Data quality analysis of phase-shift keying (PSK) modulated signals in a remote sensing satellite ground station is done by using “OFF line” quantitative measurement of end-to-end bit error rate (BER). If the E_b/N_o ratio received at the input of the PSK demodulator during the satellite-pass is more than that required for threshold BER performance (usually 1 × 10^-6), then the quality of the received data is assumed to be within acceptable limits.

This paper describes a simple method using a long loop carrier tracking phase lock loop technique, which enables the operator to continuously measure and monitor the E_b/No received from a remote sensing satellite during the satellite pass. The circuit has been implemented in a multi-mission PSK demodulator, using this configuration, and has been tested successfully.

The measurement of E_b/N_o ratio is usually done by simulating the satellite's transmitted spectrum. This is accomplished with an RF simulator unit having the ability to provide variable signal strengths to simulate different S/N values, at different data rates and spectral characteristics at the receive de-modulator input. These spectral characteristics are observed on the spectrum analyzer by switching “off” the modulation. This allows the measurement of the total signal carrier above the noise level, within the resolution bandwidth of the spectrum analyzer. The carrier-to-noise density ratios (C/N_o) and E_b/N_o are calculated from this quantity.

Measurement of the E_b/N_o, however, is not possible during a satellite pass, since there is no provision to switch off the modulation. The modulation must be switched off to enable the measurement of C/N_o on a spectrum analyzer. Since the identical simulation of the spectral characteristics of the two different modulators is impossible, this off line method cannot accurately analyze the real-time conditions.

There are other methods of measurement of the received E_b/N_o, which are based on independent measurement of the system noise and the received signal, plus noise (using narrow band filters or calibration filters). However, these are not only cumbersome, but also need the appropriate test filters to define the noise bandwidth of measurement.

The described method continuously measures and monitors the E_b/N_o ratio available at the input of the demodulator during the satellite pass. The method is useful even for a ground station that is configured for receiving data from various remote sensing satellites. The method utilizes the recovered coherent carrier required for the detection of PSK signals for measuring the real-time received signal-to-noise ratio (S/N).

Conventional carrier recovery circuits, however, cannot be used for this purpose. This is due to limitations in their ability to accommodate a wide range of data rates. The technique suggested here utilizes a long loop phase-lock loop (PLL) circuit. This circuit, with some modifications, not only recovers the carrier for a multi-mission PSK demodulator, but also can be used for accurate measurement of the received E_b/N_o ratio.

Conventional carrier recovery techniques require wideband prefiltering prior to the non linearity for modulation wipe off. This is necessary to accommodate the highest data rate modulated signals. The excess noise bandwidth in the filter, while receiving the lower data rate modulated signals, will, however, increase the squaring loss and, consequently, severely reduce the output S/N.

The analysis, and subsequent results, of a practical demodulator, using these conventional carrier recovery design procedures, shows that the unit performs reasonably well for limited range of input data rates (such as 40 Mbps to 110 Mbps QPSK) and E_b/No variations.

It is possible to use a narrow bandpass filter to improve the S/N ratio at the input of PLL. However, the pre-PLL bandwidth control is decided by the maximum Doppler offset on the received signal. For some low circular orbits this may be as high as ±400 KHz. Therefore, the pre-PLL filter passband response, must be flat for at least ±2 MHz. But, this method results in a noise bandwidth much larger than that.

Thusly, the online measurement of E_b/N_o, in the above design configurations for multi data rate reception system, faces several constraints and uncertainties in exact C/N_o estimation.

The long loop circuit described here has been developed to accommodate a wide range of data rates, providing a multi-mission feature for the measurement circuit. The circuit provides coherent carrier (without modulation) with its output S/N ratio being proportional to the input received C/N. This ratio is used for the estimation of received E_b/N_o and data quality analysis. The configuration of long loop scheme is explained with reference to the block diagram in figure 1.

The received modulated RF signal is mixed with the VCO signal of the PLL to provide an IF signal. The IF signal is passed through a three-way power divider. One output from the divider goes for the coherent detection, while the other two outputs are fed to the corresponding band pass filters with the bandwidth that is selected optimally for the minimum squaring loss, for the highest baud rate, in each of modulation schemes (quadrature/unbalanced phase-shift keying, binary phase-shift keying — QPSK/UQPSK, BPSK). The filtered signal is then passed through the non-linearity to remove the modulation.

The fourth, or second harmonic component of the IF carrier frequency after the non-linearity, is filtered through the corresponding very narrow bandpass filter. However, before passing this “signal without modulation” to such a filter, it is power divided and passed to the spectrum analyzer for continuous measurement of the C/N_o ratio received during the satellite pass. The S/N ratio at this point varies with S/N ratio of the input modulated received signal.

