Receiver for QAM Modulation

نویسنده

  • Amit Kumar Dutta
چکیده

In Internet data transmission, the throughput of 45 Mbps is achieved. Now the goal is to increase the data rate further to 100 Mbps and beyond. We designed a Frequency Division Multiple Access system for that and realized it using Q-Dot technology. The principal reason for coherent demodulation in LAN is the size of PLL in QDot circuit. It also allows higher capacity than 1⁄2 achieved so far. 1.0 INTRODUCTION High speed data communication is evolving according to advancement in device technology. Recent demand in research is for data transmission nearing 100 Mb/s rate in LAN, TV cable and fiber optics communication. In this article we find Wideband Frequency Division Multiple Access (WFDMA) as the solution to the problem along with a QAM modulation scheme. In absence of fading this is sufficient to meet the goal of a 100 million bits/second. Recently there is advancement in WCDMA over capacity. This will be possible to achieve if we can implement linear phase analog filter. To achieve higher capacity in WCDMA we need to implement m sequence as the PN sequence and delay locked loop in the receiver. Otherwise the same capacity increase will be possible in differential FDMA transmission (PAM/PSK) with lesser burden on the filter. Moreover FDMA receiver will consume much less power. The organization of the paper is as follows. Section II introduces the topology for higher capacity. Section III and Section IV introduce the Costa’s feed forward circuit to extract carrier for PAM and QAM. In Section V we discuss about the phase locked loop modified structure to recover carrier and data clock. Section VI is about the circuit detail in Q-dot technology. Section VII is about a high speed demodulator. Section VIII is about super heterodyne receiver and we conclude the paper in the next Section. 2.0 ON TOPOLOGY FOR HIGHER CAPACITY Recently it is found that design of receiver becomes possible in case the consecutive frequency bands are not used. This achieves 1⁄4 capacity. This capacity problem is addressed here. In case the transmission is using a LC oscillator, in magnetic field it will be a sinusoidal oscillation. Then the capacity can be doubled that is the capacity is 1⁄2 because of I and Q modulation. If we use a capacitor as an integrator and no sample and hold circuit, we can use all frequency bands at our will neglecting the guard band then we achieve a capacity of 1. The phase locked loop uses a VCO whose output is modulated twice the data rate. The capacity can be further improved using multi level PAM. Then the last improvement comes from the use of QAM modulation to 2. 3.0 CLOCK RECOVERY FOR PAMCOSTA’S LOOP This is a feed forward circuit as explained in Figure 1. We pass the incoming signal through a band pass filter to get rid of the interference from adjacent channels. The PAM signal is given by A(t)cos(wct) where A(t) is the data and cos(wct) is the carrier. Then we square it and get rid of dc component by passing it through a high pass filter. This results into a double of the wc frequency and we get wc term by frequency divider. Thus we get cos(wct+ ) where  is the phase delay associated with generated signal. 4.0 CLOCK RECOVERY FOR QAMCOSTA’S LOOP This is a similar feedforward cicuit as explained in Figure 2. The QAM signal is given by A(t)cos(wct)+B(t)sin(wct). We pass the signal through a bandpass filter and square it. We get cos2(wct)+sin2(wct)+2A(t)B(t)cos(wct)sin(wct) or 1+A(t)B(t)sin(2wct). We get rid of the dc component and square it again. This signal will have a frequency four times of wc frequency. Using frequency divider we get cos(wct+ ). International Journal of Enhanced Research in Science Technology & Engineering, ISSN: 2319-7463 Vol. 4 Issue 4, April-2015, pp: (181-186), Impact Factor: 1.252, Available online at: www.erpublications.com Page | 182 The Costa’s feed forward circuits have the generation of higher frequency and that causes EM radiation. Moreover we need a bandpass filter which is difficult to design. 5.0 PHASE LOCKED LOOP FOR PAM & QAM Costa’s feedforward circuit is simple but requires superheterodyne structure to reduce the higher frequency component. This forces us to look for phase locked loop structure. Unfortunately, the signal A(t)cos(wct) cannot be used to lock on wc frequency, as the error signal (the mixer output) is data dependent. This causes it to be converging or diverging even if we lock to wc by making A(t)=1 at the synchronization time. This made digital transmission almost impossible at high rate and higher frequency. This is circumvented by finding the data edges of the mixer output and setting a frequency divider of VCO output. The VCO output has the same frequency as wc and the data clock phase locks to the data edges. This makes the PLL to phase synchronize and derive the carrier frequency at the same time. Now the range at which PLL will converge will be decided by the signal at the mixer that is A(t)sin(w-wc)t term after low pass filter. If w-wc is lower in compare to A(t) signal rate, then data edge could be found. For QAM signal, the A(t) is kept equal to B(t) during synchronization time. This method of clock recovery for PAM or QAM is most suitable for LAN or TV Cable. This is shown in Figure 3. 