RHealthCare SP100 Spezifikationen

viii Contents7.2 Signal and channel models with channel estimation errors 1707.2.1 Signal and channel model 1707.2.2 Estimation errors of channel para

84 Channel estimation for high-rate systems0 5 10 15 2044.555.56SNR (dB)Rmse (dB)CM1CM2−4 −2 0 2 4 6 844.555.56SNR (dB)Rmse (dB)CM3CM4Figure 3.6 Analy

3.3 Impact of channel estimation error on performance 85Table 3.2 Required computational complexity for CFR estimation per subband in a frame (after [

86 Channel estimation for high-rate systemsmultistage estimators on the system performance by comparing their resulting averageSERs and FERs.The perfo

3.3 Impact of channel estimation error on performance 87−4 −2 0 2 4 6 8 10 12 14 1610−410−310−210−1SNR (dB)SERLSMLMultistageKnown channelFigure 3.7 An

88 Channel estimation for high-rate systems−5 0 5 10 151234(a) CM1SERMSE−5 0 5 10 151234(b) CM2SERMSE−5 0 511.522.533.5(c) CM3SNR (dB) Performance Ga

3.3 Impact of channel estimation error on performance 89−4.5 −4 −3.5 −3 −2.5 −2 −1.5 −1 −0.510−210−1100SNR (dB)FER(a) CM1 and CM2LS (2 Symbols)Multist

90 Channel estimation for high-rate systemscomplexity similar to that of the conventional LS CFR estimator in an OFDM-UWBsystem. Overall, compared wit

References 91[15] J. van de Beek, O. Edfors, M. Sandell, S. K. Wilson, and P. O. B¨orjesson, “On channelestimation in OFDM system,” in Proc. IEEE Veh.

92 Channel estimation for high-rate systems[33] O.Edfors,M.Sandell,J.vandeBeek,S.K.Wilson,andP.O.B¨orjesson, “OFDM channelestimation by singular value

4 Adaptive modulation and coding forhigh-rate systemsRuonan Zhang and Lin CaiAs wireless channels are fading and error-prone in nature, the adaptive m

Contents ix9.4 Improving WPAN’s reliability under interference:dynamic channel selection 2619.4.1 Algorithm description 2619.4.2 Simulation results 26

94 Adaptive modulation and coding for high-rate systemsFigure 4.1 Wireless transmission system with joint AMC and ARQ.4.1 Adaptive modulation and codi

4.2 AMC in MB-OFDM systems 95First, the premier requirement for AMC is the feedback path, which can be usedby the receiver to inform the transmitter o

96 Adaptive modulation and coding for high-rate systemsTable 4.1 Transmission mode implementation in MB-OFDM [10].Coded bits / Information bits /Data

4.3 WPAN link architecture in ECMA-368 97Figure 4.3 MAS reservation in a superframe: (a) MAS reservation in a superframe; (b) the timingof burst trans

98 Adaptive modulation and coding for high-rate systemsslots (MASs). An MAS lasts for 256 μs and is the minimum time unit for reservation.Each superfr

4.4 Packet-level model for UWB channels with shadowing 99TxRxDobstructingpositionxr1θ2θy3θ4θFigure 4.4 The modeling of the body shadowing effect (c 2

100 Adaptive modulation and coding for high-rate systemsX coordinate of the personY coordinate of the person−4.5−4.5−4.5−4.5−3.5−3.5−3.5−3.5−3−3−3−3−3

4.4 Packet-level model for UWB channels with shadowing 1012S1S1λ2μKS...Kμ1Kλ−2λ3μFigure 4.6 FSMC model for UWB channels with shadowing process.4.4.2

102 Adaptive modulation and coding for high-rate systemsSecond, we approximate that the time the person stays inside a zone is exponentiallydistribute

4.5 WPAN link performance analysis 103Figure 4.7 Embedded Markov chain model (c 2010 IEEE) [19, 20].model can capture the MAC protocol scheduling, ch

x Contents12.5 Limited feedback centralized relay selection 33712.5.1 Outage probability and effective rate 33912.5.2 DMT analysis 34112.6 Summary 343

104 Adaptive modulation and coding for high-rate systems2. Channel state transition Because the channel variation is caused by the mobilityof pedestri

4.5 WPAN link performance analysis 105The PMF of the random variable bt− dtcan be obtained asfbt−dt(x|nt, kt, kt+1, qt) =F−qty=0fbt(y|nt, qt) fdt(y −

106 Adaptive modulation and coding for high-rate systemsTable 4.2 Transmission modes and channel modelChannel TM SNR interval Transition rate Transiti

4.6 Simulation results 1070 10 20 30 40 50 60 70 80 900.10.20.30.40.50.60.70.80.911.1Queue Length (KB)CDF of Queue Length Distribution Max FER = 0.02

108 Adaptive modulation and coding for high-rate systems20 30 40 50 60 70 80 9044.555.5Buffer Size (KB)Throughput (Mbps) Max FER = 0.02 (Analytical)M

4.7 AMC in 60 GHz millimeter-wave radio systems 109the AMC is important for 60 GHz systems to combat the channel fading and enhancetransmission reliab

110 Adaptive modulation and coding for high-rate systemsCoordination of devices within the radio range is achieved by the transmission andreception of

References 111References[1] A. J. Goldsmith and S. Chua, “Variable-rate variable-power MQAM for fading channels,”IEEE Trans. Commun., vol. 45, no. 10,

112 Adaptive modulation and coding for high-rate systems[18] R. Zhang and L. Cai, “A packet-level model for UWB Channel with people shadowingprocess b

5 MIMO techniques for high-ratecommunicationsWasim Q. Malik and Andr´e PollokThis chapter presents an analysis of the gain in system capacity and reli

ContributorsHuseyin ArslanUniversity of South Florida, Florida, USALin CaiUniversity of Victoria, CanadaStark C. DraperUniversity of Wisconsin-Madison

114 MIMO techniques for high-rate communicationsFor a conventional narrowband NT× NRMIMO system, the received signal is givenby⎡⎢⎣y1...yNR⎤⎥⎦=7ρNT⎡⎢⎣h

5.2 MIMO for ultrawideband systems 115thus the integral can be replaced by a sum. Conceptually, this treatment can be easilyunderstood in the context

116 MIMO techniques for high-rate communications3.1 4.6 6.1 7.6 9.1 10.6−70−50−30−10Frequency, GHzMagnitude, dB0 50 100 150−80−60−40−200Time delay, ns

5.2 MIMO for ultrawideband systems 117−50 −25 0 25 5000.250.50.751Cross-range offset, cmCorrelation coefficient 25 MHz100 MHz7.5 GHz−50 −25 0 25 5000

118 MIMO techniques for high-rate communicationsmeasurements with W = 500 GHz and center frequency varying between the 3.1–10.6 GHz FCC-defined UWB ban

5.2 MIMO for ultrawideband systems 1190 3 6 9 12 15 18024681012Signal-to-noise ratio, dBCapacity, bps/Hz 1x11x21x32x23x30 3 6 900.250.50.751Capacity,

120 MIMO techniques for high-rate communications7 8 9 10 1100.250.50.751√L / NCDF N = 1, 2, 30 5 10 15 20 25 30 350369√LCapacity, bps/Hz Measurement

5.2 MIMO for ultrawideband systems 121Theoretically, TR arises as a consequence of the wave equation that describes thepropagation of an electromagnet

122 MIMO techniques for high-rate communications−100 −50 0 50 100−50−250Time, nsMagnitude, dB W = 25 MHzW = 7.5 GHz−50 −25 0 25 50−50−250Time, nsMagn

5.3 MIMO for 60 GHz systems 1235.2.6 SummaryThe analysis of UWB MIMO systems in this section has highlighted some of its keypotential applications and

xii List of contributorsAndreas F. MolischUniversity of Southern California, California, USAAria NosratiniaUniversity of Texas at Dallas, Texas, USA¨O

124 MIMO techniques for high-rate communicationsSmall form factors of 60 GHz RF components and antennas open up the possibility tointegrate multiple 6

5.3 MIMO for 60 GHz systems 125Figure 5.6 Fading map and magnitude of the spatial complex correlation coefficient in 60 GHzLOS channel. The fading map

126 MIMO techniques for high-rate communicationsRange offset [cm]Correlation coefficient0 2 4 6 8 10121416182000.20.40.60.81(a) Range direction, LOSRan

