Keywords OFDM, GNU Radio, USRP, FPGA, FFT, Wavelet based OFDM

Transcription

1 Volume 3, Issue 6, June 2013 ISSN: X International Journal of Advanced Research in Computer Science and Software Engineering Research Paper Available online at: Performance Analysis of Real Time OFDM Based Communication System Using GNU Radio and USRP Abirami M, Gandhiraj R*, Soman K P Department of Computational Engineering and Networking, *Department of Electronics and Communication Engineering, Amrita Vishwa Vidyapeetham, Coimbatore, , India Abstract In the vast repertoire of communication, the role of OFDM has its own vital importance. In particular, OFDM has been shown as the most effective technique in the field of wireless communication. There have been many platforms such as FPGA, MATLAB etc. in the field to showcase the working and performance of OFDM. In this work, the recently emerging platform- Open source GNU Radio is used to limelight the importance of OFDM communication system. A flexible, cost-effective and reprogrammable hardware such as USRP (Universal Software Radio Peripheral) is required to do a real-time communication experiment. This paper focuses on a real time implementation of OFDM Communication for different modulation schemes with different wavelet techniques. The novel idea behind this work is to portray the importance of OFDM communication using the FFT bases and wavelet bases with the aid of GNU Radio. Keywords OFDM, GNU Radio, USRP, FPGA, FFT, Wavelet based OFDM I. INTRODUCTION A Software Defined Radio (SDR) system is a radio where one or more of the physical layer functions are software defined. The same leads to better operability for switching between different ranges of frequencies. In this system the components that are typically implemented in hardware is re-placed by software [1]. In conventional radio, the system modification is done only by replacing the hardware parts. In general, SDR is mentioned as the software based communication platform in which the parameters can be reconfigured and modified with different functions. This helps the new comers, to define and develop their own modules for execution under GNU Radio Companion (GRC). The GRC platform is a rich Graphical User Interface (GUI) which helps in developing SDR modules in a better and effective way, thus enabling less complex ways of development for programmers. The GUI model of program development was devised based on the idea of giving practical exposure to communication theory like signal generation, signal manipulation, multi-rate concepts, multiplexing schemes and analog and digital modulation schemes. Furthermore, SDR has ability to analyze and visualize the signals that operates and plots the same which can never be done on hardware. The ability of SDR to accept and configure a considerable lot of daughter boards enable it to function on different frequency bands simultaneously with a wide range of operating frequencies. In order to implement SDR, a free and open source software development tool kit known as GNU Radio is available. This toolkit provides a simple drag and drop based signal processing blocks with which many communication experiments can be done. SDR consists of a library developed in C++ for signal processing and a python that act as a glue to tie these blocks together for building and developing SDR. Using SDR, for real-time implementation of communication experiments, much hardware is available. One among that hardware is USRP (Universal Software Radio Peripheral), a hardware module that provides transmission and reception capabilities over a range of frequencies with different set of daughter boards are available [2]. II. GNU RADIO AND USRP A. GNU Radio platform It is an open source software tool kit that enables building of Software Defined Radio. Different functionalities such as modulation, demodulation, filtering, encoding, decoding, source coding, channel coding etc. are provided as software codes [3]. The advantage of implementing functionalities as software modules helps provides re-configurability property to SDR. Traditionally, for example, if a modulation scheme of a radio has to be changed, the entire hardware circuitry employed for modulation has to be changed. GNU Radio provides a graphical user interface with GNU Radio companion (GRC). Experiments are done by connecting signal processing blocks written in C++. The flow graph is written and performed using python. B. USRP The advantage of using SDR is its ability to interface the software with the hardware in real time. USRP N210 hardware is used for both transmission and reception. There are several daughter boards available for USRP, whose operating 2013, IJARCSSE All Rights Reserved Page 1584

