CC BY
39
DIGITAL AGC REFERENCE LEVEL CORRECTION IN A WIDEBAND QUADRATURE AMPLITUDE MODULATED RADIO RECEIVER
DOI 10.24411/2072-8735-2018-10291
Aleksandr A. Prasolov,
Bonch-Bruevich Saint-Petersburg State University of Telecommunications, St. Petersburg, Russia, [email protected]
Keywords: digital AGC, reference level, bit error rate, quadrature amplitude modulation, control signal.
Introduction. Currently, quadrature amplitude modulation is used in a large number of diverse communication systems, including digital television broadcasting, mobile communication systems, as well as in all the other systems with OFDM. When receiving signals with varying amplitude, such as BPSK, DQPSK or QAM, it is necessary to maintain the signal amplitude at the input of the demodulator in certain predetermined limits. Communication systems today are mostly digital. One of the tasks of automatic gain control in digital receivers is to maintain analog signals at a level that prevents saturation of the analog-to-digital converter. Therefore, automatic gain control plays an important role in modern communication systems. Objective. This paper examines the effectiveness of the digital automatic gain control reference level adjustment in wideband QAM receivers. Materials and methods. In this paper, mathematical model of the QAM receiver with two blocks of the automatic gain control unit and the reference level correction is suggested. Results. Using digital automatic gain control reference level adjustment when overloading an analog-to-digital converter allows you to more precisely adjust the signal level. Digital automatic gain control parameters ware estimated. Conclusion. An algorithm which allows to reduce the bit error probability in receivers with QAM due to the digital automatic gain control reference level correction is proposed. The results of this work are relevant in the tasks of developing digital receivers for communication systems for various purposes.
Information about author:
Aleksandr A. Prasolov, Senior Lecturer, Department of Radio Communications and Broadcasting, The Bonch-Bruevich Saint-Petersburg State University of Telecommunications, St. Petersburg, Russia
Для цитирования:
Прасолов А.А. Коррекция опорного уровня цифровой АРУ широкополосного радиоприемника с квадратурной амплитудной модуляцией // T-Comm: Телекоммуникации и транспорт. 2019. Том 13. №7. С. 54-59.
For citation:
Prasolov A.A. (2019). Digital agc reference level correction in a wideband quadrature amplitude modulated radio receiver. T-Comm, vol. 13, no.7, pр. 54-59. (in Russian)
w
Introduction
Automatic gain control (AGC) plays an important role in modern communication systems. The use of automatic gain control allows you to reduce the signal power fluctuation at the radio receiver input Ml and to increase the dynamic range of the received signals ¡2-6]. At present, communication systems are mostly digital ¡7,8]. In digital receivers, an analog signal is usually sampled and quantized using an analog-to-digital converter (ADC) at an intermediate frequency [0]. AGC is used to maintain the amplitude of analog signals at the inverter at an appropriate fixed level to prevent saturation of the ADC [1(1-12]. If the power of the received signals is low, the AGC increases the gain, thereby reducing the quantization noise. If the power of the received signal is high, the AGC reduces the gain to avoid overloading the ADC in particular. The overload time and the magnitude of the quantization noise introduced by the ADC can be minimized by using digital AGC reference level correction.
In systems with QAM the AGC system provides amplification of weak signals after channel filtering to ensure that weak signals have sufficient bit resolution to operate the equalizer [13. 14].
Since the actual gain at the output of the channel liher is unknown, you need to use adaptive methods of obtaining information about the real gain error to control the gain of the AGC. One of the simple estimation methods is to collect information over a longtime interval, estimate the average power and adjust the gain of the amplifier [15,16]. This requires additional memory devices and leads to the high complexity of the hardware implementation. Another method is to compare the average amplitude with a known reference power value [17]. This may lead to additional gain error when setting the reference value manually.
This paper provides a method of building an AGC system for a wideband QAM receiver with reference level adjustment. Conclusions about the quantitative and acceptable parameters of the digital AGC are drawn.
