Title of Invention

METHOD AND APPARATUS FOR MEASURING THE FREQUENCY OF A RECEIVED SIGNAL

Abstract

Method and apparatus for measuring the frequency of a received signal are disclosed. The method comprising the steps of generating a first phase signal by digitizing the phase of the received signal, delaying the first phase signal by a predetermined amount to generate a second phase signal, calculating a phase difference between the first and the second phase signals and calculating the frequency of the input signal from the phase difference. The apparatus comprising a limiting amplifier (2, 60, 62), digitization means comprising a signal splitter, a first single bit analogue to digital converter, a second single bit analogue to digital converter, a first deserialiser (8), a second deserialiser (10), delay means and processing means.

Full Text

Field of the Invention The present invention relates to a method of measuring the frequency of a received signal. The present invention also relates to an apparatus for measuring the frequency of a received signal. It can be applied to a frequency counter. Background of the Invention There exists a need to analyse a received signal in a given frequency band and determine the frequency of a signal present at a particular frequency within the band. This is generally known as frequency measurement. A device known as a Delay Line Correlator can be used for frequency measurement. However, a Delay Line Correlator is implemented entirely by analogue components and consequently can be expensive to manufacture. It also has relatively high power consumption requirements and often has a large physical volume. The delay line correlator operates by comparing the phase of a received signal with a delayed version of the signal. Mixers are used to determine the frequency of the signal depending on the result of the comparison. Digital frequency measurement devices have also been proposed. These apply signal processing techniques such as fourier transforms to a sampled version of the received signal to derive the frequency of the received signal. In a wideband digital frequency measurement device short pulses are typically analysed over a large dynamic range. However this requires complex analogue- to-digital converters (ADCs) and a large amount of processing to cope with the frequency range, which may have an upper limit of 2GHz or more. In a narrowband digital device, a heterodyne is used to achieve increased resolution and sensitivity with reduced complexity in the ADC and processor. However such a device can only cover a narrow frequency band and therefore cannot cover wideband of interest (such as 0.5 to 2 GHz) simultaneously. EP 1450170 (Anritsu) discloses a method of measuring the frequency of a received signal comprising the steps of generating a first phase signal by digitising the phase of the received signal, delaying the first phase signal by a predetermined amount to generate a second phase signal; calculating the phase difference between the first and second phase signals; and, calculating the frequency of the input signal from the phase difference. EP 0373802 (Thorn EMI PLC) discloses a method of frequency measurement. The method comprises making a high resolution n bit measurement and then derives the frequency from the measurement by means of a best fit line In view of the above, it is an object of the invention to provide an improved method and apparatus for measuring frequency using digital components. Summary of the Invention According to a first aspect of the invention, there is provided a method of measuring the frequency of a received signal comprising the steps of: generating a first phase signal by digitising the phase of the received signal; delaying the first phase signal by a predetermined amount to generate a second phase signal; calculating a phase difference between the first and the second phase signals; and, calculating the frequency of the input signal from the phase difference. The predetermined amount of delay preferably corresponds to an integer multiple of the length of one sampling period used when generating the first phase signal. This allows digital components to be used to execute the method without requiring expensive digitisation and processing. Unlike prior digital methods, the method does not use signal processing techniques such as Fourier transforms. Unlike prior analogue methods, the method can be implemented without the need for mixers (or their digital equivalent of multipliers) in the calculation of the phase difference, reducing complexity. Preferably, the stuff of generating a first phase signal comprises: splitting the received signal into in-phase and quadrature components, thereby generating a received in-phase signal and received quadrature signal; digitising the received in-phase signal thereby generating a digitised received in-phase signal; digitising the received quadrature signal thereby generating a digitised received quadrature signal; and generating a first phase signal from the digitised received in-phase signal and the digitised received quadrature signal. This allows the phase of the received signal to be digitised using standard components. The combination of the in-phase and quadrature signal together represent the signal in complex format and allow the phase