1: \documentclass[twocolumn,a4paper,superscriptaddress,showpacs,showkeys]{revtex4}
2: \usepackage[ansinew]{inputenc}
3: \usepackage[dvips]{graphicx}
4: \usepackage{psfig}
5: \usepackage{amssymb}
6: \begin{document}
7: \title{The Phase Information Associated to Synchronized Electronic Fireflies}
8:
9: \author{J.L. Guisset}
10: \affiliation{Universit\'e Libre de Bruxelles, Belgium}
11: \author{J.L. Deneubourg}
12: \affiliation{Universit\'e Libre de Bruxelles, Belgium}
13: \author{G.M. Ram\'{\i}rez \'Avila}
14: \thanks{Partially supported by Belgium Technical Cooperation}
15: \affiliation{Universit\'e Libre de Bruxelles, Belgium}
16: \affiliation{Universidad Mayor de San Andr\'es, La Paz, Bolivia}
17:
18: \begin{abstract}
19: An electronic implementation referring to fireflies ensembles
20: flashing in synchrony in a self-organization mode, shows the
21: details of the phase-locking mechanism and how the phases between
22: the electronic oscillators are related to their common period.
23: Quantitative measurements of the timing signals link the limits of
24: a steadily established synchronization to the physics of the
25: electronic circuit. Preliminary observations suggest the existence
26: of bifurcation-like phenomena.
27: \end{abstract}
28: \pacs{05.45.Xt, 85.60.Bt, 89.75.-k}
29:
30: \keywords{synchronization; coupled oscillators; optoelectronic
31: devices}
32:
33: \maketitle
34: %--------------------------------------------------------------
35: %%%%%%%%%%%%%
36: % Section 1 %
37: %%%%%%%%%%%%%
38: \section{Introduction}
39: \label{intro} Self-organization is a widespread feature appearing
40: under a variety of living and inanimate systems. Synchronization
41: that can be understood as an adjustment of rhythms of oscillating
42: objects due to their weak interaction \cite{PIKOVSKY}, represents
43: one of the forms of self-organized matter \cite{BLEKHMAN}.
44:
45: There are numerous examples of systems of coupled oscillators able
46: to induce structured behaviors between the interacting oscillators
47: \cite{MIROLLO-STROGATZ,STROGATZ,WINFUL,WIESENFELD,HOHL}.
48:
49: The synchronized flashes of huge ensembles of fireflies in
50: south-asian countries swarm trees is one of these surprising
51: self-organization effects. The phenomenon was already mentioned
52: three centuries ago by the Dutch physician Kaempfer in 1727
53: \cite{BUCK-BUCK3}, but it is only recently that experimental (see
54: e.g.
55: \cite{BUCK,BUCK-BUCK1,BUCK-BUCK2,LLOYD1,LLOYD2,MOISEFF-COPELAND})
56: and theoretical \cite{ERMENTROUT,MIROLLO-STROGATZ} researches
57: suggested an adequate operational model \cite{CAMAZINE}.
58:
59: At the individual firefly level, the rhythm of the recurring
60: flashes is supposed to be under the control of a neural center
61: which itself may be optically influenced by the flashes of
62: neighboring fireflies. From an experimental point of view, it is
63: clear that the fireflies interact and modify each other's rhythms,
64: which automatically leads to the acquisition of synchrony.
65:
66: Although the anatomical details of the neural activity are largely
67: unknown, a model has been proposed which accounts for the
68: essential operational parameters. The model is based on a
69: relaxation oscillator in which it is possible to reset the duty
70: cycle by optical means. Moreover, the reset action is phase
71: dependent: the duty cycle is lengthened or shortened depending of
72: the time interval between the flashes of the interacting
73: fireflies.
74:
75: By constructing an electronic implementation of it, Garver and
76: Moss showed that the model worked as it was supposed to do
77: \cite{GARVER-MOSS}. They report ensemble behaviors which indeed
78: are analogous to what is observed with fireflies, although they
79: experimented on a much smaller scale.
80:
81: We constructed an ``open'' version of this electronic firefly,
82: whose free-run duty cycle can be modified and adjusted manually on
83: the spot, and on which quantitative measurements of periods and
84: phase differences may be performed with the required precision. We
85: call it ``LCO'' the acronym of Light Controlled Oscillator.
86:
87: Compared to a firefly, the workings of an LCO are without any
88: mystery, which allows for a detailed quantitative description of
89: the synchronization mechanism at least for small ensembles.