Requirements for AGC in QAM receivers
Traditional AGC systems consist of a battery, low-pass filter, and digital multiplier, as shown in figure I (18]. The output signal rOn is the input signal .v(w) multiplied by the gain factor er The gain is controlled in such a way that, although the value of a-(n) is unknown and varies, the value of i-</7) is constantly kept at a known fixed level. The output signal )in) is accumulated to calculate the signal power, and then fed to a low-pass filter to adjust the appropriate amplifier gain value. However, the disadvantages of the conventional AGC are as follows 118]:
1. In channels with a large time constant, to ensure high accuracy when building a traditional AGC, the battery requires long-term data collection. This procedure is quite time consuming. In addition, the battery must convert the amplitude values to the amount of power and it needs a large amount of memory 119].
2. Since inter symbol interference affects the output digital signal, this will reduce the accuracy of the traditional AGC. In an narrowband channel, intersymbol interference will lead to an excessive estimate of the signal power in the AGC system.
As soon as the signal exceeds the input dynamic range of the ADC, the ADC acts as a hard limiter, creating critical distortions. The intermodulation products become very large. On the other hand, for signals significantly smaller than the full input dynamic range, the level of distortion remains constant and independent of the signal level.
The operation principle of the AGC is that, average level of the output signal is compared against the required (reference) level to produce an error signal; the error signal is then used to change the amplifier gain. These functions can be implemented in both digital and analog form.
A measure of the effectiveness of any AGC system lies in its ability to minimize the change in output signal level, while ensuring its minimal distortion. While maintaining a stable average level, it is desirable that the AGC shout be insensitive to some interferences, for example, pulses.
As shown in [201. a typical QAM receiver must comprise two sets of AGC; one block consists oT a coarse AGC, which can prevent overload ADC and to maintain the signal level near 0 dBFS, and the second block performs fine adjustment level before a demodulator for positioning the constellation diagram.
Block diagram of the digital AGC for the QAM receiver
It is suggested to use a wideband AGC with reference ievel adjustment, proposed by the author in |21,22| (Fig. 2), as a coarse AGC, and an AGC described in [23] (Fig. 3) as a line AGC. figure 4 shows a modei of the wideband digital receiver, comprising: an attenuator at the input, ADC, a wideband AGC with a signal level correction unit [21.22], a DDC block with quadrature component formation, a channel filter, a narrow band AGC [23], and a demodulator [24].
Fig. 2. Block diagram of the DAGC model with tlie reference signal correction unit
Figure 2 shows a block diagram of the DAGC system with a reference signal correction unit, in which the reference signal is sw itched from the calculated value to the minimum value (R„,„ = 0.1) when the reference signal value exceeds the R„„K — 0.636, In [221 it is shown the need to adjust the reference level of the DAGC system under multi-signal impact.
At a multi-signal impact, the adjustment of the reference level leads to a decrease in the control error and allows to reduce the DAGC response time and the ADC overload time by an order of magnitude when the system is overloaded by powerful interferences.
Figure 3 shows a block diagram of a digital AGC for fine tuning the signal level. The main difference from the coarse digital AGC shown in figure 2 is that the transmission coefficient is directly proportional to the control signal amplitude, whereas in the coarse digital AGC it is inversely proportional. This is determined by the difference in the calculation method of the control signal.
For a fine digital AGC. the magnitude of the control signal will be defined as:
A(n+ I) = + for a coarse AGC as:
+ I) = ,4(/j) + -/i) •
For ADC modeling the peak-to-peak level is set to Up.p = 3,4 V, then the reference signal of the broadband AGC unit at the initial moment of lime should correspond to the level 0.636' Ur_p/2 [211 and the parameter a for an exact digital AGC is taken equal to one symbol duration (L/L00), which will allow to obtain a compromise between the accuracy of power estimation and the regulation time. The value of the parameter a for a coarse digital AGC is determined by modeling. Figure 5 shows the dependence of the bit error on the value of the parameter ct depending on the number of estimated symbols: <¿=[1/100... 1/11000]. As can be seen from figure 5, the greatest dependence of the digital AGC accuracy on the parameter a, as expected, can be seen in the case of 256-QAM. To get a compromise between accuracy and speed of operation the value of parameter a is set Lo 1 /300.
ii, relativity of symbol duration Fig. 5. Bil error versus parameter value a
Fig. 3. Block diagram ofdigital AOC
a)
a) The number of errors at 16-QAM
1.S i 29 3 3.5 Frequency.