to be derived. Preferably, the method further comprises limiting the amplitude of the signal prior to the step of splitting the received signal into in phase and quadrature components. By limiting the amplitude, the resolution required to digitised signal can be reduced. Preferably, the steps of digitising use a one-bit resolution, such that the digitised received in-phase signal and the digitised received quadrature signal each comprise a succession of single bits representing signal value at a particular instant in time. By using a one-bit resolution, so that each sample has only two possible values the processing requirements can be further reduced. Preferably, the method further comprises de-serialising the succession of single bits of the digitised received in-phase signal and the digitised received quadrature signal into words having a predetermined number of bits. If the succession of single bit samples is grouped together into words having a predetermined number of bits, these words can then be processed at a lower clock speed thereby allowing the use of a processor with a lower clock speed. Advantageously, the predetermined number can correspond to the internal word length used when processing. Likewise, the predetermined number can be of the form 2° where n is an integer greater than or equal to 1. Preferably, the method further comprises: converting the phase difference into an in-phase and quadrature components, thereby generating a phase difference in-phase signal and a phase difference quadrature signal; filtering the phase difference in-phase signal thereby generating a filtered phase difference in-phase signal; filtering the phase difference quadrature signal thereby generating a filtered phase difference quadrature signal; generating a filtered phase difference signal from the filtered phase difference in-phase signal and the filtered phase difference quadrature signal; and, using the filtered phase difference signal in the step of calculating the frequency of the received signal. By filtering the signal the effective resolution can be increased. This allows a lower sampling resolution to be used while still achieving good resolution. In one embodiment, filter is a moving average filter. Preferably, the step of calculating the frequency of the receive signal uses the formula: Where f is the frequency of the receive signal, Fs is the sampling frequency used when digitising the signal, and d is the predetermined amount of delay used in the step of delaying expressed as a number of sample periods. This formula can be implemented in a simple manner without requiring complex processing. Preferably, the method is executed at least twice using a different value for the predetermined amount of delay in the step of delaying. By implementing the method more than once with a different value for the delay, cyclic (or ambiguous) frequencies can have their frequency correctly identified. According to a second aspect of the present invention, there is provided an apparatus for measuring the frequency of received signal, the apparatus comprising: digitisation means for digitising the phase of the receive signal and generating a first phase signal; delay means for delaying the first phase signal by a predetermined amount to generate a second phase signal; and, processing means for calculating a phase difference between the first and the second phase signals and for calculating the frequency of the received signal from the phase difference. Thus, the method of the first aspect can be implemented without requiring expensive specialist hardware. Preferably, the digitisation means comprises: a signal splitter for splitting the received signal into in-phase and quadrature components, thereby generating a received in-phase signal and a received quadrature signal; a first analogue-to-digital converter for digitising the received in-phase signal thereby generating a digitised received in-phase signal; a second analog-to-digital converter for digitising the received quadrature signal thereby generating a digitised received quadrature signal; and means for resolving the phase of the received signal from the digitised received in-phase signal and the digitised received quadrature signal and for outputting the first phase signal. Preferably, the apparatus further comprises a limiting amplifier connected to the input of the signal splitter. Preferably, the first and second analog-to-digital converters are single bit devices, preferably comparators. Preferably, the apparatus further comprises a first deserialiser connected to the output of the first analogue to digital converter for deserialising the output from the first analogue-to-digital converter and for outputting words having a predetermined number of bits; and, a second deserialiser connected to the output of the first analogue to digital converter for deserialising the output from the second analog-to-digital converter and for outputting words having a predetermined number of bits. Deserialiser is used to refer to any device which can convert a serial bitstream into a parallel bitstream of a given word length. Preferably, the processing means further comprises: means for converting the phase difference into in-phase and quadrature components, thereby generating a phase difference in-phase signal and a phase