90:
91: Our aim is to experiment with LCO's and to investigate the local
92: level features of the self-organization they exhibit. At first we
93: are looking for the parameters involved in a steady
94: synchronization state achieved by two LCO's, bringing out the
95: factors leading to synchrony. It appears that the period of
96: synchronized LCO's is tightly related to the phase difference
97: between them, showing how the electronics of the synchronization
98: actually works, and why synchrony acquisition ceases outside
99: limits set by the physics of the system. Similar phases
100: relationships have been found for systems of three LCO's and more,
101: revealing interesting features.
102: %--------------------------------------------------------------------
103: %%%%%%%%%%%%%
104: % Section 2 %
105: %%%%%%%%%%%%%
106: \section{Presentation of an LCO}
107: \label{lco} Basically, our LCO (Fig. \ref{f1}) is composed of a
108: LM555 circuit wired as an astable \cite{LM555}, the alternations
109: of which are determined by a dual RC circuit in parallel with four
110: photo-sensors \cite{GARVER-MOSS}.
111: %---------Figure 1 (one LCO)--------------------------------------------
112: \begin{figure}[hbtp!]
113: \begin{center}
114: \hbox{
115: \includegraphics[scale=0.75]{fi1.eps}
116: }
117: \end{center}
118: \vskip-.3in
119: \caption{View of a single LCO.}
120: \label{f1}
121: \end{figure}
122: %-----------------------------------------------------------------
123:
124: We made nine LCO's (Fig. \ref{f2}): nine autonomous oscillators
125: coupled by their IR beams. They had much success when, disposed on
126: a table, they went to synchronize like exotic fireflies which they
127: are aimed to mimic. Each LCO is a module made of the same
128: electronic components and having the same structure. A square base
129: (11 cm $X$ 11 cm) gives the over-all horizontal dimensions of each
130: LCO in the global pattern. With nine LCO's it is possible to
131: achieve several different patterns.
132: %---------Figure 2 (9 LCO's)-------------------------------------
133: \begin{figure}[hbtp!]
134: \begin{center}
135: \hbox{
136: \includegraphics[scale=0.5]{fi2.eps}
137: }
138: \end{center}
139: \vskip-.3in
140: \caption{Group of nine LCO's.}
141: \label{f2}
142: \end{figure}
143: %-----------------------------------------------------------------
144:
145: Each base may sustain several printed circuits giving the
146: possibility of vertical extension, but keeping the same over-all
147: horizontal dimensions.
148:
149: For the time being, our LCO's have two levels (Fig. \ref{f1}). The
150: lower part consists of a 9 volt battery and its clamping system.
151: The oscillator's printed circuit with the variable resistors
152: allowing the adjustment of the period's two time intervals, makes
153: the upper part of a LCO module. The circuit is square-like too but
154: its size is smaller than the basis. Even smaller printed circuits
155: bearing each an infrared LED and a photo-sensor, are fixed
156: vertically on the sides of the upper part. Provision is made to
157: mask the sensors allowing the LCO's to oscillate "in the dark". In
158: the aim of public presentations, the upper part bears a fifth LED
159: flashing visible light in synchrony with the IR one's, just to
160: produce a ``firefly effect''.
161:
162: The RC timing components of the LM555 consist of two resistors and
163: a single capacitor (Fig. \ref{f3}). Let $R_{\lambda}$,
164: $R_{\gamma}$ and $C$ be the values of those components responsible
165: for the LCO's timing with masked photo-sensors (timing ``in the
166: dark'').
167: %---------Figure 3 (LCO block diagram)-----------------------
168: \begin{figure}[hbtp!]
169: \begin{center}
170: \hbox
171: {
172: \includegraphics[scale=0.5]{fi3.eps}
173: }
174: \end{center}
175: \vskip-.3in
176: \caption{Block diagram of the LCO.}
177: \label{f3}
178: \end{figure}
179: %----------------------------------------------------------------
180:
181: The period is made up of two states: a longer one that may be
182: changed by manually adjusting $R_{\lambda}$, a shorter one that
183: may be changed by acting on $R_{\gamma}$. The LED's are wired at
184: the output of the LM555, switching on during the shorter part of
185: the period.
186:
187: The photo-sensors act as current sources when they are receiving
188: light, shortening the charging time of the capacitor and making
189: longer the time required to discharge it.
190:
191: In our model the resistors $R_{\lambda}$ and $R_{\gamma}$ are
192: partially variable
193: \[
194: \begin{array}{lll}
195: R_\lambda \, &=&\, 68 \, \text{k}\Omega \, +\,[0, 50] \, \text{k}\Omega \\
196: R_\gamma \, &=&\, 1.2 \,\text{k}\Omega \, +\, [0.0, 1.0] \,
197: \text{k}\Omega.