2 frequency ranges from DC to GHz. So, based on our requirement real time signal can be captured. The USRP is used to allow general purpose computers to function as high bandwidth software radios. It serves as a digital baseband and IF section for a radio communication system. The basic design principle behind in the USRP has been to do the entire signal processing such as modulation and demodulation, on the host CPU. All of the high-speed general purpose operations like digital up and down conversion, decimation and interpolation are done on the FPGA. III. OFDM (ORTHOGONAL FREQUENCY DIVISION MULTIPLEXING) The modern digital communication system is developed with the aid of Orthogonal Frequency Division Multiplexing (OFDM). The technology has increased with high popularity because it provides high speed data transmission. Recent technologies like Digital Audio Broadcasting (DAB), High Definition Television (HDTV), Wireless Local Area Network (WLAN) etc. also uses OFDM. Conventional OFDM system is implemented with the aid of MATLAB, FPGA toolkit etc. Here the OFDM communication system is designed in GNU Radio platform which is cost effective when compared to conventional system [5]. The major advantage of this system is that, it is mostly focused on the software part in case of any functionalities changes. So, it is easy to understand and utilize in an effective manner can easily check your margins to see if your print area fits within the space allowed. A. OFDM System FFT Bases The block diagram of a typical OFDM system is shown in Figure 1. The main idea behind OFDM is to divide a high-rate encoded data stream into N parallel sub streams that are modulated onto N orthogonal carriers (referred to as subcarriers). This operation is easily implemented in the discrete time domain through an N -point inverse discrete Fourier transform (IDFT) unit and the result is transmitted in a serial manner. At the receiver side, the information is recovered by performing DFT on the received block of signal samples. Figure 1: Block diagram of OFDM system- using FFT bases B. OFDM System Wavelet bases The transceiver of DWT-OFDM is shown in Figure 2. The transmitter part first uses a digital modulator (i.e. QAM, QPSK) which maps the serial bits into symbols. The main task of the transmitter is to perform the discrete wavelet modulation by constructing orthonormal wavelets. Then, the signal is up-sampled and filtered by the approximated coefficients or LPF coefficients. In the receiver side, note that the HPF contains detailed coefficients or wavelet coefficients and then it perform discrete wavelet demodulation by reconstructing the orthonormal wavelet [6]. Different wavelet families have different filter length and values of approximated and detailed coefficients. Both the filters should satisfy orthonormal bases in order to operate as wavelet transform. Figure 2: Block diagram of OFDM system- using WAVELET bases C. Signal to noise ratio (SNR) Signal-to-noise ratio is the ratio of the amplitude of the desired analog or digital data signal to the amplitude of noise in a transmission channel at a specific point in time. SNR is typically expressed in decibels (db). SNR measures the quality 2013, IJARCSSE All Rights Reserved Page 1585

3 of a transmission channel or an audio signal over a network channel. If the ratio is greater, it is easier to identify and subsequently isolate and eliminate the source of noise. If the SNR value indicates as zero means the desired signal is virtually distinct from the unwanted noise. D. Bit Error Rate (BER) Bit Error Rate is the parameter which is used to represent the number of received bit that has been changed due to noise, interference or distortion. It is one of the most important parameter for the communication system. BER measures the whole end to end performance of the given system including the transmitter, receiver and the medium between the two. BER is defined as the rate in which error occur in the OFDM system. IV. EXPERIMENTS AND RESULTS The experimental setup is built to do the real time communication experiments as shown in Figure 3.A separate USRP N210 is used for transmission and reception WBX daughter board is used, which covers the frequency from 50MHz to 2.2GHz. To connect more than one USRP for the same system the D-Link Gigabit Ethernet switch is used.. Figure 3: Real time Experiment setup A. OFDM based communication system using FFT bases simulation This communication system is done by using GNU Radio. OFDM modulation and demodulation block based on FFT bases is used for transmission and reception. For simulation purpose consider the values present in random source block which is passed through the OFDM mod and demod block. Here BPSK modulation scheme is used with FFT length 512 and cyclic prefix length 128. The corresponding transmitter plot and receiver plot is shown in Figure 4. Instead of BPSK some other types of modulation like QAM, PSK, DPSK can also be used. The corresponding transmitter plot and receiver plot is shown in Figure 5 and 6. Figure 4: Simulation block diagram for OFDM communication system using FFT bases 2013, IJARCSSE All Rights Reserved Page 1586

4 Figure 5: Transmitter plot for OFDM system using FFT bases Figure 6: Receiver plot for OFDM system using FFT bases B. Real time OFDM communication system using FFT bases in USRP For real time purpose, consider the USRP sink block at the transmitter and USRP source block at the receiver. Input file is given to the File source block. A data file is given as input to the OFDM modulation block inside file source block. Similarly the same file has to be received at the demodulation block in the receiver side. Set USRP centre frequency as shown in Figure 7. Figure 7: Block diagram for real time OFDM communication system with FFT bases C. Wavelet- Simulation by using GNU Radio In conventional method, wavelet decomposition and reconstruction is done using MATLAB. This paper employs a method using GNU Radio platform for the same. Here two cosine signals are taken with different frequency and both the signals are given to the decimating FIR filter with tap value as haar transform coefficients. The signal is decomposed at n-levels using decimating FIR filter. Then the decomposed signal is passed through the interpolating FIR filter, in order to obtain reconstruction of the signal. The Figure 8 shows, one level decomposition and reconstruction. 2013, IJARCSSE All Rights Reserved Page 1587

5 Figure 8: Simulation block diagram for Wavelet decomposition and reconstruction Figure 9: Plot for the original signal Figure 10: Plot for the reconstructed signal D. Wavelet based OFDM communication system- Simulation by using GNU Radio For wavelet based OFDM communication system, pass the random source blocks via PSK modulation and the modulated signal is passed through Decimating FIR filter with different wavelet coefficients taps such as Haar, db2, db4, db8 which is used for the decomposition purpose. The decomposed signal is given as input to the interpolating FIR filter in order to reconstruct the signal. Then the signal is demodulated using PSK demod block as shown in Figure , IJARCSSE All Rights Reserved Page 1588