15 2 2.5 3 3 5 A Frequency, i (Hi)
4.5 5
Fig. 6. a) spectrum of the ADC input signal, b) spectrum of the ADC output signal with the disabled coarse AGC, c) spectrum of ADC output signal with the enabled coarse AGC
Table 1 shows the simulation results for different levels of interference at the system input with 64-QAM input signal, and table 2 - for the influence of two interferences of the same amplitude.
Table 1
Estimate of the number of errors and BER at different levels of interference
Interference The number of errors and BER
level. V w ithout AGC without AGC in AGC without AGC wilh
RF correction correction
! 19« 23 15 13
(DrR-3.3101) (BER-.1.83-10"! (BER-2,5.|ffJ) (UIER=2,0.10i>
3 21(16 327 287 63
(BER-331-Iflr1) (BKR--5.45-l(n (BLR=4.7K-ltr:> (BGll^l.O5-t0-)
j 2000 814 662 123
(BBR-ISMO-1) (BEFMJS'tO-1) (BBR=1.I2-Hr'> (BHR-2.051Cr5)
7 2005 980 W>2 205
(BER=3.34.|0") (BER=fcSM0"') {BER=U5-10'l> (£IEK=3,4 l-K)'")
Q 2(121 tose 745 269
(BF.R'3.36-101) (BER.=I.8M0'') (BER=I 24-Id1) (BFR=4 48-!0':!)
Table 2
Estimate of the number of errors and BER under the influence of two interferences of the same amplitude
Interference The number of errors and It: K
level, V without AGC without AGC in AGC without AGC with
RF correction comtclion
! 2(106 21 38 5
<BERP334I0-'> (BHR=3,5-lirvl (BEK=6.3-in'J) <BKR=5.3I0'|
3 2086 623 613 121
<BKR=3.47!0'') (BER=I 01-10''1 (BER=I,02-10'> (BER=3,01-tff®)
5 2066 1155 1045 244
(BER-3.44-10"1) (BFR-I.92I0") (Br-R=l,74-10") <BFR=4.0M0,1
7 2136 1528 1350 353
(BER=3.5fri0*1") lBER-2.54'10"1! (BLR-2.25-10"1) (BER'5.8SIO:l
9 2223 1741 1567 461
(BER-3.7H0'1) (BLR^.W-IO'1) (BER-2.61-10") <BER=7.6B-15$)
1200 1000 eoo too
400 200
ll ill.
I Without AGC * without AGC in AF >AGC without correction AGC with correction
b) The number of errors at 64-QAM
2500 2000 1500
tooo
500 0
... HI Hi. HI
1 3 5 7 9
■ without AGC «without AGC in RF «AGC without correction AGC with correction
c) The number of errors at 256-QAM
5500 3ooo 2SOO 2000
ll
■ without AGC ■ without AGC In 8F ■ AGC without correction AGC with correction
Fig. 7. The number of errors at different levels of interference on The input
Figure 7a. 7b, 7c show the bit error graphs for different levels of interference at the system input with 16-QAM, 64-QAM and 256-QAM input signals respectively. Figure 8a, 8b, 8c show the bit error graphs for two interferences of the same amplitude at the system input with 16-QAM. 64-QAM and 256-QAM input signals respectively.
As can be seen from figures 7 and 8, until the ADC is overloaded, only the lack of the fine AGC that performs precise level control before the demodulator affects the positioning of the constellation diagram.
It can be seen that the benefit from the use of die reference level correction block does not depend on the QAM positionality, the interference level, and the amount of interferences. and on average bit error rate decrease by a factor of 3.
Y
The number of errors at 16-QAM
iooo
IÏ0O JOWJ КЮ MOT 400 гио 0
Ui
t Wim QUI Л А Г * Without A6C in flf ■ A.*..' svlthiUJt ' '-'ЛIbri - M'jt. voitti f.iriiiflJon
The number of errors at 64-QAM
3500 JOQO HOO HKKl JQO
III
I Wlthöul Al J С ■ 'Л'h I «LI LU. AG С In flf ■ AG С wtthWI LI-I A'L"|., t Hit V..II corrccucm
The number of errors at 256-QAM
Э'.М 3U00 jsno гооо
1Б110 10KJ 500 0
I.I
HI.
