Electro-optic system controller and method of operation

[0002] This is a continuation-in-part of application Ser. No. 09/517,375, filed on Mar. 2, 2000, which is a continuation-in-part of application Ser. No. 09/472,709 filed on Dec. 24, 1999, which is a continuation-in-part of application Ser. No. 08/988,865, filed on Dec. 11, 1997, now U.S. Pat. No. 6,028,423. This application also claims priority under 35 U.S.C. 119(e) to provisional U.S. Patent Application No. 60/173,056 filed Dec. 24, 1999.

[0003] The present invention relates to a logic system based controller and more specifically a controller that utilizes state machines, logic and/or a microprocessor for an electro-optic system.

[0004] Electro-optic systems are known in the art for providing an interface between electronic and optically-based systems. Such electro-optic systems are used in a variety of applications including telecommunications, remote sensing, medical devices, and in other fields as well.

[0005] FIG. 1aillustrates a conventional electro-optic system 100 for use with one of the aforementioned systems. The electro-optic system 100 includes a laser 110 and an analog bias controller 120. Typically, an electronic system produces a modulating signal which is combined 112 with a laser bias current I_DC. The combined signal is injected into the laser 110 which produces a modulating optical output signal, P_O, that typically is an intensity modulated signal in which the light intensity varies as a function of the amplitude of the modulating signal. In analog systems, the bias current I_DCis selected such that it biases the laser 110 at an operating point where the laser output P_Oexhibits a linear relationship to the input bias I_DC(herein referred to as the laser's transfer function). The modulating signal varies the bias point above and below this operating point thereby producing a corresponding change in the intensity of the output power, P_O, of the laser 110.

[0006] Continuing with FIG. 1, the laser 110 may include a monitoring photodiode (not shown) either separate or integrated into the same package with the laser 110. The monitoring photodiode produces a photodiode current I_PDindicative of the laser output power P_o. The photodiode current I_PDis supplied to an analog bias controller 120 which includes circuitry designed to measure the photodiode current I_PDand to return the laser bias I_DCto a predefined current level.

[0007] FIG. 1billustrates a graph showing three transfer functions of a laser at three operating temperatures T1, T2, T3. The graph illustrates laser output power P_Oalong the y-axis, and laser bias current I_DCalong the x-axis. The first trace 152 illustrates the laser's transfer function at a first operating temperature T₁. The second trace 154 and the third trace 156 similarly illustrate the laser's transfer function at a second and a third operating temperature, T₂and T₃, respectively. Each of the transfer functions 152, 154, and 156 also include a corresponding threshold operating point I_th1, I_th2, and I_th3, respectively. The threshold operating points I_th1, I_th2, and I_th3, indicate the laser bias current level I_DCat which the laser produces appreciable output power. As can be seen, a laser's threshold operating point varies greatly with operating temperature. A change in the operating temperature of the laser shifts the transfer function laterally along the x-axis.

[0008] A shift in the operating temperature can produce a significant change in the laser output power. For example, a laser bias current I_DCselected to bias the laser at operating point I_Q1produces a laser output power P_T1in the center 162 of the linear region of the transfer function 152. At operating point I_Q1, The laser will operate as intended to produce a substantially linearly varying output power P_Owhen excited by a modulating signal. If the operating temperature of the laser operating temperature changes to T₂, the laser output power P_T2for the same bias current I_Q1drops significantly as shown by operating point 163. In this case, the linear operating region of the laser and electro-optic system 100 is limited to the linear region below operating point 163, in the case of negative current swings below I_Q1. The analog bias controller 120 contains circuits to compensate for change in temperature, but with a limited degree of accuracy. The analog bias controller 120 also requires manual adjustments, has limited control capabilities, and is subject to significant temperature drift.

[0009] In view of the aforementioned inadequacies of the prior art, the need exists for a new system and method for controlling an electro-optic system over varying temperatures of operation.

[0010] It is an advantage of the present invention to provide a system and method for controlling an electro-optic system by which the threshold operating point of the system gains and other parameters can be accurately established and the laser transfer function controlled over a variety of different operating temperatures of a laser module.

[0011] It is another advantage to provide a system for subtracting errors due to dark current and aging of a laser-monitoring photodiode.

[0012] It is yet another advantage to provide a system for driving a laser utilizing both a low frequency circuit to drive DC current in addition to a high frequency circuit for driving a high frequency AC/DC signal to set the operating point of a laser.

[0013] A further advantage is to provide a system and method for calibrating and initializing the electro-optic system upon each power up of the system.

[0014] It is another advantage to provide a system for continuously monitoring and adjusting parameters of an electro-optic system while the system is fully operational.

[0015] Still another advantage of the present invention further is to provide a system and method for enabling a link characterization process to be performed between two electro-optic systems.

[0016] Electro-Optic System Controller

[0017] In an exemplary embodiment, the present invention is an electro-optic system for driving a laser, monitoring the laser operation, and maintaining the laser parameters within acceptable limits over temperature and device variations. The electro-optic system of an exemplary embodiment includes a signal input coupled to a pre-amplifier via a test system switch (TSS). The output of the preamplifier is fed into a combiner and coupled to a laser module input. A laser module of an exemplary embodiment includes a laser bias input, a laser output, a temperature sensor output, and a monitoring photodiode sensor output. In other embodiments, the laser module may further include a power sensor and a cooler/heater component. The test system switch, the preamplifier and the laser module are controlled by a controller circuit.

[0018] The controller circuit of an exemplary embodiment may be a digital processor, a microprocessor, or an ASIC device that is capable of generating control data. The processor of the exemplary embodiment includes a communication port that may be connected to an external host computer. The processor of the exemplary embodiment further includes memory or an input for connection to a memory component. In other embodiments of the present invention the processor may include a port for storing/receiving digital data to/from an external source.

[0019] The processor controls the test system switch utilizing switch control lines. In an exemplary embodiment, the test system switch may connect one of a data signal, a signal generated from a signal generator, or a reference ground to the input of the preamplifier. The signal generator is controlled by the processor. The test system switch of an exemplary embodiment also includes a switched or dedicated line for connecting a coded input signal to a code detector of the controller circuit. The code detector identifies the reception of an encoded signal received by the electro-optic system and outputs coded data to the processor.

[0020] The preamplifier of an exemplary embodiment includes an amplifier connected to the switched input from the test system switch, a high frequency voltage variable resistor, and a high frequency voltage controlled current source (HF-VCCS) that is connected to the signal input of the laser. The preamplifier is coupled to the processor and may have its gain and offset characteristics changed by means of a dual digital to analog converter (DAC) that accepts digital control signals from the processor. The dual DAC of an exemplary embodiment dedicates one DAC for adjusting the gain of the preamplifier and a second DAC for adjusting the offset of the preamplifier.

[0021] The driving signal to the laser input of an exemplary embodiment is the combination of a high frequency DC/AC signal current I_Sand a low frequency laser bias current I_L. The biasing current I_Lis generated by the controller of the electro-optic system utilizing a low frequency voltage controlled current source controller by the processor by means of a bias adjust DAC. The controller of an exemplary embodiment may further include digital to analog converters connected to the processor for controlling the wavelength of the laser, and for controlling other laser components such as a cooler/heater.

[0022] The controller of an exemplary embodiment of the present invention further provides circuitry for monitoring the laser module components including the photodiode and the temperature sensor, the preamplifier, and also includes other signals generated by the controller such as the low frequency bias current I_L. In an exemplary embodiment, a voltage signal is provided by the temperature sensor that is indicative of the real-time temperature of the laser module. This temperature voltage signal is fed into a sensing amplifier of a package of sensing amplifiers. The outputs of the sensing amplifiers are connected to a multiplexer that is coupled to the processor via an analog to digital converter. Other signals are also supplied to the processor by means of the sensing amplifiers including a test signal from the high frequency VCCS of the preamplifier and a test signal from the low frequency VCCS. The photodiode of the laser module of the exemplary embodiment includes a photodiode current output signal that is connected to a photodiode amplifier in the controller circuit. The amplifier is controller by the processor via a dual digital to analog converter that includes an offset adjust DAC and a gain adjust DAC. The output of the photodiode amplifier is connected to an input of the multiplexer. The multiplexer of the exemplary embodiment also includes additional inputs for monitoring the various voltages utilized by the electro-optic system.

[0023] Factory Calibration

[0024] An exemplary embodiment of the present invention provides a factory calibration process for calibrating the laser module system. Test equipment utilized for the laser calibration is connected from the laser output to the laser drive control system. In one embodiment, the testing equipment may consist of an external optical power meter instrument having an I/O interface to a host computer. Once the testing equipment is connected to the laser module system, the dark current of the laser module photodiode is determined by first setting the gain of the laser preamplifier to nominal. The Test System Switch (TSS), which is connected to the input of the laser preamplifier, is connected to a circuit ground utilizing a control signal from the microprocessor of the electro-optic system.