5.3 MIMO for 60 GHz systems 127Figure 5.8 General structure of a MIMO-OFDM system with joint transmit and receive BF:(a) subcarrier-wise BF and (b) sy

128 MIMO techniques for high-rate communications(e.g., see reference [54])yn=√ρ u†nHnvnxn+ u†nwn, (5.10)where Hnis a shorthand notation for H(f =n f)

5.3 MIMO for 60 GHz systems 129freedom (compared to subcarrier-wise BF) [54]. However, our computer simulationsin Section 5.3.4 demonstrate that this

130 MIMO techniques for high-rate communicationsSpectral Efficiency [bps/Hz]Pr(I > Abscissa)MIMO WFmaxMIsc WFSISO EP0 0.5 1 1.5 2 2.5 3 3.5 44.5500.2

5.3 MIMO for 60 GHz systems 131Spectral Efficiency [bps/Hz]Pr(I > Abscissa)20 λcspacing1 λcspacing00.511.522.533.544.5500.20.40.60.81Figure 5.11 Dist

132 MIMO techniques for high-rate communicationsreference [58]. For reference, the distribution of the MIMO WF capacity is also shown.As noted earlier

References 133multiplexing gain and boost the achievable data-rates in UWB systems, but not in 60 GHzsystems. On the other hand, beamforming is more a

List of contributors xiiiSerhan YarkanTexas A&M University, Texas, USARuonan ZhangUniversity of Victoria, Canada

134 MIMO techniques for high-rate communications[16] J. Keignart, C. Abou-Rjeily, C. Delaveaud, and N. Daniele, “UWB SIMO channel measure-ments and si

References 135[36] L. Borcea, G. Papanicolaou, C. Tsogka, and J. Berryman, “Imaging and time reversal inrandom media,” Inverse Problems, vol. 18, no.

136 MIMO techniques for high-rate communications[53] D.-S. Shiu, G. Foschini, M. Gans, and J. Kahn, “Fading correlation and its effect on thecapacity

Part IILow-rate systems

6 ZigBee networks and low-rate UWBcommunicationsZafer Sahinoglu and Ismail GuvencIn this chapter, technologies and standards for low data rate communi

140 ZigBee networks and low-rate UWB communicationsrself configuration: detects addition of a new device into the network, and continuouslyupdates and

6.1 Overview and application examples 141Table 6.1 Key real-time localization systems (RTLS) a pplications, ranges, and accuracy requirements [7].Core

142 ZigBee networks and low-rate UWB communicationsFigure 6.1 Illustration of the network topologies supported by the ZigBee: (a) star topology;(b) tr

6.2 ZigBee 143Table 6.2 Available frequency bands for IEEE 802.15.4.Frequency band (MHz) Modulation Bit rate (Kbps) Number of channels Regions868–868.

144 ZigBee networks and low-rate UWB communicationsFigure 6.3 Flowchart of the slotted CSMA-CA and unslotted CSMA-CA channel accessmechanisms in the I

6.2 ZigBee 145of backoff periods that need to be clear of channel activity prior to transmission. TheBE is related to the number of backoff periods a

146 ZigBee networks and low-rate UWB communications6.2.2.2 Slotted CSMA-CAThe backoff period boundaries of different devices are not related in time t

6.2 ZigBee 147Figure 6.4 GTS packet drop rate versus Pefor an IEEE 802.15.4 beacon-enabled network atvarious GTS packet arrival rates λ (adapted from

148 ZigBee networks and low-rate UWB communicationsFigure 6.5 Illustration of the interference avoidance mechanism in ZigBee (adapted fromreference [1

6.3 Impulse-radio based UWB (IEEE 802.15.4a) 149energy is higher than the other channels, the channel with the minimum energy level,channel c∗i, is co

150 ZigBee networks and low-rate UWB communicationsTable 6.3 UWB channels for the IEEE 802.15.4a standard [16].Channel no. Center freq. (MHz) Bandwidt

6.3 Impulse-radio based UWB (IEEE 802.15.4a) 151Table 6.4 CSS channels for the IEEE 802.15.4a standard [18].Channel no. Center freq. (MHz) Channel no.

152 ZigBee networks and low-rate UWB communicationsFigure 6.8 UWB symbol structure according to the IEEE 802.15.4a standard.of the burst can be determ

6.3 Impulse-radio based UWB (IEEE 802.15.4a) 153Figure 6.9 Basic blocks of CSS PHY transmitter according to the IEEE 802.15.4a standard [18].defined as

1 Short-range wirelesscommunications and reliabilityIsmail Guvenc, Sinan Gezici, Zafer Sahinoglu, and Ulas C. KozatEven though there is no universally

154 ZigBee networks and low-rate UWB communicationsFigure 6.10 Illustration of the IEEE 802.15.4a packet structure. The data part is BPM-BPSKmodulated

6.3 Impulse-radio based UWB (IEEE 802.15.4a) 155Figure 6.11 Structure of the PHR.{16, 64, 1024, 4096} symbols. The preamble length is specified by the

156 ZigBee networks and low-rate UWB communicationsTable 6.5 System parameters for UWB PHY of the IEEE 802.15.4a standard for amean PRF of 62.4 MHz, w

6.3 Impulse-radio based UWB (IEEE 802.15.4a) 157Table 6.6 The basis preamble symbol set [16].Index SymbolS1-0000+0-0+++0 +-000+- +++00-+0-00S20 +0 +-0

158 ZigBee networks and low-rate UWB communicationsFigure 6.12 A received UWB PHY waveform, and representation of the confidence interval withrespect t

6.4 Low latency MAC for WPANs (IEEE 802.15.4e) 159beaconsuperframeMulti superframeCAP CAPCAPCAPCAPCAP(a)(b)CAPCAPFigure 6.13 Illustration of the IEEE

160 ZigBee networks and low-rate UWB communicationsTable 6.7 Logical channel numbering in IEEE 802.15.4e.PHY hopping sequence Logical channel index{1,

6.4 Low latency MAC for WPANs (IEEE 802.15.4e) 161beaconSensor time slotstimeActuatortime slotsManagementtime slotsRetransmissiontime slotsGroupACKDLG

162 ZigBee networks and low-rate UWB communicationsFigure 6.15 Illustration of the TSCH slotframe structure with five time slots. The Tiindicatestime s

6.5 IEEE 802.15.4f (active RFID) 163If a device wishing to join the network receives a valid advertisement command frame,the new device can attempt to

2 Short-range wireless communications and reliabilitycharacteristics, and reliability requirements is provided, and globally available frequencybands

164 ZigBee networks and low-rate UWB communicationsWAN (802.16)InternetHAN(802.15.4)802.15.4g802 15 4g802 15 4g802.15.4gLAN(802.11))Bluetooth(802.15.1

References 165each other (see, e.g., Figure 6.16). Such devices include meters, display systems, con-trollers, and various other infrastructure compon

166 ZigBee networks and low-rate UWB communicationsWireless medium access control (MAC) and physical layer (PHY) specifications forlow-rate wireless pe

References 167[28] N. Patwari, J. N. Ash, S. Kyperountas, A. O. Hero, R. L. Moses, and N. S. Correal, “Locatingthe nodes: Cooperative localization in

7 Impact of channel estimationon reliabilityHongsan ShengThis chapter discusses the impacts of channel estimation on the reliability of ultrawide-band

7.1 Introduction 169The application of pilot-aided channel estimation to UWB systems is discussed in ref-erence [6]. In reference [13], the performanc

170 Impact of channel estimation on reliability7.2 Signal and channel models with channel estimation errorsIn this section, the system model is presen

7.2 Signal and channel models with channel estimation errors 171different numbers of paths L, the constant δ0is determined by the procedure suggestedi

172 Impact of channel estimation on reliabilitypilot symbols [1]. The ML estimate is used in the numerical simulations discussed inSection 7.4. The cl

7.3 Reliability with channel estimation errors 173Substituting (7.5)into(7.10), the path amplitudes conditioned on the path delay estimatesbecomeˆα=

1.1 Short-range wireless communications 3much simpler frequency-domain equalization techniques can be utilized efficiently,1(ii)it is robust in frequen

174 Impact of channel estimation on reliabilitywherew=1Ep∞−∞n(t)q(t − ˆτ)dt ,= 0,...,L − 1 (7.16)is the noise term in the corresponding branch of

7.3 Reliability with channel estimation errors 175andE%η22&=N02EpL−1=0α2μ2, (7.22)respectively. By the CLT, when L is large, η3, which contain

176 Impact of channel estimation on reliabilitywhereγt E[γt]=2EpN0L−1=0α2E%μ2&, (7.28)andγ02EpN0L−1=0α2. (7.29)To obtain the bound in (7.