• L^itl.uiK • VJ ACL In M • AG L without correction AGC with correction
Irig. 8. The number oTerrors at different levels of two input interferences
Conclusion.
In QAM receivers, it is necessary to use the digital AGC performing precise level control before the demodulation for positioning the constellation diagram. When the ADC is overloaded with powerful interferences, the use of the w ideband AGC reference level correction block will decrease the bit error rate by a factor of 3. The bene lit from using the suggested reference level correction block does not depend on the QAM positionality, the interference level, and the amount of interferences.
References
1. Rappaport. Theodore S, (1996), Wireless communications; principles and practice. Vol. 2. New Jersey: prentice hall PTR,
2. Wang, Jingdian, Xiuhong Lu, and Li Zhang. (2008). "Modeling of a multiple digital automatic gain control system." Tsinghua Science and Technology, vol. 13, no.6, pp. 807-811,
3. Nguyen, Huy-Flieu, et al. (2009). "A binary-weighted switching and reconfiguration-based programmable gain amplifier." IEEE Transactions on Circuits and Systems II; Express Briefs, vol 56, no. 9. pp. 699-703.
4. Garcia-Albcrdi, Coro, et al. (2013). "Micropower class-AB VGA with gain-in dependent bandwidth." IEEE Transactions on Circuits mid Systems It: Express Briefs, vol. 60, no. 7, pp. 397-401.
5. Liu, Chang, el al. (2012). "Л 5-Gb/s automatic gain control amplifier with temperature compensation." IEEE Journal of Solid-Slate Circuits, vol. 47, no. 6, pp. 1323-1333.
6. Choi, Inyoung, Heesong Seo, and Bumman Kim, (2012). ".Accurate dВ-linear variable gain amplifier with gain error compensation," IEEE Journal of Solid-State Circuits, vol.48, no.2, pp. 456-464.
7. Ariwa, Ezendu, and Eyas El-Qawasmeh, eds. (2011 ). Digital Enterprise and Information Systems: International Conference, DEIS 2011, London, UK July 20-22. 2011, Proceedings. Vol. 194. Springer Science & Business Media.
8. Sigismondi, Paolo, (201 i). The digital glocalization of entertainment: New paradigms in the 21st century global mediascape, Vol, 3. Springer Science & Business Media.
9. Narieda. Shusuke. (2013). "AGC and ADC effects on receiver performance in FDM based narrowband wireless systems." 2013 IEEE 10th Consumer Communications and Networking Conference (CCNC).
10. Whitlow, Dana, (2003), "Design and operation of automatic gain control loops for receivers in modern communications systems," Microwave Journal, vol. 46, no. 5, pp. 254-256.
11 Khoury, John M, (1998), "On the design of constant settling time AGC circuits." IEEE Transactions on Circuits and Systems II; Analog and Digital Signal Processing, vol. 45, no. 3, pp. 283-294.
12. Pan, Hsuan-Yu Marcus, and Lawrence E, Larson. (2007), "Improved dynamic model of fast-settling iinear-in-dB automatic gain control circuit." 2007 IEEE International Symposium on Circuits and Systems. IEEE, pp. 681-684,
13. Tan. Loke Kun. et al. (1998). "A 70-Mb/s variable-rate 1024-QAM cable receiver 1С with integrated 10-b ADC and FEC decoder," IEEE Journal of Solid-Stare Circuits, vol, 33 no. 12, pp. 2205-2218.
14. Zhang, Yongxue. et al. (2003). "Practical implementation of blind equalization carrier recovery and liming recovery for QAM cable receiver chip." Proceedings, of 5th International Conference on ASIC. pp. 886-S89.
15. Green, D. (1983). "Global stability analysis of automatic gain control circuits." IEEE transactions on circuits and systems, vol. 30, no. 2, pp. 78-83.
16. Wang, Chorng-Kuang, and Po-Chiun Huang. (1997). "An automatic gain control architecture for SONET OC-3 VLSI." IEEE Transactions on Circuits and Systems II: Analog and Digital Signal Processing, vol. 44. no. 9, pp. 779-783.
