RESEARCH OpenAccess Reliabilityevaluationof5GC/U-plane … · 2017. 8. 29. · afford this huge...

Yan and Fang EURASIP Journal onWireless Communications and Networking 2014, 2014:127http://jwcn.eurasipjournals.com/content/2014/1/127

RESEARCH Open Access

Reliability evaluation of 5G C/U-planedecoupled architecture for high-speed railwayLi Yan*† and Xuming Fang†

Abstract

To facilitate the mobility of heterogeneous networks, control plane (C-plane) and user plane (U-plane) decoupledarchitecture is being considered by the fifth generation (5G) wireless communication network, in which relativelycrucial C-plane is expanded and kept at dependable lower frequency bands to guarantee transmission reliability andthe corresponding U-plane is moved to available higher frequency bands to boost capacity. Moreover, we apply thisarchitecture to future professional high-speed railway wireless communication system to fulfill the wireless accessdesire of train passengers. However, for such emerging architecture, there still exist many problems to be solved toguarantee the reliable transmission. In this article, the problem of how to appropriately evaluate the transmissionreliability of C/U-plane decoupled architecture is investigated. Due to the lack of ability to reflect the importance ofC-plane, conventional outage probability cannot properly indicate the transmission reliability of C/U-plane decoupledarchitecture whose primary design consideration is that C-plane more heavily affects the transmission reliabilitythereby being kept at dependable lower frequency bands. Based on this, a novel indicator named unreliability factor(URF) is proposed. Theoretical analysis and simulation results demonstrate that URF can exactly highlight the effects ofC-plane on the entire transmission process. Hence, it is more appropriate to employ URF as the indicator to evaluatethe transmission reliability of C/U-plane decoupled architecture.

Keywords: 5G; C/U-plane decoupled architecture; Transmission reliability; Reliability evaluation; High-speed railway

IntroductionTo cater to the exponentially increasing traffic vol-ume requirements in public mobile network, higher fre-quency bands with wider spectrum are exploited tofurther extend the capacity of Long-Term Evolution(LTE) network. Unfortunately, compared with lower fre-quency bands, higher frequency bands suffer from sev-erer propagation loss which seriously limits the coverage.Hence, cells working at higher frequency bands are calledsmall cells. On the account of mobility performance,small cells are usually overlaid on the coverage ofmacro cells that use lower frequency bands, which formsheterogeneous network. However, as deployment getsincreasingly dense, the huge redundant control signalinginteraction caused by frequent handovers between smalland macro cells reduces the efficiency of heterogeneousnetworks. In order to mitigate this situation, C/U-plane

*Correspondence: [email protected]†Equal contributorsInstitute of Mobile Communications, Southwest Jiaotong University, Chengdu610031, China

decoupled architecture is proposed for upcoming 5Gwireless communication network [1,2]. In this architec-ture, the relatively important C-plane is extended and keptat lower frequency bands of macro cells. In addition, themain capacity demander U-plane is moved to the smallcells using available higher frequency bands with widerspectrum. With considerable coverage of macro cells,much fewer handovers happen to the C-plane comparedto conventional coupled architecture of heterogeneousnetworks. Therefore, under a macro cell, the handoverprocess is just simplified to the U-plane handover, whichsaves lots of control signaling interaction.With the remarkable success of railway industry and

wireless technologies, train passengers are eager to enjoythe Internet during long-distance journey. However, exist-ing narrowband GSM-R (Global System for Mobile Com-munication for Railway) which is originally designed totransmit low-volume train control information cannotafford this huge capacity. In order to provide broadband

© 2014 Yan and Fang; licensee Springer. This is an Open Access article distributed under the terms of the Creative CommonsAttribution License (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproductionin any medium, provided the original work is properly credited.

mailto:[email protected]

http://creativecommons.org/licenses/by/4.0

Yan and Fang EURASIP Journal onWireless Communications and Networking 2014, 2014:127 of 11http://jwcn.eurasipjournals.com/content/2014/1/127

services for train passengers, railway wireless communica-tion system is confronting the evolution to its next genera-tion.With enhanced robustness, LTE is the most potentialone for the evolution [3,4]. Based on this, in our previouswork [5,6], we applied C/U-plane decoupled architectureto future professional high-speed railway wireless com-munication system to fulfill the wireless access desire oftrain passengers and the network architecture is depictedin Figure 1. With consideration of the return on invest-ment, no public mobile network operator would like tooffer a thorough coverage for most sparsely populatedrailway scenarios. Nevertheless, in the C/U-plane decou-pled railway wireless network, both train control infor-mation and passengers’ services can be transmitted. Thisprovides a better choice for passengers to get a higher-quality service. Furthermore, on account of the severechallenges faced by high-speed railway scenario such aslarge penetration loss and group handover, it is difficult forpassengers inside the train to hold a dependable connec-tion with outside base station directly. Hence, as shownin Figure 1, mobile relay (MR) and access point (AP),which are connected with each other via optical fiber,are separately mounted on the roofs outside and insidethe train [7]. This also provides a potential way to saveoperators’ infrastructure cost by converging different airinterface technologies (e.g., LTE/3G/2G/Wi-Fi) on the APand employing LTE on the link between MR and the out-side base stations. Inside the train, passengers connect tothe AP and then their data are forwarded to outside basestations via MR. In the C/U-plane decoupled architecture,C-plane signaling and U-plane data of train passengers’services are separately transmitted by the macro and smallcells. The macro cell employs lower frequency bands toprovide good connectivity and mobility of C-plane and