difference quadrature signal; a first digital filter for filtering the phase difference in-phase signal thereby generating a filtered phase difference in-phase signal; a second digital filter for filtering the phase difference quadrature signal thereby generating a filtered phase difference quadrature signal; and means for generating a filtered phase difference signal from the filtered phase difference in-phase signal and the filtered phase difference quadrature signal; wherein the filtered phase difference signal is used in the calculation of the frequency of the received signal. Preferably, the first and second digital filters are moving average filters. Preferably, the processing means is adapted to calculate the frequency using the formula: Where f is the frequency of the received signal, Fs is the sampling frequency used by the analogue-to-digital converters the signal, and d is the predetermined amount of delay used in the step of delaying expressed as a number of sample periods. Preferably, the apparatus further comprises at least two delay means each of which delays the signal by a different amount. Preferably, the delay means and the processing means are implemented in a Field Programmable Gate Array. A Field Programmable Gate Array is a standard component which can be programmed at the point of manufacture to operate in a particular way. It therefore allows a cost-effective implementation of the invention. Other devices such as programmable DSP or microprocessor could also be used for the processing means. Brief Description of the Accompanying Drawings Embodiments of the invention will now be described with reference to the accompanying drawings, in which: Figure 1 depicts a block diagram of a frequency measurement receiver according to a first embodiment of the present invention; Figure 2 depicts a block diagram of the processing to calculate a received frequency according to the first embodiment; Figures 3A, 3B and 3C illustrate the incidence of quantisation noise in the first embodiment; Figure 4 illustrates the simulated performance of the first embodiment; and The figure 5 depicts a block diagram of a second exemplary embodiment. According to a first embodiment, an apparatus (or receiver) for frequency measurement performs a frequency discrimination by measuring the differential phase between two time-separated versions of an input signal. Unlike an analogue delay line correlator, the input signal is digitised, and phase discrimination subsequently performed in the digital domain using high-speed Field Programmable Gate Array. A block diagram of the receiver according to this embodiment is shown in figure 1. Referring to figure 1, in general terms a limiting amplifier 2 compresses the input signal dynamic range such as a pair of high-speed comparators 4 and 6 can perform I-Q phase digitisation. The high-speed comparators operate at a sampling frequency Fs. This generates two serial data streams In and Q„ which are then fed into 1:16 deserialisers 8 and 10 which produced 16-bit words (Im and Qm) at 1/16th of the sampling frequency. All subsequent processing is performed by an FPGA at this reduced rate. Detailed Description of the Invention A more detailed description of each component of figure 1 will now follow. The use of a limiting amplifier 2 strips all amplitude information from the input signal and therefore the digital correlation receiver is only able to process one pulse at a time. The limiting amplifier 2 is connected to an I-Q generator 3. The I-Q generator 3 is used because a complex representation of the input signal is required. In this embodiment the I-Q generator 3 is either a distributed or a lumped-element realisation of a quadrature hybrid. Only a single hybrid is required and thus tracking errors introduced by multiple distributed element components are eliminated. The output of the I-Q generator 3 is connected to the high-speed comparators 4 and 6. A clock 5 controls the sampling frequency and operates at a rate greater than the bandwidth of the input signal. The high-speed comparators 4 and 6 must also be capable of tracking analogue signals in the band of interest. Providing the latter requirement is met, sub-Nyquist sampling can be implemented and any band (of width Fs Hz) can be analysed without ambiguity. Thus, for example, to analyse a band of 0.5 to 2 GHz sampling frequency of 2GHz is required. In this embodiment a sampling frequency of 2GHz is used. The high-speed comparators 4 and 6 produce an output of a bitstream of successive single-bit sample values at the sampling frequency; signals In and Qn respectively. The output of the high-speed comparators 4 and 6 is connected to the deserialisers 8 and 10 respectively. These reduce the data rate of the data streams I„ and Qn by combining several successive bits into longer length words output at a slower rate. This allows the signal to be processed at a slower rate than the sampling rate. Deserialisers operating at 2.5 gigabits per second are readily available and may also provide appropriately divided clocks as an output. In this embodiment the deserialisers 8 and 10 output data words having a length of 16 bits. The data streams of these 16-bit words are indicated in figure 1 by Im and Qm. Careful phase alignment of the deserialisers 8 and 10 is required. This is performed