1. Field of the Invention
Embodiments of the present invention relate generally to optical communication systems and components and, more particularly, to a photodiode array with algorithm-based crosstalk reduction.
2. Description of the Related Art
Photodiodes are used to perform optical power measurement and monitoring in a range of devices, including optical amplifiers, wavelength selective switches, and other systems that benefit from the monitoring of optical channel performance. In devices using multiple photodiodes, such as wavelength-division-multiplexing (WDM) devices, the size and cost of the relatively large numbers of photodiodes required can be problematic. Manufacturing multiple photodiodes together as an integrated detector array can greatly reduce the size and unit cost of the photodiodes, such as when a large number of semiconductor photodiodes are fabricated from a common substrate. However, such photodiode arrays are more readily subject to crosstalk between the elements of the array, which can affect the accuracy and resolution of the detector array.
Crosstalk between adjacent photodiode elements of a detector array is an important parameter of the performance of the array and can be due to either electrical or optical crosstalk. Electrical crosstalk occurs when carriers that are photo-generated by one photodiode element diffuse across a common substrate and are collected by a neighboring photodiode element in the array. Optical crosstalk occurs when the light energy itself is partially incident on neighboring elements, and includes the effects of photon refraction, reflection at boundaries, and external and internal scattering in a photodiode array.
The effect of crosstalk can be reduced between elements of a PD array through structural and other design changes to the PD devices in the array. For example, optical crosstalk can be reduced by the formation of a radiation shield or other optical energy barrier between the PD array elements during the process of fabricating the PD elements. Similarly, electrical crosstalk can be reduced by forming electrical barriers between the PD array elements. However, such techniques add complexity and cost to the manufacturing process. Accordingly, there is a need in the art for PD arrays having reduced crosstalk without the presence of additional structures in the PD array.
One or more embodiments of the present invention provide a photodiode (PD) array for measuring incident optical power and a method of accurately measuring optical power of light incident on each of the PDs in such an array. The effect of crosstalk between the individual PDs of the array is removed by measuring the current generated by each PD in the array and generating a corrected optical power value for each PD that is based on the measured current for each PD and on coupling coefficients associated with other PDs in the array, where the coupling coefficients are determined during a previous calibration procedure. In some embodiments, the corrected optical power values may be solved for simultaneously using a system of linear equations, where each equation expresses the measured optical power value for a PD in the array as a function of the corrected optical power value for the PD and the appropriate coupling coefficients.
In a photodiode array, a method of measuring optical power of light incident on photodiodes according to an embodiment of the present invention comprises measuring a current generated by a photodiode in the array in response to the light incident on the photodiode and generating an optical power value based on the measured current and correction factors associated with other photodiodes in the array.
A computer-readable storage medium according to another embodiment of the present invention comprises instructions to be executed by a processing unit of an optical power monitoring device to carry out the steps of measuring a current generated by a photodiode in a photodiode array in response to the light incident on the photodiode and generating an optical power value based on the measured current and correction factors associated with other photodiodes in the array.
According to another embodiment of the present invention, an optical power measuring system comprises a processor and a photodiode array, wherein the processor is configured to perform the steps of measuring a current generated by a photodiode in the photodiode array in response to the light incident on the photodiode, and generating an optical power value based on the measured current and correction factors associated with other photodiodes in the array.
So that the manner in which the above recited features of the present invention can be understood in detail, a more particular description of the invention, briefly summarized above, may be had by reference to embodiments, some of which are illustrated in the appended drawings. It is to be noted, however, that the appended drawings illustrate only typical embodiments of this invention and are therefore not to be considered limiting of its scope, for the invention may admit to other equally effective embodiments.
For clarity, identical reference numbers have been used, where applicable, to designate identical elements that are common between figures. It is contemplated that features of one embodiment may be incorporated in other embodiments without further recitation.
One or more embodiments of the present invention provide a low-crosstalk photodiode (PD) array for measuring incident optical power and a method for accurately measuring optical power of light incident on each of the PDs in an array that removes the effect of crosstalk. Coupling coefficients quantifying the optical and/or electrical coupling between each PD of the array are determined during a previous calibration procedure and are used to remove the effect of crosstalk on the different outputs of the PD array.