[0025] The calibration process of the exemplary embodiment continues by turning off the DC bias for the laser by supplying an appropriate digital word to a bias adjust digital to analog converter (DAC). The analog signal produced by the bias adjust DAC turns off the current normally generated by the Low Frequency Voltage Controlled Current Source (VCCS). The offset voltage of the preamplifier is canceled by adjusting an offset adjust DAC while monitoring the offset utilizing a measurement of the voltage across a resistor of the high frequency VCCS. The dark current of the laser photodiode is calibrated to avoid errors as the laser photodiode changes with temperature and aging. If the laser has no current drive, then the output light power of the laser is zero, that is, no light is coupled into the monitoring laser photodiode. The parasitic dark current I_Dis defined as the current that flows through the laser photodiode for a no-light output condition. The measured dark current is recorded and utilized to compensate the system measurements.

[0026] The threshold currents of the laser, and corresponding laser photodiode currents are determined in the next steps of the exemplary calibration method. The DC bias current to the laser I_Lis incremented gradually by changing the input of the bias adjust DAC utilizing the microprocessor control lines. The change to the input of the bias adjust DAC changes the current output of the low frequency VCCS on the DC bias path. At the same time, laser power P_Lis read with the external optical power meter, and the microprocessor computes the slope of the transfer characteristic dP_L/dI_L. In the next step of the exemplary embodiment, the laser bias current I_Lis set to a threshold I_LTHby setting the bias current I_Lto a value at which the derivative dP_L/dI_Lstops changing. This step sets the threshold I_LTHin the linear region of the laser characteristic. Once the threshold I_LTHis determined, the laser power at threshold P_LTHis read utilizing the optical power meter and recorded by the microprocessor.

[0027] The calibration method of the exemplary embodiment continues by measuring the laser photodiode current I_P1under the above threshold conditions. Specifically, the threshold values of the laser photodiode current I_P1Tare measured and recorded via a path through a photodiode amplifier to a MUX and analog to digital (A/D) converter that is connected to the microprocessor. The laser threshold current I_LTHis measured and recorded via a path through a measurement resistor to the sensing amplifiers, through the MUX and A/D converter, and to the microprocessor. The recorded value of the laser photodiode threshold current I_P1Tis adjusted utilizing the value of the dark current I_D. A precision method for measuring the laser bias current I_Lstarts with the sensing of a voltage across a resistor of the low frequency VCCS utilizing a sensing resistor. The sensed voltage is routed through an amplifier of the sensing amplifiers to the multiplexer, and is converted to a digital word and routed to the microprocessor by the A/D converter. The temperature of laser module is read and recorded at the threshold values so that the microprocessor may associate a given temperature to the laser threshold current I_LTH, to a photodiode threshold current I_P1T, and to photodiode dark current I_D.

[0028] Continuing with the calibration process of an exemplary embodiment, the input to the laser preamplifier is disconnected from circuit ground and connected to a voltage reference of a signal generator utilizing the test system switch. The voltage reference utilized in the calibration process is a precision full-scale reference voltage V_REFthat is applied to the input of the laser preamplifier. The gain G₁of the laser preamplifier is incremented while reading the power output P_Lof the laser utilizing the optical power meter. The gain G₁of the laser preamplifier is incremented by the microprocessor by applying an appropriate digital word to a gain adjust DAC. The gain adjust DAC, in turn, applies a voltage to the high frequency voltage variable resistor, which then changes the gain G₁of the laser preamplifier. The gain adjust DAC is incremented to determine a full power gain G_Fneeded to obtain full-scale laser power P_LMI. The maximum current I_PMthrough the laser photodiode is determined for the full-scale laser power P_LMIcondition.

[0029] The effective photodiode responsivity R_effof the combination of the laser module and the laser photodiode is calculated and recorded. The factory calibration process of the exemplary embodiment is completed by preparing drift compensation models, storing the models in a processor memory, and disconnecting the test equipment utilized for the laser calibration.

[0030] Power Up Calibration and Initialization

[0031] An exemplary embodiment of the present invention provides a power up calibration process for calibrating the laser module system. Upon power up of the laser module system, the preamplifier gain is set to the to the full power gain G_F1obtained during factory calibration, needed to obtain full-scale laser power P_LMIgain. The input of the laser preamplifier is connected to circuit ground using the test system switch. In a next step, the DC bias for laser is turned off by supplying an appropriate digital word to a bias adjust digital to analog converter (DAC). The offset voltage of the preamplifier is canceled by adjusting an offset adjust DAC while monitoring the offset utilizing a measurement of the voltage across a resistor of the high frequency VCCS. The dark current I_Dfrom the laser photodiode is measured and recorded. The temperature T of laser module is then measured and recorded.

[0032] The power up calibration and initialization process of an exemplary embodiment continues by determining the laser power threshold. The DC bias current I_Lto the laser is measured concurrently with the laser photodiode current I_P1. The microprocessor then utilizes the recorded values from the above measurements to compute the laser power P_L. The DC bias current I_Lis set to threshold by selecting a point at which dP_L/dI_Lstops changing. In a next step of the exemplary embodiment, the laser threshold power P_LTHis determined by reading the laser photodiode threshold current I_PTHand computing and recording the corresponding laser threshold power P_LTH. The laser threshold current I_LTHand the photodiode threshold currents I_P1Tare also recorded. The process of determining the laser power threshold is completed by determining and recording the threshold temperature of laser module T_TH.

[0033] The exemplary power up calibration process continues by calibrating the system gain. Utilizing the test system switch (TSS), the laser preamplifier input is disconnected from the circuit ground, and is connected to a precision full-scale reference voltage V_REFof the signal generator. A photodiode current I_PMresulting from the application of the reference voltage V_REFis measured, and the corresponding laser power output P_Lis determined. In a next step, the microprocessor compensates for gain errors of the laser module. The gain G₁of the laser preamplifier is incremented while determining the laser power output P_Lwith laser photodiode measurements. The gain G_Fneeded to obtain full scale output laser power P_LMis determined. The TSS path to the laser preamplifier is switched from the reference voltage V_REFto circuit ground. The DC bias current is set to an a “Q point” current I_LQ, defined as an operating point above the threshold current, using the DC bias path measurement circuit. The microprocessor of the Electro-Optic System computes drift compensation for the above measured parameter. The power up calibration and initialization is completed by connecting the input of the preamplifier to the signal driving the laser.

[0034] Electro-Optic System Monitoring

[0035] In an exemplary embodiment, the Electro-Optic System enters a monitor phase where drift compensation is implemented on a continuous basis without having to disconnect the Electro-Optic System from the input signal. Adjustments to the system parameters such as laser DC bias or system gain are made as needed to compensate for temperature changes. The drift compensation process determines the adjustments with the use of drift models and then changes the system parameters, as needed, utilizing very gradual steps that emulate analog drift.

[0036] The present invention will be better understood from the following detailed description of a preferred embodiment of the invention, taken in conjunction with the accompanying drawings in which like reference numerals refer to like parts and in which:

[0037] FIG. 1aillustrates a conventional electro-optic communication system;

[0038] FIG. 1billustrates a transfer function of the laser shown in FIG. 1a;

[0039] FIG. 2aillustrates an electro-optic communication system representing one embodiment in which the present invention may be utilized;

[0040] FIG. 2billustrates one embodiment of the elector-optic system illustrated in FIG. 2ain accordance with the present invention;

[0041] FIG. 3 illustrates one embodiment of the electro-optic system controller in accordance with the present invention;

[0042] FIG. 4 illustrates a second embodiment of the electro-optic system controller utilized for calibration methods in accordance with the present invention;

[0043] FIG. 5 illustrates the volt-ampere characteristics for an exemplary germanium photodiode in accordance with the present invention;

[0044] FIGS. 6aand 6billustrate the laser and photodiode characteristics, respectively, measured during the calibration process in accordance with one embodiment of the present invention;

[0045] FIGS. 7aand 7billustrate the laser and photodiode characteristics, respectively, in accordance with the present invention,

[0046] FIG. 8 illustrates the laser transfer characteristic in accordance with one embodiment of the present invention;

[0047] FIG. 9 illustrates one embodiment of the drive controller factory calibration process in accordance with the present invention;

[0048] FIG. 10 illustrates one embodiment of the power up calibration and initialization process in accordance with one embodiment of the present invention; and

[0049] FIG. 11 illustrates the monitor phase of one embodiment of the present invention.

[0050] I. General Overview

[0051] FIG. 2aillustrates an electro-optic communication system 200 representing one embodiment in which the present invention may be utilized In this embodiment, the electro-optic system 200 includes a first electronic system 210 connected to a first electro-optic system 230, and a second electronic system 220 connected to a second electro-optic system 240. An optical channel 235 provides a communication link between the first and second electro-optic systems 230, 240. In one embodiment, the optical channel 235 is a single fiber optic cable for carrying signals of various wavelengths. In another embodiment, the optical channel 235 includes two separate cables for bi-directional communication. During operation, the first electronic system 210 generates an electronic signal for transmission to the second electronic system 220. The first electro-optic system 230 receives the electronic signal, converts it to an optical signal, and transmits the optical signal over the optical channel 235 to the second electro-optic system 240. The second electro-optic system receives the optical signal, converts it back to substantially the original electronic signal, and provides it to the second electronic system 220.