the small cell works at available higher frequency bandsto expand capacity. On account of transmission reliability,crucial train control information is entirely kept at lowerfrequency bands without decoupling. As a result, traincontrol information and data of passengers are separatelytransmitted through different nodes. In this way, the arti-ficial interference from the passengers is avoided, therebyenhancing the security to some extent. However, this isalready out of the scope of our study. In addition, this arti-cle focuses on the effect of decoupled C/U-plane on thetransmission reliability of passengers’ services.For clarity, the booming LTE network is employed as

the benchmark system for following analysis. In LTE net-work, C-plane is responsible for essential control opera-tions such as broadcasting system information, networkattaches, paging ,andmobility management [8]. Moreover,the functionality of U-plane is to forward user data flow.Definitely, without a reliable C-plane, the U-plane can-not work properly. From the perspective of air interfaces,without reliably transmitted Physical Downlink ControlChannel (PDCCH) that accommodates control informa-tion to indicate ‘who’ the data are for, ‘what’ data aresent, and ‘how’ the data are sent on Physical Down-link Shared Channel (PDSCH), the user data cannot becorrectly decoded. Considering the above mentioned in[9,10], the crucial C-plane instead of U-plane is keptat dependable lower frequency bands so that the trans-mission reliability of C/U-plane decoupled architecturecould be well guaranteed. However, the performanceevaluation of this architecture has not been well stud-ied. Actually, for such emerging network architecture,there still exist many problems to be solved to guar-antee the reliable transmission. For instance, Dopplereffect is always a severe challenge for high-speed railway

Figure 1 C/U-plane decoupled architecture in high-speed railway.


scenario. While in the C/U-plane decoupled architecture,C-plane and U-plane of the same user are transmittedat different frequencies thereby facing different Dopplershifts and Doppler spreads. Hence, the Doppler effectmay get even worse. Fortunately, railways are mostly builtin wide suburban or viaduct environment where multi-paths can be neglected and the wireless channel canbe regarded as LOS [4]. Thus, there almost exists noDoppler spread in high-speed railway scenario. Accordingto [4,6], with a known train’s location and velocity sup-plied by the communication-based train control (CBTC)system, it can be assumed that Doppler shifts are sepa-rately compensated for C-plane and U-plane. Hence, howto appropriately evaluate the transmission reliability ofthis decoupled architecture becomes an urgent problem,which will greatly impact subsequent research directionson performance enhancement.In wireless networks, the outage probability defined as

the probability that the received signal quality is lowerthan some threshold is a popularly used indicator toreflect the transmission reliability [11,12]. Under this def-inition, the outage probability of C/U-plane decoupledarchitecture can be expressed as the complementary prob-ability of an event that both signal qualities of C-planeand U-plane are larger than the outage threshold. Obvi-ously, from the view of air interface reliability, the effectsof C-plane and U-plane on outage probability are virtu-ally equal. That is to say, due to the lack of ability toreflect the importance of C-plane, the conventional out-age probability cannot properly indicate the transmissionreliability of C/U-plane decoupled architecture whose pri-mary design consideration is that C-plane more heavilyaffects the transmission reliability thereby being kept atdependable lower frequency bands.To facilitate the presentation, except for special behav-

iors such as paging and handover, we only take the generalcommunication process as an example to qualitativelyanalyze the reliability relationship between C-plane andU-plane. In terms of theoretical analysis based on signalquality of air interface, the analysis procedures and resultsof uplink and downlink are the same. Hence, for simplicity,the following analysis is just on the basis of downlink. Inthe general communication process of C/U-plane decou-pled architecture, PDSCH served by the small cell carriesuser date and PDCCH provided by the macro trans-mits corresponding control information such as transmis-sion format to help receiver correctly decode the dataon PDSCH. For PDCCH, its symbol error rate (SER) isdirectly caused by its poorly received signal quality. Nev-ertheless, these errors of PDCCH will badly affect thedecoding correctness of data on PDSCH [13]. As a con-sequence for PDSCH, its errors result from two aspects,some of which are caused by its own poor signal qual-ity and others are induced from the errors of PDCCH.