6.0 CIRCUIT ELEMENTS The Q-Dot analog PLL circuits will be a reality which could be found in reference [5] or variations of that. We know that Q-Dot oscillator works at very high frequency which means if controlled properly we can realize very high frequency Voltage Controlled Oscillator (VCO) and mixer for local oscillator. At first we look at the high frequency CMOS mixer circuit as we find it in reference [3] which is a Gilbert cell or a voltage switching mixer. We find the Q-Dot mixer circuit as proposed in Figure 4, is a variation of voltage switching mixer. The mixer circuit could be explained as shown in Figure 4, considering a bipolar signal to be demodulated using a bipolar clock signal. If we multiply we get the signal as +,-,-,+ and if frequency matches the integration of the area will be zero. The mixer has four branches, each injecting charge or taking charge out of a filter depending on polarity and magnitude of two signals. The mixer is of two kinds, one using Q-Dot devices for 2 level PAM, the other using QFET devices for multilevel PAM. The QFET devices work in linear region for multilevel PAM. The VCO is a current steered oscillator with switched dc voltage as can be found in [3]. The circuit is a modified version of the Q-Dot VCO first reported in reference [5]. Here we bias the circuit by switching the Vref dc voltage to the oscillating two nodes. When one node is high, the other node is switched to the Vref dc voltage and vice versa. The VCO is shown in figure 5. The filter is a RC filter whose time constants are decided by the speed of the sample and hold circuit. The switch is implemented by HEMT device for high speed or by NMOS for low speed. The resistance value and the capacitance value of sampler is very critical as it should be accurate as well as fast and integrator in behavior . We need a frequency divider circuit and we choose the divide-by-2 circuit using D flip/flop. The D flip/flop is clocked by VCO output and the Q_bar is connected to the Data_in. The Q is connected to the CLOCK of next D flip/flop. Now the D flip/flop is set by the edge of the data and the set signal could the repetitive or missing some pulses sometime. We have to generate same clock period as the data period and continuously repeat it. It gets set repetitively and when synchronized the VCO output frequency will be in phase and same as carrier frequency. 7.0 DEMODULATOR CIRCUIT The demodulator circuit is similar to PLL, except there is no loop here. The signal is taken from the mixer output. This is integrated, compared and latched at the data clock. Because of integration, adjacent channel interference will be attenuated. Hence noise and correlated noise is integrated. We assume that the differential signaling is done at the transmitter. The Figure 6 shows the demodulator, which is true for two levels PAM. It has two mixers for I and Q branches. After multiplication they are low pass filtered. They will be X* cos(wct+ )=1/2[Acos B sin   and X*sin(wct+ )=1/2[Asin  + B cos   . We find  by making A=B. International Journal of Enhanced Research in Science Technology & Engineering, ISSN: 2319-7463 Vol. 4 Issue 4, April-2015, pp: (181-186), Impact Factor: 1.252, Available online at: www.erpublications.com Page | 183 If the carrier is wavering slowly, then there will be a sinusoidal term in the output which we cannot get rid of as A(t) or B(t) is of very high frequency. We find the wavering signal by Costa’s method and vary the Vref accordingly in the comparator of the detector or do a matrix inversion in analog as shown in Figure 10. 8.0 SUPER HETERODYNE RECEIVER We can have a super heterodyne receiver instead of using direct down conversion receiver. The RF front end will be as shown in Figure 9. We do not need any band pass filter. The receiver is shown in Figure 10. We use the same principle to get rid of the frequency variation around a stable operating point of DPLL like it Direct Down Conversion. We put the integrator at the end which works both as LPF and as integrator. 9.0 CONCLUSIONA design principle for high speed data transmission is presented here. If we use 25 MHz channel bandwidth, we maytransmit 100 Mbps, which could be a possibility provided the circuits mentioned here work. We here mentioned a clockrecovery circuit for QAM for level 4, which doubles the capacity from existing one. REFERENCES[1]. R. Hanson, L P Kouwenhoven, J R Petta et al, “Spins in few-electron Quantum Dots,” Physics Review.[2]. R. Gregorian and G. C. Temes, “Analog MOS integrated Circuits for signal processing,” Wiely series on Filter.[3]. P. R. Gray and R. G. Meyer, ”Analysis and Design of Analog Integrated Circuit,” John Wiley and Sons, Inc., 3 rdEdition.[4]. A.S. Sedra, K.C. Smith, “Microelectronic Circuits”.[5]. A.K. Dutta, “UHF RF Transmitter and Receiver,” IJERSTE ISSN 2319-7463 Vol. 3 Issue 11, November 2014. Figure 1. Costa’s feed forward circuit to recover carrier. Figure 2. Costa’s feed forward circuit to recover carrier.SquareHighpassfilterSquareBandpassfilter HighpassfilterFrequencydividerby 4A(t)cos(wct)+B(t)sin(wct) Cos(wct+ )Cos(wct)SquareHighpassfilterFrequencydivider by2BandpassfilterA(t)cos(wct) International Journal of Enhanced Research in Science Technology & Engineering, ISSN: 2319-7463Vol. 4 Issue 4, April-2015, pp: (181-186), Impact Factor: 1.252, Available online at: www.erpublications.com Page | 184Figure 3. Phase locked loop to recover carrier. Figure 4. Mixer using Q-Dot or QFET devices. Figure 5. Voltage Controlled Oscillator. The Vref is connected by Q-dot / QFET switching devices.Vin (Radiofrquencysignal)HEMTQDOTfoVrefVrefVV

