Professional Documents
Culture Documents
3
4
20
21
I. I NTRODUCTION
22
constraints,
EP
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
ro
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
of
IEE
Q1
Manuscript received September 12, 2015; accepted February 10, 2016. The
associate editor coordinating the review of this paper and approving it for
publication was H. H. Nguyen.
K. P. Peppas is with the Department of Informatics and Telecommunications,
University of Peloponnese, Tripoli 22100, Greece (e-mail: peppas@uop.gr).
P. T. Mathiopoulos is with the Department of Informatics and
Telecommunications, National and Kapodistrian University of Athens,
Athens 15784, Greece (e-mail: mathio@di.uoa.gr).
J. Yang is with the School of Information Engineering, Yangzhou University,
Yangzhou, China (e-mail: jingyang905@163.com).
Digital Object Identifier 10.1109/LWC.2016.2530787
2162-2337 2016 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission.
See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
(1)
IEE
where end is the end-to-end instantaneous SNR of the AF CSIassisted multi-hop transmission over arbitrary fading channels,
= T f B/N ln(2) where T f is the fading block length, B the
system bandwidth and the exponentially decay rate of the QoS
violation probability. Note that a large value for corresponds
to a fast decaying rate thus having more stringent QoS requirements, while a smaller one refers to slower decaying rate and
thus looser QoS requirements.
The end-to-end instantaneous SNR, end , is characterized by
the normalized harmonic mean of the instantaneous SNRs at
each hop as [7]
end
N
1
1
=
n
(2)
n=1
130
131
132
133
134
135
136
137
138
139
140
141
142
146
exp [(x + 1)s] s 1 ds = ()(1 + x) .
(4)
0
1
1
ln
R( ) =
exp(s)s 1 M end (s) ds .
Tf B
() 0
(5)
152
153
154
155
d
1 J (2 su), the MGF of
1
end can be 157
d u J0 (2 su) = su
expressed in terms of the MGF of end as
158
M end (u)
du.
(6)
M end (s) =
J0 (2 su)
u
0
EP
R( ) =
119
120
121
122
123
124
125
126
127
128
129
it will be shown next, by using the MGF of the reciprocal end- 143
to-end SNR, end , a generic expression for the exact evaluation 144
of R( ) can be obtained.
145
of
ro
94
(u)
M
end
du ds
exp(s)s 1
J0 (2 su)
u
0
0
M end (u)
1
=
exp(s)s
J0 (2 su)ds du.
u
0
0
The inner integral, with respect to s, can be evaluated in closed
form by employing [5, Eq. (6.643/1)] and [5, Eq. (9.220/2)]
yielding (3), which completes the proof.
It is noted that (3) is valid for arbitrary and correlated fading
channels, as long as the MGF of end exists. For the special
case of uncorrelated n , the EC of the considered system can be
readily evaluated by employing the following corollary.
Corollary: When there is no correlation among all hops, the
EC of AF CSI-assisted multi-hop system is deduced as
1
ln
R( ) =
1 F1 (; 1; u)
Tf B
0
N
N
Mn (u)
Mk (u)du .
(7)
u
161
162
163
164
165
166
167
168
169
170
171
172
173
174
n=1
k=1k=n
176
177
178
179
180
181
182
183
184
185
186
187
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
Fig. 1. EC of multi-hop CSI-assisted AF systems over independent and identically distributed generalized gamma fading channels as a function of the
average SNR per hop for m n = m = 1.5, n = = 2.4, and various numbers
of hops, N .
Q2
EP
197
198
199
200
201
202
IEE
188
189
190
191
192
193
194
195
196
ro
175
of
(9)
where kn and
m n are the shaping parameters of the distribution
and n = kn m n /gn . Similarly to the first case study, substituting (9) and [10, Eq. (15)] in (7), the EC performance has
been obtained and is illustrated in Fig. 2, assuming m 1 = 2,
m 2 = 1.5 and for values for the parameters kn , k1 = k2 = k
that account for two different shadowing scenarios, namely,
frequent heavy shadowing (k = 1.0931) and average shadowing (k = 38.0809) [11]. The accuracy of the analysis is again
Fig. 2. EC of dual-hop CSI-assisted AF systems over independent and nonidentically distributed generalized-K fading channels as a function of the first
hop average SNR, g1 , for m 1 = 2, m 2 = 1.5, and various shadowing scenarios.
225
226
227
228
229
230
exp {1 /[(1 )
1 ] 2 /[(1 )
2 ]}
(m)
1
2 (1 )(
1
2 )m1
2 1 2
Im1
(1 2 )(m1)/2
1
2 (1 )
(10)
231
232
233
234
k
u m+k
4
(m)
k!(m + k)(
1
2 )m+k 1
k=0
u
u
K m+k 2
.
