + 2
322: \sum_{k = 1}^{k = \infty} J_k ({\cal Z}) i^k \cos k (a\xi - \phi )\right]
323: \nonumber \\
324: && \times\,\left[J_o( {\cal N} ) + 2 \sum_{m = 1}^{m = \infty} J_m ({\cal N}) i^m
325: \cos m (\xi - \delta )\right] d\xi
326: \end{eqnarray}
327:
328: \noindent After performing the integration and proceeding in a straight-forward
329: manner we have
330:
331: \begin{equation}
332: \label{eq:hft}
333: \tilde{h}(f) \simeq \frac{\nu}{2 w_{rot}} \sum_{k = -
334: \infty}^{k = \infty} \sum_{m = - \infty}^{m = \infty} e^{ i {\cal A}}{\cal
335: B}[ \tilde{{\cal C}} - i\tilde{{\cal D}} ] \; ; \;
336: \end{equation}
337:
338: \begin{equation}
339: \left.\begin{array}{lcl}
340: \vspace{0.2cm}
341: {\cal A}& = &{(k + m)\pi\over 2} - {\cal R} - {\cal Q} \\
342: \vspace{0.2cm}
343: {\cal B} & = & {J_k({\cal Z}) J_m({\cal N})\over {\nu^2 - (a k + m)^2}} \\
344: \vspace{0.2cm}
345: \tilde{{\cal C}} &=& \sin \nu\xi_o \cos ( a k \xi_o + m\xi_o - k \phi - m \delta ) -
346: { a k + m \over \nu}\{\cos\nu\xi_o \sin ( a k \xi_o + m\xi_o - k \phi - m \delta ) \nonumber \\
347: \vspace{0.2cm}
348: &&+ \sin ( k \phi + m \delta )\}\nonumber \\
349: \vspace{0.2cm}
350: \tilde{{\cal D}} & = & \cos \nu\xi_o \cos ( a k \xi_o + m\xi_o - k \phi - m \delta ) +
351: {k a + m \over \nu}\sin \nu \xi_o \sin ( a k \xi_o + m\xi_o - k \phi - m \delta ) \nonumber \\
352: \vspace{0.2cm}
353: && - \cos ( k \phi + m \delta )
354: \end{array} \right\}
355: \end{equation}
356:
357: \noindent The FT of the two polarisation states of the wave can now be written as
358: \begin{eqnarray}
359: \label{eq:hfpt}
360: {h}_+(f)&=&h_{o_+}\tilde{h}(f) \nonumber \\
361: &\simeq & \frac{\nu h_{o_+}}{2 w_{rot}} \sum_{k = - \infty}^{k =
362: \infty} \sum_{m = - \infty}^{m = \infty} e^{ i {\cal A}}{\cal B}[ \tilde{{\cal C}}
363: - i\tilde{{\cal D}} ] \; ;
364: \end{eqnarray}
365: \begin{eqnarray}
366: \label{eq:hfct}
367: \tilde{h}_\times (f) &=&- i h_{o_\times}\tilde{h}(f) \nonumber \\
368: &\simeq & \frac{\nu h_{o_\times}}{2 w_{rot}} \sum_{k = - \infty}^{k =
369: \infty} \sum_{m = - \infty}^{m = \infty} e^{ i {\cal A}}{\cal B}[ \tilde{{\cal D}}
370: - i\tilde{{\cal C}} ]
371: \end{eqnarray}
372:
373: \noindent Now it is simple matter to obtain the FT of complete response. One gets
374: \begin{eqnarray}
375: \label{eq:rft}
376: \tilde{R}(f) & = & e^{-i 2 \beta_o}\tilde{h}( f
377: + 2 f_{rot})/2\left[ h_{o_+}( F_{1_+} + i F_{2_+} )
378: + h_{o_\times} ( F_{2_\times} - i F_{1_\times} )\right] +\nonumber \\
379: && e^{i2\beta_o}\tilde{h}( f - 2 f_{rot})/2\left[ h_{o_+}
380: ( F_{1_+} - i F_{2_+} )
381: - h_{o_\times} ( F_{2_\times} + i F_{1_\times} )\right] +\nonumber \\
