Optimization of the Weight Processing Algorithm in Multichannel Doppler Filtering

Introduction. The use of non-equidistant pulse sequences as probing radar signals makes it possible to eliminate blind spots in speed and range. However, the implementation of multi-channel Doppler filtering (MDF) based on the classical fast Fourier transform (FFT) algorithm of non-equidistant signal samples in the signal detection problem is associated with energy losses. The use of modified FFT algorithms increases the efficiency of MDF against the background of white Gaussian noise, while reducing the efficiency of signal accumulation in the part of signal processing channels blocked by the narrow-band clutter. To eliminate this drawback, the authors previously proposed using combined classical and modified FFT algorithms. However, the use of the combined method does not lead to an optimal solution in terms of MDF efficiency.

Aim. Optimization of weight processing of non-equidistant signals to improve the efficiency of MDF.

Materials and methods. An MDF synthesis was carried out using optimization procedures, and the effectiveness of the algorithms was assessed using computer calculations.

Results. The results show that the Kaiser Bessel window with a window parameter of 4.42 provides the highest signal-(clutter+noise) ratio improvement coefficient averaged over frequency channels equal to 30.06 dB and the highest probability of correct signal detection averaged over MDF channels equal to 0.5 at processing of non-equidistant pulse sequences. Optimization of the weight processing of MDF under the specified conditions increased the average efficiency characteristics used of up to 53.18 dB and 0.92, respectively.

Conclusion. Separate optimization of weighting processing for each frequency channel can significantly improve the average efficiency characteristics of a multichannel Doppler filter and eliminate all the shortcomings of the classical and modified FFT algorithms when processing non-equidistant pulse sequences. However, these advantages are achieved at the cost of not using the FFT, i.e., implemented within the framework of the discrete Fourier transform (DFT) algorithm.

1. Bakulev P. A. Radiolokatsionnye sistemy [Radar Systems]. Moscow, Radiotekhnika, 2015, 437 p. (In Russ.)

2. Bakulev P. A., Stepin V. M. Metody i ustroistva selektsii dvizhushchikhsya tselei [Methods and Devices of Selection of Moving Targets]. Moscow, Radio i svyaz', 1986, 288 p. (In Russ.)

3. Roy R., Lowenschuss O. Design of MTI Detection Filter with Nonunifoum Interpulse Periods. IEEE Transactions on Circuit Theory. 1970, vol. 17, iss. 4, pp. 604–612. doi: 10.1109/TCT.1970.1083195

4. Thomas H. W., Abram T. M. Stagger Period Selection for Moving-Target Radar. Proc. of the Institution of Electrical Eng. 1976, vol. 123, iss. 3, pp. 195– 199. doi: 10.1049/piee.1976.0045

5. Tang T., Wu C., Elangage J. A. Signal Processing Algorithm of Two-Phase Staggered PRI and Slow Time Signal Integration for MTI Triangular FMCW Multi-Target Tracking Radars. Sensors. 2021, vol. 21, iss. 7, p. 2296. doi: 10.3390/s21072296

6. Novoseltsev L. Ya., Flyagin A. E. Processing of Radar Signals at the Wobble of the Frequency of Repetition of Probing Pulses. Izv. of Higher Educational Institutions. Radioelectronics. 1975, vol. 20, no. 3, pp. 40–45. (In Russ.)

7. Murakami T., Jonson R. S. Clutter Suppression by Use of Weighted Pulse Trains. RCA Review. 1971, vol. 32, no. 3, pp. 402–428.

8. Ispir M., Candan C. On the Design of Staggered Moving Target Indicator Filters. IET Radar Sonar Navig. 2016, vol. 10, iss. 1, pp. 205–215. doi: 10.1049/ietrsn.2015.0175

9. Tuszynski M., Wojtkiewicz A., Klembowski W. Bimodal Clutter MTI Filter for Staggered PRF Radars. IEEE Intern. Conf. on Radar, Arlington, USA, 07–10 May 1990. IEEE, 1990, pp. 176–180. doi: 10.1109/RADAR.1990.201158

10. Bogatov A. D., Kostrov V. V., Tersin V. V. Algorithm for Joint Estimation of the Doppler Frequency and Its Derivative By a Bundle of Non-Equidistant Radio Pulses. J. Radioengineering. 2007, no. 6, pp. 55–59. (In Russ.)

11. Popov D. I. Synthesis and Analysis of Doppler Signal Detectors and Meters. Digital Signal Processing. 2023, no. 2, pp. 32–37. (In Russ.)

12. Doerry A. W. Radar Doppler Processing with Nonuniform PRF. Proc. of the SPIE 10633. Radar Sensor Technology XXII, Orlando, USA, 2018, vol. 10633, p. 19. doi: 10.1117/12.2303453

13. Anju P., Bazil Raj A. A., Shekhar C. Pulse Doppler Processing A Novel Digital Technique. 4th Intern. Conf. on Intelligent Computing and Control Systems (ICICCS), Madurai, India, 13–15 May 2020. IEEE, 2020, pp. 1089–1095. doi: 10.1109/ICICCS48265.2020.9120950

14. Koshelev V. I. Multichannel Doppler Filtering of Radar Signals. J. Radioengineering. 2012, no. 3, pp. 30–35. (In Russ.)

15. Belokurov V. A., Koshelev V. I., Loginov S. N. Realization Algorithms of the Doppler Filtration by the Instrumentality of DSP Analog Device. Voprosy radioelektroniki. 2010, vol. 2, no. 3, pp. 65–76. (In Russ.)

16. Koshelev V. I. Coherent Filtering Nonequidistant Pulse Sequences in Primary Processing of Radar Systems. Successes of Modern Radio Electronics. 2014, no. 10, pp. 16–22. (In Russ.)

17. Koshelev V. I., Trinh N. H. Efficiency of Multichannel Doppler Filtering of Non-Equivalent Pulse Sequences. Digital Signal Processing. 2023, no. 2, pp. 3–8. (In Russ.)

18. Marple Jr. S. L. Digital Spectral Analysis with Applications. New Jersey, Prentice-Hall, 1987, 492 p.