Optical power monitor 220 includes PD array 250 and a processor card 260. Optical power monitor 220 is a device that may be used for monitoring optical power of multiple light sources, and may be incorporated into any system or device that includes the function of optical power monitoring. For example, optical power monitor 220 may be an integrated component of an optical amplifier card, a wavelength selective switch, or other performance-monitoring devices used in a WDM application, and may be used to quantify the performance of WDM data channels, optical supervisor channels, etc.
PD array 250 includes a plurality of N PD elements, each of which is configured to monitor the optical power of light incident thereon. In some embodiments, PD elements 1-N may be part of a monolithically integrated semiconductor sensor array and fabricated on a common substrate, such as an InP or InSb substrate. When light is incident on one of the PD elements of PD array 250, the PD element integrates the incident optical energy and produces an electrical output 251 that is proportional to the optical energy in said bandwidth. Thus, in operation, PD array 250 may be utilized as an optical power monitor for a plurality of light sources, such as individual WDM data channels that have been demultiplexed from a WDM optical signal.
Processor card 260 includes electronic components and other hardware for measuring the electrical output 251 of each PD element 1-N and is configured to control the normal operation of optical power monitor 220. Processor card 260 is also configured to perform calibration measurements and the calculation of coupling coefficients between PD elements 1-N based on said measurements. Processor card 260 further includes a processor 270, which is configured to perform calculations and execute instructions for the operation and calibration of optical power monitor 220. Processor 270 can be any appropriate processor including but not limited to general-purpose processors such as microprocessors and digital signal processors (DSP), or a special-purpose processor such as an application specific integrated circuit (ASIC) or field-programmable gate array (FPGA).
In operation, calibration apparatus 200, when coupled to optical power monitor 220, determines coupling coefficients between PD elements 1-N of a PD array 250.
The method begins in step 401, in which laser source 201 is coupled to a single PD element via optical switching element 202 and optical path OP1. Thus, one PD element is illuminated with a known quantity of optical power and the remaining PD elements are completely “dark,” i.e., no incident light energy is directed thereto. For clarity, the illuminated PD element is herein referred to as PD element M, where M may be any of the 1 to N PD elements in PD array 250.
In step 402, processor 270 measures electrical output 251 of each of PD elements 1-N. Electrical output 251 of the illuminated PD element, i.e., PD element M, is approximately proportional to the optical power of light thereon, but includes significant inaccuracy due to optical and electrical crosstalk from some or all of the other PD elements in PD array 250. Electrical output 251 of each dark PD element is, within measurement error, exactly proportional to the crosstalk between the PD element M and the dark PD element. The value of each electrical output 251 measured in step 402 can be used as coupling coefficients that together can be used to solve for the actual incident optical power on PD elements 1-N. The use of coupling coefficients to solve for actual incident power is described below in steps 403-405 of method 400 and step 503 of method 500.
In step 403, processor 270 stores the coupling coefficient values measured in step 402 in memory block 303. Because the coupling coefficient values stored are each proportional to the crosstalk between the dark PD elements and PD element M, these coupling coefficient values may be used in Equation 1. Equation 1 relates the actual optical power Pincident that is incident on PD element M to the measured optical power Pmeasured that is reported by PD element M, where Pmeasured includes the effect of crosstalk from all other PD elements in PD array 250 on PD element M:
where M is a specific PD element, N is the total number of PD elements in PD array 250, and cMj is the coupling coefficient that represents the effect of crosstalk from PD element j on PD element M. It is noted that Equation 1 holds true even when PD element M is not the only PD element in PD array 250 that is being illuminated. By way of example, for the first PD element (M=1) in a 4-element PD array, substitution of the appropriate values into Equation 1 yields Equation 2:
P
measured(1)=C11Pinc(1)+C12Pinc(2)+C13Pinc(3)+C14Pinc(4) (2)
In practice, the contribution to crosstalk to PD element M by other than the closest PD elements in PD array 250 may be negligible. In some embodiments, for simplicity, only the coupling coefficients of PD elements proximate and/or adjacent to PD element M are stored in step 403, and all other coupling coefficient values are assumed to be zero.