[0052] The electro-optic system 230, 240 of the present invention may be implemented in other systems outside of the communications field as well. For example, in medical, research, industrial or other applications, the electro-optic system 230, 240 may be used in systems that require precision control of a laser for high levels of reliability.

[0053] FIG. 2billustrates one embodiment of the electro-optic systems 230, 240, as illustrated in FIG. 2a,of the present invention. Each electro-optic system 230, 240 includes an emulator circuit 255 coupled to a test system switch 6. The test system switch 6 has an output coupled to a laser driver amplifier 8, and an input coupled to a photo-detector amplifier 265. The laser driver/amplifier 8 is coupled to a laser module 270 which emits an optical signal onto the optical channel 235 which is a fiber optic cable in the illustrated embodiment. In one embodiment of the present invention, the laser module 270 may include a cooling/heating element 271, an operating temperature sensor 272, a laser output power sensor 273, and a laser 274, which may be a laser diode or a light emitting diode (LED). The term “laser” may be used interchangeably for an opto-electronic device such as a Laser device or an LED. The photo-detector amplifier 265 has an input coupled to a photo-detector circuit 275 configured to receive an optical signal transmitted to the system 230, 240 along the fiber optic cable 235.

[0054] The electro-optic system 230, 240 further includes a controller circuit 300 for managing the functions of the components in the system 230, 240. In one embodiment of the present invention, the controller circuit 300 is coupled to the test system switch 6, the laser driver/amplifier 8, and to the laser module 270. The controller circuit 300 may additionally be coupled to the photo-detector 275, the photo-detector amplifier 265, the emulator 255, or other system components for controlling the operation of these components, as well.

[0055] The electro-optic system 230, 240 of a preferred embodiment may further include an input protection circuit (not shown) as disclosed in U.S. Pat. No. 6,029,423, entitled “Isolation Instrument for Electrical Testing, incorporated herein by reference. The input protection circuit protects the electro-optic system 230, 240 from over voltages, power surges and electro-static discharge.

[0056] II. Processor Circuit Architecture

[0057] FIG. 3 illustrates a preferred embodiment of the controller circuit 300, as shown in FIG. 2, in accordance with the present invention. As illustrated, the controller circuit 300 includes a processor 305 that is programmable using a local or remotely located host input/output (I/O) device 390. The processor 305 may include state machines, logic and a microcontroller. Depending upon the application, one or more of these logic circuits may be utilized in the processor 305, and may be part of an ASIC. The input/output device 390 communicates with the controller circuit 300 via an I/O communication port 310. In one embodiment, the processor 305 may store and/or retrieve programming signals to/from a memory storage device 315, such as an integrally formed erasable read only memory (EAROM). In an alternated embodiment, the memory storage device 315 may be a local or remotely located RAM, ROM, PROM storage device, a fixed or removable hard, floppy, or CD-ROM disc, or a similar medium which is readable and/or recordable via a corresponding drive unit (not shown). Other embodiments of the present invention may utilize other memory storage units that are operably compatible with the controller circuit 300 of the present invention. A digital input/output interface 317 may be used to generate digital control signals for transmission to other components within the electro-optic system 230, 240.

[0058] The processor 305 includes one or more switch control lines 305acoupled to the test system switch (TSS) 6 to communicate a switch control signal, SW_CTRL, for controlling the state of the switch 6. In a preferred embodiment, the switch control signals are CMOS logic level signals having voltages in the range of +/−5 VDC. The control line 305amay further include one or more power lines, typically +/−5 VDC, and a ground line to provide these signals to the switch 6.

[0059] The controller circuit 300 further includes a signal generator 320 that provides an analog waveform or a serial stream of digital data to the test system switch 6. The signal generator 320 receives a processor reference signal Ref_p, and, in response, produces a signal generator reference signal Ref_sthat is suppled to the test system switch 6 via line 320a.The reference signal Ref_sis utilized to perform a transmitter calibration or a link characterization process, further described in the inventor's co-pending application Ser. No. 09/772,709, entitled “Electro-optic Interface System and Method of Operation,” and incorporated herein by reference. The processor reference signal Ref_pand/or the signal generator reference signal Ref_smay be a DC voltage or an AC signal, in either digital or analog form. In a specific embodiment, the processor reference signal Ref_pis a serial stream of bits representing a digital word, and the signal generator reference signal Ref_sis an analog waveform. The signal generator 320 may include modulation, multiplexing, and/or conditioning circuitry, such as amplifiers, filters, and the like. In an alternate embodiment, the signal generator 320 produces a digital-to-analog reference signal Ref_sthat is applied to the preamplifier 8 via the TSS 6 to calibrate the laser transmitter system.

[0060] The processor 305 may further include a signal receiver (not shown) coupled to the test system switch 6 and configured to receive a returned reference signal which is communicated from the second electro-optic system 240, as shown in FIG. 2a,during the link characterization process, further described in the above referenced “Electro-optic Interface System and Method of Operation” application. The receiver may be configured to receive and process analog and/or digital signals and may include modulation, multiplexing, and/or conditioning circuitry, such as amplifiers, filters, digital-to-analog converters, analog-to-digital converters and the like. In one embodiment of the present invention, the waveforms generated by the signal generator 320 may be used to determine signal to noise ratios and bit error rates of the fiber optic communications channel shown in FIG. 2a,and as described in the link characterization process of the “Electro-optic Interface System and Method of Operation” application.

[0061] The processor 305 and test system switch 6 are further coupled together by means of a code detector 325 operable to identify the reception of an encoded signal received by the electro-optic system 230 or 240, as shown in FIGS. 2aand 2b.The code detector 325 is coupled to the test system switch 6 via a line 325ato receive a coded signal C_s. In a preferred embodiment, the coded signal C_sis a binary word of a predefined length to indicate that a link characterization process is underway. A link characterization process of a preferred embodiment determines the existence and quality of the communication link, and is further described the above referenced “Electro-optic Interface System and Method of Operation” application.

[0062] In response to the coded signal C_s, the code detector 325 outputs coded data C_Dto the processor 305 via lines 305c.The coded signal C_sand coded data C_Dmay be DC voltages and/or an AC signals, in either analog or digital form. In a preferred embodiment, the code detector 325 is a digital comparator having two registers. A first register is loaded with a predetermined pattern of binary bits containing a code. A second register is a universal asynchronous receiver transmitter (UART) register in which the incoming bits are received and shifted. The digital comparator includes a second stage in which each bit in the first and second registers are compared. If all of the bits in the first and second registers match, the code signal C_sis deemed received, and the coded data C_D, a high logic level in one embodiment, is output to the processor 305. Some or all of the described functions may reside within the processor 305 as hardware, software, and/or firmware.

[0063] In other embodiments of the present invention, each signal line 305a,320aand 325amay comprise an integrated bus line in addition to the aforementioned control and data signals, Ref_S, C_S, and SW_CTRL, respectively. The integrated bus lines may include power and ground potentials, as well as other digital and/or analog signals.

[0064] In a preferred embodiment, the controller circuit 300 includes a current source or digital-to-analog converter (DAC) 330 coupled to the processor 305 via line 305d.In an alternate embodiment, a pulse width modulator (PWM) with a filter may be utilized in place of the DAC 330. The DAC 330 provides current to drive the laser 274. In response to a suppled digital signal D₁from the processor 305, the DAC 330 generates a laser biasing current I_dc. During operation, the processor 305 outputs a digital signal D₁having a multitude of different values, depending upon the desired level of laser output power, P_O. The processor of the present invention also generates a multitude of different digital signals D₁to calibrate the laser 274, the process of which is further described below.

[0065] The controller circuit 300 also includes a dual digital-to-analog converter 335 for controlling the output signal level and/or the gain of a laser driver/amplifier 8. The dual DAC 335 is coupled to the processor 305 via line 305e,and receives digital control signals D₂for each one of the DACs in dual DAC 335 to set the output power level and the signal gain of the laser driver/amplifier 8 via lines 335aand 335b.The value of the digital control signal D₂may be determined responsive to operating parameters such as output signal level of the test system switch 6, input handling of the laser 274, and/or a sensed laser output power. Other operating parameters of the laser module 270, such as the sensing of the laser wavelength that may be utilized for other embodiments of the laser control system, may also be used for determining the value of D₂.

[0066] In an embodiment in which the laser 274 is tunable, the controller circuit 300 may further include a DAC 340 for controlling the wavelength of the laser 274. The DAC 340 is coupled to the processor 305 via line 305f,and receives a digital control signal D₃corresponding to the desired wavelength of the laser 274. The DAC 340 generates an analog signal that is supplied to a wavelength tuning input of the laser 274. The emitted wavelength of the laser 274 may vary in response to a variety of signal characteristics, for example, the amplitude, phase, or frequency of the signal output from the DAC 340. In other embodiments or the present invention, the signal output from the DAC 340 may be used in conjunction with other electro-optic devices to tune the laser. Those skilled in the art will appreciate that circuits other than a DAC circuit may be used to tune the wavelength of the Laser 274.