K m+k 2
(1 )
1
(1 )
2
(11)
(m)2
m m
243
244
(m)
(13)
245
246
247
248
1
1,1
[13, Eq. (8.4.45/1)] and K m 2 u/
G 1,2 u 0,0
0
K m1 2 u/
= 0.5 G 3,0
1,3 4u/
1/2,m1/2,m+1/2
[13, Eq. (8.4.24/29)]. Then, by employing [13, Eq. (2.24.1/1)]
and [13, Eq. (8.2.2/15)], R( ) is deduced in closed form as
2m+1
2
3,1 4 1/2+m,1
1
ln
.
R( ) =
G
Tf B
()(m)2 2,3
,m,2m
250
251
252
260
261
262
263
264
265
266
267
268
269
R EFERENCES
270
[1] D. Wu and R. Negi, Effective capacity: A wireless link model for support of quality of service, IEEE Trans. Wireless Commun., vol. 2, no. 4,
pp. 630643, Jul. 2003.
[2] C.-S. Chang, Stability, queue length, and delay of deterministic and
stochastic queueing networks, IEEE Trans. Autom. Control, vol. 39,
no. 5, pp. 913931, May 1994.
[3] S. Efazati and P. Azmi, Effective capacity maximization in multirelay networks with a novel cross layer transmission framework and
power allocation scheme, IEEE Trans. Veh. Technol., vol. 63, no. 4,
pp. 16911702, Nov. 2013.
[4] G. G. Ozcan and M. Gursoy, Effective capacity analysis of fixedgain and variable-gain AF two-way relaying, in Proc. IEEE 78th Veh.
Technol. Conf. (VTC Fall), Las Vegas, NV, USA, 2013, pp. 15.
[5] I. Gradshteyn and I. M. Ryzhik, Tables of Integrals, Series, and Products,
6th ed. New York, NY, USA: Academic, 2000.
[6] M. A. Chaudhry and S. M. Zubair, On a Class of Incomplete Gamma
Functions With Applications, 1st ed. London, U.K.: Chapman &
Hall/CRC, 2002.
[7] M. O. Hasna and M.-S. Alouini, Outage probability of multihop transmission over Nakagami fading channels, IEEE Commun. Lett., vol. 7,
no. 5, pp. 216218, May 2003.
[8] M. D. Renzo, F. Graziosi, and F. Santucci, A unified framework for
performance analysis of CSI-assisted cooperative communications over
fading channels, IEEE Trans. Commun., vol. 57, no. 9, pp. 25512557,
Sep. 2009.
[9] F. Yilmaz, O. Kucur, and M.-S. Alouini, Exact capacity analysis of multihop transmission over amplify-and-forward relay fading channels, in
Proc. IEEE Int. Symp. Pers. Indoor Mobile Radio Commun. (PIMRC),
2010, pp. 22932298.
[10] K. Peppas, F. Lazarakis, A. Alexandridis, and K. Dangakis, Momentsbased analysis of dual-hop amplify-and-forward relaying communications systems over generalised fading channels, IET Commun., vol. 6,
no. 13, pp. 20402047, Sep. 2012.
[11] K. Peppas, Accurate closed-form approximations to generalised-k sum
distributions and applications in the performance analysis of equal-gain
combining receivers, IET Commun., vol. 5, no. 7, pp. 982989, May
2011.
[12] M. Nakagami, The m-distributionA general formula of intensity
distribution of rapid fading, in Statistical Methods in Radio Wave
Propagation. New York, NY, USA: Pergamon, 1960, pp. 336.
[13] A. P. Prudnikov, Y. A. Brychkov, and O. I. Marichev, Integrals and Series
Volume 3: More Special Functions, 1st ed. New York, NY, USA: Gordon
and Breach, 1986.
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
IEE
EP
235
236
237
238
239
240
241
242
249
V. C ONCLUSIONS
ro
M end (u) =
253
254
255
256
257
258
259
of
Substituting the derivative of M end (u) in (3), the EC performance has been obtained and is depicted in Fig. 3, for
the cases where m = 2 and {0, 0.8}. The infinite series in
(11) are truncated to 40 terms for numerical accuracy. Again
here numerically evaluated results and computer simulations
are in excellent agreement. It is also noted that the impact of
correlation on the EC performance is rather small.
QUERIES
IEE
EP
ro
of
Q1: Please be advised that per instructions from the Communications Society this proof was formatted in Times Roman font and
therefore some of the fonts will appear different from the fonts in your originally submitted manuscript. For instance, the
math calligraphy font may appear different due to usage of the usepackage[mathcal]euscript. We are no longer permitted to
use Computer Modern fonts.
Q2: Note that if you require corrections/changes to tables or figures, you must supply the revised files, as these items are not
edited for you.