The output from the narrow bandpass filter is then phase locked with the x4 or x2 multiplied component of a very highly stable crystal oscillator source.

The crystal oscillator source frequency is centered at the IF carrier frequency and becomes the coherent carrier, reference for data detection, after phase locking. The source frequency, after multiplication, is passed through a bandpass filter with identical characteristics as the one in the signal path to the input of the PLL. This ensures the correct phase referencing in the PLL. The PLL tracks the Doppler frequency variations of the input signal by correcting the voltage-controlled oscillator (VCO) output frequency accordingly, while maintaining the IF output of the mixer at a constant frequency.

The VCO is swept for initial signal acquisition and designed for acquiring the signal with a Doppler shift as high as ±500 KHz. The Doppler effect is compensated for prior to the carrier recovery nonlinearity. This bandwidth control is made possible by using a narrow bandpass filter at the input to the carrier referencing PLL. This configuration maximizes the (S/N)out at output of the nonlinearity (or S/N input to PLL) to improve the overall lock acquisition threshold of the PLL.

The bandwidth of the narrow band pass filter, however, must be at least twice the loop bandwidth to maintain the loop dynamics.

The signal-to-noise ratio at the output of the pre-PLL narrow bandpass filter is given by:

Where:

B_n = the noise bandwidth of pre-PLL narrow band filter.

C/N = the input C/N of the received signal.

B = noise bandwidth of the pre-filter applied to the non-linearity.

Similarly, the equation of (S/N)out for the QPSK non linearity is given by:

Equations (1) and (2) are valid for the case where the C/N values are low or the signal is non band-limited. Equation (1) for the output signal-to-noise ratio of the bandwidth B_n can be rewritten as:

Hence, the output noise power spectral density becomes equal to 4N_o+[(2N_o)/(C/N)] (see figure 2). It can be seen that the power spectral density of the C × N (the cross-product of the components) component is half that of the power spectral density of N × N component after squaring. The distribution, however, is rectangular. The reduction in loop SNR is primarily due to the noise spectral densities being closer to the recovered carrier frequency 2ω_c. The contribution of the N × N noise component is considered only within the input noise bandwidth B Hz of the pre- squaring filter. This effectively halves the contribution of the N × N component owing to it having twice the bandwidth to C × N component, after squaring.

The above analysis shows the noise spectral density characteristics seen at the output of the non linearity. In order to reduce the excess noise effects (i.e, C × N and N × N contribution as explained above), a narrow bandpass filter is used in front of the PLL. It is obvious that if the signal is not pre-filtered to the PLL, the signal to noise ratio entering the PLL depends entirely on the input filter noise bandwidth B and keeps reducing with the reduction of data rate and the E_b/N_o of the received signal.

This reduction in turn results in the increase in N × N and C × N components around the 2ω_c carrier component, which increases the loop acquisition threshold requirement. When the loop encounters very large variations in the input S/N ratios, it is important to improve the output S/N ratios. Such is is the case when the demodulator receives very low data rate signals with low E_b/N_o values.

The fixed noise bandwidth of the very narrow bandpass pre-PLL filter effectively facilitates the noise estimation to the lowest of the data rate in the range. It does this by nearly annulling the squaring effects in the signal yet providing the noise bandwidth scaling for accurate input E_b/N_o measurement.

This circuit configuration has been built and successfully tested for differing data rates and modulation schemes. Sample results pertaining to one QPSK data rate of 105 Mbps are shown in table 1.

The table data consists of various measured values of C/N_o within the bandwidth of the pre-PLL narrow bandpass filter and corresponding calculated values of output S/N ratio and input E_b/N_o ratio using Equations (1) and (2). The measurements are done using an RF QPSK simulator unit to verify and authenticate the results.

The accuracy of the results depends upon the accuracy with which C/N_o measurements are made, within the bandwidth of the pre-PLL narrow band pass filter. If a spectrum analyzer is used for the measurement, its correction factor (required due to noise power bandwidth and log display mode correction) must be determined accurately.

Similarly, the noise bandwidth measurement (other than the 3 dB bandwidth) of both the pre-filter and pre-PLL narrow filter must be precise. Figure 3 consists of plots of the S/N ratio measured at the input of the PLL with different values of the input E_b/N_o ratio, for various data rates. The effect of the excess noise bandwidth of the input filter on lower data rates, in terms of reduced S/N ratio to the PLL can also be assessed from these plots.

It has been shown that the long loop configuration is a viable solution to problems with measurement and quality analysis of suppressed carrier PSK modulated data during a satellite pass. The circuitry (the demodulator) can be interfaced to a station control computer (SCC) at the ground station for the overall automation of the data reception process. The SCC takes the input, on line, from the demodulator in the form of the measured C/N_o ratio at the output of the non-linearity. This provides a continuous online display of the received E_b/N_o.