7.3 Reliability with channel estimation errors 177Now, denote X αμ+ eand Y αμ+ w. Substituting back in (7.32),D =L−1=0XY. (7.33)Conditio

178 Impact of channel estimation on reliabilityand use the alternative form of Q1(a, b)with finite limits [34, p. 79, (4.28)], to obtainQ1(a, b)= Q1(ζ

7.3 Reliability with channel estimation errors 179Using (7.25)in(7.45), we haveMγt(s)=L−1?=0M(s), (7.47)whereM(s)= Eexps2EpN0α2μ2, (7.48)and

180 Impact of channel estimation on reliabilityand the unconditional BER is given byPe=1ππ/20Mγt−12sin2θdx . (7.54)As expected, this is the BER exp

7.4 System optimization with channel estimation errors 1810 2 4 6 8 10 12 1410−610−510−410−310−210−1100Eb/N0 (dB)BERSimulation, M = 5 Simulation, M =

182 Impact of channel estimation on reliability0 10 20 30 40 50 60 70 8010−410−310−210−1100Percent of symbols allocated to the pilotBEREb/N0=3 dBEb/N0

7.4 System optimization with channel estimation errors 1830.5 1 1.5 2 2.5 3 3.578910111213141516Signal bandwidth W (GHz)Required E b/N0 (dB)M = 5 M =

This page intentionally left blank

4 Short-range wireless communications and reliabilitysystems. On the other hand, multihop and cooperative communications may be consid-ered as importa

184 Impact of channel estimation on reliability0 0.5 1 1.5 210−410−310−210−1Signal bandwidth W (GHz)BERml = 0.5 ml = 1 ml = 2Figure 7.4 BER versus the

7.4 System optimization with channel estimation errors 1850 10 20 30 40 50 60 7010−510−410−310−210−1Number of rake fingersBEREb/N0 = 12 dB Eb/N0 = 10

186 Impact of channel estimation on reliability5 10 15 20 25 30 35 40 45 5078910111213141516Number of rake fingersRequired E b/N0 (dB)BER = 10−4 BER =

References 187of paths to be processed by the Rake receiver is determined to attain minimum error ratein the presence of imperfect CSI. For the 2 GHz

188 Impact of channel estimation on reliability[16] L. Huang and C. C. Ko, “Performance of maximum-likelihood channel estimator for UWBcommunications,

References 189[33] J. G. Proakis, “Probabilities of error for adaptive reception of M-phase signals,” IEEE Trans.Commun. Technol., vol. COM-16, no. 1,

8 Interference mitigation andawareness for improved reliabilityHuseyin Arslan, Serhan Yarkan, Mustafa E. Sahin, and Sinan GeziciWireless systems are c

8.1 Mitigation of multiple-access interference (MAI) 191from user k is expressed ass(k)tx(t) =(EkNf∞j=−∞d(k)jb(k)j/Nfptxt − jTf− c(k)jTc− a(k)j/N

192 Interference mitigation and awareness for improved reliabilityr (t)prx (–t)b (1)DetectorˆFigure 8.1 A receiver structure with chip-rate sampling.w

8.1 Mitigation of multiple-access interference (MAI) 193r(t)b(1)ˆstemp,1 (–t)rl1,jrl2,jrlM,j(1)stemp,2 (–t)Detector(1)stemp,M (–t)(1)Figure 8.2 A rece

1.1 Short-range wireless communications 5Table 1.1 Example applications for short-range wireless communications.Low-rate systems High-rate systemsTele

194 Interference mitigation and awareness for improved reliabilitywithAl, j={(n, m):n ∈{1,...,L}, m ∈ Fi, m = j,mTf+ c(1)mTc+ nTc= jTf+ c(1)jTc+lTc}

8.1 Mitigation of multiple-access interference (MAI) 195Since IR-UWB systems transmit pulses with a low duty cycle, signals from some ofthe users may

196 Interference mitigation and awareness for improved reliabilitywhere θ represents a weighting vector, and r is the vector of received signal sample

8.1 Mitigation of multiple-access interference (MAI) 197where˜sdecorrepresents the first column ofST1S1−1with S1denoting the signaturematrix in (8.16

198 Interference mitigation and awareness for improved reliabilityl ∈ L ={l1,...,lM}and j ∈{1,...,Nf}, and let r represent an N × 1 vector consistingo

8.1 Mitigation of multiple-access interference (MAI) 199complexity when the number of frames and/or the number of receiver branches (Rakefingers) is la

200 Interference mitigation and awareness for improved reliability1. S ={1,...,N }2. for i = 1:N1− 13. Choose a random sample s from S4. S = S −{s}5.˜

8.1 Mitigation of multiple-access interference (MAI) 2016 8 10 12 14 1610−510−410−310−210−1100SNR (dB)Bit Error ProbabilityOptimal Combining2−step MMS

202 Interference mitigation and awareness for improved reliability6 8 10 12 14 1610−510−410−310−210−1100SNR (dB)Bit Error ProbabilityOptimal Combining

8.1 Mitigation of multiple-access interference (MAI) 2030 5 10 15 2010−510−410−310−210−1100SNR (dB)Bit Error ProbabilityOptimal CombiningOptimal Multi

6 Short-range wireless communications and reliabilityin such scenarios. While such techniques may also be applied to certain high-rate com-munication

204 Interference mitigation and awareness for improved reliabilityprx (–t)ll,j ,j,…,11M1prx (–t)prx (–t)rr(1)(1)ll,j,j,…,22M1rr(2) (2)ll,j,j,…,KKM1rr(

8.1 Mitigation of multiple-access interference (MAI) 205for j = 1,...,Nfand k = 1,...,K , where f˜r(k)j|b(k)j= iis the likelihood of thejth combined

206 Interference mitigation and awareness for improved reliability0 2 4 6 8 10 1210−510−410−310−210−1100SNR (dB)Bit Error ProbabilityLC 1st iter.LC 2n

8.1 Mitigation of multiple-access interference (MAI) 2070 2 4 6 8 10 12 14 1610−410−310−210−1100SNR (dB)Bit Error ProbabilityLC 1st iter.LC 2nd iter.S

208 Interference mitigation and awareness for improved reliabilitysignal subspace spanned by the eigenvectors associated with the largest eigenvalues

8.1 Mitigation of multiple-access interference (MAI) 209of MAI are investigated. In particular, the design of TH sequences and/or polarity codesin (8.

210 Interference mitigation and awareness for improved reliabilityFigure 8.9 Block diagram of the transmitter for user k in a PCTH system.where S ={1,

8.1 Mitigation of multiple-access interference (MAI) 211Figure 8.10 Block diagram of the receiver for user k in a PCTH system.where bi∈{0, 1}, and x ∈

212 Interference mitigation and awareness for improved reliabilityUWB (-41 dBm/MHz)FCC Part 15 LimitIEEE 802.11bBluetoothIEEE 802.11gHome RFCordless P

8.2 Mitigation of narrowband interference (NBI) 213frequency bands. In CDMA systems, NBI is partially handled by the processing gainas well as by empl

1.2 Definition of reliability 7Table 1.2 Review of the ISM/U-NII bands, and the spectrum used for UWB and 60 GHz systems inthe USA.ISM bands Power limi

214 Interference mitigation and awareness for improved reliabilityDepending on its type, the NBI can be modeled in various ways. For example, it canbe

8.2 Mitigation of narrowband interference (NBI) 215subcarriers depending on the level of interference. The NBI models that can be consid-ered for OFDM

216 Interference mitigation and awareness for improved reliabilityThe feedback information can be various, including the interfered subcarrier index,i

8.2 Mitigation of narrowband interference (NBI) 2173 4 5 6 7 8 9 10−90−80−70−60−50−40−30−20−10010Normalized Spectrum Magnitude (dB)Frequency (GHz)Spec

218 Interference mitigation and awareness for improved reliability4.5 5 5.5−0.500.5Time (ns)Amplitude6th order Gaussian pulse0 5 10 1510−1510−1010−510

8.2 Mitigation of narrowband interference (NBI) 219frames, which last for Tf= Ts/Nfand are divided into chips with a duration of Tc.The pseudo-random

220 Interference mitigation and awareness for improved reliabilityinterferers, methods such as employing notch filters or changing the parameters of th

8.2 Mitigation of narrowband interference (NBI) 221notch filters can be used to suppress NBI. The appealing fact about this method is thatit can be uti

222 Interference mitigation and awareness for improved reliabilityprimarily reflect the NBI rather than the UWB signal. This fact leads to the conseque

8.3 Interference awareness 2232. Interference from other users, which can be further categorized as– Multiuser interference, which is the interference

8 Short-range wireless communications and reliabilityitself. For some applications (e.g., data transfer), reliability is about data integrity andall t

224 Interference mitigation and awareness for improved reliabilityproperties, interference conditions change depending on the propagation characterist

8.3 Interference awareness 225estimation techniques have been popularly used for optimal receiver designs (such aschannel estimation and soft informat

226 Interference mitigation and awareness for improved reliabilitymentioning that with the increasing services and applications, nodes are expected to

References 227[5] M. L. Welborn, “System considerations for ultrawideband wireless networks,” in Proc.IEEE Radio and Wireless Conf., Boston, MA, Aug.