6 Figure 11: Simulation block diagram for Wavelet (Haar) based OFDM E. BER Vs SNR results for different modulation schemes for OFDM system- FFT bases For real time BER Vs SNR calculations consider the random source block is passed to the OFDM mod and demod block with different modulation schemes. BER values can be achieved by using a block called Error rate which is given to the number sink where the corresponding bit error values are displayed. In Error rate block, bits per symbol value are given according to the modulation scheme. For example, on considering the modulation scheme such as BPSK, the corresponding bits per symbol is 1, similarly for QPSK it is 2, for QAM16 it is 3, so the bits per symbol value is obtained as power of 2 in accordance with the modulation scheme. By double clicking the OFDM demod block the SNR value can be varied according to our convenience. For different values of SNR the corresponding BER values are achieved with different modulation scheme in real time. Figure 12: BER Vs SNR calculation block in real time using FFT bases Figure 13: BER Vs SNR plot for different modulation schemes in real time 2013, IJARCSSE All Rights Reserved Page 1589

7 V. CONCLUSION AND FUTURE WORK The working and performance of OFDM has been carried out using the most commonly existing platforms such as MATLAB etc. As most of the tools provide only simulation output, the real time OFDM based communication system seems to be a tedious task. To bring out the importance of OFDM communication system in real time, a recently emerging platform- Open source GNU Radio and USRP is used which has been portrayed in this paper. The BER Vs SNR plot value is estimated for different modulation schemes in real time. Simulation part is done for wavelet bases, in OFDM communication system by using a block called decimating and interpolating FIR filter block. The real time wavelet based communication system is possible by creating a block for wavelet decomposition and reconstruction by using C++ and Python codes respectively. REFERENCES [1] Markus Dillinger, Kambiz Madani, and Nancy Alonistioti, "Software-Defined Radio Basics, " review of Software Defined Radio: Architectures, vol. 6, no. 10, [2] Matt Ettus "USRP User s and Developer s Guide," Ettus Research LLC matt@ettus.com. [3] GNU Radio Companion: GNU Radio companion. [4] Mohammed Aboud Kadhim, Widad Ismail Implementation of Wimax IEEE802.16d Baseband Transceiver Based Wavelet OFDM on Multi-Core Software-Defined Radio Platform. European Journal of Scientific Research 2010 [5] Arief Marwanto, Mohd Adib Sarijari, Norsheila Fisal, Sharifah Kamilah Syed Yusof, Rozecha A.Rashid Experimental study of OFDM implementation Utilizing GNU Radio and USRP-SDR International Conference on Communication Malaysia [6] Mohamed H. M. Nerma1, Nidal S. Kamel and Varun Jeoti OFDM Based on Complex wavelet Transform Electrical & Electronic Engineering Department, University Technology PETRONAS. [7] Mohammed Aboud Kadhim, Widad Ismail Implementation of Wimax IEEE802.16d Baseband Transceiver Based Wavelet OFDM on Multi-Core Software-Defined Radio Platform. European Journal of Scientific Research 2010 [8] Nelly M. Shafik Wavelet Transform Effect on MIMO-OFDM System Performance. Journal of Selected Areas in Telecommunications (JSAT), September Edition, 2011 [9] Shravan Sriram, Gunturi Srivasta, Gandhiraj R and Soman K P. "Plug-ins for GNU Radio Companion", International Journal of Computer Applications 52(16):11-16, August Published by Foundation of Computer Science, New York, USA [10] Ranjini Ram, R.Gandhiraj, K.P.Soman, Analog and Digital Modulation Toolkit for Software Defined Radio, Elsevier- Procedia Engineering, Volume 30, 2012, Pages , Dec 7-9, [11] Sruthi M B,Abirami M, Akhil Manikkoth, Gandhiraj R, Soman K P, "Low cost digital transceiver design for Software Defined Radio using RTL-SDR", IEEE International Multi Conference on Automation, Computing, Control, Communication and Compressed Sensing, organised by St. Joseph s College of Engineering and Technology, Palai, Kerala, on March [12] Aavind.H, R.Gandhiraj, K.P.Soman, M.Sabarimalai Manikandan, Rakesh Peter, Spectrum Sensing Implementations for Software Defined Radio in Simulink, Elsevier- Procedia Engineering, Volume 30, 2012, Pages , Dec 7-9, 2011 [13] K.P. Soman, K.L Ramachandran, N.G. Resmi, Insight into Wavelets, Third Edition, Prentice Hall of India, , IJARCSSE All Rights Reserved Page 1590