at power-up to ensure proper operation of the receiver. The signals Im and Qm are provided to a Field Programmable Gate Array (FPGA) 12. The FPGA 12 processes the signals Im and Qm using a clock frequency equal to 1/16* that of the clock 5 because each 16-bit word contain 16 samples. Thus, it also produces updated frequency data 18 at a rate of Fs/16. In order to supply the correct clock frequency to the FPGA 12, a divider 14 divides the signal from the clock 5 by 16 for supply to the FPGA 12. Therefore, in this embodiment, the FPGA is clock at 125Mz. Although in this embodiment the frequency discrimination is performed in a continuously clocked fashion, asynchronous triggered operation is possible in alternative embodiments due to the relatively high processing rate of me FPGA. The processing carried out by the FPGA 12 will now be described. A block diagram showing the processing required is depicted in figure 2. The use of single bit comparators 4 and 6 results in the two digitised signals Im and Qm effectively resolving the phase of the input signal into four states. As depicted in figure 2A the phase φ can be resolved into any of the states 20, 22, 24, 26. The phase φ is resolved by the block 27. The quantised phase representation φ is delayed by a number of sample clock cycles in the delay block 28. The phase difference between the undelayed and the delayed signals is then calculated in block 30, producing a two-bit, low resolution estimate of the differential phase Acp. (Once again, this is resolved into four possible states). This estimate is converted to IQ format by block 32 to produce signals ΔI and ΔQ. The signals ΔI and ΔQ are then filtered by filters 34 and 36. This has the beneficial result of increasing the resolution of the phase estimate, and an improved differential phase estimate Δφ' is generated by resolving the phase from the filtered signals ΔI' and ΔQ' in block 38. In this embodiment the filters 34 and 36 are both moving average filters, although other filter types could also be used. The frequency, f, of the input signal can be derived by block 40 from the differential phase estimate Δφ' as follows: Where d is the number of sample delays in the delay block 28, and Fs is input sample rate. Improved resolution can be achieved by increasing the delay in delay block 28. Cyclic (and therefore ambiguous) frequency estimates can be resolved by using multiple correlators each using varying delays. All correlators are implemented within a single FPGA , unlike prior proposed frequency measurement apparatus, and therefore an extremely compact solution is realised. In this embodiment the correlation process uses the principle of over-sampling an input signal using low-bit resolution sampling devices (the high-speed comparators 4 and 6). The digital data is subsequently filtered to reduce the quantisation noise introduced during the sampling process. The performance of the system will now be discussed, particularly in relation to the noise introduced. The quantisation noise introduced during the digitisation process in the high- speed comparators 4 and 6 is of the same order as the noise present in the input signal when the correlator operates in poor signal-to-noise ratio (SNR) environments. Both the noise present in the input signal and the quantisation noise contribute to the output noise and as a result define the frequency accuracy performance of the apparatus. An analysis of both these noise components is given below. Figure 3A depicts the four states into which the phase of the signal can be resolved. Figure 3B illustrates how the quantisation noise changes, as the absolute phase φ of the input signal is resolved into one of the four phase states depicted in figure 3A. Assuming the quantisation noise samples are statistically independent (i.e. uncorrelated with the sample clock and with each other), the noise is uniformly distributed as shown in figure 3C and can be shown to have a standard deviation or RMS value of Using standard phase noise theory, the phase noise at the output of a phase detector for a given SNR, S is: These two noise components, Gqnoise and oSnoisc are statistically independent; noise present at the output of the phase discriminator (before filtering by filters 34 and 36) is therefore: In this embodiment the filters 34 and 36 are moving average filters, which is optimal for this embodiment. The filters 34 and 36 average N adjacent samples are averaged, reducing the RMS noise by VN. In this embodiment the value of N is 64 and therefore the RMS noise is reduced by a factor of 8. Different values of N may also be used. The assumption in the calculations above that successive quantisation noise samples are uncorrected is vital for all but a few cases of input frequencies (and phase relative to the sample clock). It can be shown mat within four bands of input signal frequency defined by: Where N is the order of the filter, Fs is the sample rate and k is an integer k=1..4; quantisation noise samples are correlated resulting in a localised loss of sensitivity. In practice, system noise will tend to de-correlate this noise therefore reducing its effect. A specific example of a hardware implementation of this embodiment will now be described. This example hardware implementation has the following target specification and