[0067] The controller circuit 300 may further include a DAC 345 and driver 350 configured to control the operation of a heating and/or cooling element 271, which may be integrally formed with the laser 274 in the laser module 270. The DAC 345 has an input 305gwhich is coupled to the processor 305, and receives a digital control signal D₄corresponding to the desired laser temperature setting. The value of the digital control signal D₄may be determined responsive to operating parameters such as a sensed operating temperature that indicates that the laser is operating outside of a predefined temperature window. Alternatively, or in addition, a sensed operating temperature may indicate that the laser has undergone a relative change in operating temperature beyond a predefined window. Other operating conditions of the laser module 270 of alternate embodiments may also be included in determining the value of D₄.

[0068] In a preferred embodiment of the present invention, the controller circuit 300 may include a dual DAC 355 that is utilized to modify the gain and offset of an amplifier 391. The amplifier 391 is connected to a photodiode circuit 3 that outputs photodiode characteristics in response to the light detected from the laser. A calibrated voltage V_S1is provided at the output of the amplifier 391.

[0069] The controller circuit 300 further includes sensing amplifiers 360, a multiplexer 365, and an analog to digital converter (ADC) 370 coupled to the processor 305. In other embodiments, the analog to digital converter 370 may be a companding (logarithmic) analog to digital converter which may offer the advantage of having a higher degree of resolution for low level signals. The sensing amplifiers 360 are used to amplify a variety of operating signals such as a laser temperature voltage V_s2from a temperature sensor 272 which indicates the operating temperature of the laser 274. In one embodiment of the present invention, one or both of the voltages V_s1and/or V_s2may be provided as outputs of the laser module 270. Other sensing amplifiers 361 may be used to sense additional operating parameters such as the laser current and voltage to determine power.

[0070] The outputs of the sensing amplifiers 360 are coupled to the multiplexer 365. The multiplexer selects an output of one of the sensing amplifiers in response to a digital control signal D₆from the processor 305 transmitted along control line 305_i. The output of the multiplexer 365 is coupled to an ADC 370 which converts the selected signal, such as analog voltage V_s1, or V_s2, to a digital signal. The digital signal is supplied to the processor 305 via line 305j.In this manner, the processor 305 can monitor operating parameters of the electro-optic system 230, 240, including the output power P_Oand operating temperature of the laser 274. Those of skill in the art will appreciate that the same or similar arrangement may be used to monitor other operating parameters of the laser 274, laser module 270, laser driver/amplifier 8, test system switch 6, or other components of the electro-optic system 230, 240. Responsive to the monitored operating parameters, e.g., the output power P_Oand/or operating temperature of the laser 274, the processor 305 may adjust any of the values of the digital signals, including signal D₁for controlling the laser biasing current I_DC, signal D₂for controlling the dual D/A 335, or signal D3 for controlling the light emitted by the laser 274.

[0071] In one embodiment of the present invention in which analog signals modulate the input to the laser 274, the processor 305 senses the operating temperature of the laser 274 by means of the sensing amplifiers 361. As is further described below, the laser 274 undergoes a calibration routine to determine the laser transfer function and optimum bias condition for laser operation at one or more laser operating temperatures. The processor 305 retrieves parameters from memory 315 to determine the optimum laser operating currents, and outputs a corresponding digital signal D₁to the pulse width modulator (PWM) or DAC 330 and a corresponding digital signal D₂to the DAC 335. In an alternate embodiment, in which digital signals modulate the laser frequency utilizing DAC 340, the processor 305 senses the operating conditions and performs a similar compensation as described above by adjusting the controller circuit DACs.

[0072] The controller circuit 300 has been described in terms of the components and interconnections for controlling a single electro-optic system 230. Alternatively, in a larger electro-optic system such as a wavelength division multiplexing (WDM) system which includes multiple channels, each channel having the aforementioned components as shown in FIG. 2B, the controller circuit 300 may operate to control the operations of some or all of the components within each channel. The architecture of an exemplary WDM electro-optic system 230 is further described in applicant's co-pending patent application “Electro-Optic Interface System and Method of Operation.”

[0073] III. Controller Circuit Operation

[0074] An electro-optic system 230, 240 of the present invention includes a controller circuit 300, as shown in FIG. 3, for controlling and calibrating the electro-optic system 230, 240. The controller circuit 300 drives the laser 274, monitors the operation of the laser 274 utilizing measurement circuits, maintains operating parameters of the laser 274 for ensuring operation within acceptable limits over temperature and device variations, and controls operation of related circuits such as signal amplifiers 8, 391 and host computer 390 interfaces. A model showing a preferred embodiment of the electro-optic system 230, 240 used to perform a factory calibration process 900 and a power up calibration process 950 is shown in FIG. 4. FIG. 9 is a flowchart summarizing a preferred embodiment of a method for a factory calibration process 900 in accordance with the present invention. FIG. 10 is a flowchart summarizing a preferred embodiment of a power up calibration and initialization process 950 in accordance with the present invention.

[0075] a. A Factory Calibration Process

[0076] Referring to FIG. 9 and FIG. 4, a factory calibration process 900 is performed by connecting testing equipment used for laser calibration to the laser drive control system 460, as shown in step 902. In a preferred embodiment, the testing equipment may consist of an external optical power meter instrument 407 having an I/O interface to a host computer 433, as shown in FIG. 4. The Host Computer 433 of a preferred embodiment is a personal computer (PC) that connects to the laser drive control system utilizing any suitable input/output (I/O) interface such as a universal serial bus (USB) interface. An optical power meter 407 is connected to the output of the laser 406 in a manner that represents the physical configuration of the target system, i.e. the system that utilizes the calibrated laser. For example, a target system may be a laser system utilized in medical applications or a laser system for fiber optic communications.

[0077] Once the testing equipment is connected to the laser module system 902, the dark current of a photodiode 409 is determined in steps 904 through 912, by first setting the preamplifier 403 gain to nominal, step 904. The signal preamplifier 403 normally has a given amount of offset voltage. This offset voltage translates into an offset current being applied by the High Frequency Voltage Controlled Current Source (HF-VCCS) to the laser diode 406. The methods described herein corrects for the offset current caused by the signal path 454. This offset current will cause an error for applications where the electro-optic system 230, 240 is required to drive the laser 406 with a DC signal in addition to an AC signal. The gain of the laser driver amplifier 456 is set to a predetermined nominal amount in order to amplify the offset by the amount expected during normal operation. After this adjustment is made, any drift of the offset voltage with temperature or with a small change of amplifier gain is accounted for by setting an appropriate margin in the amount of DC bias current I_Lto the laser 406, as discussed further below for step 990 in the power up calibration sequence 950 as shown in FIG. 10.

[0078] The appropriate selection of resistor values R1 and R2 for the preamplifier 403 allows an offset adjustment 437 to be highly independent of a gain adjustment 421. A preferred way to achieve independent adjustments for offset and gain for the laser driver amplifier 456, and to retain a high input impedance for the signal input 401, is to apply an offset current to the minus input of the preamplifier 403 with a relatively large value of resistor R1 (413) as compared to the resistance value of the high frequency voltage variable resistor 415. In an alternate embodiment, independent offset and gain adjustments may be accomplished utilizing an offset adjust D/A converter 437 that contains a high impedance current source output. Another option to adjust offset is the use of a separate amplifier (not shown) inserted between amplifiers 403 and 404.

[0079] In step 906, zero volts is connected to the input of the laser driver amplifier 456 by connecting the Test System Switch (TSS) 402 to circuit common 450. Zero volts is connected to the high frequency amplifier 456 to ensure that a high frequency DC/AC signal current I_Sis not applied to the laser 406 via the signal path 454. The calibration process continues by turning off the DC bias for the laser, step 908, by supplying an appropriate digital word to the digital to analog converter 425. The analog signal produced by the digital to analog converter 425 turns off the low frequency bias current I_Lnormally generated by the Low Frequency Voltage Controlled Current Source (VCCS) comprised of amplifier 424, transistor 423 and resistor 430. With the high frequency signal drive I_Sand the low frequency bias drive I_Lset to an off condition, no light is emitted from the laser 406, and the dark current in the photodiode 409 may be measured.

[0080] In step 910, the offset voltage of the laser driver amplifier 456 is canceled by changing the offset adjust digital to analog converter 437 while monitoring the offset by a measurement of the voltage across resistor R3 (416). To null the offset voltage of the laser driver amplifier 456, the laser drive control system 460 adds an offset current to the minus input of amplifier 403. Utilizing the offset adjust DIA converter 437, the laser drive control system 460 sets an appropriate voltage value to resistor R1 (413), which then applies the appropriate value of current to the minus input of the preamplifier 403. The offset D/A converter 437 is incremented or decremented until the offset of the high frequency signal path 454 is zero. The high frequency signal path 454 is driven by amplifier 403 and the High Frequency VCCS 404, 405, 416. The offset measurement is carried out by first sensing the voltage across R3 (416) with an offset sensing resistor R4 (417) that is coupled to the processor 431 via a measurement path of a sensing amplifier package 436 connected to a multiplexer 435 that is connected to an analog to digital converter 432. One amplifier in the sense amplifier IC 436 amplifies the sensed voltage, and the processor 431 sends the appropriate control signal to the multiplexer 435 to select the input of the sensed voltage across the offset sensing resistor R4. The resistor R4 and the sense amplifier IC 436 provide high impedance isolation for the laser driver amplifier circuit 456. The digital processor 431 then determines if the word received corresponds to a zero current.