198: \end{array}
199: \]
200:
201: We use the same LCO's for two different missions: firstly, for
202: demonstrations where it is required to carry out synchronization
203: at a pace of about one flash per second (very impressive), and
204: secondly for observing the synchronization by an oscilloscope,
205: which requires a period of about 30 ms. This change of period
206: range is made by modifying nothing else but the capacitor value,
207: which has the advantage of leaving the lighting percentage per
208: period unchanged (less than 2\%) because the
209: $R_{\lambda}/R_{\gamma}$ ratio is not modified:
210: \[
211: \begin{array}{lll}
212: C \ =\, 10 \ \mu \text{F} & \rightarrow & {\mathrm period} \ \ T \ =\ 0.7 \ \text{s} \\
213: C \ =\, 0.47 \ \mu \text{F} & \rightarrow & {\mathrm period} \ \ T \ =\ 30 \
214: \text{ms}
215: \end{array}
216: \]
217:
218: As the illumination of the photo-sensors modifies the period of an
219: LCO, it is useful to distinguish the functioning of it in the dark
220: from its functioning when receiving light pulses from its
221: neighbors or diffuse light from the surroundings.
222:
223: When all the photo-sensors of an LCO are masked, its period
224: depends only from its electronics. We took this particular period
225: as a reference for each LCO.
226: %---------Figure 4 (definitions)-----------------------------------
227: \begin{figure}[hbtp!]
228: \begin{center}
229: \hbox
230: {
231: \includegraphics[scale=0.5]{fi4.eps}
232: }
233: \end{center}
234: \vskip-.3in
235: \caption{Definition of parameters in an LCO.}
236: \label{f4}
237: \end{figure}
238: %----------------------------------------------------------------
239:
240: In the framework of this article, we use the following parameters
241: related to the synchronization of the LCO (Fig. \ref{f4}):
242:
243: \begin{itemize}
244: \item $\lambda=(R_{\lambda}+R_{\gamma})C\ln 2$, the larger
245: alternation time, it corresponds with a capacitor's charging time
246: between $1/3$ and $2/3$ of the total charge. \item
247: $\gamma=R_{\gamma}C\ln 2$, the shorter alternation time, it
248: corresponds with a capacitor's discharging time within the same
249: limits. \item $\lambda^-, \lambda^+$, the beginning and the end of
250: a long alternation. \item $\tau^o$, the instant coinciding with
251: the transition from $\lambda$ to $\gamma$ ; here after we will use
252: this parameter as the ``reference moment'' in the period of an
253: LCO. \item $T_s$, the common period of a set of synchronized
254: LCO's. \item $T_A, \tau^o_A, \lambda_A, \gamma_A, T_B, \tau^o_B,
255: \lambda_B, \gamma_B, \ldots$, the periods, the ``reference
256: moments'', and the duration of the alternations of LCO$_A$,
257: LCO$_B$, $\ldots$ in illumination situations. \item $T^d_A, T^d_B,
258: T^{\text{dark}}_C,\ldots, \gamma^{\text{dark}}_A,
259: \gamma^{\text{dark}}_B, \gamma^{\text{dark}}_C, \ldots$, the
260: periods and the durations of the short and long alternations ``in
261: the dark'' of LCO$_A$, LCO$_B$, LCO$_C$, $\ldots$ i.e. when all
262: their photo-sensors are masked.
263: \end{itemize}
264: %--------------------------------------------------------------------
265: %%%%%%%%%%%%%
266: % Section 3 %
267: %%%%%%%%%%%%%
268: \section{An LCO coupled to a short pulse blind LCO}
269: \label{bf} In order to investigate the mechanisms inducing the
270: synchronization, we have placed an LCO, namely LCO$_B$, in
271: interaction with the equivalent of an LCO ``kept in the dark''.
272:
273: Let LCO$_A$ be this blind LCO; its IR pulse (lasting during
274: $\gamma_A^{\text{dark}}$) has been reduced to a quarter of
275: $\gamma_B$. In practice it is obtained from a low frequency
276: generator controlling a monostable producing a $\gamma_A$ pulse of
277: constant duration and sufficiently short.
278:
279: The signals are picked up at the low impedance output of the
280: LM555; the measurements of phases and periods are carried out with
281: an oscilloscope {\em Tektronix TDS 3012} according to well-known
282: procedures; observations are done without difficulty with a
283: precision of about 0,1\%. To be coherent with the measurements
284: presented further on in this article, the triggering is provided
285: by the LCO$_A$ considered as the reference LCO.