In practice, if SER of PDCCH exceeds some threshold,then no matter how well the signal quality of PDSCH is,the receiver cannot correctly decode the data on PDSCH.Based on this, a more appropriate indicator named unre-liability factor (URF) which can highlight the importanceof C-plane is proposed to evaluate the transmission relia-bility of C/U-plane decoupled architecture.The rest of this article is arranged as follows. ‘Radio

propagation model’ section gives the propagation modelfor C/U-plane decoupled architecture. ‘System outageprobability of C/U-plane decoupled architecture’ sectionproves that conventional outage probability cannot prop-erly indicate the transmission reliability of this archi-tecture. ‘System reliability-based reliability evaluationmethod’ section describes the proposed indicator URFand its appropriateness in evaluating the transmis-sion reliability of this architecture. Finally, ‘Conclusions’section concludes the whole article.

Radio propagationmodelSIRmodelSince the communication system in high-speed railwayscenario has a linear topology, the macro and small cellsof C/U-plane decoupled architecture are deployed on astraight line with vertical distance ofD to the rail as shownin Figure 2. Suppose the base stations’ radiation is omni-directional, and the coverage radiuses of macro and smallcell are RC and RU respectively. Besides, the abscissa axiscoinciding with the driving direction is set to facilitate theexpression of train travel distance d. Without loss of gen-erality, two macro cells are considered in the followinganalysis, i.e., the analysis scope of d is from 0 to 4RC-2aC, where aC is the overlapping area distance of twomacro cells and as shown in Figure 2, the overlapping areadistance of two small cells is denoted by aU.Suppose the train starts from the origin and travels

through distance of d, then the C-plane signal propagationdistance between the train and macro cell is

xC(d)=√(

d−⌊

d2RC−aC

⌋·(2RC−aC)−

(RC− aC

2

))2+D2

(1)

where �·� denotes rounding down operation.Correspondingly, the U-plane signal propagation dis-

tance between the train and current serving small cell canbe expressed as

xU(d)=√(

d−⌊

d2RU−aU

⌋·(2RU−aU)−

(RU− aU

2

))2+D2

(2)


. . . . . . . .0

RC

DRU . . . . . . . .

d/kmaU aC

4RC-2aC

Figure 2 Geometric sketch for analysis.

Generally, the propagation attenuation ismodeled as theproduct of path loss and a log-normal component repre-senting shadow fading [14]. Then, with transmit power ofPt, the received signal-to-interference ratio (SIR) can beexpressed as

SIR(x) = Pt · PL(x) × 10−ε/10

I(3)

where PL(x) is the path loss with propagation distance ofx; ε is the decibel attenuation due to shadow fading withzero mean and standard variance σ ; and the co-channelinterference I is given by

I =∑NeNB

n=1Pt · PL(xn) (4)

where NeNB represents the number of co-channeleNodeBs.Consequently, the received signal quality of C-plane and

U-plane can be separately computed as

SIRC(xC(d)) = Pt,C · PLC(xC(d)) × 10−εC/10

IC(5)

SIRU(xU(d)) = Pt,U · PLU(xU(d)) × 10−εU/10

IU(6)

For clarity, subscripts of parameters that relate to C-plane are set to C. And for U-plane they are set to U.However, in fact Pt, PL(x), ε, and I are determined by cur-rent serving base station while not the plane, that is, Pt,C,PLC(x), εC and IC are the properties of the macro cellwhich serves the C-plane. With subscript U, they are theproperties of the small cell which provides U-plane. Theabove also applies to the following expressions.

Cross-correlated shadow fadingShadow fading is a large-scale phenomenon whichdepends on the surrounding communication environ-ment. Although in C/U-plane decoupled architecture themacro and small cell are deployed at different loca-tions, they simultaneously serve the same train user.Hence, there exist some common components betweenthe propagation paths of macro cell and small cell sig-nals, especially for the area near the train. As a conse-quence, the shadow fading of macro and small cells arecross-correlated. Based on [15,16], cross-correlation ofshadow fading can be explained by a partial overlap ofthe large-scale propagation medium as shown in Figure 3,and non-overlapping propagation areas are consideredindependent.

