Lth band filter design and performance analysis for wimax application
- 1. Lth Band Filter Design and Performance Analysis for WIMAX Application International Journal of Computer Applications Presented By Ayushi Gagneja
- 2. Abstract Nyquist Filters are generally used for equalization of video signals, are an essential part of communication systems. In this paper two Nyquist filters have been compared and analyzed having L=2 and L=3 and the results are claimed for WiMax applications. The design and analysis have been developed with the help of MATLAB with order 10, sampling frequency 91.392MHz, transition width of 4.75MHz and roll off factor of 0.25. RESULT : It is concluded that Nyquist filter with L=2 is much more cost effective as compared to the Nyquist filter with L=3.
- 3. INTRODUCTION Multirate systems play a major role in communication systems by offering relatively lower sampling rates thereby reducing the order of system and minimizing the hardware requirements, hence reducing the complexity of the product. These filters are powerful structures which concurrently perform digital filtering, spectral translation, interpolation and/or decimation in both non-recursive and recursive structures. Digital conversion is a fast medium but still undergo many losses, so digital filters are used for eliminating noise, shaping spectrum, and reducing inter-symbol interference in communication systems.Their main function is alteration of rate of the discrete-time signals, by different processes. DSP algorithms have contributed to reduce the circuit complexity and cost of products.
- 4. NYQUIST FILTER Nyquist filter is the electronic filter used inTV receivers to equalize the video characteristics. Basic function being equalizing the low and high frequency components of theVF (video signal) signal. Used before demodulation and acts as a low-pass . It plays an essential role in the generation of n bandlimited pulse for minimum inter-symbol interference in wired as well as wireless communication systems. Found its applications mainly as the pulse shaping filter. Also low pass Nyquist filter can be used as the raised cosine Nyquist fir filter. The L-bands are varied for the study i.e. the values are kept as 2 and 3 and hence the performance is analyzed for both in case of WiMAX digital down convertor.
- 5. WIMAX (IEEE802.16) WiMAX (worldwide interoperability for microwave access) is a family of wireless communications standards initially designed to provide enhanced data rates i.e. 30-40 Mbps. It is an emerging technology which states air limitations of fixed broadband wireless access (BWA) systems supporting multimedia services. It can provide Internet access across regions of large boundaries. The stop band attenuation for practical implementations is usually required to in the order of 90 dB, for successful operation in interference prone situations. Demand of high speed internet access let to development of Broadband wireless access (BWA) as an alternative wireless technology to DSL lines and cable modems. Designer of WIMAX system needs to meet some requirements like processing speed, flexibility and time to market.
- 6. DIGITAL DOWN CONVERTOR DDC converts a digitized real signal positioned at an intermediate frequency to a base band complex signal positioned at zero frequency and decimates to a lower sampling rate. It receives the digital IF signal and modifies the signal into a baseband complex signal, resulting in an in-phase signal and also a Quadrature signal as output. It’s design can be implemented using FPGAs. The digital down converter (DDC) performs the reverse function compared to that of a DUC. It converts a signal from the IF band to the base band. It has three sub-parts: 1. A direct digital synthesizer (DDS) 2. A low-pass filter (LPF) 3. A down sampler 0 20 40 60 80 100 120 140 160 -1 -0.5 0 0.5 1 Sample number Original 0 5 10 15 20 25 30 35 40 -1 -0.5 0 0.5 1 Sample number Decimated
- 7. FILTER RESPONSE & CHARACTERISTICS Nyquist filter for L-bands is used for decimation as per the requirement for WiMAX applications. Practical digital communication systems use a roll off factor between 0.1 and 0.5, thus the roll off factor is selected as 0.25. The stop band attenuation is required in approximation of 90 dB while a sharp transition band is set between 4.75 MHz and 5.45 MHz . The order is fixed at 10, sampling frequency is equal to 91.392MHz. The Impulse Response of the filter shows that an alternative zero occurs on both sides of the graph taking Fc=0Hz as the center. The step response of the filter shows variation from positive to negative and then again reach back to positive within the range of 0-40dB.
- 8. FILTER PERFORMANCE COMPARISON Nyquist filters have been developed especially forWIMAX based applications. Digital down conversion is carried out in both the filters with same order of 10. The sampling frequency, roll off factor as well as transition width is kept same for both the Nyquist filters. Impulse response: both results have been brought up on a single plot for comparison. In the Figure, blue plot represents Nyquist filter with L=2 which has zero at every alternate frequency around Fc on both the sides, while the green plot which represents Nyquist filter with L=3 has zero at every second frequency on both the sides around the same Fc. Step response: The blue plot (L=2) has positive-negative-positive relation in 0-40dB range, while in the same range, the green plot (L=3) travels from negative to positive.
- 9. 0 5 10 15 20 25 30 35 40 45 -60 -50 -40 -30 -20 -10 0 Frequency (MHz) Magnitude(dB) Magnitude Response (dB) Nyquist Filter with L=2 Nyquist Filter with L=3 0 10 20 30 40 50 60 70 80 90 100 -0.1 0 0.1 0.2 0.3 0.4 0.5 Time (nseconds) Amplitude Impulse Response Nyquist Filter with L=2 Nyquist Filter with L=3 Magnitude Comparison of Nyquist Filter with L=2 and L=3 Impulse Response Comparison of Nyquist Filters with L=2 and L=3
- 10. 0 10 20 30 40 50 60 70 80 90 100 0 0.2 0.4 0.6 0.8 1 Time (nseconds) Amplitude Step Response Nyquist Filter with L=2 Nyquist Filter with L=3 Step Response Comparison of both Nyquist Filters with L=2 and L=3 Implementation Cost Nyquist Filter with L=2 Nyquist Filter with L=3 No. of Multipliers 7 9 No. of Adders 6 8 No. of States 8 8 Multiplication per Input sample 1.75 2.25 Additions per Input sample 1.5 2 Resource Comparison of both the Nyquist Filters with L=2 and L=3
- 11. CONCLUSION Nyquist filter with L=2 is much more cost effective as compared to the Nyquist filter with L=3 because ▪ It uses less number of multipliers ▪ It has less number of adders ▪ It is faster as it uses less number of additions per sample and also multiplication per sample This Nyquist Filter (L=2) is hence proved to be better than the other in terms of hardware complexity and cost.