382: & & e^{-i\beta_o}\tilde{h}( f + f_{rot})/2\left[ h_{o_+}( F_{3_+} + i F_{4_+} )
383: + h_{o_\times} ( F_{4_\times} - i F_{3_\times} )\right] +\nonumber \\
384: && e^{i\beta_o}\tilde{h}( f - f_{rot})/2\left[ h_{o_+}( F_{3_+} - i F_{4_+} )
385: - h_{o_\times} ( F_{4_\times} + i F_{3_\times} )\right] +\nonumber \\
386: &&\tilde{h}(f)\left[ h_{o_+}F_{5_+} - i h_{o_\times}F_{5_\times}\right]
387: \end{eqnarray}
388:
389: \noindent The FT of FM signal of a detector for a data of 120 days for
390:
391: \begin{equation}
392: \left.\begin{array}{lll}
393: \vspace{0.2cm}
394: f_o = 25\; Hz\, , & h_{o_+} = h_{o_\times} = 1 & \\
395: \vspace{0.2cm}
396: \alpha = \pi /6\, , & \beta_o = \pi /3\, , & \gamma = 2\pi /3 \, , \\
397: \theta = \pi /9\, , & \phi = \pi /4\, , & \psi = \pi /4. \\
398: \end{array} \right\}
399: \label{eq:120loc}
400: \vspace{0.4cm}
401: \end{equation}
402:
403: \noindent is plotted in Fig.~(\ref{fig:120realfm})
404: with a resolution of $\,1/2T_o \approx 9.67 \times 10^{-8}$ Hz. The Power
405: spectrum of the complete response is plotted in Fig.~(\ref{fig:120cr}). In
406: this case also the most of the power of the signal lies in the frequency $f_o + 2f_{rot}$
407: and the least with $f_o - f_{rot}.$
408: \section{Spin down}
409: \label{sec:fts}
410: Pulsars loose their rotational energy by processes like electro-magnetic breaking,
411: emission of particles and emission of GW. Thus, the rotational
412: frequency is not completely stable, but varies over a time scale which is of the
413: order of the age of the pulsar. Typically, younger pulsars have the largest spin
414: down rates. Current observations suggest that spin down is primarily due to
415: electro-magnetic breaking (Manchester, 1992 and Kulkarni, 1992).
416: Over the entire observing time, $T_o$ the frequency drift would be small
417: but it may be taken into account for better sensitivity.
418: To account this aspect we consider the evaluation of FT in a sequence of
419: time windows by splitting the interval $0-T_o$ in M equal
420: parts, each of interval $\bigtriangleup t$ $(T_o = M\bigtriangleup t$)
421: such that the signal over a window may be treated as monochromatic. The
422: strategy is to evaluate the FT over the window and finally to add the result.
423: This process has been suggested by Brady and Creighton (2000) and Schutz (1998)
424: in numerical computing and
425: has been termed as {\it stacking\/} and {\it tracking\/}. For any such window
426: the initial and final times may be taken respectively as $t_o + n\bigtriangleup t$ and
427: $t_o + (n + 1)\bigtriangleup t$ where $t_o$ represents the starting of the data
428: set and $0 \le n \le M - 1$.