performance attributes: • The sample rate is 2 Gsps, with an unambiguous bandwidth of 2GHz • 125 MHz correlator processing rate • 50ns pulse-width capability • 7-bit resolution for each discriminator • 4-correlators are used in the designed to give a robust solution • there is 125MHz unambiguous bandwidth for a 'fine discriminator', resulting in an approximately 1MHz resolution • measurement delay: 120ns A VHDL design of this embodiment has been implemented and full timing driven gate level simulations performed. The results of the simulation of a single correlator are shown in figure 4. Figure 4 shows the RMS phase error for different input SNR conditions for two digital delay correlator sample delays; a 1 sample delay (curve 50) and a 16- sample delay (curve 52). The theoretical performance is also shown as curve 54, which is calculated from equation (4). It is clear that the simulation performs better than the theory suggests. It is thought that the improved performance is a result of in-phase correlation effects experienced in phase discriminator correlator designs. The fact that the short delay design appears to perform better man the long delay correlator supports this assumption. It is possible to estimate the frequency accuracy performance of the design for a given input SNR thus: The 16 sample delay 'fine discriminator'(with performance depicted by curve 52) defines the frequency accuracy. This will have an unambiguous bandwidth of 2GHz/16 =125 MHz. From figure 4, the RMS phase error for 3dB SNR is approximately 5°. The frequency error is therefore 5/360*125 = 1.7MHz RMS. The design described above was targeted towards FPGAs which are commercially available from Xilinix is sold under the mark Spartan-3. These are low-cost, high performance parts. These devices also have a large amount of on- board memory, which can be used as fast look-up-tables carrying calibration data to overcome non-ideal behaviour of the analogue components. A four-correlator design when targeted to an XC3S400 device (which is a member of the Xilinix Spartan-3 family) uses less than 12 % of the available logic resources. Gate-level timing simulations show that this design can be clocked at over 150MHz (as opposed to thel25MHz required). This margin can be exploited by reducing the amount of pipelining used in the design to decrease the measurement delay, or indeed the system clock rate could be increased to improve the resolution {i.e. increase the over-sampling rate). Simulation results state that the power dissipation of the FPGA design is around 1.2W. Dissipation of other components such as the de-serialisers, digitizers and limiting amplifier suggest that the dissipation of the complete receiver would be around 3.3W. This is approximately one third of that typically dissipated by existing analogue frequency measurement devices. Furthermore, a double-sided PCB design can be realised with approximate the mentions of 100*100*15 mm which is a quarter of the volume occupied by existing 2-18GHz analogue frequency measurement devices. Figure 5 depicts a block diagram of a second embodiment of the invention. It shows how a static microwave frequency divider could be used to implement a compact 2-18GHz receiver. The embodiment uses two limiting amplifier 60 and 62. Limiting amplifier 62 is operative for frequencies in the range 0.5 to 2GHz and limiting amplifier 60 is operative for frequencies in the range 2 to 18GHz. The output of limiting amplifier 60 is connected to a prescaler 64 which divides by 8 and produces an output in the range 0.25 to 2.25 GHz. A multi-octave I-Q generator 66 generates an in-phase and quadrature component signals to supply to the digital correlator 68. The multi-octave I-Q generator 66 could be implemented digitally using high- speed flip-flops and absorbed into the frequency division operation, this would provide for an extremely robust receiver design with 'zero' alignment. In an alternative embodiment the in-phase and quadrature signals are digitised with a resolution higher than one bit. WE CLAIM : 1. A method of measuring the frequency of a received signal comprising the steps of: generating a first phase signal by digitizing the phase of the received signal, said step of generating a first phase signal comprising limiting the amplitude of the received signal, splitting the amplitude limited received signal into in-phase and quadrature components, thereby generating a received in-phase signal and a received quadrature signal, digitizing the received in-phase signal using a one bit resolution, thereby generating a digitized received in-phase signal having a succession of single bits representing the signal value at a particular instant in time, digitising the received quadrature signal using one bit resolution, thereby generating a digitised received quadrature signal having a succession of single bits representing the signal value at a particular instance in time, and deserialising the succession of single bits of the digitized received in-phase signal and the digitized received quadrature signal into words having a predetermined number of bits; delaying the first phase signal by a predetermined amount to generate a second phase signal; calculating a phase difference between the first and the second phase signals; and, calculating the frequency of the input signal from the phase difference. 