Ultra Wideband Communications Novel Trends System, Architecture and Implementation Part 3 doc

4.2 Comparisons of characteristics for the number of selective RAKE fingers Under CM4 and CM1 environments, receiving performance for the number of RAKE fingers when RR sequence is combi

A Proposal of Received Response Code Sequence in DS/UWB 41

A Proposal of Received Response Code Sequence in DS/UWB 43

A Proposal of Received Response Code Sequence in DS/UWB 45

13RR

Fig 9 BER characteristics of MF reception under the CM1 environment

4.2 Comparisons of characteristics for the number of selective RAKE fingers

Under CM4 and CM1 environments, receiving performance for the number of RAKE fingers when RR sequence is combined with LMS-RAKE reception system [9] is discussed by using the BER characteristics Table 2 shows the specification of simulations 2 Figure 10 shows the BER characteristics when 6RR sequence is used under the CM4 environment Figure 11 shows the BER characteristics when 3RR sequence is used under the CM1 environment In this section the BER characteristics using M sequence also is shows for comparison In each figure, the curve that the number of RAKE fingers is one means that it is the same results with the MF reception

At first, in Figure 10 of the BER characteristics adopting CM4, as the number of RAKE fingers of 6RR sequence and M sequence is increased, it can be confirmed that the BER characteristics are improved And an amount of improvement becomes small as the number

of RAKE fingers of the combined system is increased When the number of RAKE fingers is increased from 10 to 20 in 6RR sequence, the BER characteristics are improved only a little The BER characteristics are saturated On the other hand, when the number of RAKE fingers

is 20 in M sequence, the BER characteristics are not yet saturated Therefore, it is necessary

to increase more the number of RAKE fingers From the above, the number of RAKE fingers

of 6RR sequence has fewer than that of M sequence, so that, the BER characteristics can be improved to a saturated condition In other words, the energy scattering under the multipath environment is captured efficiently by using RR sequence, and the almost part of the scattering energy can be captured with about 10 fingers

Next, the BER characteristics under CM1 environment in Figure 11 show similar with that of Figure 10 Even in the case of M sequence, the property approaching the saturated condition

is shown according to increment of the number of RAKE fingers Additionally when BER characteristics of the case of 20 fingers in 3RR sequence, which is approaching the saturated condition, is compared with that in M sequence, the difference of the performance of 3 [dB] can be obtained, that is, the difference of performance between 3RR sequence and M sequence is shown by using the LMS-RAKE reception method

Consequently, RR sequence has better performances than that of M sequence in the number

of a few RAKE fingers And RR sequence can be approach the saturated condition of the BER characteristics Therefore, a circuit scale in the receiver is reduced by using RR sequence, and a cost of the system can be reduced

Table 2 Specification of simulations 2

A Proposal of Received Response Code Sequence in DS/UWB 47

5 Conclusions

In this chapter, in order to solve the ISI problem caused by the multipath environments, we have proposed the received response sequence (ternary code sequence) in DS/UWB which is generated by using the channel information of the multipath environment, and have shown the generating method By using the proposed sequence, it has been shown that the BER characteristics have been improved greater than that of M sequence in a conventional sequence when the number of pulses has been selected properly And the receiving energy has been captured efficiently even if the number of selective RAKE fingers has been a few Therefore, the circuit scale in the receiver has become small and the cost of the system can be reduced

For further studies, it will be necessary that the effectiveness of the received response is discussed by using a pilot signal which is estimated the channel information in the transmitter practically

6 References

[1] Marubayashi, G.; Nakagawa, M & Kohno, R (1988) Spread Spectrum Communications

and its Applications, The Institute of Electronics, Information and Communication Engineers (IEICE), Corona-sha, May 1998

[2] Tsuzuku, A (1999) OFDM Modulation and Demodulation method, The Journal of The

Institute of Electronics, Information and Communication Engineers (J IEICE), Vol.79,

No.8, pp.831-834, Aug 1999

[3] Kohno, R (2004) Ultra Wideband(UWB) Wireless Technology and Its Contribution in

Future Intelligent Wireless Access, The Journal of The Institute of Electronics, Information and Communication Engineers (J IEICE), Vol.87, No.5, pp396-401, May

2004

[4] Xiao, Z.; Su, L.; Jin, D & Zeng, L (2010) Performance Comparison of RAKE Receivers in

SC-UWB Systems and DS-UWB Systems, The Institute of Electronics, Information and Communication Engineers (IEICE) Trans Communications., Vol.E93-B, No.4, pp.1041-