429: The window under consideration is the $n_{th}$ window. Let
430: \begin{eqnarray}
431: I& =& \int_{t_o + n\bigtriangleup t}^{t_o + (n+1)\bigtriangleup t} h(\bar{t})
432: e^{-i2\pi f\bar{t}}d\bar{t} \nonumber \\
433: \label{eq:s1}
434: & = &\int_0^{\bigtriangleup t} h(t+t_o\;+
435: n\bigtriangleup t ) e^{-i2\pi (t+t_o+n\bigtriangleup t )f}dt\; ;\\
436: \bar{t} & = & t + t_o + n\bigtriangleup t
437: \label{eq:s2}
438: \end{eqnarray}
439:
440: \noindent Hence the FT for the data of the time-window under consideration is given via
441:
442: \begin{equation}
443: \label{eq:s3}
444: \left[\tilde{h}(f)\right]_s \;= \;\int_0^{\bigtriangleup t}\cos [\Phi (t + t_o
445: + n\bigtriangleup t )]e^{-i2\pi (t + t_o + n\bigtriangleup t ) f} dt
446: \end{equation}
447:
448: \noindent Taking the initial time of the data set
449: \begin{equation}
450: \label{eq:s4}
451: t_o = 0
452: \end{equation}
453:
454: \noindent and proceeding as in the previous section we obtain
455: \begin{eqnarray}
456: \left[\tilde{h}(f)\right]_s &\simeq & \frac{1 }{2 w_{rot}} e^{i \left[
457: 2\pi n(f_o - f)\bigtriangleup t
458: - {\cal R} - {\cal Q}\right]} \int_0^{\bigtriangleup t}
459: e^{i\nu\xi} \left[ J_o( {\cal Z} ) +
460: 2\sum_{k = 1}^{k = \infty} J_k ({\cal Z}) i^k \cos k (a\xi - \lambda )\right]
461: \;\times \nonumber \\
462: && \left[J_o( {\cal N} ) + 2 \sum_{m = 1}^{m = \infty} J_m ({\cal N}) i^m
463: \cos m (\xi - \zeta )\right] d\xi
464: \end{eqnarray}
465:
466: \noindent After integration we get
467: \begin{eqnarray}
468: \label{eq:hfs}
469: \left[\tilde {h}(f)\right]_s& =& {\nu \over 2 w_{rot}} \sum_{k = - \infty}^{k =
470: \infty} \sum_{m = - \infty}^{m = \infty} e^{ i {\cal A}_s}{\cal B} \left[
471: {\cal C }_s - i {\cal D}_s \right]\, ;
472: \end{eqnarray}
473: \begin{equation}
474: \left.\begin{array}{lcl}
475: \vspace{0.2cm}
476: {\cal A}_s& = &{(k + m)\pi\over 2} + 2\pi n\bigtriangleup t(f_o - f) - {\cal R} - {\cal Q}\\
477: \vspace{0.2cm}
478: {\cal B} & = & {J_k({\cal Z}) J_m({\cal N})\over {\nu^2 - (a k + m)^2}} \, , \\
479: \vspace{0.2cm}
480: {\cal C}_s &= & \sin (\nu \tau )\cos ( a k \tau + m \tau - k \lambda - m \zeta )
481: - { a k + m \over \nu}\{\cos (\nu \tau ) \sin ( a k \tau + m \tau - k \lambda - m \zeta ) \\
482: \vspace{0.2cm}
483: && + \sin ( k \lambda + m \zeta )\}\, , \\
484: {\cal D}_s & = & \cos (\nu \tau )\cos ( a k \tau + m \tau - k \lambda - m \zeta )
485: + {a k + m \over \nu}\sin (\nu \tau ) \sin ( a k \tau + m \tau - k \lambda - m \zeta )\\
486: \vspace{0.2cm}
487: && - \cos ( k \lambda + m \zeta) \, , \\
488: \vspace{0.2cm}
489: \lambda & = & \phi - a n \tau \, , \qquad \zeta \; = \; \delta - n\tau \, ,\\
490: \tau &= &w_{rot}\bigtriangleup t \, , \qquad n \; = \; 0, 1, 2, 3,......., M - 1.