286: %-----------Figure 5 (LCO with a LFG)---------------------------
287: \begin{figure}[hbtp!]
288: \begin{center}
289: \hbox{
290: \includegraphics[scale=0.45]{fi5.eps}
291: }
292: \end{center}
293: \vskip-.3in
294: \caption{(a) Time difference as a function of the short
295: impulsion oscillator period. (b) Duration of the LCO$_A$'s short
296: alternation as a function of the short impulsion oscillator period.}
297: \label{f5}
298: \end{figure}
299: %------------------------------------------------------------------
300:
301: From the very first observation it is obvious that the
302: synchronization implies a phase relation between the two
303: oscillators.
304:
305: Fig. \ref{f5}(a) represents $\Delta \tau=\tau^o_B-\tau^o_A$ (i.e.
306: the position of $\tau^o_B$ related to $\tau^o_A$, the latter being
307: taken as reference) as a function of the period
308: $T_A^{\text{dark}}(=T_s)$ of the blind LCO$_A$.
309:
310: When $\Delta\tau >0$, i.e. when $\tau^o_A$ appears before
311: $\tau^o_B$ (Fig. \ref{f6} cases 1 and 2), the phase-control is
312: stable and $T_s$ can be measured easily. However, for
313: $\Delta\tau<0$ (Fig. \ref{f6} cases 4 and 5), the stability is
314: much more precarious, even if also in that situation there is a
315: synchronization of LCO$_B$ to the blind LCO$_A$.
316: %----------Figure 6 (signals shape LCO-LFG)--------------------------
317: \begin{figure}[hbtp!]
318: %\vskip-1.8in
319: \begin{center}
320: \hbox{
321: \includegraphics[scale=0.7]{fi6.eps}
322: }
323: \end{center}
324: \vskip-.3in
325: \caption{Signal shape for several phase differences between LCO$_B$
326: and the short pulse blind LCO$_A$.}
327: \label{f6}
328: \end{figure}
329: %----------------------------------------------------------------------
330:
331: The shape of $\Delta \tau$ as a function of
332: $T_A^{\text{dark}}(=T_s)$ can be fairly well schematized by two
333: straight lines intersecting at $\Delta\tau=0$ for the abscissa
334: $T_{\text{break}}$.
335:
336: The plot suggests two phase-control modes, situated at either side
337: of $\Delta\tau=0$ (Fig. \ref{f6}, case 3), distinguishing
338: themselves by two parameters which are easy to measure:
339: \begin{itemize}
340: \item[-] The domain of the phase-locking, that is to say the
341: interval limited by the periods for which there is
342: synchronization. \item[-] The slope of the straight lines, which
343: represents somehow the gain of a servo-system's feedback control.
344: \end{itemize}
345: These two modes correspond to different control mechanisms:
346: \begin{itemize}
347: \item For $\Delta \tau<0$, there can be only widening of the
348: period because the illumination and thus the photo-current are
349: totally included in the interval $\gamma_B$ (Fig. \ref{f6} cases 4
350: and 5), under these circumstances the photo-current adds to the
351: discharging current of the capacitor $C$ through $R_{\gamma}$. As
352: the extension of $\gamma$ corresponds with an increase of the
353: period, the phase-control is stable. Nevertheless the influence of
354: the photo-current on the extension of the period is of little
355: importance, first because the discharge current is by two orders
356: of magnitude superior to the charge current, and secondly because
357: $\gamma$ represents only 2\% of the period. The upper limit of the
358: phase-locking is reached as soon as the photo-current shortens the
359: alternation $\lambda_B$ of the following period, that is to say as
360: soon as the positive feedback resulting from this situation makes
361: the phase-control unstable. \item For $\Delta \tau>0$, Only a
362: shortening of the free-run period $T_B$ is possible. The
363: shortening takes place during $\lambda^+$ the end of the interval
364: $\lambda$, by increasing the charging speed of capacitor $C$ due
365: to the photo-current brought in, in parallel with $R_{\lambda}$.
366: Due to the photo-current, the voltage of $C$ reaches more rapidly
367: the value $V_C=2V_M/3$ which triggers the switching from the
368: $\lambda$ alternation towards the $\gamma$ alternation. The lower
369: limit of the phase-locking domain is reached when the duration
370: $(\Delta \tau)_{\text{min}}=\gamma_A$ (Fig. \ref{f6}, case 1)
371: during which the photo-current speeds up the charging of the
372: capacitor, is not sufficient anymore to reach the switching point
373: of $\lambda$ towards $\gamma$. As a consequence, for the shortest
374: periods, the synchronization depends strongly on the intensity of
375: the light received by phase-locked LCO$_B$. \item For
376: $0<\Delta\tau<\gamma_A$, This constitutes the intermediate
377: situations (Fig. \ref{f6}, case 2): the excess of photo-current
378: acts in accordance with the mode $\Delta\tau<0$ already described,
379: achieving a not so important lengthening of the synchronized
380: period as compared to its shortening.