2. A method as claimed in claim 1, comprising the steps of converting the phase difference into in-phase and quadrature components, thereby generating a phase difference in -phase signal and a phase difference quadrature signal; filtering the phase difference in-phase signal thereby generating a filtered phase difference in-phase signal; filtering the phase difference quadrature signal thereby generating a filtered phase difference quadrature signal; generating a filtered phase difference signal from the filtered phase difference in-phase signal and the filtered phase difference quadrature signal; and, using the filtered phase difference signal in the step of calculating the frequency of the received signal. 3. A method as claimed in claim 2, wherein the filter used in the step of filtering is a moving average filter. 4. A method as claimed in claim 1, wherein the step of calculating the frequency of the received signal uses the formula Where f is the frequency of the received signal, Fs is the sampling frequency used when digitising the signal and d is the predetermined amount of delay used in the step of delaying expressed as a number of sample periods. 5. A method as claimed in claim 1, wherein the method is executed at least twice using a different value for predetermined amount of delay in the step of delaying. 6. An apparatus for measuring the frequency of a received signal, the apparatus comprising a limiting amplifier for limiting the amplitude of the received signal, digitization means for digitizing the phase of the received signal and generating a first phase signal, the digitization means comprising a signal splitter having an input connected to the output of the limiting amplifier for splitting the received signal into in-phase and quadrature components, thereby generating a received in-phase signal and a received quadrature signal, a first single bit analogue to digital converter for digitising the received in-phase signal thereby generating a digitized received in -phase signal; a second single bit analogue to digital converter for digitizing the received quadrature signal thereby generating a digitized received quadrature signal, a first deserialiser connected to the output of the first analogue to digital converter for deserialising the output from the first analogue to digital converter and for outputting words havng a predetermined number of bits; and a second deserialiser connected to the output of the second analogue to digital converter for deserialising the output from the second analogue to digital converter and for outputting words having a predetermined number of bits; delay means for delaying the first phase signal by a predetermined amount to generate a second phase signal; and, processing means for calculating a phase difference between the first and second phase signals and for calculating the frequency of the received signal from the phase difference. 7. An apparatus as claimed in claim 6, wherein the processing means comprises means for converting the phase difference into in-phase and quadrature components, thereby generating a phase difference in-phase signal and a phase difference quadrature signal; a first digital filter for filtering the phase difference in-phase signal thereby generating a filtered phase difference in-phase signal; a second digital filter for filtering the phase difference quadrature signal thereby generating a filtered phase difference quadrature signal; and, means for generating a filtered phase difference signal from the filtered phase difference in-phase signal and filtered phase difference quadrature signal; wherein the filtered phase difference signal is used in the calculation of the frequency of the received signal. 8. An apparatus as claimed in claim 7, wherein the first and second digital filters are moving average filters. 9. An apparatus as claimed in claim 8, wherein the processing means is adapted to calculate the frequency using the formula where f is the frequency of the received signal, Fs is the sampling frequency used by the analogue to digital converter and d is the predetermined amount of delay used in the step of delaying expressed as a number of sample periods. 10. An apparatus as claimed in claim 6, comprising at least two delay means each of which delays the signal by a different amount. 11. An apparatus as claimed in claim 6, wherein the delay means and the processing means are implemented in a Field Programmable Gate Array. Abstract Method and Apparatus for Measuring the Frequency of a Received Signal Method and apparatus for measuring the frequency of a received signal are disclosed. The method comprising the steps of generating a first phase signal by digitizing the phase of the received signal, delaying the first phase signal by a predetermined amount to generate a second phase signal, calculating a phase difference between the first and the second phase signals and calculating the frequency of the input signal from the phase difference. The apparatus comprising a limiting amplifier (2, 60, 62), digitization means comprising a signal splitter, a first single bit analogue to digital converter, a second single bit analogue to digital converter, a first deserialiser (8), a second deserialiser (10), delay means and processing means.