1044, April 2010

[5] Win, M Z.; Chrisikos, G & Sollenberger, N R (2000) Performance of Rake reception in

dense multipath channlels:implications of spreading bandwidth and selection

diversity order, IEEE JSAC, vol.18, pp.1516-1525, August 2000

[6] Terashima, Y.; Sasaki, S.; Rahman, M A.; Zhou J & Kikuchi, H (2005) A study on Rake

reception for DS-UWB communications, The Institute of Electronics, Information and Communication Engineers (IEICE) Technical Report, WBS2005-3 pp.13-18, June 2005

[7] Rahman, M A.; Sasaki, S.; Zhou J.; Muramatsu, S & Kikuchi, H (2004) Evaluation of

Selective Rake Receiver in Direct Sequence Ultra Wideband Communications in the

Presence of Interference, The Institute of Electronics, Information and Communication Engineers (IEICE) Trans Fundamentals., Vol.E87-A, No.7, pp.1742-1746, July 2004 [8] Foerster, J (2003) Channel modeling sub-committee report final, IEEE P802.15-02/490r1-

SG3a, Feb 2003

[9] Yokota, M & Tachikawa, S (2006) LMS-RAKE Reception in DS/UWB System against

Long Delay-Path Channel, The Institute of Electronics, Information and Communication Engineers (IEICE) General Conference, pp.147, Mar 2006

Genetic Algorithm based Equalizer

for Ultra-Wideband Wireless Communication Systems

Nazmat Surajudeen-Bakinde, Xu Zhu, Jingbo Gao,

Asoke K Nandi and Hai Lin

Department of Electrical Engineering and Electronics, University of Liverpool

United Kingdom

1 Introduction

the Federal Communications Commission (FCC) on an unlicensed basis The ultrawidebandwidth and ultralow transmission power density (-41.25 dBm/MHz for indoorapplications) make UWB technology attractive for high-speed, short-range (e.g.,indoor)

short-range networking is in support of a variety of potential low-cost, low-power multimediatransport applications in home and enterprise environments Typical scenario is provisioningwireless data connectivity between desktop PC and associated peripherals like keyboard,mouse, printer, etc Additional driver applications relates to streaming of digital mediacontent between consumer electronics appliances such as TV sets, VCRs, audio CD/DVD andMP3 players Roy et al (2004)

In an impulse-based DS-UWB system, the transmitted data bit is spread over multipleconsecutive pulses of very low power density and ultra-short duration This introducesresolvable multipath components having differential delays in the order of nanoseconds.Thus, the performance of a DS-UWB system is significantly degraded by the inter-chipinterference (ICI) and inter-symbol interference (ISI) due to multipath propagation Liu &Elmirghani (2007)

In a frequency-selective fading channel, a RAKE receiver can be used to exploit multipathdiversity by combining constructively the monocycles received from the resolvable paths.Maximum ratio combining (MRC)-RAKE is optimum when the disturbance to the desiredsignal is sourced only from additive white Gaussian noise (AWGN), therefore it haslow computational complexity However, the presence of multipath fading, ISI, and/ornarrowband interference (NBI) degrades the system performance severely Sato & Ohtsuki(2005) The maximum likelihood detection (MLD) is optimal in such a frequency selectivechannel environment as UWB channel but its computational complexity grows exponentiallywith the constellation size and the number of RAKE fingers

The high computational complexity of MLD motivates research for suboptimal receivers withreduced complexity such as linear and non-linear equalizers In Kaligineedi & Bhargava(2006), performance of non-linear frequency domain equalization schemes viz decisionfeedback equalization (DFE) and iterative DFE for DS-UWB systems were studied Eslami

et al in Eslami & Dong (2005) presented the performance of joint RAKE and minimum mean

4

square error (MMSE) equalizer receiver for UWB communication systems Parihar et al intwo different papers, Parihar et al (2005) and Parihar et al (2007) gave thorough analysis oflinear and non-linear equalizers for DS-UWB systems considering two different modulationtechniques, binary phase shift keying (BPSK) and 4-ary bi-orthogonal keying (4BOK).Known channel state information (CSI) has been assumed in previous work but practicallythis is not feasible, because the wireless environment is always changing Channel estimation

is of particular importance in future broadband wireless networks since high data-ratetransmissions lead to severe frequency-selective channel fading, which necessitates the use

of channel estimation/equalization techniques to combat significant the ISI Sun & Li (2007).Lots of research work has been done on channel estimation techniques using both the trainingbased and blind approaches In Sato & Ohtsuki (2005), Mielczarek et al (2003) and Chu

Sato & Ohtsuki (2005) used data-aided approach based on using known pilot symbols toestimate the channel impulse response The sliding window (SW) and successive cancellation

a pilot-channel-assisted log-likelihood-ratio selective combining (PCA-LLR-SC) scheme forUWB systems in Chu et al (2008) In another set of data-aided approaches based on maximumlikelihood (ML) scheme, Wang, Xu, Ji & Zhang (2008) proposes a ML approach to channelestimation using a data-aided simplified ML channel estimation algorithm In Lottici et al.(2002), Lottici et al proposed data-aided (DA) and non-data aided (NDA) scenarios based

on the ML criterion Frequency domain channel estimation were reported in Takanashi et al.(2008) where an iterative frequency domain channel estimation technique was proposed formultiple-input multiple-output (MIMO)-UWB communication systems