[0081] The dark current of photodiode 409 is calibrated, as shown in step 912, in order to avoid errors as the photodiode 409 changes with temperature and aging. If the laser 406 has no current drive, then the output light power of the laser 406 is zero, that is, no light is coupled into monitoring photodiode 409. The parasitic dark current I_Dis defined as the current that flows through the photodiode 409 for a no-light output condition. The measured dark current I_Dis utilized to compensate the system measurements. FIG. 5 shows the volt-ampere characteristics for a germanium photodiode. These characteristics are clearly nonlinear for reverse voltages below 10 volts. A typical photodiode current I_Ois given by

[0082] I_O=I_R+I_D(1−e^V/nV)  equation 1

[0083] where I_O, I_R, and I_Drepresent the magnitude of the reverse current, V is negative for a reverse bias, and n varies based upon the material used in the process. V_Tis a voltage equivalent of temperature, where V_T=T/11,600, and T is in degrees Kelvin. I_Dis the reverse saturation current which is also known as dark current. The photodiode current, as shown in equation 1, is highly dependent on temperature. Additional leakage of current over the surface of the photodiode contributes to this current beyond that shown in equation 1. Additional experimental data places the temperature drift of the reverse saturation current I_Dat 7% per degree C. In addition, the volt-ampere characteristics will drift somewhat with photodiode age. Finally, I_Ris the current proportional to the light intensity obtained with low values of a series resistance. The series resistance R9 (410), as shown in FIG. 4, is represented by the load line of FIG. 5.

[0084] There are many other types of photodiodes and photo detectors. For example, some systems utilize avalanche photodiodes. These other devices, which exhibit similar issues in terms of temperature drift and aging, also can be described by equations. Thus, other embodiments of the present invention may be calibrated in a similar manner to what is shown in the preferred embodiment utilizing an appropriate model.

[0085] An example of the significance of the dark current error is illustrated utilizing the characteristics of the monitor photodiode used with the Mitsubishi MLXX2 series laser diode. This device has a dark current I_Dof 0.5 μA at 25° C. At 55° C., the dark current will have been increased by 7% factor for each additional change of temperature in degree C., i.e. a change of temperature of 30° C., resulting in a dark current I_D=(0.5 μA) (1.07)³⁰=3.806 μA. For this example, a laser output of 3 mW may have a photodiode current that is 150 μA. Thus, the change in the dark current from 25° C. to 55° C. is approximately 2.5% of the photodiode output current. Generally, errors of no more than a fraction of a percent are desired in order to allocate error budget for the other areas of the electro-optic system. The foregoing discussion demonstrates the need to account for dark current. A laser drive controller 460 that utilizes a photodiode 409 to monitor and control a laser must account for the dark current in order to avoid significant errors due to temperature drift and aging of the photodiode.

[0086] A preferred embodiment of the present invention for measuring and subtracting the dark current from the laser drive controller 460 is shown in FIG. 4. This embodiment includes of a current amplifier consisting of an operational amplifier 420 and a voltage variable resistor 412. Matching photodiodes 409 and 418 are connected to the summing junction of the operational amplifier 420 through series resistors R9 (410) and R8 (419), respectively. Typically, matching photodiodes are photodiodes from the same manufacturing process, i.e. that are made at the same time. The operational amplifier 420 is a precision amplifier with low bias currents and a current-to-voltage conversion gain determined by the magnitude of the voltage variable resistor 412. The photodiodes 409, 418 are mounted on a circuit card of the laser drive controller 460 in such way that the temperatures of each photodiode match and track with changes. For example, the photodiodes 409, 418 may be positioned against each other as part of the same module or in the same integrated circuit to ensure a close match of temperatures. Each photodiode 409, 418 is reverse biased by the application of a voltage +V1 or −V2 as needed. In a preferred embodiment, the top photodiode 409 receives a portion of the laser output light, whereas the bottom photodiode 418 receives no light. Thus, the total current into the summing junction of the amplifier 420 is given by:

[0087] I_O=I_p1−I_p2=I_R+I_D(1−e^V/nV)−I_D(1−e^V/nV)=I_R  equation 2

[0088] In this manner, one of the branches to the summing junction subtracts the dark current, I_D. The only current remaining is the current I_Rresulting from the light intensity being coupled from the laser diode 406 to the photodiode 409.

[0089] Another common problem in photo detection is that the photodiode devices may exhibit non-linearities in portions of their characteristics. An important feature of the circuit of the preferred embodiment is the presence of a current amplifier 420 that allows the use of low values, e.g. tens of ohms, for resistors R8 and R9. The low value of resistance causes a load line, as shown in FIG. 5, to have a steep slope, which allows photodiode operation in the area where the current I_Rchanges linearly with the light intensity. This is crucial in a preferred method of calibration since the matching photodiode circuit simplifies the calibration of the photodiode 409, and allows the use of only two data points to determine the line representing the photodiode characteristic.

[0090] In second embodiment of the present invention, it is also possible to utilize only one photodetector 409 and simply subtract the dark current I_D. The remaining discussion of the methods for calibration assumes that the second photodetector 462, as shown by the by dotted line in FIG. 4, is not present. The second embodiment for subtracting the dark current I_Dutilizes a compensating current I_Capplied to the summing junction of the operational amplifier 420. The compensating current Ic is generated by changing the value of a digital word sent to the offset adjust D/A converter 426 by the processor 431. The D/A converter 426 provides the compensating current I_Cthrough resistor R10 (411). By this process, the dark current I_Dis treated as an additional offset to the offset adjustment required by the amplifier 420.

[0091] A third embodiment for subtracting dark current I_Derror may be accomplished by measuring the magnitude of the dark current I_D, and subtracting the measured dark current value from all subsequent photodiode measurements. This approach may be used in factory calibration processes for low cost systems where there is a need to eliminate the offset adjustment circuits. For the remainder of the calibration discussions, it is assumed that the approach used simply subtracts dark current and no offset adjusting circuit is used.

[0092] The threshold currents of the laser 406 and corresponding photodiode currents 409 are determined in steps 914 through 922 as shown in FIG. 9. These steps are preformed to calibrate the photodiode current at the laser threshold, and to set the laser current to the threshold during calibration. FIG. 6 illustrates the various parameters utilized for calibration of the photodiode. Specifically, FIG. 6aillustrates a laser gain characteristic 602 and a photodiode gain characteristic 604. The Optical Power Meter 407 is utilized to obtain the laser power P_L. In step 914, the DC bias current to the laser 406 is incremented gradually by changing the input of the bias adjust digital to analog converter 425 utilizing the digital processor 431. The change to the input of the digital to analog converter 425 applies a voltage to a precision low frequency Voltage Controlled Current Source (VCCS) that changes the current output of the low frequency VCCS on the DC bias path 452. The low frequency VCCS consists of an amplifier 424, a transistor 423, and a resistor R7 (430). The low frequency VCCS of the present invention has the capability to increment the bias current I_Lof the laser diode 406. Gradually incrementing the laser bias current to avoid an overcurrent switching of the laser is recommended by manufacturers to avoid damaging the laser due to the drastic changes that may occur in the laser characteristic with a change in temperature. Additionally, the level of power output P_Lis not known prior to power up of any given laser 406. Each time the laser bias current I_Lis incremented, the laser power P_Lis read by the host computer 433 with external optical power meter 407 as shown in FIG. 4.

[0093] In step 916, the DC bias current I_Lis set to a threshold I_LTHby setting the bias current I_Lto a value at which the derivative dP_L/dI_Lstops changing. This method is guarantees that the threshold I_LTHwill be set in the linear region of the laser characteristic. In other words, by computing dP_L/dI_L, the derivative of laser power with respect to laser current, it is possible to know whether the laser has fully transitioned from a region of spontaneous emission, to a region of stimulated emission. The transition is known to have occurred when the derivative dP_L/dI_Lreaches a maximum constant value and the laser current I_Lhas gone past the threshold I_LTH. Thus, during the calibration process of the preferred embodiment, the laser current I_Lis incremented beyond the threshold current I_LTH, and then gradually decreased until a change in the derivative is detected again. Depending on the specific laser drive control system 460 used in the electro-optic system 230, 240, there may be a need to implement a low pass filter on the data returned from the measurement circuits if there is noise present, or if there are minor remaining non-linearities in the gain characteristic of the laser 406. Once the threshold current I_LTHis determined, the laser drive control system 460 records the laser threshold current value I_LTH. Any further changes in the threshold current I_LTHdue to conditions such as a change in temperature may be treated as a system offset by the processor 431.