491: \end{array} \right\}
492: \label{eq:sloc}
493: \end{equation}
494:
495: \noindent The FT of the complete response would now be given via
496: \begin{eqnarray}
497: \label{eq:rfs}
498: \left[\tilde{R}(f)\right]_s & = & e^{-i 2 \beta_o}\left[\tilde{h}( f
499: + 2 f_{rot})/2\right]_s\left[ h_{o_+}( F_{1_+} + i F_{2_+} )
500: + h_{o_\times} ( F_{2_\times} - i F_{1_\times} )\right] +\nonumber \\
501: && e^{i2\beta_o}\left[\tilde{h}( f - 2 f_{rot})/2\right]_s\left[ h_{o_+}
502: ( F_{1_+} - i F_{2_+} )
503: - h_{o_\times} ( F_{2_\times} + i F_{1_\times} )\right] +\nonumber \\
504: & & e^{-i\beta_o}\left[\tilde{h}( f + f_{rot})/2\right]_s\left[ h_{o_+}( F_{3_+} + i F_{4_+} )
505: + h_{o_\times} ( F_{4_\times} - i F_{3_\times} )\right] +\nonumber \\
506: && e^{i\beta_o}\left[\tilde{h}( f - f_{rot})/2\right]_s\left[ h_{o_+}( F_{3_+} - i F_{4_+} )
507: - h_{o_\times} ( F_{4_\times} + i F_{3_\times} )\right] +\nonumber \\
508: &&\left[\tilde{h}(f)\right]_s\left[ h_{o_+}F_{5_+} - i h_{o_\times}F_{5_\times}\right]
509: \end{eqnarray}
510:
511: \section{N-component signal}
512: \label{sec:ftn}
513: The FT in Eqs.~(\ref{eq:hft}) and~(\ref{eq:rft}) are valid for a
514: pulsar which emits GW signal at single frequency $f_o$. But there are known
515: physical mechanisms which generate GW
516: signals consisting of many components. An axially symmetric pulsar
517: undergoing free precession, emits quadrupole GW at
518: two frequencies, one equal to the sum of the spin frequency
519: and the precession frequency, and the other twice of it (Zimmermann and
520: Szedenits, 1979).
521: The quadrupole GW from a triaxial ellipsoid rotating
522: about one of its principal axes consists of one component only (Thorne, 1987).
523: In this case the signal has frequency
524: twice the spin frequency of the star. In general, if a star is
525: non-axisymmetric and precesses, the GW signal consists
526: of more than two components. Recently, new mechanisms e.g.
527: r-mode instability of spinning neutron stars (Anderson, 1998;
528: Lindblom et al., 1998; Owen, et al., 1998) and temperature asymmetry in the
529: interior of the neutron star with miss-aligned spin axis (Bildsten, 1988)
530: have been discussed in the literature.\\
531:
532: \par In view of the above discussion CGW signal may
533: consist of frequencies which are multiple of some basic frequencies. An
534: analysis of the GW data of N-component of signal has been made recently by
535: Jaranowski and Kr\'{o}lak (2000). We do not continue this analysis as the requisite formalism
536: is analogous to what we have presented in paper-I and the earlier sections of this paper
537:
538: \section{Discussion and conclusions}
539: \label{sec:concl4}
540: The analysis and results obtained in paper-I regarding FT of the
541: response of a Laser Interferometer have been generalised in this
542: paper. In this context following points must be noted.
543: \renewcommand{\theenumii}{\roman{enumii}}
544: \begin{enumerate}
545: \item For 120 days observation time data the resolution provided by
546: FFT, which is equal to $1/T_o$, is sufficient to represent the structure of side bands.
547:
548: \item In every case discussed, it turned out that the maximum power lies
549: in the frequency $f_o + 2f_{rot}$. However, this is not established conclusively
550: if this result is generic.
551:
552: \item Throughout our analysis in paper-I and this paper we have employed following
553: conditions.
554: \begin{enumerate}
555: \item The phase of the wave is zero at $t = 0$.
556:
557: \item The observation time of the data set is from $t = 0$ to $t = T_o$.
558: \end{enumerate}
559:
560: \item By trivial modifications we could get the general results with non-zero
561: initial time and phase.
562: \end{enumerate}
563:
564: \par It is pointed out that in choosing the parameters for the source and the detectors
565: we have been constrained by the computer memory and efficiency. Accordingly the choice of
566: the parameters has been a matter of convenience. The application of the
567: results obtained in the present and the earlier paper to the problem of all
568: sky search for continuous gravitational wave sources will be made in our
569: forthcoming article (Srivastava and Sahay, 2002 b).