Then, the shadow fading can be decomposed as

εC = W + WC

εU = W + WU(7)

whereW ,WC, andWU are independent Gaussian randomvariables with zero mean and standard variances a, b, andc, respectively, and they satisfy

σC2 = a2 + b2

σU2 = a2 + c2

E [εCεU] = a2 = ρC,UσCσU

(8)

According to [13], the cross-correlation coefficient ofshadow fading can be modeled as

ρC,U(d)=ρU,C(d)=

⎧⎪⎨⎪⎩√

min(xC(d),xU(d)))max(xC(d),xU(d))

, 0≤�≤�T

(�T

�

)γ√min(xC(d),xU(d))max(xC(d),xU(d))

, �T≤�≤π

(9)

where γ is referred to as a parameter determined by thesize and height of terrain and the height of base stationand is generally set to 0.3 [17]; �T corresponds to theangle threshold that depends on the serial de-correlationdistance dcor and is defined as

�T = 2 arctan(

dcor2min (xC(d), xU(d))

)(10)

Moreover, � of (9) is the angle between the propagationpaths of C-plane and U-plane signals as shown in Figure 3.As the site-to-site distance between macro and small cellscan be calculated via the following equation

xsep(d)=

⎧⎪⎪⎨⎪⎪⎩(2RU−aU)·

(⌊√xC2(d)−D2

2RU−aU

⌋−1

2

), d=0, d=4RC−2aC

(2RU−aU)·(⌊√

xC2(d)−D2

2RU−aU

⌋+ 1

2

), otherwise

(11)

the included angle � can be derived as

�(d) = arccos(xC2(d) + xU2(d) − xsep2(d)

2xC(d)xU(d)

)(12)


W

WUWC

xsep(d)

xU(d)

xC(d)

Figure 3 Cross-correlated shadow fading.

Then, substitute (12) into (9), ρ(C,U) is obtained, withwhich the standard variances, a(d), b(d), and C(d) in (8)can be worked out.

System outage probability of C/U-plane decoupledarchitectureGenerally, outage probability is used to evaluate transmis-sion reliability of wireless communication networks. It isdefined as the probability that the received SIR is lowerthan some threshold. Based on this, the system outageprobability of C/U-plane decoupled architecture can bederived as

Pout,cov = 1 − P[SIRC > SIRC

th, SIRU > SIRUth]

= 1−P[10lg

Pt,C ·PLC(xC(d))

IC−W−WC>SIRC

th,

10 lgPt,U · PLU(xU(d))

IU− W − WU > SIRU

th]

= 1 − 1(√2π)3a(d)b(d)c(d)

∫ +∞

−∞e−

W22a2(d)

×⎛⎝∫ 10 lg Pt,C ·PLC(xC(d))

IC−W−SIRCth

−∞e−

WC2

2b2(d) dWC

∫ 10 lg Pt,U ·PLU(xU(d))

IU−W−SIRUth

−∞e−

WU2

2c2(d) dWU

⎞⎠ dW

= 1 − 1√2π

∫ +∞

−∞e−

t22 �

×⎛⎝10 lg Pt,C ·PLC(xC(d))

IC − a(d)t − SIRCth

b(d)

⎞⎠

× �

⎛⎝10 lg Pt,U·PLU(xU(d))

IU − a(d)t − SIRUth

c(d)

⎞⎠ dt

(13)

where SIRCth and SIRUth are the decibel outage thresholds

of C-plane and U-plane, respectively, and

�(x) = 1√2π

∫ x

−∞e−

t22 dt (14)

Considering the fairness of comparison, outage prob-abilities of coupled macro and small cell architecturesare also computed as the complementary probabilityof an event that both the signal qualities of C-planeand U-plane are larger than the outage thresholds,that is,

Pout,m = 1 −∫ +∞

−∞e− t2

2√2π

�

×⎛⎝10 lg Pt,C·PLC(xC(d))

IC − a(d)t − SIRCth

b(d)

⎞⎠

2

dt

(15)

Pout,s = 1 −∫ +∞

−∞e− t2

2√2π

�

×⎛⎝10 lg Pt,U·PLU(xU(d))

IU − a(d)t − SIRUth

c(d)

⎞⎠

2

dt

(16)

Based on the above modeling, a numerical simulation isconducted and the results are shown in Figures 4 and 5,where Figure 4 depicts the outage probabilities of differ-ent network architectures and Figure 5 corresponds tothe trend in Figure 4. Detailed simulation parameters arelisted in Table 1.From the above theoretical analysis and simulation, it

is easy to find that in terms of the air interface reliability,the effects of C-plane and U-plane on outage probabil-ity are virtually equal. As shown in Figures 4 and 5,the transmission performance of C/U-plane decoupled


0 0.5 1 1.5 2 2.5 3 3.50

0.02

0.04

0.06

0.08

0.1

0.12

0.14

0.16

distance d /km

outa

gepr

oabi

lity

C/U plane decoupled architecturemacro cell of coupled architecturesmall cell of coupled architectureC/U-plane reversed

Ad=1.35km

Bd=2.25km

Figure 4 Reliability evaluation under conventional outage probability.