[0094] The foregoing method of determining the laser current threshold I_LTHis different from approaches employed by laser manufacturers. Typically, manufacturers define the current threshold as the intersection of a tangent line in the region below the transition region of a laser characteristic to a tangent line in the region above transition region. This approach will derive a value of the threshold current I_LTHcentered in the transition region, i.e. the nonlinear region of the laser characteristic. The method of the present invention utilizes a point at which dP_L/dI_Lreaches a stable value to determine a laser current threshold I_LTH. This method guarantees that a laser 406 will operate in the linear region of the laser characteristic for a measured threshold current I_LTH.

[0095] Continuing with FIG. 9, once the threshold I_LTHis determined, the laser power at threshold P_LTHis read and recorded, as shown in step 918, utilizing the optical power meter 407. In step 920, the photodiode current I_P1is measured and recorded under the threshold conditions set in step 916. The recorded value of the photodiode threshold current I_P1Tis corrected by the microprocessor 431 by subtracting the value of the dark current I_D, as determined in step 912, from the photodiode current measurement in order to obtain a calibrated value for the photodiode current I_PTHat the laser threshold I_LTH. In this embodiment of the present invention, the laser drive control system 460 does not use the dark current compensation circuits 462, as shown in FIG. 4. Specifically, the threshold value of the photodiode current I_P1Tis measured and recorded via the path through amplifier 420 to the MUX 435 and A/D converter 432, and to the digital processor 431. A precision method for measuring the laser bias current I_Lstarts with the sensing of a voltage across resistor R7 (430) by a sensing resistor R6 (429). The voltage is routed through an amplifier of the sense amplifier package 436 to the multiplexer 435, and is converted to a digital word by A/D converter 432. The processor 431 examines the digital word from the A/D converter 432.

[0096] In step 922, the temperature of the laser temperature sensor 408 is read and recorded at the threshold calibration. The temperature at threshold calibration T_STCis utilized by the processor 431 to associate a given temperature with the threshold laser current I_LTH, the threshold photodiode current I_PTH, and the photodiode dark current I_D. The voltage of the temperature sensor 408 is amplified by one of the sensing amplifiers 436, routed by the multiplexer 435, converted into a digital word by the A/D converter 432, and then recorded in memory by processor 431. By a similar process, the processor 431 determines the power consumption in the laser 406 by measuring the DC bias current I_Lwith sense resistor R6 (429), and by determining a voltage across the laser diode by measuring the voltage +V1 and the voltage at the collector of transistor 423 with the use of sense resistor R5 (422). Utilizing the thermal model as described below in the section entitled “Drift Compensation Methods and Laser Thermal Model”, the temperature of the laser 406 is determined and recorded by the Processor 431.

[0097] Continuing with FIG. 9, the system gain calibration is determined in steps 924 through 928. Using the test system switch 402, the input to preamplifier 403 is disconnected from the circuit common 450 and connected to the voltage reference 428 of the signal generator 427. A precision full-scale reference voltage V_REFis applied to the input of the signal preamplifier 403. FIGS. 7aand 7bare graphical representations of a model utilized for this step. FIG. 7aillustrates an ideal system gain characteristic 704 and an actual system gain characteristic 706 for a laser signal current I_S. The actual characteristic 706 has a different slope than the gain represented by the ideal characteristic 704. Differences in the ideal and the actual gain characteristics 704, 706 result from errors in the gain of the preamplifier 403 as well as errors in the slope efficiency of the laser 406.

[0098] The high frequency AC/DC signal path, which includes the laser, has an overall gain G_Sthat is defined as the output optical power P_Lper unit of input signal current I_Sto the laser 406. The input signal current I_Sis related to the input voltage signal 401 that is used to modulate the laser 406. Further, the signal system gain G_Sis a multiplier to the input signal I_S, and is not related to the DC bias of the laser 406. One way to determine the system signal gain G_Sis to determine the response generated by an input signal I_s. By the principle of superposition, any additional current driving the laser 406 is simply added to the value of I_LTH, and may be analyzed independently from the bias current I_Lof the DC path. Thus, the DC bias I_Lis shown as an offset that translates the origin of the coordinate system of FIG. 7ato the threshold point I_LTH, and any additional current to the laser 406 is driven from the high frequency signal path 454. As shown FIG. 7a,for gain calculation purposes, the origin of the laser characteristic 706 is placed at the point (I_LTH, P_LTH). Similarly, the origin of the photodiode characteristic 708 is placed at the point (I_PTH, P_LTH).

[0099] In step 926, the gain G of the high frequency AC/DC signal path for the electro-optic system is compensated to account for the deviation in the ideal and actual gain characteristics as described above and as shown in FIG. 7a.While applying a nominal full-scale voltage V_REFto the laser driver amplifier 456 via the test system switch 402, a voltage of V_REFG_1Ais produced at the output of the preamplifier 403, where G_1Ais the gain of the preamplifier 403. This produces a maximum laser current I_SMof

[0100] I_SM=V_REFG_1A/R_1A  equation 3

[0101] and a laser optical power output of:

[0102] P_LMA=V_REFG_1AG_LA/R_1A  equation 4

[0103] where G_LAis the gain of the laser transfer characteristic (i.e. the slope efficiency).

[0104] The system gain transfer characteristic exhibits an actual gain G_1AG_LA/R_1A, that is different from the nominal or ideal gain. At this point in the process, the value of the maximum laser current I_SMand the value of the laser optical power output P_LMAare read and recorded. The Optical Power Meter 407 is utilized to measure the laser optical power output P_LMAin this process. For a given laser current I_SM, different readings of power are obtained depending upon the slope of the laser characteristic 706 and the actual gain of the laser driver amplifier 456.

[0105] FIG. 8 illustrates a general approach to calibrating gain errors. In this example, the values obtained for power are:

[0106] P_I=I_MG_I, and P_A=I_MG_A.  equation 5

[0107] where the subscript A represents actual measurements and the subscript I represents the ideal values we should obtain. Solving for P_I

[0108] P_I=P_AG_I/G_A=(G_I/G_A)G_AI_M  equation 6

[0109] Equation 6 indicates that to obtain the ideal amount of power starting from an actual amount with a gain error that is yielded by the system, the actual gain must be multiplied by the factor (G_I/G_A), which is the ratio of the slope of the ideal gain characteristic 804 to the slope of the actual gain characteristic 802. Applying this concept to a preferred embodiment as shown in FIG. 7a

[0110] P_LMA=(G_1AG_LA/R_1A)V_REFand P_LMI=(G_1IG_LI/R_1I)V_REF  equation 7

[0111] By utilizing the gain compensation technique given in the discussion leading to equation 6, P_LMI=[[(G_1IG_LI/R_1I)V_REF]/[(G_1AG_LA/R_1A)V_REF]]P_LMA, and we obtain,

[0112] P_LMI=[(G_1IG_LI/R_1I)/(G_1AG_LA/R_1A)]P_LMA  equation 8

[0113] Equation 8 indicates that by increasing the gain of the signal I_SMby the ratio of the slopes [(G_1IG_LI/R_1I)/(G_1AG_LA/R_1A)], the corrected gain of the system may be obtained. Using this concept, an approach of the preferred embodiment of the present invention consists of effecting changes in the laser driver amplifier circuit 456 in order to compensate for the gain error. The model also illustrates that there is no need to utilize a costly specialized circuit such as a multiplier to compensate for the error in the slope efficiency of the laser 406 since the preamplifier driver 403 may be utilized for gain compensation. Further, for high frequency applications, only one high frequency circuit is required, namely, the high frequency driver amplifier circuit 456, and a high frequency multiplier is not required. Any time a signal is applied on the high frequency path, the signal is automatically incremented to produce a corresponding laser signal current Is that is free of gain errors.

[0114] The circuit compensation for the system gain is realized by changing the value of the gain adjust D/A converter 421, as shown in FIG. 4. The adjusted output changes the effective resistance of the High Frequency Voltage Variable Resistor 415 of the preamplifier 403 until the desired reading of the output power P_Lis read on the Optical Power Meter 407. This procedure sets the gain of the high frequency preamplifier 404. An example of a device that may be used to change the gain is a high frequency, digitally controlled gain attenuator such as an AMP-MACOM AT65-0107. The use of this part does not require a gain adjust D/A converter 421. However, because the gain adjustment of this part may be incremented in larger steps, additional components for a feedback circuit may be required depending upon the specific embodiment of the invention. Another alternative for changing the gain is to use a voltage variable absorptive attenuator such as the AT-309 manufactured by AMP-MACOM. These types of devices will operate from DC to several GHz. A combination of resistors and other devices in conjunction with these above-mentioned parts may be needed in the preamplifier circuit in order to obtain a circuit that changes resistance for a control input that may be either an analog voltage or a digital word. The circuit needs to exhibit appropriate high frequency response. Note that during all of the above process, the laser DC bias is maintained at the threshold value I_LTH. Thus, if the system has not been fully warmed up prior to calibration, depending on the embodiment, there may be a need to re-adjust the low frequency bias current I_LTHto the threshold before continuing with the system gain calibration.