381: \end{itemize}
382: %--------------------------------------------------------------------
383: %%%%%%%%%%%%%
384: % Section 4 %
385: %%%%%%%%%%%%%
386: \section{An LCO coupled to a blind LCO flashing large pulses}
387: \label{bfl} The experimental setup that has been used to get Fig.
388: \ref{f7}, differs from the previous one, only by the width of the
389: flashes emitted by the blind LCO: they have been widened to
390: increase the ratio $\gamma_A/\gamma_B$ from $1/4$ to 2, allowing
391: the blind LCO flashes to overlap the synchronized LCO flashes.
392:
393: The observation of this second type of synchronization is
394: important because it looks more like that of an actual LCO pair in
395: mutual interaction, for which differences between the $\gamma$
396: alternations are the rule with occurrences of overlapping.
397:
398: In the phase-locking with short pulses ($1/4$ ratio) one
399: distinguishes five unambiguous situations of synchronization.
400: However, when the alternation $\gamma_A$ is significantly larger
401: than $\gamma_B$, it is possible that it overrides $\gamma_B$ by
402: illuminating the end $\lambda^+$ and the beginning $\lambda^-$ of
403: the neighboring $\lambda_B$ intervals of $\gamma_B$ (Fig.
404: \ref{f8}, case 4).
405:
406: The graphs $\Delta \tau$ and $\gamma_B$ as a function of
407: $T_s=T_A^{\text{dark}}$ in Figs. \ref{f7}(a) and (b), show a
408: singular value $T_A^{\text{dark}}=T_{\text{break}}$ for which
409: there is a change of slope.
410: %-------------Figure 7 (long impulsions)----------------------
411: \begin{figure}[hbtp!]
412: \begin{center}
413: \hbox{
414: \includegraphics[scale=0.45]{fi7.eps}
415: }
416: \end{center}
417: \vskip-.3in
418: \caption{(a) Time difference as a function of the long impulsion
419: blind LCO period. (b) Duration of the LCO$_B$'s short alternation
420: as a function of the long impulsion blind LCO period.}
421: \label{f7}
422: \end{figure}
423: %---------------------------------------------------------------
424: $T_{\text{break}}$ corresponds to a situation in which $\gamma_A$
425: covers $\gamma_B$ completely and is about to start the covering of
426: $\lambda^-_B$ (the start of the following interval $\lambda_B$
427: (Fig. \ref{f8}, case 3).
428: %----------Figure 8 (signals shape long impulsions)--------------
429: \begin{figure}[hbtp!]
430: \begin{center}
431: \hbox{
432: \includegraphics[scale=0.62]{fi8.eps}
433: }
434: \end{center}
435: \vskip-.3in
436: \caption{Signal shape under different conditions for the coupling
437: of a LCO$_B$ with a long pulse blind LCO$_A$.}
438: \label{f8}
439: \end{figure}
440: %-------------------------------------------------------------------
441:
442: We observe that :
443: \begin{itemize}
444: \item[a)] For $T_A<T_{\text{break}}$, the interval $\gamma_A$
445: covers over two alternations of LCO$_B$ (Fig. \ref{f8}, case 2);
446: the phase-control acts as in the case $\gamma_A \ll \gamma_B$
447: (Fig. \ref{f6}, case 2). In the same way, the lower limit of the
448: synchronization domain depends of the total width $\gamma_A$.
449: \item[b)] For $T_A>T_{\text{break}}$, the interval $\gamma_A$
450: covers three alternations of LCO$_B$ (Fig. \ref{f8}, case 4) and
451: the phase-control functions differently:
452: \begin{description}
453: \item[In Fig. \ref{f7}(a)]: $\gamma_B$ is constant because it
454: remains entirely covered by $\gamma_A$; however $\gamma_A$ is
455: sufficiently large to illuminate either sides of $\gamma_B$, that
456: is to say to ensure the phase-control by truncating $\lambda^+$
457: (the end of alternation $\lambda_B$) while at the same time
458: shortening $\lambda^-$) the beginning of the following
459: $\lambda_B$.