architecture is just the averaging of that of coupled macrocell and small cell architectures. For instance, at the scope(A, B), although C-plane cannot be reliably transmitted,the entire system performance of C/U-plane decoupledarchitecture is not so poor. While at the center of macrocell, no matter how reliably C-plane performs, the entire

transmission reliability of C/U-plane decoupled architec-ture is badly affected by the poor U-plane. Besides, ifC-plane and U-plane are reversed, i.e., U-plane is kept atmacro cell and C-plane is moved to small cell, the resultof outage probability has not changed. This thoroughlyreveals that the conventional system outage probability

0 0.5 1 1.5 2 2.5 3 3.50

0.02

0.04

0.06

0.08

0.1

0.12

0.14

0.16

distance d /km

tren

d of

out

age

prob

abili

ty

C/U plane decoupled architecture

macro cell of coupled architecturesmall cell of coupled architecture

C/U-plane reversed

Figure 5 Trends of outage probability in Figure 4.


Table 1 Simulation parameters [18]

Parameters Values

Frequency of macro cell 2 GHz

Frequency of small cell 5 GHz

Path loss model of macro cell PLC Hata

Path loss model of small cell PLU M.2135

Transmit power of macro cell Pt,C 43 dBm

Transmit power of small cell Pt,U 33 dBm

Radius of macro cell RC 1 km

Radius of small cell RU 0.25 km

Overlapping area distance of macro cells aC 0.2 km

Overlapping area distance of small cells aU 0.05 km

Distance between base station and rail D 30 m

C-plane SER threshold thC 10−6

U-plane SER threshold thU 10−2

α(thC) 10−4

Correlation distance dcor 100 m [19]

Modulation scheme of U-plane 16QAM

Standard variance of C-plane shadowing σC 6 dB

Standard variance of U-plane shadowing σU 8 dB

cannot convey the primary design consideration of C/U-plane decoupled architecture that C-plane more heavilyaffects transmission reliability than U-plane thereby beingkept at dependable lower frequency bands. Hence, it isnot proper to employ simple outage probability as theindicator to evaluate the transmission reliability of thisdecoupled architecture.

System reliability-based reliability evaluationmethodReliability relationship between C-plane and U-planeAs mentioned before, in the general communication pro-cess, PDSCH served by the small cell carries user data,and the corresponding PDCCH served by the macro celltransmits control information such as modulation andcoding scheme of PDSCH to help the receiver correctlydecode the data on PDSCH. For PDCCH, its transmis-sion errors are just caused by its poor SIRC. However, forPDSCH there are two aspects resulting in data errors. Asshown in Figure 6, some errors of PDSCH are due to thepoor received signal quality, while others are induced fromthe errors of PDCCH.Maybe, these parts of symbols are ofhigh signal quality, but they cannot be correctly decodedbecause of the inaccurate transmission format indicationon PDCCH. Hence, it is reasonable to believe that if SERof PDCCH exceeds some tolerable value, then no matterhow well the signal quality of PDSCH is, the data cannot

be correctly received. This exactly reveals why C-plane isregarded more crucial than U-plane. Although how theerrors of PDCCH affects the data receiving on PDSCHis beyond the scope of this study, a mapping function isrequired to describe the relationship between SERC andSERU/C that is

SERU/C = α (SERC) (17)

where α function is supposed to be monotone increas-ing with definition field of SERC from 0 to 1 and range ofSERU/C from 0 to 1 as well. However, the exact expressionof α function depends on the system settings and is out ofour study scope.

Unreliability factorThrough the above discussion, it can be concluded thatfor C/U-plane decoupled architecture, a proper indicatoris needed that can exactly highlight the importance of C-plane. Based on this, a novel indicator named unreliabilityfactor, URF, is proposed to appropriately evaluate thesystem transmission reliability of C/U-plane decoupledarchitecture, which is defined as

URF ={P(SERU > thU), SERC ≤ thC1, SERC > thC

(18)

It is obvious that URF has the ability to reflect theimportance of C-plane. When the SER of crucial C-planeis beyond the threshold thC, in spite of the transmissionperformance of PDSCH, URF is directly set to 1. Thisexactly conforms to the previous analysis result that if SERof PDCCH exceeds some tolerable value, then no matterhow well the signal quality of PDSCH is, the data cannotbe correctly received. While if C-plane is reliably trans-mitted, the value of URF will depend on the SER outageprobability of PDSCH which is defined as the probabilitythat the SER is higher than some threshold [20]. Practi-cally, the SER outage probability of PDSCH is much lowerthan 1. Hence, at the point of SERC = thC, URF is notrightly continuous thereby not a probability function. As amatter of fact, URF can be interpreted as a kind of indica-tor, which equals to a complex and reasonable probabilityvalue.From the SIR modeling in ‘Radio propagation model’

section, it can be found that due to the cross-correlationof shadow fading between the macro and small cells, SIRCand SIRU are not absolutely independent. Then, it seemsthat SERU/C and SERU/SIRU are correlated. Fortunately,there exists the principle of conditional independence thattwo random variables X and Y are conditionally indepen-