[0115] Referring again to FIG. 9, step 928 determines and records the resulting preamplifier 403 gain G_Fneeded to produce a full-scale output of the laser power P_LMI. When the output power P_Lreaches a predetermined value for full-scale output P_LMI, as indicated by the Optical Power Meter 407, the processor 431 records the value of the gain G_Frequired to achieve the full-scale signal gain of the system. In the preferred embodiment, the laser output power P_LMIis the ideal target output. Once the gain calibration is complete, the subscript “I” may be dropped from the power term, and the equation for the full-scale output power may be written as:

[0116] P_LM=V_REFG_FG_LA/R_1A  equation 9

[0117] In step 930, the maximum current I_PMthrough the photodiode 409 is determined for the full-scale laser power P_LMIcondition. When the output power from the laser 406 is at full scale, the monitoring photodiode 409 should also exhibit a full scale current I_PM. Because of gain errors in the characteristic of the photodiode 409, the photodiode 409 will yield a value of current that is different from the expected ideal. By the methods described in step 928, and by superposition, the gain adjustment of the photodiode can be made independently of the dark current adjustment. Compensation for the gain error of the photodiode 409 from the characteristic shown in FIG. 7b,is realized in one preferred method by increasing the gain of amplifier 420 by an amount equal to the ratio of the ideal to the actual value of the slope of the photodiode characteristic 708. Accordingly, the gain adjust D/A converter 438 is changed in value by the processor 431. An increment or decrement of the output value of the D/A converter 438 changes the resistance of the voltage variable resistor 412, which, in turn, changes the gain characteristics of the photodiode 409. As the gain is incremented, the photodiode current is monitored until the nominal full-scale value of the photodiode current I_PFis obtained. Once the photodiode current I_PFis obtained and recorded, all subsequent measurements of power made with the photodiode 409 will be accurate at the measured temperature.

[0118] In an alternate embodiment, the gain calibration of the photodiode 409 is accomplished utilizing a fixed resistor in place of the voltage variable resistor 412. This embodiment does not require a gain adjust D/A converter 438, and the photodiode current gain is adjusted by a calculation carried out in the processor 431. This method of gain calibration is more complex, and is used for illustration of the remainder of the method of calibration discussions. Accordingly, to calibrate for gain, the calibration method includes the step of finding a second data point on the photodiode characteristic 708, as shown in FIG. 7b,in order to calculate the photodiode gain. The gain may also be referred to as the effective responsivity R_eff. When the laser diode 409 produces a full-scale output, the corresponding photodiode current is measured. After subtracting the dark current I_D, a value of the maximum photodiode current I_PMis obtained. The Optical Power Meter 407 simultaneously reads a value of the maximum power output P_LMIof the laser.

[0119] As shown in step 932 of FIG. 9, the effective photodiode responsivity R_effof the laser 406 and the photodiode 409 combination is calculated and stored in memory utilizing the equation

[0120] R_eff=(P_LMI−P_LTH)/(I_PM−I_PTH).  equation 10

[0121] As illustrated in step 934, the factory calibration process 900 is completed by preparing drift compensation models, storing the models in a processor 431 memory, and disconnecting the optical power meter 407 and host computer 433. In this step all of the parameters previously obtained during calibration, which are needed to compensate for temperature drift, are computed and then stored by the processor 431. Additional parameters previously obtained from the characterization of the laser 406, the photodiode module 409, and the amplifier system are utilized for preparing drift compensation models. These additional parameters, which include drift coefficients, are discussed further the section below entitled “Drift Compensation Methods and Laser Thermal Model”.

[0122] A first equation utilized for drift compensation models is a threshold compensation formula given by

[0123] I_LT(T)=I_LTH+D_TH(T)ΔT  equation 11

[0124] where I_LT(T) is a new laser DC threshold as a function of temperature, and D_TH(T) is a function of temperature representing the temperature drift coefficient of the threshold. A gain compensation formula is given by

[0125] G_F2(T)=G_F+D_G(T)ΔT  equation 12

[0126] where G_F2(T) is a new amplifier gain as a function of temperature, G_Fis the gain obtained in the gain calibration, and D_G(T) is a function of temperature representing the temperature drift coefficient of the laser and amplifier gain. A responsivity compensation formula is given by

[0127] R_effT=R_eff+D_Reff(T)ΔT  equation 13

[0128] where R_efTis a new responsivity of the photodiode as a function of temperature, R_effis the responsivity obtained in the gain calibration of the photodiode, and D_Reff(T) is a function of temperature representing the temperature drift coefficient of the photodiode gain.

[0129] The calibration parameters as well as the parameters related to the thermal model are stored in the memory of the processor 431. The final step of the factory calibration process 900 includes disconnecting the optical power meter 407 and host computer 433 from the laser drive control system 460. No further use of this equipment is required unless there is a need to re-calibrate the system due to significant aging degradation of the laser 406 or laser diode 409.

[0130] b. A Power Up Calibration and Initialization Process

[0131] FIG. 10 illustrates the power up calibration and initialization process 950 utilized with an electro-optic system as shown in FIG. 4 in accordance with a preferred embodiment of the present invention. During power up, the laser drive control system 460 utilizes the equations and the obtained calibration parameters of the factory calibration process 900. The laser drive control system 460 also utilizes the drift compensation models discussed below.

[0132] The steps of the power up calibration and initialization process 950 are similar to the steps used in the factory calibration process 900. Thus, the discussions of the preferred embodiment of this method focus on the method steps which have not been introduced in the factory calibration process 900. Referring to FIG. 10, step 952 sets the preamplifier gain is to the full-scale signal gain G_Fof the system that was obtained during factory calibration. Because the gain G_Falso produces gain for an offset of the amplifier 403, a representative value of the offset for the amplifier 403 must be obtained to eliminate the offset. In step 954, the input of preamplifier 403 is connected to circuit ground 450 using the Test System Switch (TSS) 402. This step connects zero volts to the input of the preamplifier 403. In step 956, the DC bias current I_Lfor the laser 406 is turned off. The method continues by cancelling the offset voltage of the preamplifier 403, step 958, and reading and recording the dark I_Dcurrent for the photodiode 409, step 960. In step 962, the temperature T of the laser module 406 is read utilizing an output of the temperature sensor 408.

[0133] Continuing with FIG. 10, step 964 starts the process to determine laser threshold current I_LTHby gradually increment the DC bias current I_Lfor the laser 406. The photodiode current I_P1is measured via a path through an amplifier 420 to a multiplexer 435 that is coupled to the processor 431 by means of an A/D converter 432. In a next step 966, the laser power P_Lis computed using the equation

[0134] P_L=(I_P1−I_D)R_effT  equation 14

[0135] Equation 14 first subtracts the dark current I_Dfrom the measured photodiode current I_P1and multiples the result by the effective responsivity over temperature R_effTobtained in step 934 using equation 13 of the factory calibration. In step 968 of the power up calibration and initialization method 950 of a preferred embodiment, the DC bias current I_Lis set to threshold by selecting a point of I_Lat which dP_L/dI_Lstops changing. The laser power P_Lis determined at the DC bias current threshold I_LTH, as shown in step 970, by first reading the corresponding photodiode current I_PT, subtracting the dark current, and computing the corresponding power utilizing equations 15 and 16 as follows:

[0136] I_PTH=I_PT−I_D  equation 15

[0137] P_LTH=(I_PTH−I_D)R_effT  equation 16

[0138] The threshold current I_PTHof the photodiode 409 and the threshold output power P_LTHof the laser 406 are recorded in the processor memory 431 in step 972.

[0139] Continuing, step 974 of the power up calibration and initialization 950 includes determining and recording the temperature of laser module T_THat the threshold condition. This step assumes that the laser 406 is warmed-up to stabilize its operating temperature. Depending upon the laser characteristics of a specific embodiment implemented utilizing the method of the present invention, a wait time may be required in this step.

[0140] Step 976, as shown in FIG. 10, represents the start of the system gain calibration procedure. The Test System Switch 402 is switched from circuit common 450 to a voltage reference V_REF428 generated by a signal generator 427 to simulate a full scale input signal 401. In step 978, the corresponding photodiode current I_PMis determined for above full-scale voltage reference V_REF. In steps 980 through 984, a corresponding laser power output P_LMAis determined using the photodiode equation 14. The processor 431 compensates for gain errors by incrementing gain G₁of preamplifier 403 while determining laser power output P_Lutilizing photodiode measurements. The gain G_Fneeded to obtain full-scale output of laser power P_LMis determined and recorded.

[0141] In steps 986 and 988, the reference voltage is disconnected from the preamplifier input 402, and reconnected to the circuit ground 450. Continuing with the preferred method in step 990, the DC bias current I_Lis set to the operating point utilizing the measurement circuit comprised of sense resistor R6 (429), an amplifier of the sensing amplifiers 436, the multiplexer 435, and the A/D converter 432 that is connect to the processor 431. The DC bias operating point is given by the equation

[0142] I_LQ=I_LTH+Δ_LTH  equation 17

[0143] wherein the DC bias operating point I_LQ, also referred to as the “Q point”, is set at a value above the laser threshold current I_LTH. The exact bias is dependent upon the specific use of components. For digital applications of the Electro-Optic System Controller of the present invention, Δ_LTHincludes a sufficient margin of error to prevent operation near the threshold level region which is nonlinear and exhibits a relatively high level of noise. The margin Δ_LTHcalled here “threshold margin” allows the laser 406 to operate in the target region even when there are small changes in temperature between calibrations.