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

Phase Noise Jitter Synchronization for Coherent Optical OFDM via Pilot-Data-Aided and Wiener Filter

We investigate a carrier phase jitter synchronization technique for square M-ary quadrature amplitude modulation (M-QAM) coherent optical orthogonal frequency division multiplexing (CO-OFDM) signal employing a unique pilot’s system design, Feed forward maximum likelihood phase estimator as well as Wiener filter-type Minimum Mean square error (MMSE) interpolator. The wiener filter relies upon Ko...

متن کامل

Hardware-Efficient Coherent Digital Receiver Concept With Feedforward Carrier Recovery for -QAM Constellations

This paper presents a novel digital feedforward carrier recovery algorithm for arbitrary -ary quadrature amplitude modulation ( -QAM) constellations in an intradyne coherent optical receiver. The approach does not contain any feedback loop and is therefore highly tolerant against laser phase noise. This is crucial, especially for higher order QAM constellations, which inherently have a smaller ...

متن کامل

Quadrature Amplitude Modulation All Optical Orthogonal Frequency Division Multiplexing-dense Wavelength Division Multiplexing-optical Wireless Communication System under Different Weather Conditions

This paper proposes an analytical model for evaluating the performance of dense wavelength division multiplexing (DWDM) for all optical orthogonal frequency division multiplexing (AO-OFDM) optical wireless channel. The investigated performance for proposed system is evaluated for the parameters  bit error rate (BER) and Q factor .The constellation diagrams, and bit error rate (BER) of the recei...

متن کامل

64-qam Carrier Synchronization by an Extended Kalman Filtering

High Order Quadrature Amplitude Modulation (QAM) is a high-bandwidth-efficient modulation scheme employed in the wireless communications. For this type of modulation, one of the most important problems is carrier recovery to allow a coherent demodulation. To extract correctly the transmitted symbols as the carrier frequency offset between the transmitter and receiver is inevitable, the receiver...

متن کامل

Performance Analysis and Comparison of Zero - forcing SIC and MMSE SIC for MIMO Receivers using BSPK and 16- QAM Modulation methods

This paper investigates the benefits of successive interference cancellation (SIC) based detectors over linear detectors. MIMO receiver with SIC, is simulated using two different modulation schemes namely BPSK and 16QAM modulation methods. The receiver design is less complex in nature. The simulation results obtained at Signal processing laboratory shows that by combining SIC with MMSE or SIC w...

متن کامل

Study of Tranmission Characteristics of 2x2 Mimo System for Different Digital Modulation Using Ofdm and Stbc Multiplexing Techniques and Zf Equalizer Receiver

A detailed analysis of the performance of 2×2 Multiple Input Multiple Output (MIMO) antenna system has been carried out by determining the transmit diversity using Orthogonal Frequency Division Multiplexing (OFDM) techniques. The transmission characteristics are determined for BPSK, QPSK and 16-QAM modulation. Additive White Gaussian Noise (AWGN) has been used presuming flat fading Rayleigh cha...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

عنوان ژورنال:

دوره   شماره 

صفحات  -

تاریخ انتشار 2015