# High Data Rate Mobile Communication Applications Biology Essay

Published:

This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.

Mobile WiMAX is a promising standard for high data rate mobile communication applications. As receiver is mobile the multipath fading and doppler frequency shift of carrier are major factors which degrade the performance of the system. Hence to mitigate impacts of these two effects, we need to perform effective channel estimation at receiver side. In this paper we have shown comparative results of mobile WiMAX Downlink system with and without consideration of channel estimation block. The pilot arrangements employed is based on Downlink Partial Usage of Subchannels (DL-PUSC) method. We have performed linear interpolation based on low complexity LS channel estimation with QPSK and 16-QAM under the consideration of different Doppler shifts (100Hz, 200Hz and 300Hz) in rayleigh fading channel and compared the results for the same with respect to BER performance parameter. The simulation results shows that 6-7dB SNR difference is noted when Doppler shift changes 100Hz to 300Hz with increased mobility of receiver. Also the comparative performance of QPSK and 16-QAM modulation schemes for LS-Based Channel Estimation are analyzed and shown that QPSK requires 4-5dB less SNR to achieve same BER as 16-QAM.

Keywords: Mobile WiMAX, Channel Estimation, DL-PUSC, Doppler frequency.

Received:; Revised Accepted.

*Corresponding Author

Satish Patel

Department of Electronics and Communication Engg., Sarvajanik College of Engineering and Technology,Surat, Gujarat, India

Email: satish.sdp@gmail.com

INTRODUCTION

WIRELESS metropolitan area network (Wireless MAN) or worldwide interoperability for microwave access (WiMAX), which is defined in IEEE Std. 802.16d/e (Prashad et al., 2004, Phoung et al., 2010), is technology that provides wireless access for both fixed and mobile users. This system is also referred as Mobile WiMAX which uses orthogonal frequency division multiple access technique as a modulation method. This technique is adopted from the powerful orthogonal frequency division multiplexing (OFDM) which effectively mitigates the impairment of the time-variant frequency selective fading channel (Hwang et al., 2009, Tolochko et al., 2005). A baseband OFDM system is shown in Figure-1.

In practice, the relative motion between the base and mobile station results in time variations of the channel and high speed movement causes large Doppler shift and rapid fluctuation of channel. Also, channel is frequency selective and noise is added, so it becomes rather difficult to estimate channel information. It can be assumed that while intersymbol interference (ISI) is negligible due to insert guard interval, intercarrier interference (ICI) caused by Doppler shift cannot be reduced. It is difficult to estimate channel transfer function, so it is preferable to estimate channel information based on pilots in practice. The most commonly used pilot pattern for mobile WiMAX is cluster based pilot allocation method according DL-PUSC subchannelization (2006).

There are many channel estimation techniques and they are classified according to Block type and Comb type pilot insertion techniques (Morelli et al., 2001, Coleri et al., 2002). Block type pilot pattern is used for slow fading channel while comb type pilot pattern is used for fast fading channel. Mobile WiMAX employs 2D channel estimation by interpolation method. The most common techniques for channel estimation are LS, MMSE, LMMSE etc (Yu et al., 2012, Yucek et al., 2007, Mohammad, et al., 2008) which are basically block type pilot channel estimations. And different interpolation techniques are constant, linear, spline cubic, low pass interpolation etc[6] which are comb type pilot estimations

In this paper we have performed linear interpolation based on LS estimator considering effect of different doppler frequency shifts and compared the BER performance. We have used pilot insertion according to DL-PUSC subchannelization.

The remainder of this paper is organized as: system model, shannel estimation method, simulation results and Discussions on results.

## SYSTEM DESCRIPTION

The data bits provided from the source are converted from serial to parallel to form parallel data of some subchannels. Each parallel subchannel modulated to complex QAM symbols of Nu active subcarriers. The modulated data with other null carrier as guard band and DC form N subcarriers. This data sequence of length N {X(k)} are then fed into IDFT block symbol by symbol to transform them into time domain and generate an OFDM signal {x(n)} with the following equation:

x(n)= IDFT {X(k)} =..........................................(1)

Where N is the DFT length or the number of subcarriers.