460: \item[In Fig. \ref{f7}(b)]: $T_s=T_A^{\text{dark}}$
461: varies more slowly indicating a change of phase-control mode,
462: indeed the increase of the period results from the combination of
463: a constant extension of $\gamma_B$ and of a shortening of the two
464: neighboring alternations $\lambda^-_B$ and $\lambda^+_B$.
465: \end{description}
466: \end{itemize}
467:
468: Finally, Fig. \ref{f7}(a) shows that the upper limit of the
469: synchronization domain is reached when $\Delta\tau=0$ (Fig.
470: \ref{f8}, case 5), i.e. when the time-marks $\tau^o_A$ and
471: $\tau^o_B$ are superposed. This observation is of major importance
472: for the analysis of the mutual synchronization between two
473: interacting LCO's , neither being blind. Indeed the loss of
474: synchronization at $\Delta\tau=0$, means that shortening
475: $\lambda^-_B$ at the start of a alternation is not sufficient to
476: maintain the phase-control, the latter working only when it is the
477: end of $\lambda$ which is truncated, i.e. $\lambda^+_B$.
478: %--------------------------------------------------------------------
479: %%%%%%%%%%%%%
480: % Section 5 %
481: %%%%%%%%%%%%%
482: \section{Synchronization between two interacting LCO's. Measure and analysis}
483: \label{2int} When two interacting LCO's synchronize, their short
484: alternations $\gamma_A$ and $\gamma_B$ cover each other mutually,
485: including their time-mark $\tau^o_A$ and $\tau^o_B$. As a
486: consequence, the fractions of the alternations $\gamma_A$ and
487: $\gamma_B$ which are not superposed illuminate the preceding
488: $\lambda^+$ and the following $\lambda^-$ (Fig. \ref{f10}, case 2
489: and its symmetrical case which is not represented).
490:
491: This situation is similar to that described at the end of the
492: preceding paragraph (Fig. \ref{f8}, case 4 and 5); it allows us to
493: deduce that of the two parts, $\lambda^+$ and $\lambda^-$ of an
494: alternation $\lambda$, it is only $\lambda^+$ that controls the
495: synchronization in a decisive way. Indeed (Fig. \ref{f8}, case 5)
496: shows that phase-locking and synchronization stops as soon as
497: $\lambda^+$ ceases to be illuminated.
498:
499: The plots of $\Delta\tau$, $\gamma_A$ and $\gamma_B$, and $T_s$,
500: as a function of $T_A^{\text{dark}}$ in Figs. \ref{f9}(a), (b) and
501: (c) show the binary structure of the interaction between two
502: LCO's: In Fig. \ref{f9}(a), may be of two polarities in an
503: equivalent way, indicating that the two LCO's are interchangeable.
504: In Fig. \ref{f9}(b), the time intervals $\gamma_A$ and $\gamma_B$
505: change in the same manner when they are overlapping. When
506: $\gamma_A$ and $\gamma_B$ are exactly superimposed, i.e. when
507: $\Delta\tau=0$ (Fig. \ref{f10}, case 3), their length is at a
508: maximum as does $T_s$ the common period of the synchronized LCO's
509: (Fig. \ref{f9}(c)).
510: %--------------Figure 9 (two LCO's are interacting)---------
511: \begin{figure}[hbtp!]
512: \begin{center}
513: \hbox{
514: \includegraphics[scale=0.45]{fi9.eps}
515: }
516: \end{center}
517: \vskip-.35in
518: \caption{Variation of some magnitudes as a function of the LCO$_A$
519: period in the dark. (a) Time difference. (b) Short alternations
520: $\gamma_A$ y $\gamma_B$ of both LCO's. (c) Synchronization
521: period.}
522: \label{f9}
523: \end{figure}
524: %------------------------------------------------------------------
525: %----------Figure 10 (signals shape both LCO's interacting)----------
526: \begin{figure}[hbtp!]
527: \begin{center}
528: \hbox{
529: \includegraphics[scale=0.6]{fi10.eps}
530: }
531: \end{center}
532: \vskip-.3in
533: \caption{Signal shape under different conditions for two coupled
534: LCO's.}
535: \label{f10}
536: \end{figure}
537: %------------------------------------------------------------------
538:
539: On both sides of the maximum of $T_s$, the light associated with
540: the one or the other alternation $\gamma$ truncates the
541: corresponding $\lambda^+$ (Fig. \ref{f10}, case 1, 2, 4 and 5). As
542: observed earlier, the photo-effect on $\lambda^-$ is not as strong
543: as on $\lambda^+$. In fact two interacting LCO's have not the same
544: status: one is a leader who cuts the $\lambda^+$ part of his
545: coupled partner, the latter being the follower who cuts the
546: $\lambda^-$ part of his leader. However this status may change,
547: depending of the sign of $\Delta\tau$, i.e. the relative positions
548: of $\tau^o_A$ and $\tau^o_B$.