The series of coupling coefficients stored in step 403, i.e., cM1, cM2, cM3, . . . cMN, can be used to populate the Mth row of an N×N matrix, herein referred to as a “crosstalk matrix.” As described below, the N×N crosstalk matrix can be used to simultaneously solve for the actual incident optical power Pincident for each of PD elements 1-N. In some embodiments, prior to storing the coupling coefficients for PD element M measured in step 402, PD element M is set as the reference element and the coupling coefficients are normalized to 1 for the self-coupling element in the series. In other words, the coupling coefficient representing the coupling of PD element M to itself is set to one and the remaining coupling coefficient values are adjusted proportionally, i.e., when M=j, cMj=1. This normalization process facilitates later matrix inversion of the crosstalk matrix when solving for the actual incident power Pincident.
In step 404, steps 401-403 are repeated for each PD element in PD array 250, so that each of PD elements 1-N is treated as element M. In this way, a series of coupling coefficients is collected for each PD element in PD array 250, so that the N×N crosstalk matrix can be constructed.
In some applications, crosstalk between PD elements may vary significantly as a function of wavelength. Consequently, the coupling coefficients measured in method 400 may not be accurate when different wavelengths of light are monitored by optical power monitor 220. In some embodiments, multiple sets of coupling coefficients are measured, where a different set of coupling coefficients is measured for each wavelength or wavelength band of interest. In such embodiments, method 400 is repeated for each wavelength or wavelength band. Subsequently, when optical power monitor 220 is in operation, the crosstalk matrix that is used to solve for the actual incident power Pincident is constructed by selecting the appropriate coupling coefficients for each PD element from the different wavelength-dependent sets of coupling coefficients, based on what particular wavelength of light is incident on each of the PD elements.
In operation, optical power monitor 220 determines the actual incident power on each of PD elements 1-N by measuring optical power at each PD element and applying a crosstalk matrix of coupling elements to solve the system of equations that expresses the measured optical power for all PD elements as a function of actual incident power on each PD element and the predetermined coupling coefficients. The predetermined coupling coefficients are found using method 400 and the system of equations is solved by constructing a crosstalk matrix of said coupling coefficients. Thus, application of the appropriate coupling coefficients eliminates the effect of crosstalk on the electrical outputs of PD array 250.
The method begins in step 501, in which processor 270 measures electrical output 251 of each PD element of PD array 250. It is noted that electrical output 251 of each PD element includes significant inaccuracy due to optical and electrical crosstalk from some or all of the other PD elements in PD array 250.
In step 502, processor 270 retrieves the appropriate coupling coefficients from memory block 303. In embodiments in which different wavelengths or wavelength bands are incident on different PD elements, the coupling coefficients may be selectively retrieved from multiple coupling coefficient sets.
In step 503, processor 270 solves the system of equations represented by Equation 3:
[Pmeasured]=[C][Pincident] (3)
where [Pmeasured] is the vector representing the electrical outputs 251 of PD elements 1-N measured in step 501, [C] is the crosstalk matrix constructed in step 502, and [Pincident] is the vector representing the actual incident power of light incident on each of the PD elements 1-N.
In step 504, processor 270 generates a corrected optical power value for each PD element based on the solution of Equation 3 for [Pincident].
In sum, embodiments of the invention provide a PD array for measuring incident optical power and a method of accurately measuring optical power of light incident on each of the PDs in such an array. During operation, the effect of crosstalk between the individual PDs of the array is removed by measuring the current generated by each PD in the array and generating a corrected optical power value for each PD that is based on the measured current for each PD and on coupling coefficients associated with other PDs in the array. Thus, crosstalk can be greatly reduced in a PD array without modifying the structures of the PD elements.
While the foregoing is directed to embodiments of the present invention, other and further embodiments of the invention may be devised without departing from the basic scope thereof, and the scope thereof is determined by the claims that follow.