Other parameters (such as the encountered data rate, modulation scheme, noise bandwidth of the both pre-filter to the non-linearity and narrow bandpass filter etc. required for the calculation) are fed to SCC before the satellite pass. Suitable analog-to-digital (A/D) converters and other minor interfaces are incorporated in the demodulator to make it compatible to SCC.

The method suggested provides distinct advantages and the ability to work in a multi-mission data reception environment. Moreover, with some additional circuitry and some minor PLL design changes, the circuit can also be used to recover the coherent carrier with an improved lock acquisition threshold. The data quality assessment circuitry, hence, becomes the part of the demodulator unit to justify its cost of implementation.

1. D.V. Raju, U.D. Limaye and P.M.Rao, “Designing the Remote Sensing Satellite Ground Station for Multi-mission Operations,” Asia Pacific Microwave Communications Confer-ence (Japan, 1990).

2. P.M.Rao, G. Umadevi and P.K.Jain, “Design of Multifunction Digital Data Demodulator and Bit Synchronizer for Remote Sensing Satellite Data Reception,” IETE Technical Review, Vol. 13 No. 4-5 (July-October 1996), pp. 233-240.

3. “Spectrum Analyzer Measure-ments and Noise: Measuring Noise and Noise-like Digital Communication Signals With a Spectrum Analyzer,” Application Note 1303 (Hewlett-Packard).

4. M.K. Simon, “Tracking Performance of Costas Loop With Hard Limited in Phase Channel,” IEEE Transactions on Communications, Vol. COM, No. 4 (April 1978), pp. 420-432.

5. Floyd M Gardner, “Phase Lock Techniques,” 2nd ed. (John Wiley & Sons: New York, 1979).

6. C.R. Cahn, “Improving Frequency Acquisition of a Costas Loop,” IEEE Transactions on Communications, Vol. COM 25, No. 12 (December 1977), pp. 1453-1459.

7. W.C. Lindsey and M.K. Simon, “Telecommunication System Engin-eering,” (Prentice-Hall: Englewood Cliffs, N.J., 1973).

8. J.W. Layland, “An Optimum Squaring Loop Filter,” IEEE Trans-actions on Communication Technology, Vol. COM 18, No. 5 (October 1970), pp. 695-697.

9. P.M. Rao, P.K. Jain and Padmavathi C.S, “Coherent Carrier Regeneration Using a Long Loop PLL Technique,” Applied Microwave & Wireless (May 1998), pp. 28-42.

10. Tri T Ha, “Digital Satellite Communication,” (McGraw Hill Inter-national Editions: New York, 1990).

P. K. Jain has worked as a scientist/engineer at the national remote sensing agency, Hyderabad, departmemt of space, government of India, for the last 14 years. He recently joined the space application centre, Ahmedabad. His history includes work in the design and development of remote sensing satellite ground station systems such as PSK mod/demodulators, PLL beacon tracking receivers, up/down converters etc. He possesses the degrees of B.E., M.Tech. and is registered for a Ph.D, due to submit his thesis in the next year. Currently, he is working for the establishment of C-band, Ku-band and Ka-band satcom ground stations. He can be reached at pkjainshubhra@yahoo.com

Dr. S. Pal, an alumnus of BITS, Pilani and IISc, Bangalore, is presently the deputy director, digital & communication area, Indian Space Research Organization Satellite Centre, Bangalore. He joined ISRO in 1971, after a brief tenure at TIFR. Dr. Pal is responsible for starting Antenna & Microwave activities in the ISRO Satellite Centre at the inception. He has been a consultant to INMARSAT/ICO (UK) for LEOSAT definitions, responsible for the development of satellite handheld telephone antenna system for ICO (UK) & (ii) ITU for defining Regional African Satellite Communication System. Presently he is the chairman of information infrastructure working group of ministry of information technology. Dr. Pal is a Distinguished Fellow of IETE, Fellow of INAE, INASc, sr. member, IEEE (USA), MAMTA (USA), MASI has received several awards, including NRDC awards for various developments and inventions. Besides, he holds Indian, European & International patents for his various inventions. He has published more than 135 papers in international and national journals of repute. He can be reached at pal_surendra@hotmail.com

V.M.Pandharipande has a B.E. From Nagpur University and an M.Tech and Ph. D fromI.I.T.KHARAGPUR. He Worked as faculty in IIT from 74-83, where he engaged in research and teaching in the area of microwave circuits, phased array antennas and electronically scanned radar. He joined Osmania Univ, Hyderabad as Professor in 1983 where he worked as department head and chairman of the board of studies. He won the best teacher award (Visvesvaraya Award) and Telecommn Award for 2001/2. He can be reaced at vijaympande@yahoo.com.

RSS    Save to Del.icio.us  Digg This