549: %--------------------------------------------------------------------
550: %%%%%%%%%%%%%
551: % Section 6 %
552: %%%%%%%%%%%%%
553: \section{Synchronization between several LCO's}
554: \label{3int} As long as the interacting system has a binary
555: symmetry, the choice of the reference LCO is irrelevant and is
556: without importance for the quantitative observation of the
557: synchronization: the reference LCO is simply that one whose period
558: $T_A^{\text{dark}}$ is modified manually. On the opposite, for
559: sets of more than two LCO's it is mandatory to specify the
560: position of the reference LCO in that of the interacting LCO's.
561:
562: We have been able to observe phase-locking in sets made up of 5
563: LCO's, using two {\em Tektronix TDS 3012} oscilloscopes triggered
564: simultaneously by the output signal of the reference LCO. However
565: those last measurements were rather difficult to perform, due to
566: the intrinsic instability of the LM555 oscillators and/or the
567: presence of multiple states.
568: %------------Figure 11 (three interacting LCO's)--------------------
569: \begin{figure}
570: \begin{center}
571: \hbox{
572: \includegraphics[scale=0.45]{fi11.eps}
573: }
574: \end{center}
575: \vskip-.3in
576: \caption{Variation of some magnitudes as a function of the
577: LCO$_A$ (reference, in the mid position) period in the dark in
578: the case of three LCO's interacting in line.(a) Synchronization
579: period. (b) Time difference of LCO$_B$ and LCO$_C$ with respect
580: to LCO$_A$.}
581: \label{f11}
582: \end{figure}
583: %------------------------------------------------------------------
584:
585: Fig. \ref{f11}(a) and (b) have been obtained with a set of 3 LCO's
586: put in line: LCO$_B$--LCO$_A$--LCO$_C$. The reference was
587: LCO$_A^{\text{dark}}$ in the mid position. Synchronization has
588: been achieved for $T_A^{\text{dark}}$ varying between 33.98 ms and
589: 35.2 ms. The ``periods in darkness'' of the other are:
590: $T_B^{\text{dark}}$=33.98 ms and $T_C^{\text{dark}}$=33.99 ms.
591:
592: Fig. \ref{f11}(a) shows a bifurcation-like phenomenon: the 3-LCO
593: system synchronizes according to two modes brought out by two
594: significantly different values of $T_s$ as a function of
595: $T_A^{\text{dark}}$.
596:
597: Fig. \ref{f11}(b) shows the same behavior by plotting the
598: time-marks $\tau^o_B$ and $\tau^o_C$ as a function of
599: $T_A^{\text{dark}}$, taking $\tau^o_A$ as the reference time-mark.
600:
601: There is obviously a symmetry in this 3-LCO synchronization: once
602: the LCO's are synchronized, the phases may equally well have one
603: or the other polarity (Fig. \ref{f11}(b)). Moreover, a closer look
604: at the data producing Fig. \ref{f11}(b) shows that
605: $\Delta\tau^o_B$ and $\Delta\tau^o_C$ are always of opposite
606: polarity; this suggests that LCO$_B$ and LCO$_C$ are in some way
607: interchangeable. However, before the synchronization has been
608: settled, it is not possible to foresee the mutually exclusive
609: polarities of $\tau_B$ and $\tau_C$, confirming that a
610: bifurcation-like phenomenon is present.
611: %--------------------------------------------------------------------
612: %%%%%%%%%%%%%%
613: % Conclusion %
614: %%%%%%%%%%%%%%
615: \section{Conclusion}
616: \label{conc} Obviously, synchronization is tightly linked to
617: phase-locking and the domain associated with it. This is why we
618: considered as a criterion that coupled LCO's are synchronized only
619: if they exhibit a stable dependency of their phase as a function
620: of their period differences.
621:
622: Our measurements show that LCO's synchronize in as much that the
623: phases involved in the phase-locking feedback do not exceed values
624: related to the widths of the coupling light pulses.
625:
626: Finally we suggest using the criterion associating synchronization
627: and phase-locking domain as a test to compare theoretical models
628: of interacting LCO's to their experimental implementations. In a further
629: publication, we analyze this situation.
630: %--------------------------------------------------------------------
631: %%%%%%%%%%%%%%%%
632: % Bibliography %
633: %%%%%%%%%%%%%%%%
634: \begin{thebibliography}{10}
635:
636: \bibitem{PIKOVSKY}
637: A. Pikovsky, M. Rosenblum and J. Kurths, {\em Synchronization a
638: universal concept in nonlinear sciences\/} (Cambridge University
639: Press, Cambridge, 2001).