17. Rhodes, Charles W. (1992). "Measuring peak and average power of digitally modulated advanced television systems," IEEE transactions on broadcasting, vol. 38, no, 4, pp. 197-201.
18. Deng, Qing, et al. (2007). "A novel AGC scheme for QAM demodulator applications." 2007 9th International Symposium on Signal Processing and Its Applications. IEEFJ, pp. 1-4,
19. Nam. Hyoungsik, et al. (2004). "A 300-mW programmable QAM transceiver for VDSL applications." IEICE transactions on electronics. vol. 87 no, 8, pp. 1367-1375,
20. Brand. S., and Philips Semiconductors, "QAM Demodulation," Wireless Communications.
21. Prasolov A. A., Shpak S. (2013). A. Correction of the reference level of digital automatic gain control with a multi-signal effect (in Russian). Actual problems of information and telecommunications in science and education, pp 315-320.
22. Prasolov, A. (2018). "Modeling of digital AGC with multisignal impact and adaptation of the reference level," 2018 Moscow Workshop on Electronic and Networking Technologies (MWENT). 1F.EE, pp. I-4.
23. Prasolov A. A., Shpak S. (2012). A. Modeling the transient process of digital automatic gain control. Control systems and information technologies, no, 4. Vol. !. No. 50. pp. 197-200,
24. Harada. Hiroshi, and Ramjee Prasad. (2002). Simulation and software radio for mobile communications. Artech House.
7Тл
Y
КОРРЕКЦИЯ ОПОРНОГО УРОВНЯ ЦИФРОВОЙ АРУ ШИРОКОПОЛОСНОГО РАДИОПРИЕМНИКА
С КВАДРАТУРНОЙ АМПЛИТУДНОЙ МОДУЛЯЦИЕЙ
Прасолов Александр Александрович, Санкт-Петербургский государственный университет телекоммуникаций им. проф. М.А.
Бонч-Бруевича, г. Санкт-Петербург, Россия, [email protected]
Аннотация
Введение. В настоящее время, квадратурная амплитудная модуляция используется в большом количестве разнообразных систем связи, в том числе в цифровом телевизионном вещании, мобильных системах связи, а также в системах с OFDM. При приеме сигналов с изменяющейся амплитудой, таких как BPSK, DQPSK или QAM, возникает необходимость поддерживать амплитуду сигнала на входе демодулятора в определенных, заранее заданных пределах. Системы связи на сегодняшний день в основном являются цифровыми. Одной из задач автоматической регулировки усиления в цифровых приемниках является поддержание аналоговых сигналов на уровне, не допускающем насыщения аналого-цифрового преобразователя. Поэтому автоматическая регулировка усиления играет важную роль в современных системы связи. Цель работы. Целью данного исследования является анализ эффективности подстройки опорного уровня цифровой автоматической регулировки усиления в широкополосных приемниках с QAM. Материалы и методы. В рамках исследования разработана математическая модель приемника QAM с двумя блоками автоматической регулировки усиления и блоком коррекции опорного уровня. Результаты. Показана эффективность подстройки опорного уровня цифровой автоматической регулировки усиления при перегрузке аналого-цифрового преобразователя. Произведена оценка параметров цифровой автоматической регулировки усиления. Заключение. В результате данного исследования предложен алгоритм позволяющий уменьшить вероятность битовой ошибки в приемниках с QAM за счет коррекции опорного уровня цифровой автоматической регулировки усиления. Результаты этой работы актуальны в задачах разработки цифровых приемников для систем связи различного назначения.
Ключевые слова: цифровая АРУ, опорный уровень, битовая ошибка, квадратурная амплитудная модуляция, управляющий сигнал. Литература
1. Rappaport T. S. et al. Wireless communications: principles and practice. New Jersey: prentice hall PTR, 1996. Т. 2.
2. Wang J., Lu X., Zhang L. Modeling of a multiple digital automatic gain control system // Tsinghua Science and Technology. 2008. Т. 1 3. №. 6. С. 807-811.
3. Nguyen H. H. et al. A binary-weighted switching and reconfiguration-based programmable gain amplifier // IEEE Transactions on Circuits and Systems II: Express Briefs. 2009. Т. 56. №. 9. С. 699-703.
4. Garcia-Alberdi C. et al. Micropower class-AB VGA with gain-independent bandwidth // IEEE Transactions on Circuits and Systems II: Express Briefs. 2013. Т. 60. №. 7. С. 397-401.
5. Liu C. et al. A 5-Gb/s automatic gain control amplifier with temperature compensation // IEEE Journal of Solid-State Circuits. 2012. Т. 47. №. 6. С. 1323-1333.
6. Choi I., Seo H., Kim B. Accurate dB-linear variable gain amplifier with gain error compensation // IEEE Journal of Solid-State Circuits. 2012. Т. 48. №. 2. С. 456-464.
7. El-Qawasmeh E. A. E. Digital Enterprise and Information Systems // International Conference. 2011.
8. Sigismondi P. The digital glocalization of entertainment: New paradigms in the 21st century global mediascape. Springer Science & Business Media, 2011. Т. 3.
9. Narieda S. AGC and ADC effects on receiver performance in FDM based narrowband wireless systems // 2013 IEEE 10th Consumer Communications and Networking Conference (CCNC). IEEE, 2013. С. 449-454.
10. Whitlow D. Design and operation of automatic gain control loops for receivers in modern communications systems // Microwave Journal. 2003. Т. 46. №. 5. С. 254-256.
11. KhouryJ. M. On the design of constant settling time AGC circuits // IEEE Transactions on Circuits and Systems II: Analog and Digital Signal Processing. 1998. Т. 45. №. 3. С. 283-294.
12. Pan H. Y. M., Larson L. E. Improved dynamic model of fast-settling linear-in-dB automatic gain control circuit // 2007 IEEE International Symposium on Circuits and Systems. IEEE, 2007. С. 681-684.
13. Tan L. K. et al. A 70-Mb/s variable-rate 1024-QAM cable receiver IC with integrated 10-b ADC and FEC decoder // IEEE Journal of Solid-State Circuits. 1998. Т. 33. №. 12. С. 2205-2218.
14. Zhang Y. et al. Practical implementation of blind equalization carrier recovery and timing recovery for QAM cable receiver chip // Proceedings. of 5th International Conference on ASIC. 2003. С. 886-889.
15. Green D. Global stability analysis of automatic gain control circuits // IEEE transactions on circuits and systems. 1983. Т. 30. №. 2. С. 78-83.
16. Wang C. K., Huang P. C. An automatic gain control architecture for SONET OC-3 VLSI // IEEE Transactions on Circuits and Systems II: Analog and Digital Signal Processing. 1997. Т. 44. №. 9. С. 779-783.
17. Rhodes C. W. Measuring peak and average power of digitally modulated advanced television systems // IEEE transactions on broadcasting. 1992. Т. 38. №. 4. С. 197-201.
18. Deng Q. et al. A novel AGC scheme for QAM demodulator applications // 2007 9th International Symposium on Signal Processing and Its Applications. IEEE, 2007. С. 1-4.
19. Nam H. et al. A 300-mW programmable QAM transceiver for VDSL applications // IEICE transactions on electronics. 2004. Т. 87. №. 8. С. 1367-1375.
20. Brand S., Semiconductors P. QAM Demodulation // Wireless Communications.
21. Прасолов А. А., Шпак С. А. Коррекция опорного уровня цифровой автоматической регулировки усиления при многосигнальном воздействии //А ктуальные проблемы инфотелекоммуникаций в науке и образовании. 2013. С. 315-320.
22. Prasolov A. Modeling of digital AGC with multi-signal impact and adaptation of the reference level // 2018 Moscow Workshop on Electronic and Networking Technologies (MWENT). IEEE, 2018. С. 1-4.
23. Прасолов А. А., Шпак С. А. Моделирование переходного процесса цифровой автоматической регулировки усиления //Системы управления и информационные технологии, № 4. 2012. Т. 1. №. 50. С. 197.
24. Harada H., Prasad R. Simulation and software radio for mobile communications. Artech House, 2002. Информация об авторе:
Прасолов Александр Александрович, Санкт-Петербургский государственный университет телекоммуникаций им. проф. М.А. Бонч-Бруевича, старший преподаватель, г. Санкт-Петербург, Россия