[0144] For analog applications of the Electro-Optic System Controller of the present invention, Δ_LTHcontains an additional biasing term in addition to the previous threshold margin as discussed above. This second term is defined as the “range bias”. The range bias will send the DC operating point I_LQto a location near the center of the laser characteristic for the laser in order to maximize the dynamic range and the signal to noise ratio. For digital applications of the Electro-Optic System Controller, an equivalent equation to equation 17, but given in terms of power is:

[0145] P_LQ=P_LTH+ΔP  equation 18

[0146] The term ΔP is the amount of laser power that corresponds to the threshold margin. For analog applications of the Electro-optic System Controller, an equivalent equation to equation 17, but given in terms of power is:

[0147] P_LQ=P_LTH+(P_LM−P_LTH−2ΔP)/2  equation 19

[0148] where P_LMis the maximum laser power that can be utilized for a specific laser. The term 2ΔP, called here “power margin”, provides a margin to avoid operation near both extreme ends of the laser characteristic. This margin ensures that the maximum power capability of the laser is not exceeded and that the laser does not operate near the threshold. In addition, the laser characteristic may become nonlinear in the maximum power region, especially as the laser ages. Equation 18 provides the margins to avoid operating near threshold in the case of applications of the drive controller such as digital signal transmission. Equation 19 provides margins to avoid operating near threshold, and to maximize signal to noise ratio for analog applications such as analog signal transmission.

[0149] Continuing with the power up calibration and initialization 950 method of FIG. 10, the system computes drift compensation parameters in step 992. In addition, the slope efficiency of the transfer characteristic G_LAis computed and stored to monitor any degradation of the laser 406. The system also prepares drift compensation equations as discussed below. The compensation parameters are stored in the processor 431 memory.

[0150] In the final step 994 of the power up calibration and initialization method 950, the processor 431 sends control signals to reconfigure the system test switch 402 to connect the input of the preamplifier 403 to the signal driving the laser 401. This marks the end of the power up calibration and initialization process 950 of a preferred embodiment of the present invention, and the electro-optic system 230, 240 is ready for use.

[0151] c. Drift Compensation and Laser Thermal Model

[0152] Referring to FIGS. 4 and 11, upon completion of the power up calibration and initialization process 1002, the electro-optic system enters a monitor phase 1006. During the monitor phase 1006, the laser drive control system 460 performs drift compensation utilizing drift models, and determines how to adjust critical controller circuit parameters. The temperature of the laser 406 is monitored for any changes that require re-adjustment of critical system parameters such as laser DC bias I_Lor system gain. Drift compensation is performed in a continuous manner, and without the need to disconnect the electro-optic system 230, 240 from the input signal 401. The method determines the adjustments with the use of drift models, and then changes the system parameters, as needed, utilizing very gradual steps that emulate analog drift. A key advantage of the compensation method of the present invention is that the compensation adjustments are made without taking the laser off-line or connecting any equipment.

[0153] There are several types of drift compensation methods carried out by the system during the monitoring phase 1006 including laser threshold drift compensation 1008, amplifier and laser gain drift 1010 compensation, and photodiode effective responsivity drift compensation 1012. In addition, a thermal model 1014 for the laser 406 determines whether the temperature has changed by a significant amount, requiring compensation of the laser 406 parameters. The monitor phase continues as long as the electro-optic system 230, 240 is operating 1016.

[0154] 1. Laser Threshold Drift Compensation

[0155] The formula used for laser threshold drift compensation 1008 is given by:

[0156] I_LTH2=I_LTH+D_TH(T)ΔT  equation 20

[0157] where I_LTH2is a new laser DC bias, D_TH(T) is a function of temperature representing the temperature drift coefficient of the bias current, I_LTHis the laser current threshold calculated at the last power up calibration, and ΔT is the change in temperature from the value at the power up calibration of the threshold current. The drift coefficient D_TH(T) is the best-suited approximation to the drift behavior of the laser, and may be a polynomial, a constant or a set of data points stored in the processor memory 431.

[0158] Referring to FIG. 4, in one embodiment of the present invention, the laser bias current adjustments are made utilizing the bias adjustment D/A converter 425 that is a 12-bit D/A converter with a 5V reference voltage. This device provides a resolution of ΔV=5V/2¹²=1.22 mV/count. If the low frequency VCCS uses a 10Ω resistor R7 (430), then the device has a resolution of ΔI=ΔV/10Ω=0.122 mA/count. In one embodiment of the laser bias current adjustment circuit, the bias adjustment D/A converter 425 is a DAC1218manufactured by National Semiconductor. For a laser diode 409 such as the ML3XX1 made by Mitsubishi, the drift of the threshold current is specified as 0.317 mA/° C. Therefore, to make an adjustment for a 1° C. change in temperature, the bias adjust D/A converter word must be adjusted by approximately 2 counts.

[0159] For a specific implementation requiring a relatively high adjustment resolution and accuracy, it is possible to use a 16-bit D/A converter with a 5V reference. This device provides a resolution of ΔV=5V/2¹⁶=76.29 μV/count. If the low frequency Voltage Controlled Current Source uses a value of 100Ω for resistor R7 (430), then a resolution of ΔI=ΔV/100Ω=0.763 μA/count is obtained. Utilizing this device, a 1° C. change in temperature may be compensated in approximately 415 steps. Because the current adjustment capability is of such high resolution, the offset current drift may be continuously adjusted in a manner that simulates analog drift without taking the laser system out of operation.

[0160] 2. Gain Drift of Amplifier and Laser

[0161] The formula used to compensate for gain drift of the amplifier (1010) and laser is given by:

[0162] G_F2(T)=G_F+D_G(T)ΔT  equation 21

[0163] where G_F2(T)=new amplifier gain as a function of temperature, G_Fis the preamplifier gain obtained in the gain calibration, D_G(T) is a function of temperature representing the temperature drift coefficient of the gain for both the laser and the preamplifier. The drift coefficient is the best-suited approximation to the drift behavior of the total gain for the amplifier and laser; it can be a general equation, a polynomial, a constant or a set of data points stored in the Processor memory 431.

[0164] As an example, for the laser diode ML2XX1 made by Mitsubishi, the drift of the laser gain (slope efficiency) is specified as −0.0028 mW/mA/° C. For the preamplifier, a significant contributor to the gain of amplifiers is the drift of the resistors that determine the gain. For high performance amplifiers, the drift for the gain can be typically 0.01% per degree C. For some applications, these factors, along with the drift of the gain adjust circuits, may be combined into a single constant drift coefficient for the gain. Also, for this specific embodiment, a high resolution D/A converter may be used to adjust for the gain drift without taking the laser system out of operation.

[0165] 3. Photodiode Effective Responsivity Drift

[0166] Although in many applications the photodiode drift can be neglected, there may be some instances in which a correction for the drift of the effective responsivity of the photodiode 1012 is needed. Correction can be performed utilizing the equation:

[0167] R_effT=R_eff+D_Reff(T)ΔT  equation 22

[0168] where R_effTis a new photodiode effective responsivity as a function of temperature, R_effis the photodiode effective responsivity obtained in the gain calibration, and D_Reff(T) is a function representing the drift as a function of temperature. The D_Reff(T) function may be a polynomial, a constant or a set of data points stored in the processor memory 431.

[0169] 4. Thermal Model

[0170] A thermal model 1014 equation may represent the laser temperature in terms of the characteristic thermal resistances of the mechanical system that comprises the laser module 406. An example of this equation is

[0171] T_J=T_S+P_L(θ_JC+θ_CSH+θ_HS)  equation 23

[0172] where T_Jis the laser junction temperature, T_Sis the sensor temperature, P_Lis the laser power, θ_JCis the laser junction to laser case thermal resistivity, θ_CSHis the thermal resistivity of the laser case to the laser module holder/heat sink, and θ_HSis the thermal resistivity from the heat sink to the sensor. Methods for determining laser thermal resistance are disclosed by Mitsubishi Corporation in the “Optical Semiconductor Devices and Optical Fiber Communication Systems” data book. Temperature sensors with a wide variety of precision may be utilized to determine laser temperature. Temperature sensors include, but are not limited to, sensors that are part of an integrated circuit, which includes the temperature sensor and the laser in the same integrated circuit. Other more complex thermal arrangements for mechanical systems involving the laser may be modeled in a similar manner with more complex thermal circuit models. These models are then characterized for a given design to determine the values of the parameters.

[0173] Variation of the calibration methods presented may be utilized depending on the constraints for a specific application. Depending upon the specific implementation of an embodiment of the present invention, the power up calibration and initialization processes 950 described above may need to insert wait times to allow for the laser 406 to warm-up to a stable temperature

[0174] The present invention has been described with reference to specific embodiments. Embodiments such as utilizing similar calibration techniques to control and stabilize the wavelength emitted by the laser are possible. Other embodiments will be apparent to those skilled in the art in view of the foregoing description. It is therefore intended that this invention not be limited except as indicated by the appended claims.