640:
641: \bibitem{BLEKHMAN}
642: I. Blekhman, {\em Synchronization in science and technology\/}
643: (Asme Press, New York, 1988).
644:
645: \bibitem{MIROLLO-STROGATZ}
646: Mirollo RE, Strogatz SH, Synchronization of pulse-coupled
647: biological oscillators , {\em SIAM Journal of Applied Mathematics}
648: {\bf 50} (1990) 1645-1662.
649:
650: \bibitem{STROGATZ}
651: S.H. Strogatz, R.E.Mirollo and P.C. Matthews, Coupled nonlinear
652: oscillators below the synchronization threshold: Relaxation by
653: generalized Landau damping, {\em Phys. Rev. Lett.\/} {\bf 68}
654: (1992) 2730--2733.
655:
656: \bibitem{WINFUL}
657: H.G. Winful and L. Rahman, Synchronized chaos and spatiotemporal
658: chaos in arrays of coupled lasers, {\em Phys. Rev. Lett.\/}{\bf
659: 65} (1990) 1575--1578.
660:
661: \bibitem{WIESENFELD}
662: K. Wiesenfeld, P. Colet and S.H. Strogatz, Synchronization
663: transitions in a disordered Josephson series array, {\em Phys.
664: Rev. Lett.\/} {\bf 76} (1996) 404--407.
665:
666: \bibitem{HOHL}
667: A. Hohl, A. Gavrielides, T. Erneux and V. Kovanis, Localized
668: synchronizationin two coupled nonidentical semiconductor lasers,
669: {\em Phys. Rev. Lett.\/} {\bf 78} (1997) 4745--4748.
670:
671: \bibitem{BUCK-BUCK3}
672: J. Buck and E. Buck, Mechanism of rhythmic synchronous flashing of
673: fireflies, {\em Science\/} {\bf 159} (1968) 1319--1327.
674:
675: \bibitem{BUCK}
676: J. Buck, E. Buck, J.F. Case and F.E. Hanson, Control of flashing
677: in fireflies. V. Pacemaker synchronization in \textit{Pteroptyx
678: cribellata}, {\em J. Comp. Physiol.\/} {\bf A 144} (1981)
679: 287--298.
680:
681: \bibitem{BUCK-BUCK1}
682: J. Buck and E. Buck, Synchronous fireflies, {\em Sc. Am.\/} {\bf
683: 234} (1976) 74--85.
684:
685: \bibitem {BUCK-BUCK2}
686: J. Buck, E. Buck, Toward a functional interpretation of
687: synchronous flashing by fireflies, {\em Am. Naturalist.\/} {\bf
688: 112} (1978) 471--492.
689:
690: \bibitem {LLOYD1}
691: J.E. Lloyd, Fireflies of Melanesia: Bioluminiscence, mating
692: behavior and synchronous flashing (\textit{Coleoptera:
693: Lampyridae}), {\em Env. Entom.\/} {\bf 2} (1973) 991--1008.
694:
695: \bibitem {LLOYD2}
696: J.E. Lloyd, Model for the mating protocol of synchronously
697: flashing fireflies, {\em Nature\/} {\bf 245} (1973) 268--270.
698:
699: \bibitem{MOISEFF-COPELAND}
700: A. Moiseff and J. Copeland, A new type of synchronized flashing in
701: a North America firefly, {\em J. Ins. Behavior\/} {\bf 13} (2000)
702: 597--612.
703:
704: \bibitem{ERMENTROUT}
705: G.B. Ermentrout, An adaptive model for synchrony in the firefly
706: \textit{Pteroptyx malaccae} {\em J. Math. Biol.\/} {\bf 29} (1991)
707: 571--585.
708:
709: \bibitem{CAMAZINE}
710: S. Camazine, J.L. Deneubourg, S. Franks, J. Sneyd, G. Theraulaz
711: and E. Bonabeau {\em Self-organization in biological systems\/}
712: (Princeton University Press, Princeton, 2001).
713:
714: \bibitem{GARVER-MOSS}
715: Garver W, Moss F, Electronic fireflies, {\em Sc. Am.\/} {\bf 269}
716: (1993) 128-130.
717:
718: \bibitem{LM555}
719: LINEAR Databook, (1982), {\em LM555/LM555C Timer}. National
720: Semiconductor Corporation (9): 33-38.
721:
722: \end{thebibliography}
723:
724: \end{document}
725: