1096 IEEE COMMUNICATIONS LETTERS, VOL. 20, NO. 6, JUNE 2016
Spatial Modulation via 3-D Mapping
Shuaishuai Guo, Student Member, IEEE, Haixia Zhang, Senior Member, IEEE,
Shi Jin, Member, IEEE, and Peng Zhang
Abstract—This letter proposes a jointly mapped spatial
modulation (JM-SM) scheme to break through the constraint
on the number of transmit antennas in traditional SM systems.
It is realized via jointly mapping the transmit information bits to
3-D constellation points. A 3-D constellation design scheme for
a JM-SM is analyzed and established by minimizing the system
average bit error probability (ABEP). In addition, the extension
of the joint 3-D mapping to generalized spatial modulation is
discussed. The simulation results show that the proposed schemes
can be adapted to multiple-input multiple-output systems with an
arbitrary number of transmit antennas and offer better ABEP
performance than those existing schemes.
Index Terms— Spatial modulation, multiple-input multiple-
output (MIMO), 3-D mapping.
I. INTRODUCTION
S
PATIAL modulation (SM) [1]–[3] is a new 3-dimension
(3-D) modulation technique developed for multiple-input
multiple-output (MIMO) systems, which exploits both antenna
index and signal modulation for data transmission. Due
to its inherent advantages, e.g., low-complexity and high
energy-efficiency [3], it has drawn intensive attention and
been extensively investigated on transceiver design [4]–[7],
error performance analysis [8], [9], extensions like gener-
alized spatial modulation (GSM) [3] and enhanced spatial
modulation (ESM) [10], etc.
Conventionally, SM is fulfilled in two steps. First, the
information bits are split into two parts; second, part of the
split bits are mapped to the activated antennas and the rest
bits are mapped to traditional modulated symbols [1]. In such
separate mapping processes, the number of bits mapped to
antenna indices must be an integer, and thus the number of
transmit antennas should be a power of two [3]. To break
through this constraint on antenna deployment, fractional bit
encoded spatial modulation (FBE-SM) [11] and bit padding
method [12] have been proposed. However, they have caused
new problems. For instance, FBE-SM suffers from error prop-
agation [11], and the bit padding method might cause bits
loss or bits increase during its depadding process when the
data symbol is detected incorrectly.
Manuscript received February 22, 2016; revised April 10, 2016; accepted
April 17, 2016. Date of publication April 20, 2016; date of current version
June 8, 2016. This work was supported in part by the International S&T
Cooperation Program of China under Grant 2014DFA11640; by the National
Natural Science Foundation of China under Grant 61371109, Grant 61271229,
Grant 61531011, and Grant 61471269; and by the Distinguished Young
Scientists Foundation of Shandong Province under Grant JQ201315. The
associate editor coordinating the review of this letter and approving it for
publication was M. Di Renzo.
S. Guo and H. Zhang are with the Shandong Provincial Key Laboratory of
Wireless Communication Technologies, Shandong University, Jinan 250100,
China (e-mail: shuaiguosdu@gmail.com; haixia.zhang@sdu.edu.cn).
S. Jin is with the National Mobile Communications Research Laboratory,
Southeast University, Nanjing 210096, China (e-mail: jinshi@seu.edu.cn).
P. Zhang is with the School of Computer Engineering, Weifang University,
Weifang 261061, China (e-mail: sduzhangp@163.com).
Digital Object Identifier 10.1109/LCOMM.2016.2556685
Fig. 1. System model of JM-SM MIMO transmission systems.
In this letter, a jointly mapped spatial modulation (JM-SM)
is proposed, with which the above mentioned constraint on
SM can be released. In addition, a 3-D constellation design
for JM-SM is analyzed and established. It is demonstrated
that employing exhaustive search method for 3-D constella-
tion design can achieve optimal performance but may cause
prohibitively high computational complexity. Assuming that
M-ary PSK (MPSK) is employed, an efficient 3-D con-
stellation design algorithm with much lower computational
complexity is proposed. Furthermore, the extension of the joint
3-D mapping to GSM is also discussed. Simulation results and
comparisons show that the proposed schemes outperform those
existing transmission schemes in terms of the average bit error
probability (ABEP).
II. S
YSTEM MODEL AND PROBLEM FORMULATION
A. System Model With 3-D Mapping
We consider a MIMO system with N
t
transmit antennas and
N
r
receive antennas, as illustrated in Fig. 1. It is assumed that
the system transmission rate is n bits per channel use (bpcu).
At the transmitter, the information bits are divided into groups,
and each group has n bits. For each channel use, a group
of n bits is directly mapped to a transmit vector, x
q
∈ C
N
t
,
q = 1, 2, ··· , 2
n
, which is referred as a 3-D constellation
point [3]. x
q
is composed of two parts and can be expressed
as x
q
= e
i
s
i
j
,inwhiche
i
is an N
t
-dimensional vector
whose i-th element is one and the rest are zeros, and s
i
j
is a complex constellation point drawn from a given M-ary
signal constellation S
M
. For clarity, let L
i
={s
i
1
, s
i
2
, ··· , s
i
m
i
},
(0 ≤ m
i
≤ M) represent the sub-signal-constellation used
for transmission when the i-th transmit antenna is activated.
Then the transmit vector set T ={x
q
|q = 1, 2, ··· , 2
n
}
(i.e., 3-D constellation) can be expressed as (1). For simplicity,
the JM-SM MIMO system using such a 3D constellation
as (1) is denoted by JM-SM (N
t
, N
r
, S
M
, n). The 3-D con-
stellation in (1) shows that JM-SM is feasible if and only if
N
t
i=1
m
i
= 2
n
. This relaxes the constraint of the conventional
SM that N
t
M = 2
n
, where both N
t
and M should be a
power of two. Besides, as it is shown that in each JM-SM
constellation point, i.e., one column of (1), there is only
1558-2558 © 2016 IEEE. Translations and content mining are permitted for academic research only. Personal use is also permitted,
but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.