228 Interference mitigation and awareness for improved reliability[24] Y. C. Yoon and R. Kohno, “Optimum multi-user detection in ultrawideband (UWB)mu

References 229[40] J. Foerster, “Channel modeling sub-committee report final, IEEE802.15-02/490,” 2002.[Online]. Available: http://ieee802.org/15[41] D

230 Interference mitigation and awareness for improved reliability[57] P. Liu and Z. Xu, “Performance of POR multiuser detection for UWB communication

References 231[74] G. Durisi and S. Benedetto, “Performance evaluation of TH-PPM UWB systems in thepresence of multiuser interference,” IEEE Commun. L

232 Interference mitigation and awareness for improved reliability[90] Y. Zhang and J. Dill, “An anti-jamming algorithm using wavelet packet modulated

References 233[109] C. W. Rhodes, “Reduction of NTSC co–channel interference by referencing carrier fre-quencies to the LORAN–C signal,” IEEE Trans. o

1.2 Definition of reliability 9DataModulation and CodingRF OscillatorAmplifierTransmitter AntennaTransmitterLNADownConversionDemodulationand DecodingCh

9 Characterization of Wi-Fiinterference for dynamic channelallocation in WPANsFederico Penna, Claudio Pastrone, Hussein Khaleel, Maurizio A. Spirito,

9.1 Towards adaptive WPANs 235Figure 9.1 Channel occupation of IEEE 802.11 and IEEE 802.15.4.environments. This study is meant as a basis for the deve

236 Characterization of Wi-Fi interference for dynamic channel allocation in WPANsbands. Being a key aspect for the implementation of CR systems, spec

9.2 WPANs under Wi-Fi interference 237features of the IEEE 802.15.4 radio chips and then used to estimate the level of channeloccupation.Compared to p

238 Characterization of Wi-Fi interference for dynamic channel allocation in WPANs(a)(b)Figure 9.2 Configuration of the test-beds: (a) first setup, used

9.2 WPANs under Wi-Fi interference 239Figure 9.3 RSSI sampling scheme (from [27],c 2009 IEEE).windows per channel in the total sensing time. Then the

240 Characterization of Wi-Fi interference for dynamic channel allocation in WPANsof the PHY payload to be sent over the air is a parameter as well. T

9.2 WPANs under Wi-Fi interference 241was considered, in order to observe the combined effects of multipath propagations andmultiple sources of interf

242 Characterization of Wi-Fi interference for dynamic channel allocation in WPANsFigure 9.4 Probability distribution of the received energy fW(x ) an

9.2 WPANs under Wi-Fi interference 243By definition of the Bernoulli process, the variable k has a binomial distribution withsuccess rate equal to p1,

10 Short-range wireless communications and reliabilityalso be explained, along with referrals to the related chapters in the book for a morecomplete t

244 Characterization of Wi-Fi interference for dynamic channel allocation in WPANshave an observedˆp1in the interval[p1− 0.02, p1+ 0.02]with a confiden

9.3 Interference characterization and performance degradation 245rSpectrograms, to observe the behavior of the interfering traffic jointly in the timea

246 Characterization of Wi-Fi interference for dynamic channel allocation in WPANsFigure 9.5 Anechoic chamber: energy PDFs of the four interfered IEEE

9.3 Interference characterization and performance degradation 247Time [min]Channel numberAverage RSSI [dBm],540 kb/s2 4 6 8 10121416182011121314151617

248 Characterization of Wi-Fi interference for dynamic channel allocation in WPANs(a)(b)kbps)kbps)kbps)kbps)kbps)kbps)kbps)kbps)kbpskbpskbpskbpsFigure

9.3 Interference characterization and performance degradation 249In this expression, the mean definition is indeed an approximation, since the distinct

250 Characterization of Wi-Fi interference for dynamic channel allocation in WPANsFigure 9.8 Relative throughput versus data rate in the anechoic cham

9.3 Interference characterization and performance degradation 251Time [min]Channel numberAverage RSSI [dBm]2 4 6 8 10 12 14 16 18111213141516171819202

252 Characterization of Wi-Fi interference for dynamic channel allocation in WPANsFigure 9.10 Indoor 1: energy PDFs of the four interfered IEEE 802.15

9.3 Interference characterization and performance degradation 253Time [min]Channel numberAverage RSSI [dBm],108 kb/s2 4 6 8 10121416181112131415161718

1.2 Definition of reliability 11received powers at the receiver. The power may also be focused along a certain beamdirection using beamforming techniqu

254 Characterization of Wi-Fi interference for dynamic channel allocation in WPANsFigure 9.12 Indoor 1: analysis of the results, (a) global mean (μ) a

9.3 Interference characterization and performance degradation 255Figure 9.13 Relative throughput versus data rate for Indoor 1 scenario.pulse shape fil

256 Characterization of Wi-Fi interference for dynamic channel allocation in WPANsTime [min]Channel numberAverage RSSI [dBm]2 4 6 8 10 12 14 16 18 201

9.3 Interference characterization and performance degradation 257Figure 9.15 Indoor 2: energy PDFs of the four interfered IEEE 802.15.4 channels (14–1

258 Characterization of Wi-Fi interference for dynamic channel allocation in WPANsTime [min]Channel numberAverage RSSI [dBm],108 kb/s2 4 6 8 101214161

9.3 Interference characterization and performance degradation 259Figure 9.17 Indoor 2: analysis of the results; (a) global mean (μ) and mean of the si

260 Characterization of Wi-Fi interference for dynamic channel allocation in WPANsFigure 9.18 Relative throughput versus data rate for Indoor 2 scenar

9.4 Improving WPAN’s reliability under interference 261Spectrograms provide a temporal visualization of the spectrum occupancy state, theyare obtained

262 Characterization of Wi-Fi interference for dynamic channel allocation in WPANsAlgorithm 9.1 Channel selection based on outage probability.1: M: nu

9.4 Improving WPAN’s reliability under interference 263and reactivity in the detection of interference, etc., according to Section 9.2.4 andSection 9.

12 Short-range wireless communications and reliabilitymultiuser and narrowband interference for short-range wireless communication systemswill be disc

264 Characterization of Wi-Fi interference for dynamic channel allocation in WPANs20 40 60 80 100 120 14000.020.040.060.080.10.120.140.160.180.2Sensin

9.4 Improving WPAN’s reliability under interference 26520 40 60 80 100 120 14000.020.040.060.080.10.120.140.160.180.2Sensing window numberEstimated ou

266 Characterization of Wi-Fi interference for dynamic channel allocation in WPANs50 100 150 200 250051015202530Sample numberInstantaneous throughput

9.5 Conclusion 267In order to discuss further the issue of the throughput reduction due to spectrumsensing as described in Section 9.2.5, it can be se

268 Characterization of Wi-Fi interference for dynamic channel allocation in WPANsThis work was partially supported by the European Commission in the

References 269[17] Q. Zhao, L. Tong, A. Swami, and Y. Chen, “Decentralized cognitive MAC for opportunisticspectrum access in ad hoc networks: A POMDP

10 Energy saving in low-rate systemsTae Rim Park and Myung J. LeeIn low-rate wireless networks, energy saving has been one of the recent importantrese

10.1 Background on energy efficiency 271Table 10.1 Ambient energy sources and harvested power [1].Source Source energy density Harvested energy density

272 Energy saving in low-rate systemsFigure 10.1 Energy per frame.Figure 10.2 Comparision of two energy control methods: transmission power control (P

10.1 Background on energy efficiency 273Figure 10.3 Bit error probability curves for four different modulation options.Assume that the required BER is

1.3 Review of related wireless standards 13changes in wireless channel quality, changes in traffic demand, etc. Depending on theparticular scenario, fe

274 Energy saving in low-rate systemsFigure 10.4 Example time line of channel activity in view of energy consumption.network devices. Exact transmissi

10.1 Background on energy efficiency 275is used for large data exchange. However, it consumes significantly higher energy formonitoring the channel. Thu

276 Energy saving in low-rate systemsActive time/active ratio a measure to present the effects of energy saving algorithmson the sleeping time. Althou

10.2 Energy saving MACs 277AsymmetricSynchronous AsynchronousTransmitter notification Receiver queryAutomatic deliveryFigure 10.5 Asymmetric single-h

278 Energy saving in low-rate systemsIdeally, automatic delivery is the best energy-saving algorithm because it does nothave any control frame exchang

10.2 Energy saving MACs 279Table 10.2 Parameters for analysis.Symbol Parameter ValueTTRTurnaround time 0.192 msTBCNBeacon frame time 0.608 ms (19 byte

280 Energy saving in low-rate systemsAPSTA AFigure 10.7 IEEE 802.11 power save mode.method, since it requires fairly large energy consumption at each

10.2 Energy saving MACs 281Figure 10.8 Average active time per wake-up interval in transmitter notification.average backoff time for a beacon and an RT

282 Energy saving in low-rate systemsAPBeaconFigure 10.9 Example time line of unscheduled-automatic power save delivery.of latency constraints. The tr

10.2 Energy saving MACs 283Active duration utilization rate (%)Figure 10.10 Average active time per wake-up interval in receiver query.SymmetricSynchr

Reliable Communications for Short-range Wireless SystemsEnsuring reliable communication is an important concern in short-range wireless com-munication

14 Short-range wireless communications and reliabilitySource ASource BSource CDestination ADestination BDestination CXXCh. 12Ch. 1Source D Destination

284 Energy saving in low-rate systemsactive duration and to resynchronize the active durations. A device having a frame totransmit notifies this by tra

10.2 Energy saving MACs 285Figure 10.12 Example timeline of the X-MAC (c 2008 IEEE) [4].traffic rate. In other words, if the number of transmissions i

286 Energy saving in low-rate systemsframes instead of a short preamble and an early ACK to facilitate the comparison withother protocols. We assume t

10.2 Energy saving MACs 287Figure 10.13 Example time line of IEEE 802.15.5 asynchronous energy saving (AES) mode(c 2008 IEEE) [4].by the WN is long e

288 Energy saving in low-rate systemsActive duration utilization rate (%)Figure 10.14 Comparison of symmetric MAC algorithms.The active time for data

References 28910.3 SummaryIn this chapter, we have presented issues for saving energy in low-rate networks andexplained how MAC protocols play a criti

290 Energy saving in low-rate systems[12] N. M. Pletcher, S. Gambini, and J. Rabaey, “A 52 W wake-up receiver with 72 dBm sensitivityusing an uncertai

Part IIISelected topics forimproved reliability

11 Cooperative communicationsfor reliabilityAndreas F. Molisch, Stark C. Draper, and Neelesh B. MehtaChapter 11 describes how teams of wireless nodes

1.3 Review of related wireless standards 15Table 1.3 Task groups (TGs) in IEEE 802.15 Working Group for WPAN [29].Name Description IEEE standardTG1 Bl

294 Cooperative communications for reliabilitythat in cellular communications, which until now has been the dominant wireless appli-cation, the reliab

11.1 Introduction 295with high attenuation, which requires a high transmit power to succeed, can be avoided.These two aspects are actually two sides o

296 Cooperative communications for reliabilityFull CSIT The nodes know both the amplitude and the phase of the channel to thereceiving node. In this c

11.2 Cooperative communication using virtual beamforming 297transmitting nodes. This functionality is transparent to the receiver. All complicationis

298 Cooperative communications for reliabilitywith low power. One of the first suggestions of virtual beamforming can be found inreference [22].A numbe

11.2 Cooperative communication using virtual beamforming 299h2h1hNPtPtPtN21RelaysDestinationSourceTransmissionPower PSDecode successfullyUnable to dec

300 Cooperative communications for reliability2. Training: Only the M relays that receive data successfully from the source sendtraining sequences at

11.2 Cooperative communication using virtual beamforming 301We model only the energy required for radio transmission and not the energy consumedfor re

302 Cooperative communications for reliabilityobtainPf(K (M), M) =N0B(2r− 1)K (M) + 1E1g[K (M)]+K (M)i=11g[i]. (11.4)The term (K (M) + 1) in the de

11.2 Cooperative communication using virtual beamforming 3032 4 6 8 10 12 14 16681012141618number of relaysenergy/messageoptimal relay selectionsingle

16 Short-range wireless communications and reliabilityTable 1.4 Different classes for Bluetooth devices.Class Maximum power (mW) Range (m)Class 1 100

304 Cooperative communications for reliability1234657Figure 11.3 Illustration of wireless network graph, G, with seven nodes (c 2008 IEEE) [2].is not

11.2 Cooperative communication using virtual beamforming 305As before, all transmissions are at a constant rate, r, and each message is d symboldurati

306 Cooperative communications for reliabilitychannel gains to j, and feeds back to each selected node, k ∈ K(D(i, j)), the gainand phase of the chann

11.2 Cooperative communication using virtual beamforming 30712346570.270.180.180.180.180.180.180.180.180.180.360.360.36Figure 11.4 Computation of opti

308 Cooperative communications for reliabilityusing the Bellman–Ford algorithm, the minimum cost route from the source (node 1) todestination (node 7)

11.3 Cooperative communication using rateless codes 309pair of independent erasure channels each having erasure probability pefrom two relaysto a sing

310 Cooperative communications for reliabilityIn performing mutual information accumulation, the receiver must be able to distin-guish the signals tra

11.3 Cooperative communication using rateless codes 311different codes, then the destination can perform mutual information accumulation.The second ph

312 Cooperative communications for reliability11.3.3.1 Analysis of two-phase protocolComputing the performance (i.e., the energy consumed and delay) p

11.3 Cooperative communication using rateless codes 313This follows from the PDF of the SNR in Rayleigh fading and Shannon’s capacityequation for AWGN

1.3 Review of related wireless standards 17Figure 1.3 (a) Full mesh topology (b) partial mesh topology.Recently, Bluetooth 3.0 specification has been a

314 Cooperative communications for reliability11.3.3.3 ShadowingWe now turn to the computation of transmission time and energy expenditure in theprese

11.3 Cooperative communication using rateless codes 315average energy expenditurenumber of used relay nodes LFigure 11.5 Mean energy expenditure as a

316 Cooperative communications for reliabilitymean energy expenditurecorrelation coefficientFigure 11.6 Mean energy expenditure as a function of the c

11.3 Cooperative communication using rateless codes 317(using different fountain codes) is denoted as τ2=Gτ2− τ1, and so on. Generally, thetime until

318 Cooperative communications for reliabilitymean transmission energymean transmission timenumber of relay nodes NFigure 11.7 Mean transmission time

11.3 Cooperative communication using rateless codes 319pdf of transmission energynormalized transmission energyFigure 11.8 PDF of transmission energy

320 Cooperative communications for reliabilityWe minimize this linear objective function subject to the following constraints. First,i≥ 0 for all i.

11.3 Cooperative communication using rateless codes 321Figure 11.9 Location of nodes in a 50-node network. The minimum-energy and minimum-delaycoopera

322 Cooperative communications for reliabilityis selected using Dijkstra’s shortest-path algorithm. First, we consider the situationwhere each node de

References 323[4] S. C. Draper, L. Liu, A. F. Molisch, and J. Yedida. “Routing in cooperative wireless networkswith mutual-information accumulation.”

18 Short-range wireless communications and reliabilityFigure 1.4 Network topologies in the IEEE 802.15.4-2006 standard, where circles represent thePAN

324 Cooperative communications for reliability[24] J. N. Laneman, D. N. C. Tse, and G. W. Wornell. “Cooperative diversity in wireless networks:Efficien

References 325[44] M. Z. Win and J. H. Winters. “Analysis of hybrid selection/maximal-ratio combining inRayleigh fading.” IEEE Trans. Commun., vol. 47

12 Reliability through relay selection incooperative networksRamy Abdallah Tannious and Aria NosratiniaThis chapter first presents an overview of the p

12.2 Signaling in multiple-relay networks 327Figure 12.1 Sensor network with a group of nodes clustered around the source.designer recruits the help o

328 Reliability through relay selection in cooperative networksdivision multiple access (TDMA/FDMA) system. To improve the spectral efficiency,the seco

12.3 Motivations for relay selection 3292. The data stream of the source is not known aprioriat the candidate relay nodes.Thus, acquiring such informa

330 Reliability through relay selection in cooperative networksTable 12.1 Comparison between signaling protocols for multiple-relay networks.Protocol

12.4 Overview of relay selection 331that relay selection is a rich problem that will continue to draw more interest due to itspracticality and simplic

332 Reliability through relay selection in cooperative networksSNR gains. If DSTC is used, xmwill be transmitted based on a space-time code structurew

12.4 Overview of relay selection 3331 2 3 4 5 6 7 81820222426283032343638MReceive SNR (dB)Distributed beamformingDistributed STCRelay selectionFigure

1.3 Review of related wireless standards 19Table 1.5 Different modulation and coding types in IEEE802.15.3, where TCM refers to trellis coded modulati

334 Reliability through relay selection in cooperative networkswhereas the other smoothes the difference between both links via a harmonic meanoperati

12.4 Overview of relay selection 335In reference [30], two relevant questions are posed about the relay selection problem.The first question is about w

336 Reliability through relay selection in cooperative networksof large overhead of CSI across the network. A relay node is feasible to participate in

12.5 Limited feedback centralized relay selection 337and switching to the selected relay occurs ifh∗m= max{hm} > T . (12.16)It is obvious that the

338 Reliability through relay selection in cooperative networksTable 12.2 The incremental transmission with relay selection (ITRS) protocol (c 2008 I

12.5 Limited feedback centralized relay selection 339therefore, the usage of channel resources may be inefficient. The details of the feedbacksignaling

340 Reliability through relay selection in cooperative networks0 5 10 15 20 25 3010−410−310−210−1100SNR (dB)Outage ProbabilityHARQ−simulationHARQ−anal

12.5 Limited feedback centralized relay selection 341rounds of transmission for which the following outage expression can easily be derived:Pout,HARQ=

342 Reliability through relay selection in cooperative networks0 0.2 0.4 0.6 0.8 10123456Multiplexing gain rDiversity gain d(r)DirectDSTC, ORDDFITRS

References 343its power resources. Under these conditions, one may use a variation of ITRS, where thesource will retransmit only if all relays have fa

20 Short-range wireless communications and reliabilityFigure 1.5 Illustration of a piconet, where the circle represents the PNC. The dashed linesindic

344 Reliability through relay selection in cooperative networks[8] R. Pabst, B. Walke, D. Schultz, P. Herhold, H. Yanikomeroglu, S. Mukherjee, H. Visw

References 345[28] A. Bletsas and A. Lippman, “Implementing cooperative diversity antenna arrays with com-modity hardware,” IEEE Commun. Mag., vol. 44

346 Reliability through relay selection in cooperative networks[47] Y. Ge, S. Wen, and Y.-H. Ang, “Analysis of optimal relay selection in IEEE 802.16

13 Fundamental performance limits inwideband relay architectures¨Ozg¨ur Oyman13.1 IntroductionThe design of large-scale distributed wireless networks

348 Fundamental performance limits in wideband relay architecturesThe power-limited wideband regime serves a practically relevant mode of operationfor

13.1 Introduction 349Option 1:DirectOption 2:DistributedRelaysWWWSourceSourceP/3P/3P/3DestinationDestinationTXPRXTXRXENCENCR2R1DECWDEC^^Figure 13.1 Po

350 Fundamental performance limits in wideband relay architectures1041051061071081010109108107106105104Bandwidth (Hz)Power (W)DirectDistributed Relays

13.1 Introduction 351bandwidth is large and the main concern is the limitation on power. Similarly, the caseof C  1 corresponds to the bandwidth-limi

352 Fundamental performance limits in wideband relay architecturesCommunicate in N hops1 2N N+1Range = DD/NFigure 13.4 Linear multihop network model f

13.2 Power–bandwidth tradeoff in serial relay architectures 353N = 4SIMULTANEOUS LINKSPHASE 1PHASE 2Λ = 2Δ = 21 52 3 41 52 3 4Figure 13.5 Linear multi

1.3 Review of related wireless standards 21include the wireless monitoring of electroencephalogram (EEG), electrocardiogram(ECG), electromyography (EM

354 Fundamental performance limits in wideband relay architecturestone a narrowband receiver can be employed. We assume that the length of the cyclicp

13.2 Power–bandwidth tradeoff in serial relay architectures 355block length, i.e., slow fading assumption. Although we assume that each receivingtermi

356 Fundamental performance limits in wideband relay architectureshop n = (m − 1) + k. The codeword error probability for transmission m over thekth

13.2 Power–bandwidth tradeoff in serial relay architectures 357andS0=a.slimSNR→02%˙I (SNR)&2−¨I (SNR), (13.3)where˙I and¨I denote the first and sec

358 Fundamental performance limits in wideband relay architectures13.2.2.1 Fixed-rate multihop relayingA suboptimal strategy that yields a lower bound

13.2 Power–bandwidth tradeoff in serial relay architectures 359then μ belongs to one of the three families of extreme-value distributions above[32]. T

360 Fundamental performance limits in wideband relay architecturesthe channel-fading parameters through the following relationships:5EbN0min=w.p .1Dp

13.2 Power–bandwidth tradeoff in serial relay architectures 3610 0.5 1 1.5 200.10.20.30.40.50.60.70.80.91Cumulative distribution function (CDF)End-to-

362 Fundamental performance limits in wideband relay architecturesgains. In other words, our results show that multihop diversity gains remain viableu

13.3 Power–bandwidth tradeoff in parallel relay architectures 363Time Slot 1W1WLSLrKtKrktkFK,lyLyly1Ek,LS1S1D1DlDLEk, 1r1F1,lt1Hk,1Hk,LRkRKR1GK,lG1,lT

22 Short-range wireless communications and reliabilityPeer-to-peermasterslaveStarBroadcastFigure 1.7 Illustration of the three network topologies supp

364 Fundamental performance limits in wideband relay architecturesspatio-temporally i.i.d. (i.e., assuming full spatial multiplexing [34] for all mult

13.3 Power–bandwidth tradeoff in parallel relay architectures 36513.3.1.3 Coding frameworkFor any block length Q,a({2QRl,m: l = 1,...,L, m = 1,...,Ms}

366 Fundamental performance limits in wideband relay architecturesexpressed as Eb/N0= SNR/C(SNR).11In this context, the power–bandwidth tradeoffis bet

13.3 Power–bandwidth tradeoff in parallel relay architectures 36713.3.2 Upper-limit on MRN power–bandwidth tradeoffIn this section, we derive an upper

368 Fundamental performance limits in wideband relay architecturesS1R1RKSLD1DLSourceterminalscooperateRelay anddestinationterminalscooperateW1W1WL^WL^

13.3 Power–bandwidth tradeoff in parallel relay architectures 369as K →∞. Since our application of the cut-set theorem through the broadcast cut leads

370 Fundamental performance limits in wideband relay architecturesTable 13.1 Practical LDMRB schemes for multi-user MRNs.Relay link channel matrix MF

13.3 Power–bandwidth tradeoff in parallel relay architectures 371bit at a finite spectral efficiency given by C∗≈ 1.15 LMsand consequentlyEbN0LDMRBmin≈(

372 Fundamental performance limits in wideband relay architecturesterminal Dlcorresponding to spatial stream sl,mis given byyZFl,m='Kk=1dk,l,m)s

13.3 Power–bandwidth tradeoff in parallel relay architectures 373Letting β = PR/PS, we find that SIR-maximizing power allocation (for fixedSNR) is achie

1.3 Review of related wireless standards 23I/O deviceGateway,System managerSecurity managerRouting devicecontrolsystemplant networkFigure 1.8 Illustra

374 Fundamental performance limits in wideband relay architecturesvectors hk,l,m(mth column of Hk,l) to yieldˆsMFk,l,m=8Ek,lAAhk,l,mAA2sl,m+(p,q)=(l

13.3 Power–bandwidth tradeoff in parallel relay architectures 375Eb/N0analysis of the ZF-LDMRB scheme. We apply the same steps as in theproof of (13.2

376 Fundamental performance limits in wideband relay architecturessince the signal power grows faster than the interference power as K →∞. Thus,while

13.3 Power–bandwidth tradeoff in parallel relay architectures 377for an MRN under the slow fading (non-ergodic) channel model [31] and thus,our asympt

378 Fundamental performance limits in wideband relay architecturesthough SNR  1, the SIR for each stream in (13.21) simplifies to (note the additional

13.3 Power–bandwidth tradeoff in parallel relay architectures 379scheme based on the ZF algorithm and compare with the performance under directtransmi

10−210−110010110210310400.10.20.30.40.50.60.70.80.91SIR per streamCDFDirectLDMRBIncreasing K = 1,2,4,8,16 Figure 13.9 CDF of SIR for direct transmissi

13.3 Power–bandwidth tradeoff in parallel relay architectures 3810 2 4 6 8 10 12 14 16 1810−210−1100101102103104105Eb/N0Spectral efficiency (b/s/Hz)ZF

382 Fundamental performance limits in wideband relay architectures10−310−210−110010110−1100101102103104Spectral efficiency (b/s/Hz)Eb/N0cutset bound 0

References 383(K−1rather than K−1/2) is achievable with LDMRB compared to previous work inreference [18] and we verify the optimality of the K−1energy

24 Short-range wireless communications and reliabilitySlottedhoppingSlowhoppingtimeChannelsFigure 1.9 Illustration of the hybrid channel hopping opera

384 Fundamental performance limits in wideband relay architectures[15] G. Caire and S. Shamai, “On the achievable throughput of a multiantenna Gaussia

References 385[35] T. M. Cover and J. A. Thomas, Elements of Information Theory,NewYork,NY,JohnWiley,1991.[36] R. J. Serfling, Approximation Theorems o

14 Reliable MAC layer andpacket schedulingUlas C. KozatMedium access control (MAC) is of paramount importance in wireless systems: itorchestrates how

14.1 Introduction 387channels at what time. Therefore, it directly impacts the access delays, the success oftransmissions, as well as the achievable c

388 Reliable MAC layer and packet schedulingtransmission rates at the PHY layer possible at the same reliability level. The latterbenefit is achieved b

14.2 Opportunistic scheduling/multiuser diversity 389Time (slot number)Rate (Kbps)Figure 14.1 Representation of how channel rates fluctuate over time,

390 Reliable MAC layer and packet schedulingapproach and instead, directly start from a utility maximization problem to find out theappropriate schedul

14.2 Opportunistic scheduling/multiuser diversity 391ABCDXYZWGFigure 14.2 Scheduling multiple multicast groups.14.2.2 Multicast caseUnlike in the unic

392 Reliable MAC layer and packet schedulingBlock-1 Block-2 Block-3 Block-4 Block-5 Block-6R[1]Slot-1 Slot-2 Slot-3R[2]R[3]A, B, D A, B, CB, DFigure 1

14.2 Opportunistic scheduling/multiuser diversity 3931 2 3 4 5 6 7 8 9 1000.511.522.53Normalized throughputNumber of targeted usersI.I.D. Rayleigh fad

References 25[12] A. M. Kuzminsky and H. R. Karimi, “Multiple-antenna interference cancellation for WLANwith MAC interference avoidance in open access

394 Reliable MAC layer and packet schedulingSimilar to the unicast case, it is also possible to define the PFS rule for single andmultiple multicast gr

14.3 Coding and scheduling 395a repeat request is sent back to the sender using negative acknowledgement (NACK). Ifno error is detected, a positive ac

396 Reliable MAC layer and packet schedulingWGP1 P2 P3ABCP2P3ABCP1P3P1P2P:=XOR(P1,P2,P3)PFigure 14.5 Coding and scheduling can be used together efficie

14.3 Coding and scheduling 397Block-1 Block-2 Block-3 Block-lOriginal Source BlocksErasureEncoderEncodingBlock-1EncodingBlock-2EncodingBlock-3Encoding

398 Reliable MAC layer and packet schedulingblocks [17]. As opposed to fixed-rate codes, a rateless code can generate as manyencoding blocks as needed

14.3 Coding and scheduling 399is to focus on the user orderings and channel conditions that makes it possible tocompletely characterize the distributi

400 Reliable MAC layer and packet scheduling1 2 3 4 5 6 7 8 9 1000.10.20.30.40.50.60.70.80.911.11.21.31.41.51.61.71.81.92Normalized throughputNumber o

14.4 Media quality driven scheduling 401further processing. Some relatively recent coding techniques [17] achieve the minimumpossible recovery time fo

402 Reliable MAC layer and packet schedulingScheduling and medium access layer have a unique role in the communication stackssince they make the ultim

14.4 Media quality driven scheduling 403no gap between throughput and goodput. Nonetheless, when we operate over finiteframe lengths or under non-i.i.d

26 Short-range wireless communications and reliability[30] “Bluetooth SIG.” [Online]. Available: http://www.bluetooth.org[31] IEEE standard for inform

404 Reliable MAC layer and packet scheduling14.5 SummaryIn this chapter we have covered reliability from the perspective of the scheduling layer.Relia

References 405[8] H. J. Kushner and P. A. Whiting, “Convergence of proportional-fair sharing algorithmsunder general conditions,” IEEE Trans. on Wirel

406 Reliable MAC layer and packet scheduling[28] M. Sharif and B. Hassibi, “A delay analysis for opportunistic transmission in fading broadcastchannel

Index60 GHz radio, 31achievable region, 249ACI, 223ACK, 98, 110active RFID, 163adaptive modulation and coding, see AMCadditive white Gaussian noise (A

408 IndexDCM, 37, 96decode-and-forward, 298, 333, 334, 336, 342decorrelator, 196deinterleaver, 203differential QPSK (DQPSK), 154differential quadratur

Index 409ISA SP-100 standard, 16ISI, 194, 222ISM band, 6iterative MUD, 202knapsack problem, 336Kolmogorov condition, 368L-MMSE, 369Laplacian distribut

410 Indexpseudo-chaotic time-hopping, 210pulse discarding detector, 196pulse repetition frequency, see PRFquantization of feedback, 298quasi-decorrela

Index 411video, 401virtual beamforming, 295–299, 304, 308–310virtual branch analysis, 302visible light communication (VLC), 22wake-up interval, 276, 2

References 27for high-rate wireless personal area networks (WPANs),” Sep. 2003. [Online]. Available:http://standards.ieee.org/getieee802/download/802.

Part IHigh-rate systems

2 High-rate UWB and 60 GHzcommunicationsSinan Gezici and Ismail GuvencIn this chapter, two technologies for high data-rate communications systems for

32 High-rate UWB and 60 GHz communications100101−80−75−70−65−60−55−50−45−40Frequency (GHz)EIRP Emission Level (dBm) 0 961.611.993.1 10.6Figure 2.1 FC

2.1 Overview and application scenarios 33Figure 2.2 A commercial wireless USB product.The frequency spectrum from 57 GHz to 64 GHz is allocated for MM

Reliable Communications forShort-range Wireless SystemsEdited byISMAIL GUVENCDOCOMO C ommunications Laboratories USA, Inc.SINAN GEZICIBilkent Universi

34 High-rate UWB and 60 GHz communicationsFigure 2.3 Wireless transfer of HD video/audio from a DVD player to an HDTV, and from alaptop to a projector

2.2 ECMA-368 high-rate UWB standard 35Table 2.1 Allocation of frequency bands in the ECMA-368 standard.Band index Center frequency (GHz) Band group1 3

36 High-rate UWB and 60 GHz communicationsTable 2.2 Seven TFCs for band group 1 [2].TFC-1123123TFC-2132132TFC-3112233TFC-4113322TFC-5111111TFC-6222222

2.2 ECMA-368 high-rate UWB standard 37Figure 2.5 Basic blocks of an MB-OFDM UWB transmitter according to the ECMA-368 [2].the rate 1/3 convolutional e

38 High-rate UWB and 60 GHz communicationswhere Tsis the symbol length, Nsis the number of symbols in the packet, si(t)isthecomplex baseband signal re

2.2 ECMA-368 high-rate UWB standard 39Table 2.3 Various data rate options and corresponding parameters in the ECMA-368 standard [2].Data rate (Mbps) M

40 High-rate UWB and 60 GHz communicationsTable 2.4 Systems parameters for the MB-OFDM UWB transmitter accordingto the ECMA-368 standard [2].Parameter

2.3 ECMA-387 millimeter-wave radio standard 41Table 2.5 Band allocation in ECMA 387 [17].Band ID Channel bonding fL(GHz) fC(GHz) fU(GHz)1 No 57.24 58.