SERC

PDCCH

SIRCSERU/C

UU/SIR

SER SIRU

PDSCH

Figure 6 Reliability relationship between C-plane and U-plane.

dent if given Z as shown in Figure 7 [21], that is, with givenZ, for any real number x, y, and z, the following equationis satisfied:

PX,Y/Z (x, y/z) = PX/Z (x/z)PY/Z (y/z) (19)

Based on this, since the relationship between SIRC andSERC is known, SIRC and SERU/C are conditionally inde-pendent. Therefore, SERU/C and SERU/SIRU are condition-ally independent and the total SER of U-plane can bederived as

SERU = SERU/C+SERU/SIRU−SERU/C·SERU/SIRU (20)

In LTE network, the modulation scheme for C-plane isQPSK, and its SER can be expressed as [22]

SERC = 2Q(√

2SIRC)(

1 − 12Q(√

2SIRC))

(21)

where Q function is defined by

Q(x) = 1√2π

∫ ∞

xe−

t22 dt (22)

Figure 7 Conditional independence.

Then, the transition condition in the definition of URF canbe further derived as

2Q(√

2SIRC)(

1 − 12Q(√

2SIRC))

≤ thC

⇒ SIRC ≥(Q−1 (1 − √

1 − thC))2

2

⇒ εC≤−10 lg(IC(Q−1 (1−√

1−thC))2

2Pt,CPLC(xC(d))

)= C (xC(d))

(23)

where Q−1 is the inverse function of Q function, andC (xC(d)) is defined to simplify the following expressions.For U-plane with modulation scheme of M-QAM,

SERU/SIRU is given by [22]

SERU/SIRU = 4(1 − 1√

M

)Q(√

3log2MM − 1

SIRU

)(24)

Then, the SER outage probability of U-plane in thedefinition of URF can be further derived as

SERU/C + SERU/SIRU − SERU/C · SERU/SIRU ≥ thU

⇒ α (SERC) + (1 − α (SERC)) SERU/SIRU ≥ thU

⇒ SERU/SIRU ≥ thU − α (SERC)

1 − α (SERC)= thU − 1

1 − α (SERC)+ 1

(25)

As in practice , thU < 1, then

thU − 11 − α (SERC)

+ 1 ≥ thU − 11 − α (thC)

+ 1 = thU − α (thC)

1 − α (thC)

(26)


Under more stringent condition, SERU/SIRU of (19) can befurther expressed as

SERU/SIRU ≥ thU − α (thC)

1 − α (thC)

⇒ 4(1 − 1√

M

)Q(√

3log2MM − 1

SIRU

)≥ thU − α (thC)

1 − α (thC)

⇒ SIRU≤ (M−1)3log2M

⎛⎝Q−1

⎛⎝ thU − α (thC)

4(1 − 1√

M

)(1−α (thC))

⎞⎠⎞⎠2

⇒ εU ≥ −10 lg

⎛⎜⎝ IU (M − 1)3log2M · Pt,U · PLU(xU(d))

×⎛⎝Q−1


4(1 − 1√

M

)(1 − α (thC))

⎞⎠⎞⎠

2⎞⎟⎠

⇒ εU ≥ U(xU(d))

(27)

where U (xU(d)) is defined to simplify the followingexpressions.Hence, the definition of URF in (18) can be rewritten as

URF ={P (WU>U (xU(d))−W ) , WC≤C (xC(d))−W

1, WC > C (xC(d)) − W(28)

Then, the average URF can be obtained as

URFave = 1(√2π)3a(d)b(d)c(d)

∫ +∞

−∞e−

W22a2(d)

×(∫ C(xC(d))−W

−∞P (WU > U (xU(d)) − W )

× e−WC

2

2b2(d) dWC +∫ ∞

C(xC(d))−We−

WC2

2b2(d) dWC

)dW

= 1√2πa(d)

∫ +∞

−∞e−

W22a2(d)

(Q(U (xU(d)) − W

c(d)

)

×�

(C (xC(d))−W

b(d)

)+Q(C (xC(d))−W

b(d)

))dW

(29)

Under the proposed indicator, the average URF of cou-pled macro cell network architecture can be presented as

URFmave= 1√2πa(d)

∫ +∞

−∞e−

W22a2(d)

(Q(UC (xC(d))−W

b(d)

)

×�

(CC(xC(d))−W

b(d)

)+Q(CC(xC(d))−W

b(d)

))dW

(30)

where

UC(xC(d))=−10 lg

⎛⎜⎝ IC (M − 1)3log2M · Pt,C · PLC(xC(d))

×⎛⎝Q−1


4(1 − 1√

M

)(1 − α (thC))

⎞⎠⎞⎠

2⎞⎟⎠

×CC(xC(d))=−10lg(IC(Q−1(1−√

1−thC))2

2Pt,CPLC(xC(d))

)

(31)

And for coupled small cell network architecture, theaverage URF is

URFsave= 1√2πa(d)

∫ +∞

−∞e−

W22a2(d)

(Q(UU (xU(d))−W

c(d)

)

×�

(CU(xU(d))−W

c(d)

)+Q(CU(xU(d))−W

c(d)

))dW

(32)

where

UU (xU(d)) = −10 lg

⎛⎜⎝ IU (M − 1)3log2M · Pt,U · PLU(xU(d))

×⎛⎝Q−1


4(1 − 1√

M

)(1 − α (thC))

⎞⎠⎞⎠

2⎞⎟⎠

CU (xU(d)) = −10 lg(IU(Q−1 (1 − √

1 − thC))2

2Pt,UPLU(xU(d))

)

(33)

On the basis of above theoretical analysis, numericalsimulation is performed as shown in Figures 8 and 9.The simulation parameters are set as in Table 1. FromFigure 8, at the scope (A, B), the entire performance ofC/U-plane decoupled architecture is badly degraded bythe poor transmission of C-plane, which just like that of


0 0.5 1 1.5 2 2.5 3 3.50

0.01

0.02

0.03

0.04

0.05

0.06

distance d /km

UR

Fave


Ad=1.35km

Bd=2.25km

Figure 8 URFave of different network architectures.

coupled macro cell architecture. While at the center ofmacro cell, thanks to the well-transmitted C-plane, theentire performance of C/U-plane decoupled architectureis not so badly affected by the poor U-plane. Besides,for the reversed case that C-plane is moved to small celland U-plane is kept at macro cell, the entire performancegets absolutely different. For example, at the scope (A, B)the reversed C/U-plane decoupled architecture performsbetter. But in most situations, it possesses much poorer

transmission reliability than normal C/U-plane decoupledarchitecture. From the trend of URF in Figure 9, it is alsoobvious that normal C/U-plane decoupled architecturecan provide more reliable transmission than the reversedcase, which exactly conforms to the consideration dur-ing the design of C/U-plane decoupled architecture thatC-plane more heavily affects the transmission reliabilitythan U-plane thereby being kept at dependable lower fre-quency bands. Therefore, under the proposed indicator

0 0.5 1 1.5 2 2.5 3 3.50

0.005

0.01

0.015

0.02

0.025

0.03

0.035

0.04

0.045

0.05

distance d /km

trend

of U

RFav

e


Figure 9 Trends of URFave in Figure 8.


URF, the importance of C-plane is completely reflectedand it is more proper to use URF as the indicator toevaluate the transmission reliability of C/U-plane decou-pled architecture.

ConclusionsFor upcoming 5G wireless communication system, C/U-plane decoupled architecture is a potential way to not onlyexpand capacity but also to prevent unnecessary controlsignaling overhead. In addition, we apply this architec-ture to future professional high-speed railway wirelesscommunication system to fulfill the wireless access desireof train passengers during long-distance journey. How-ever, how to properly evaluate the system transmissionreliability of C/U-plane decoupled architecture becomesan urgent problem, which will impact the subsequentresearch direction on performance enhancement. It hasbeen proved that the conventional outage probabilitycannot convey the primary design consideration of thisdecoupled architecture that C-plane more heavily affectsthe entire transmission reliability than U-plane therebybeing kept at lower frequency bands. Based on this, anovel indicator URF is proposed. The theoretical analy-sis and numerical simulation results have confirmed thatURF performs more properly in evaluating the entiresystem transmission reliability of C/U-plane decoupledarchitecture.

Competing interestsThe authors declare that they have no competing interests.

AcknowledgementsThe work of the authors was supported partially by the 973 Program underGrant 2012CB316100, NSFC under Grant 61032002, and the Program forDevelopment of Science and Technology of China Railway Corporation underGrant 2013X016-A.

Received: 4 April 2014 Accepted: 27 July 2014Published: 9 August 2014

References1. T Nakamura, S Nagata, A Benjebbour, Y Kishiyama, T Hai, S Xiaodong,

Y Ning, L Nan, Trends in small cell enhancements in LTE advanced. IEEECommun. Mag. 51(2), 98–105 (2013)

2. BA Bjerke, LTE-advanced and the evolution of LTE deployments. IEEEWireless Commun. Mag. 18(5), 4–5 (2011)

3. B Dusza, C Ide, P-B Bok, C Wietfeld, Optimized cross-layer protocol choicesfor LTE in high-speed vehicular environments, in Proceedings of the 9thIEEE International Wireless Communications andMobile ComputingConference (Sardinia, 1–5 July 2013), pp. 1046–1051

4. W Luo, X Fang, M Cheng, Y Zhao, Efficient multiple-groupmultiple-antenna (MGMA) scheme for high-speed railway viaducts. IEEETrans. Veh. Technol. 62(6), 2558–2569 (2013)

5. L Yan, X Fang, Decoupled wireless network architecture for high-speedrailway, in Proceedings of IEEE International Workshop on HighMobilityWireless Communications, (Shanghai, 1–3 Nov 2013), pp. 96–100

6. H Song, X Fang, L Yan, Handover scheme for 5G C/U plane splitheterogeneous network in high-speed railway. IEEE Trans. Veh. Technol(2014). doi:10.1109/TVT.2014.2315231

7. X Zhu, S Chen, H Hu, X Su, Y Shi, TDD-based mobile communicationsolutions for high-speed railway scenarios. IEEE Wireless Commun. 20(6),22–19 (2013)

8. E Dahlman, S Parkvall, J Sköld, LTE/LTE-Advanced for Mobile Broadband.(Academic Press, Oxford, 2011)

9. X Xiuqiang, H Gaoning, Z Shunqing, C Yan, X Shugong, On functionalityseparation for green mobile networks: concept study over LTE. IEEECommun. Mag. 51(5), 82–90 (2013)

10. H Ishii, Y Kishiyama, H Takahashi, A novel architecture for LTE-B:C-plane/U-plane split and phantom cell concept, in Proceedings of IEEEGlobecomWorkshopsWorkshop on Emerging Technologies forLTE-Advanced and Beyond-4G (Anaheim, 3–7 Dec 2012), pp. 624–630

11. AJ Viterbi, AM Viterbi, KS Gilhousen, E Zehavi, Soft handoff extends CDMAcell coverage and increases reverse link capacity. Sel Area Commun. 12,1281–1288 (1994). doi:10.1109/49.329346

12. MO Hasna, MS Alouini, Outage probability of multihop transmission overNakagami fading channels. IEEE Commun. Lett. 7(5), 216–218 (2003)

13. J Liu, R Love, K Stewart, ME Buckley, Design and analysis of LTE physicaldownlink control channel, in Proceedings of the 69th IEEE internationalConference on Vehicular Technology (Barcelona, 26–29 April 2009), pp. 1–5

14. T Rappaoort,Wireless Communications: Principles and Practice.(Prentice-Hall, Englewood Cliffs, 2001)

15. SS Szyszkowicz, H Yanikomeroglu, JS Thompson, On the feasibility ofwireless shadowing correlation models. IEEE Trans. Veh. Technol. 59(9),4222–4236 (2010)

16. J Weitzen, T Lowe, Measurement of angular and distance correlationproperties of log-normal shadowing at 1900 MHz and its application todesign of PCS systems. IEEE Trans. Veh. Technol. 51(2), 265–273 (2002)

17. X Yang, S Ghaheri-Niri, R Tafazolli, Downlink soft handover gain in CDMAcellular network with cross-correlated shadowing, in Proceedings of the54th IEEE international Conference on Vehicular Technology (Atlantic City,7–11 Oct 2001), pp. 276–280

18. W Luo, R Zhang, X Fang, A CoMP soft handover scheme for LTE systemsin high speed railway. EURASIP J. Wireless Commun. Netw. 1, 196 (2012).doi:10.1186/1687-1499-2012-196

19. J Salo, L Vuokko, HM El-Sallabi, P Vainikainen, An additive model as aphysical basis for shadow fading. IEEE Trans. Veh. Technol. 56(1), 13–26(2007)

20. P Wang, P Kam, Feedback power control with bit error outage probabilityQoS measure on the Rayleigh fading channel. IEEE Trans. Commun. 61(4),1621–1631 (2013)

21. AP Dawid, Conditional independence in statistical theory. J. R. Stat. Soc.Series B (Methodological). 41(1), 1–31 (1979)

22. J Proakis, M Salehi, Digital Communications. (McGraw Hill, New York, 2000)

doi:10.1186/1687-1499-2014-127Cite this article as: Yan and Fang: Reliability evaluation of 5G C/U-planedecoupled architecture for high-speed railway. EURASIP Journal onWirelessCommunications and Networking 2014 2014:127.

Submit your manuscript to a journal and benefi t from:

7 Convenient online submission

7 Rigorous peer review

7 Immediate publication on acceptance

7 Open access: articles freely available online

7 High visibility within the fi eld

7 Retaining the copyright to your article

Submit your next manuscript at 7 springeropen.com

RESEARCH OpenAccess Reliabilityevaluationof5GC/U-plane … · 2017. 8. 29. · afford this huge...

Documents

Transcript of RESEARCH OpenAccess Reliabilityevaluationof5GC/U-plane … · 2017. 8. 29. · afford this huge...