Figure : Baseband OFDM System

To prevent inter-symbol interference (ISI), a cyclic prefix of Ng samples is inserted at the beginning of every symbol. After D/A conversion, the signal is transmitted through the frequency selective time varying fading channel with additive noise.

Assumed that the impulse response of the multipath fading channel is given by:

h(t,) = ......................................................(2)

Where hr(t) and τr are the gain and delay of the rth path, respectively. The path gains h(t) are wide sense stationary (WSS) narrow-band complex Gaussian process and are mutually independent. The received signal, which has been corrupted by the multipath fading channel and contaminated by the additive white Gaussian noise can be formulated as:

y( = ..............................................(3)

Where is the continuous-time representation of the transmitted discrete-time signal, x(n). The received continuous time signal then convert back to a discrete time signal y(n), the receiver do synchronization, downsampling, and removes the cyclic prefix. The simplified baseband model of the received samples takes the form of:

Y(n)=.................................................(4)

Where L is the number of sample-spaced channel taps, w(n) is additive white Gaussian noise (AWGN) sample with zero mean and variance of σn2 and h(l) is the time domain channel impulse response (CIR) for the current OFDM symbol. It is assumed that time and frequency synchronization is perfect.

FFT transforms y(n) to the frequency domain received base band data:

Y(k)= FFT(y(n)) = X(k)H(k) + W(k) ................................................(5)

Where H and W are FFT of h and w respectively.

Following FFT block, the pilot signals are extracted and the Channel Estimation is carried out to obtain estimated channel response Ä¤(k) for the data sub-channels. Then the transmitted data is estimated by equalization process:

.................................................................................(6)

After signal demapping, the source binary information data are re-constructed at the receiver output.

## CHANNEL ESTIMATION

Using LS estimate the channel impulse can be calculated received symbols and known pilot symbols as follow (Mohammad et al., 2008, Babapour et al., 2010):

Ä¤ p= Xp-1Y............................................................................................. (7)

Where Xp is known pilot symbol and Y is the received symbol.

The advantages of the LS estimate are lower complexity and implemented easily without knowing the channel statistics compared to advance algorithm: Minimum Mean Square Error (MMSE) (Galin et al., 2004).

We have performed the channel estimation based on Downlink Partially Used Subchannelization (DL-PUSC). According to this concept the carriers are divided in clusters. Each cluster having 14 subcarriers out of which two are allocated for the pilot symbols and others are for data symbols (2006). Figure-2 shows the cluster structure and the position of pilot symbols.

Figure 2 : DL-PUSC clusters

After this allocation the channel estimation is performed in three steps which are illustrated in figure 3. Three steps can be listed as:

Step 1: LS estimate at pilot position

Step 2: Linear interpolation on time axis

Step 3: Linear interpolation on frequency axis

Figure 3: Channel estimation in WiMAX DL-PUSC

Figure-4 can be treated as step-1 for the channel estimation. Dark blue blocks can be considered as CIR at pilot position using equation-7

Figure 4: LS estimate at pilot position

The linear interpolation method is used to calculate channel response at kth subcarrier as (Galin et al., 2009)

Ä¤k(k)= Ä¤ (mL+l)

= Ä¤ p(m)+l/L(Ä¤ p(m+1) - Ä¤ p(m)), 0l L..........................(8)

Where m=0,1, ..Np-1, Np= number of pilots, L=N/Np, where N= total number of subcarriers.

Using equation-8 we have applied linear interpolation on time axis and frequency axis to get estimated values of CIR at remaining all data subcarriers. Time and frequency axis interpolation is illustrated in figure-5 and figure-6 respectively.

Figure 5: Linear interpolation on time axis

Figure 6: Linear interpolation on frequency axis

## SIMULATION

## Description of Simulation

The simulations have been made for WiMAX (IEEE 802.16e standard) to see system performance for Rayleigh Channel with and without channel estimation block considering effect of multipath fading and Doppler shift of carrier frequency as receiver is mobile. The system performance is analyzed from BER versus SNR plot and also effect of different Doppler shift analyzed on performance of system. The Mobile WiMAX Simulation parameter is listed in Table 1 (2006):

Table 1: Simulation Parameters

## Parameters Value

## Parameters Value

FFT size (NFFT)

1024

Bandwidth

10MHz

Modulation

QPSK, 16-QAM

Number of Guard Subcarriers

91+92

Number of Used Subcarriers (Nused) including all possible allocated pilots and DC subcarrier.

841

Cyclic Prefix ratio

1/8

Fading channel

Rayleigh

Multipath(LOS+NLOS)

3

Doppler Frequency Shift

100Hz,200Hz,300Hz

Carrier Frequency

3.5GHz

## Simulation Results

A Rayleigh fading channel has been simulated and the data is passed through it, followed by addition of AWGN noise. The various simulation parameters are shown in Table-1.

Figure 6: BER performance of the analyzed estimation algorithm

(QPSK , Fd=100,200,300Hz).

Figure 7: Comparison of BER performances of No Channel Estimation ,

LS Based Channel Estimation and Perfect channel (QPSK, Fd=100,200,300Hz).

Figure-6 shows the BER performance of mobile WiMAX using LS based linear interpolation channel estimation algorithms with QPSK modulation for 100Hz, 200Hz and 300Hz doppler shift corresponding to mobility speed of receiver 30kmph, 60kmph and 90kmph. Its clear that analyzed estimation algorithm perform well in all three cases but degradation in performance of system is noted with increase in mobility speed. Similar analysis is done for 16-QAM modulation scheme which is shown in figure-8.

The comparative analyses of BER performance of the Linear interpolation, no channel estimation and perfect channel conditions is shown in figure-7 with QPSK modulation. Similar comparative analysis is shown in figure-9 for 16-QAM. In both cases it can be seen that without channel estimation the BER is as high as around 10-0.6 -10-0.9 dB which is not suitable for the practical application. Hence channel estimation is essential for the practical system implementation.

QPSK and QAM are the two leading modulation schemes for WiMAX. In general the greater the number of bits transmitted per symbol, the higher the data rate is for a given bandwidth. Thus, when very high data rates are required for a given bandwidth, higher-order QAM systems, such as 16-QAM is used. Finally performance of QPSK and 16-QAM are compared for 100Hz doppler shift which is shown in figure-11. This analysis shows that QPSK perform better than 16-QAM which shows QPSK is more tolerant of interference than either 16-QAM. For this reason, where

Figure 8: BER performance of the analyzed estimation algorithm

(16-QAM, Fd=100,200,300Hz).

Figure 9: Comparison of BER performances of No Channel Estimation , LS Based Channel Estimation and Perfect channel (16-QAM, Fd=100,200,300Hz).

Figure 10: BER performance of analyzed estimation QPSK vs 16-QAM ( Fd=100Hz)

signals are expected to be resistant to noise and other impairments over long transmission distances, QPSK is the normal choice.

## DISCUSSION

In this paper, a review of LS channel estimation along with linear interpolation and DL-PUSC frame structure is described. By considering the Mobile WiMAX parameters simulation is done on MATLAB tool for OFDM system and BER performance is analyzed for Rayleigh fading channel with and without channel estimation block considering QPSK and 16-QAM modulation schemes. The simulation result shows that in case of no channel estimation the BER attains high value (10-0.7 to 10-0.9) which cannot be suitable for practical implementation of the WiMAX system. On other hand BER is achieved with linear interpolation based on LS estimate is acceptable with low complexity design of estimator. In addition, the effect of different mobility speed of receiver is analyzed on BER performance by varying the doppler shift of the carrier frequency. Simulation results show that as the Doppler shift gets higher the BER performance degrades i.e. 6-7dB SNR difference can be seen for between Doppler shifts of 100Hz and 300Hz. Also the performance of QPSK and 16-QAM are compared and conclude that QPSK is less susceptible to ISI and noise but 16-QAM is advantageous when higher data rate is requirement.