42 High-rate UWB and 60 GHz communicationsTable 2.7 Transmit spectral mask requirements in ECMA-387 for Type A, Type B, andType C devices (in MHz) [17

2.3 ECMA-387 millimeter-wave radio standard 43Figure 2.7 Block diagram of the SCBT PHY baseband of Type A devices in ECMA-387 (withEEP) [17].at the ce

CAMBRIDGE UNIVERSITY PRESSCambridge, New York, Melbourne, Madrid, Cape Town,Singapore, S˜ao Paulo, Delhi, Tokyo, Mexico CityCambridge University Press

44 High-rate UWB and 60 GHz communicationsFigure 2.8 Constellation of QPSK modulation with (a) EEP, and (b) UEP.Figure 2.9 An example of the SCBT symb

2.3 ECMA-387 millimeter-wave radio standard 45Figure 2.10 Block diagram of the OFDM PHY baseband of Type A devices in ECMA-387 [17].NCP∈{0, 32, 64, 96

46 High-rate UWB and 60 GHz communicationsFigure 2.11 An example for SC symbol structure.symbols are mapped onto separate subcarriers, all of the modu

2.3 ECMA-387 millimeter-wave radio standard 47Figure 2.12 Encoding and mapping for DAMI devices [17].Figure 2.13 Encoding procedure for Type C devices

48 High-rate UWB and 60 GHz communicationsunified way as follows [17]sRF(t) = ReNf−1n=0snt − nTsymexp( j2π fct), (2.11)where Re{.} captures the re

2.3 ECMA-387 millimeter-wave radio standard 49Table 2.8 Discovery modes with different data rates [17].Mode NDISCREPData rate (Mbps)D0 128 2.255D1 64

50 High-rate UWB and 60 GHz communicationsTable 2.9 Mode dependent parameters for Type A devices [17].Base data rate (Gbps)Mode NB= 1 NB= 2 NB= 3 NB=

2.3 ECMA-387 millimeter-wave radio standard 51Table 2.10 Mode-dependent parameters for Type B devices [17].Base data rate (Gbps)Mode NB= 1 NB= 2 NB= 3

52 High-rate UWB and 60 GHz communicationsTable 2.12 Timing-related parameters for SCBTs of Type A devices, and SC transmissions of Type B andType C d

2.4 IEEE 802.15.3c millimeter-wave radio standard 53Table 2.14 Frame-related parameters for ECMA-387 transmissions (all time units in nanoseconds) [17

ContentsList of contributors page xi1 Short-range wireless communications and reliability 1Ismail Guvenc, Sinan Gezici, Zafer Sahinoglu, and Ulas C. K

54 High-rate UWB and 60 GHz communicationsf1= 0.94 GHz, f2= 1.1 GHz, f3= 1.6 GHz, and f4= 2.2 GHz. For OOK transmis-sions, up to 40 dB transmission po

2.4 IEEE 802.15.3c millimeter-wave radio standard 55Table 2.15 MCS dependent parameters for SC PHY MCS [18].Data rate Data rateMCS MCS (Mbps), (Mbps),

56 High-rate UWB and 60 GHz communicationsFigure 2.15 General transmitter structure for SC PHY in IEEE 802.15.3c [18].Table 2.16 MCS-dependent paramet

2.4 IEEE 802.15.3c millimeter-wave radio standard 57Figure 2.16 General transmitter structure for HSI PHY in IEEE 802.15.3c [18].is composed of eight

58 High-rate UWB and 60 GHz communicationsTable 2.17 MCS-dependent parameters for AV PHY (HRP) [18].Inner code rate CodingMCS Data rate Modulationinde

References 59Figure 2.18 General transmitter structure for AV PHY in IEEE 802.15.3c (LRP) [18].bit interleaving is applied. The output bit sequence is

60 High-rate UWB and 60 GHz communications[12] N. Guo, R. C. Qiu, S. S. Mo, and K. Takahashi, “60-GHz millimeter-wave radio: Principle,technology, and

3 Channel estimation forhigh-rate systemsZhongjun Wang, Yan Xin, and Xiaodong WangIn this chapter, we consider the channel estimation issue in orthogo

62 Channel estimation for high-rate systemsFor example, application requirements and channel environments may become equallyaccountable for properly s

3.1 Channel models for high-rate systems 63wireless propagation channel. It can be evaluated by the ratio of the transmitted power Ptto the received p

vi Contents2.4 IEEE 802.15.3c millimeter-wave radio standard 532.4.1 Single-carrier PHY 552.4.2 High-speed interface PHY 562.4.3 Audio/visual PHY 573

64 Channel estimation for high-rate systemsVariation in received signal power due to multipath occurs over very short distances,on the order of the si

3.1 Channel models for high-rate systems 65MPCsRelative Power of MPCsTo A¯AoA0σθ1/Cluster 0Cluster 1Cluster LFigure 3.1 Illustration of clustered MPCs

66 Channel estimation for high-rate systemsIn the angular domain, on the other hand, the conditional distribution of lgiven l−1(or p(l|l−1)forl &g

3.1 Channel models for high-rate systems 67Correspondingly, the root-mean-square (RMS) delay spread that is defined as the squareroot of the second cen

68 Channel estimation for high-rate systemsTable 3.1 Multipath characteristics for UWB channel modeling providedby the IEEE 802.15.3 Study Group 3a [5

3.1 Channel models for high-rate systems 69where NG=GτQ−1+ 0.5 and G is a sufficiently large integer. A rule of thumb forchoosing G is to ensure that

70 Channel estimation for high-rate systemsFrequencyFrequencyTimeTimeBlock-type Pilot ArrangementComb-type Pilot ArrangementDataPilotFigure 3.2 Block-

3.2 Review of channel estimation techniques 71use of pilots, for achieving high spectral efficiency. This is achieved at the cost of higherimplementati

72 Channel estimation for high-rate systems1716151413121195010876431OFDM SymbolsMFrame Header12 OFDM Symbols6 OFDM SymbolsSequencedaolyaP emarFgniniar

3.2 Review of channel estimation techniques 7361 7372686766636258575621 1270 IndexSubcarrierGuard TonesData TonesNullData Tones Guard TonesDC. . ..

Contents vii5 MIMO techniques for high-rate communications 113Wasim Q. Malik and Andr´e Pollok5.1 Principles of MIMO systems 1135.2 MIMO for ultrawide

74 Channel estimation for high-rate systemswhere [·]∗denotes complex conjugation. Such a spreading maximizes frequency diver-sity by transmitting the

3.2 Review of channel estimation techniques 75data-subcarrier-related subsets of Yn, H, and Vn, respectively, i.e.,ˇYn= [Yn(N −R0), Yn(N −R0+1),...,Yn

76 Channel estimation for high-rate systems3.2.3 LMMSE channel frequency response estimatorThe LMMSE is based on the Bayesian approach to statistical

3.2 Review of channel estimation techniques 77where U is a unitary matrix containing the singular vectors, and Λ is a diagonal matrixcontaining the si

78 Channel estimation for high-rate systemsrequired complexity becomes necessary in the actual implementation of a low-rankLMMSE estimator.3.2.4 ML ch

3.2 Review of channel estimation techniques 79residual error in the initial LS estimate can be further reduced in time-domain as longas Nmis selected

80 Channel estimation for high-rate systemswhere CRis the R-point DCT matrix and WRis an R × R matrix with the formWR=⎛⎜⎜⎜⎝INm0 ··· 000··· 0...

3.2 Review of channel estimation techniques 81In the second step, we apply a simple frequency domain smoothing operation toˆH1and obtainˆH2asˆH2(k) =

82 Channel estimation for high-rate systems00.050.10102001SNR (dB)(a) CM1αhR1,2mse (dB)00.050.10102001SNR (dB)(b) CM2αhR1,2mse (dB)00.050.1−5051001SNR

3.2 Review of channel estimation techniques 83ˆSn(k) = c[ˆun(k) + j ˆvn(k)], where c =√2/2 andˆun(k), ˆvn(k) ∈{+1, −1}. Thus, from(3.12) and (3.13),ˆu