The genetic algorithm (GA) works on the Darwinian principle of natural selection called

"survival of the fittest" GA possesses an intrinsic flexibility and freedom to choose desirableoptima according to design specifications GA presumes that the potential solution of anyproblem is an individual and can be represented by a set of parameters regarded as thegenes of a chromosome and can be structured by a string of values in binary form Man et al.(1999) GA is a well studied and effective search technique used in lots of work in CDMAcommunication systems as can be found in Erguin & Hacioglu (2000); Yen & Hanzo (2001)and Al-Sawafi (2004) In Erguin & Hacioglu (2000), a hybrid approach that employs a GAand multistage detector for the multiuser detection in CDMA system was proposed Yenand Hanzo in Yen & Hanzo (2001) applied GA as a joint channel estimation and multiusersymbol detection in synchronous CDMA systems A micro GA was developed in Al-Sawafi(2004) as a multiuser detection technique in CDMA system GA has also been applied toUWB communication systems in Gezici et al (2005); Wang et al (2004) and Wang, Yang & Wu(2008) In Gezici et al (2005), a GA-based iterative finger selection scheme, which depends onthe direct evaluation of the objective function, was proposed T.Wang et al in Wang et al (2004)formulated an optimization problem aiming to reduce multiband jam interference power onUWB THSS IR system with 2-PPM which belongs to the class of nonlinear combinationaloptimization UWB pulse design method was carried out in Wang, Yang & Wu (2008) usingthe GA optimization However, to the best of our knowledge, no work has been done, using

GA for channel equalization with pilot-aided channel estimation in DS-UWB communicationsystems

In this chapter, we propose an equalization approach using GA in DS-UWB wirelesscommunication, where GA is combined with a RAKE receiver to combat the ISI due to thefrequency selective nature of UWB channels for high data rate transmission We also compareour proposed RAKE-GA equalization approach with the MMSE based linear equalizationapproach and the optimal MLD approach to demonstrate a trade-off between performance

Genetic Algorithm based Equalizer for Ultra-Wideband Wireless Communication Systems 3

and computational complexity Moreover, we employ a data aided approach to estimate thechannel amplitudes and delays using a sliding window method, which has lower complexitythan ML based channel estimation methods Sato & Ohtsuki (2005)

Simulation results show that the proposed GA based structure significantly outperforms

(BER) performance to the optimal RAKE-MLD approach, while requiring a much lowercomputational complexity The impact of the number of RAKE fingers on the RAKE-GAalgorithm and the speed of convergence in terms of the BER against the number of generationsare investigated by simulation, while the number of training overhead, that is the percentage

of pilot symbols size compared to the number of transmitted data, is also presented with aplot of BER against the number of training symbols

Section 2 is the system model We propose a RAKE-GA equalization approach in Section 3.The data-aided channel estimation for all the receivers are presented in Section 4 Section

5 presents the computational complexity of the RAKE-GA Simulation results are shown inSection 6 Section 7 draws the conclusion

where the transmit pulse vTR(t), is generated by using the ternary orthogonal code sequence

as specified in the IEEE standard ? due to its orthogonality and is of the form given in (2) E c

is the energy per transmitted pulse, d k ∈ {±1} is the k th transmit symbol, T sis the interval

of one symbol or frame time, each frame is subdivided into N cequally spaced chips giving

i=0

where b i ∈ {−1, 0, 1} is the i th component of the spreading code, T c is the chip width, g(t)

represents the transmitted monocycle waveform which is normalized to have unit energy and

N cis the length of the spreading code sequence

2.2 Channel model

According to Molisch & Foerster (2003), a reliable channel model, which captures theimportant characteristics of the channel, is a vital prerequisite for system design Towardthis end, the IEEE 802.15.3a task group has evaluated a number of popular indoor channelmodels to determine which model best fits the important characteristics from realistic channelmeasurements using UWB waveforms The goal of the channel model is to capture themultipath characteristics of typical environments where IEEE 802.15.3a devices are expected

to operate The model should be relatively simple to use in order to allow PHY proposers

to use the model and, in a timely manner, evaluate the performance of their PHY in typicaloperational environments

A log-normal distribution rather than a Rayleigh distribution for the multipath gainmagnitude is used because the log-normal distribution fits the measurement data better In

51

Genetic Algorithm based Equalizer for Ultra-Wideband Wireless Communication Systems