Documents:

00943-kolnp-2007- correspondence-1.1.pdf

00943-kolnp-2007- g.p.a.pdf

00943-kolnp-2007- priority document.pdf

00943-kolnp-2007-assignment.pdf

0943-kolnp-2007-abstract.pdf

0943-kolnp-2007-claims.pdf

0943-kolnp-2007-correspondence others.pdf

0943-kolnp-2007-description (complete).pdf

0943-kolnp-2007-drawings.pdf

0943-kolnp-2007-form1.pdf

0943-kolnp-2007-form3.pdf

0943-kolnp-2007-form5.pdf

0943-kolnp-2007-international publication.pdf

0943-kolnp-2007-international search authority report.pdf

943-KOLNP-2007-(04-12-2013)-CORRESPONDENCE.pdf

943-KOLNP-2007-(04-12-2013)-FORM-3.pdf

943-KOLNP-2007-(13-11-2013)-ABSTRACT.pdf

943-KOLNP-2007-(13-11-2013)-ANNEXURE TO FORM 3.pdf

943-KOLNP-2007-(13-11-2013)-CLAIMS.pdf

943-KOLNP-2007-(13-11-2013)-CORRESPONDENCE.pdf

943-KOLNP-2007-(13-11-2013)-DESCRIPTION (COMPLETE).pdf

943-KOLNP-2007-(13-11-2013)-DRAWINGS.pdf

943-KOLNP-2007-(13-11-2013)-FORM-18.pdf

943-KOLNP-2007-(13-11-2013)-FORM-2.pdf

943-KOLNP-2007-(13-11-2013)-OTHERS.pdf

943-KOLNP-2007-(13-11-2013)-PETITION UNDER RULE 137.pdf

943-KOLNP-2007-ASSIGNMENT.pdf

943-KOLNP-2007-CANCELLED PAGES.pdf

943-KOLNP-2007-CORRESPONDENCE.pdf

943-KOLNP-2007-EXAMINATION REPORT.pdf

943-KOLNP-2007-FORM 18-1.1.pdf

943-kolnp-2007-form 18.pdf

943-KOLNP-2007-GPA.pdf

943-KOLNP-2007-GRANTED-ABSTRACT.pdf

943-KOLNP-2007-GRANTED-CLAIMS.pdf

943-KOLNP-2007-GRANTED-DESCRIPTION (COMPLETE).pdf

943-KOLNP-2007-GRANTED-DRAWINGS.pdf

943-KOLNP-2007-GRANTED-FORM 1.pdf

943-KOLNP-2007-GRANTED-FORM 2.pdf

943-KOLNP-2007-GRANTED-FORM 3.pdf

943-KOLNP-2007-GRANTED-FORM 5.pdf

943-KOLNP-2007-GRANTED-SPECIFICATION-COMPLETE.pdf

943-KOLNP-2007-INTERNATIONAL PUBLICATION.pdf

943-KOLNP-2007-INTERNATIONAL SEARCH REPORT & OTHERS.pdf

943-KOLNP-2007-OTHERS.pdf

943-KOLNP-2007-PETITION UNDER RULE 137.pdf

943-KOLNP-2007-REPLY TO EXAMINATION REPORT.pdf

abstract-00943-kolnp-2007.jpg