570:
571: \section*{Acknowledgments}
572: The authors are thankful to Prof. S. Dhurandhar, IUCAA, Pune for stimulating
573: discussions. The authors are extremely thankful to the anonymous referee
574: for his hardwork in pinpointing the errors and making the detailed suggestions
575: which resulted in the improvement of the paper. The authors are also thankful to
576: IUCAA for providing hospitality
577: where major part of the work was carried out. This work is supported through
578: research scheme vide grant number SP/S2/0-15/93 by DST, New Delhi.
579:
580: \begin{thebibliography}{99}
581:
582: \bibitem{kn:4} Andersson, N., Astrophys. J, 1998, {\bf 502}, 708.
583:
584: \bibitem{kn:13} Bildsten, L., 1998, Astrophys. J. Lett., 501, L89.
585:
586: \bibitem{b3} Brady, P.R. and Creighton, T., 2000, \prd , 61, 082001.
587:
588: \bibitem{b6} Jaranowski, P. and Kr\'{o}lak, A., 2000, \prd , 61, 062001.
589:
590: \bibitem{kn:73} Kulkarni, S.R., Philos. Trans. R. Soc. London, 1992, {\bf 341}, 77.
591:
592: \bibitem{kn:80} Manchester, R.N., Philos. Trans. R. Soc. London, 1992, {\bf 341}, 3.
593:
594: \bibitem{kn:86} Owen, B.J., Lindblom, L., Cutler, C., Schutz, B.F., Vecchio,
595: A. and Andersson, N., 1998, \prd {\bf 58}, 084020.
596:
597: \bibitem{kn:77} Lindblom, L., Owen, B.J. and Morsink, S.M., 1998, \prl {\bf 80},
598: 4843.
599:
600: \bibitem{kn:5gnus} Schutz, B.F., 1998, \pp 9802020.
601:
602: \bibitem{b1} Srivastava, D.C. and Sahay, S.K., 2002 a, in revision, MNRAS, Data analysis of
603: continuous gravitational wave: Fourier transform-I.
604:
605: \bibitem{b1} Srivastava, D.C. and Sahay, S.K., 2002 b, in revision, MNRAS, Data analysis of
606: continuous gravitational wave: All sky search and study of templates.
607:
608: \bibitem{kn:18gnus} Thorne, K.S., in {\it Three Hundred Years of Gravitation.\/} eds. Hawking, S.W.,
609: Israel, W., ( Cambridge University Press, Cambridge, 1987).
610:
611: \bibitem{kn:38gnus} Zimmermann, M. and Szedenits, E., 1979, \prd {\bf 20}, 351.
612:
613: \end{thebibliography}
614: \newpage
615: \begin{figure}
616: \centering
617: \epsfig{file=yrealfm.eps,height=16.0cm}
618: \caption{FT of a FM signal of frequency, $f_o = 50$ Hz from a source located at $(\pi /18 , 0)$
619: with a resolution of $3.17 \times 10^{-8}.$ }
620: \label{fig:yrealfm}
621: \end{figure}
622: \begin{figure}
623: \centering
624: \epsfig{file=ycr.eps,height=16.0cm}
625: \caption{Power spectrum of the complete response of a modulated signal of frequency, $f_o = 50$
626: Hz from a source located at $ (\pi /18 ,0)$ with a resolution of $3.17 \times 10^{-8}.$}
627: \label{fig:ycr}
628: \end{figure}
629: \begin{figure}
630: \centering
631: \epsfig{file=120realf.eps,height=16.0cm}
632: \caption{FT of a FM signal of frequency, $f_o = 25$ Hz from a source located at $(\pi /9 , \pi /4)$
633: with a resolution of $9.67 \times 10^{-8}.$}
634: \label{fig:120realfm}
635: \end{figure}
636: \begin{figure}
637: \centering
638: \epsfig{file=120cr.eps,height=16.0cm}
639: \caption{Power spectrum of the complete response of a Doppler modulated signal of frequency,
640: $f_o = 25$ Hz from a source located at $(\pi /9 , \pi /4)$ with a resolution of
641: $9.67 \times 10^{-8}.$}
642: \label{fig:120cr}
643: \end{figure}
644:
645: \end{document}
646: