Scholarly article on topic 'CDMA Transmission with Complex OFDM/OQAM'

CDMA Transmission with Complex OFDM/OQAM Academic research paper on "Computer and information sciences"

Share paper

Academic research paper on topic "CDMA Transmission with Complex OFDM/OQAM"

Hindawi Publishing Corporation

EURASIP Journal on Wireless Communications and Networking Volume 2008, Article ID 748063, 12 pages doi:10.1155/2008/748063

Research Article

CDMA Transmission with Complex OFDM/OQAM

Chrislin Lélé, Pierre Siohan, Rodolphe Legouable, and Maurice Bellanger

France Telecom, Research & Development Division, RESA/BWA, 4 rue du Clos Courtel, 35512 Cesson-Sevigne, Cedex, France Correspondence should be addressed to Pierre Siohan, Received 15 May 2007; Accepted 10 August 2007 Recommended by Arne Svensson

We propose an alternative to the well-known multicarrier code-division multiple access (MC-CDMA) technique for downlink transmission by replacing the conventional cyclic-prefix orthogonal frequency division multiplexing (OFDM) modulation by an advanced filterbank-based multicarrier system (OFDM/OQAM). Indeed, on one hand, MC-CDMA has already proved its ability to fight against frequency-selective channels thanks to the use of the OFDM modulation and its high flexibility in multiple access thanks to the CDMA component. On the other hand, OFDM/OQAM modulation confers a theoretically optimal spectral efficiency as it operates without guard interval. However, its orthogonality is limited to the real field. In this paper, we propose an orthogonally multiplex quadrature amplitude modulation (OQAM-) CDMA combination that permits a perfect reconstruction of the complex symbols transmitted over a distortion-free channel. The validity and efficiency of our theoretical scheme are illustrated by means of a comparison, using realistic channel models, with conventional MC-CDMA and also with an OQAM-CDMA combination conveying real symbols.

Copyright © 2008 Chrislin Lele et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.


Multicarrier code-division multiple access (MC-CDMA) systems have been initially proposed in [1, 2]. This technique constitutes a popular way to combine CDMA and orthogonal frequency division multiplexing (OFDM) with cyclic prefix (CP). Nowadays, MC-CDMA is considered as one of the possible candidates for the downlink of B3G communication systems. Indeed, on one hand, this technique proposes a good way to fight against frequency-selective channels thanks to the OFDM modulation and, on the other hand, it has a high flexibility in the multiple access scheme thanks to the CDMA component. However, the insertion of the CP leads to spectral efficiency loss since this "redundant" symbol part does not carry useful data information. In addition, the conventional OFDM modulation is based on a rectangular windowing in the time domain which leads to a poor (sinc(x)) behavior in the frequency domain. Thus, CP-OFDM gives rise to 2 drawbacks: loss of spectral efficiency and sensitivity to frequency dispersion (e.g., Doppler spread). Both of them can be counteracted using a variant of OFDM introduced in [3, 4] known as orthogonally multiplex quadrature amplitude modulation (OQAM) [5] or more recently as OFDM/OQAM [6], where OQAM then stands for Offset

QAM. Here for concision, we will call it the OQAM modulation.

OQAM has many common features with OFDM. Indeed, in OQAM, the basic principle is also to divide the total transmission bandwidth into a large number of uniform subbands. As for OFDM systems, the transmitter and receiver implementations can also benefit of fast Fourier transform (FFT) algorithms. However, instead of a single FFT or inverse fast Fourier transform (IFFT), a uniform filter bank is used. So, one can get a better frequency separation between subchannels, reducing the intercarrier interference (ICI) in the presence of frequency shifts. It is also of interest to examine if these attractive features can also be efficiently exploited when OQAM is used in combination with spread spectrum techniques and also if this combination leads to some new advantages.

If a CDMA spreading is applied to OQAM in the frequency domain, leading to OQAM-CDMA, we get a transmission scheme similar to MC-CDMA, both being of a particular interest in a multiuser downlink transmission context. It is shown in [7] that, not surprisingly, we can keep the inherent advantage of OQAM over CP-OFDM of a better spectral efficiency. Furthermore, as for OQAM, the orthogonality only holds in the real field, that is, for the transmission of real

symbols, it is suggested in [7], instead of simply discarding them, to use the imaginary parts of the demodulated and de-spread signals for resynchronization. In [8] it is also shown, with a wavelet-based OFDM-CDMA system, that a pulse-shaped CDMA multicarrier system can also bring improvements with respect to the multiuser interference. In [7, 8], the data symbols transmitted over each subcarrier are real-valued. In this paper, we show that for OQAM-CDMA, a transmission of complex-valued data symbols, keeping the same symbol rate, is possible if the spreading codes are appropriately selected.

The mathematical foundations of the OQAM scheme with spread spectrum are presented in Section 2. Then, in the following sections, we analyze for a distortion-free channel the OQAM-CDMA scheme considering Walsh-Hadamard (W-H) codes. An analysis of the imaginary component, in the single user case, is provided in Section 3. In Section 4, we present a construction rule about the W-H spreading code selection that in the multiuser case leads to a perfect cancellation of the imaginary interference created by the transmission of complex-valued data with OQAM. Section 5 provides a global analysis of the main features of the complex version of OQAM-CDMA with respect to the real version and to MC-CDMA. Finally, in Section 6, some comparisons in terms of bit error rate (BER) and regarding to the systems load are carried out, using realistic channel models, between the real and complex version of OQAM-CDMA and also with MC-CDMA.


We can write the baseband equivalent of a continuous-time multicarrier OQAM signal as follows [6]:

s(t) = X Tam,ng(t - nto)ej2nmFotVm>n

m=0 neZ

gm,n ( t)

with M = 2N an even number of subcarriers, F0 = 1/T0 = 1/2to the subcarrier spacing, g the pulse shape, and vm,n an additional phase term. Here, as in [9], we set vm,n = jm+n( — i)mn. The prototype filter g is real-valued and we also assume that its length is a multiple of M such that L = bM = 2bN, with b an integer. The transmitted data symbols am,n are real-valued. They are obtained from a 22K-QAM constellation, taking the real and imaginary parts of these complex-valued symbols of duration T0 = 2t0, where t0 denotes the time offset between the two parts [5, 6, 9, 10].

Assuming a distortion-free channel, the perfect reconstruction of the real data symbols is obtained owing to the following real orthogonality condition:

R{(gm,n I gpq}} = R{fgm,n(t)g*q(t)dtj = 8m,p8„q, (2)

where Sm,p = 1 if m = p, and Sm,p = 0 if m=p. To express the complex inner product, it may be convenient to use the ambiguity function Ag of the prototype function g. Defining it as follows:

Ag(n, m) = g(u - nTo)g(u)e2jnmF°udu

J —o

and taking into account the limited duration of g with the indicating function Iin—noi<2b, equal to 1 if In — n0| < 2b and 0 elsewhere, it can be easily shown that

{gm,n, gp,no) $m-p,n-no + 1


where y^no) is given by

eo) = I{( — 1)m(n+no) jm+n—p—noAg(n — no, m — p)}. (5)

The block diagram illustrating the OQAM transmission scheme is depicted in Figure 1. Compared to conventional CP-OFDM, real-data symbols are transmitted via an OQAM modulator involving an IFFT operation followed by a filtering operation polyphase with the polyphase components of g [9, 10]. At the receiver side, the dual operations are carried out; and thanks to the real orthogonality demodulation, followed by one-tap equalization, the data symbols are recovered. Different kinds of prototype functions can be implemented as the isotropic orthogonal transform algorithm (IOTA) prototype [6] or some other prototypes directly optimized in discrete time using the time-frequency localization (TFL) criterion [11].

Let us now present the CDMA component of the proposed transmission scheme. We denote by Nc the length of the CDMA code used and assume that N0 = M/Nc is an integer number. Let us denote by cu = [C0 the code used by the uth user. Then, for a user u0 at a given time n0, N0 different data are transmitted denoted by

du0,n0,0, du0,n0,1,..., du0,n0,N0—1. Then by spreading with the Cu

codes, we get the real symbol am0,n0 transmitted at frequency m0 and time n0 by

= X Cmo/Nc,udu,noXmo/Nc\'

where U is the number of users, / the modulo operator, and I • J the floor operator. From the am0,n0 term, the reconstruction of du,n0,p (for p <E [0,N0 — 1]) is insured thanks to the orthogonality of the code, that is, cT1cu2 = Su1>u2 (see [12] for more details). Therefore, the despreading operator leads to


X cm,uapNc+m,n0• m=o

In [7], it is shown that, thanks to the real orthogonality of the OQAM modulation, the transmission of these spread real data (du,n0,p) can be insured at a symbol rate which is more than twice the one used for transmitting complex MC-CDMA data as no CP is inserted. Figure 2 depicts the real OQAM-CDMA transmission scheme where after the de-spreading operation, only the real part of the symbol is kept whereas the imaginary component is not detected.

We now propose to consider the transmission of complex data, denoted by d^^p, using U well-chosen Walsh-Hadamard codes. In order to establish the theoretical features of this complex OQAM-CDMA scheme, we suppose that the transmission channel is free of any type of distortion. Also

Figure 1: Conventional OQAM transmission scheme.

Figure 2: Real OQAM-CDMA transmission scheme.

for simplicity reasons, we assume a maximum frequency diversity, M = 2N = Nc. Then we can denote by d^u the transmitted complex data and by the complex symbol transmitted at time nT0 over the carrier m and for the code u. As usual, the length of the W-H codes are supposed to be a power of 2, M = 2N = 2q with q an integer.

The corresponding transmission scheme is depicted in Figure 3. This complex OQAM-CDMA transmission case has similarities with the MC-CDMA one. However, the modulation and demodulation operations include a specific mapping and demapping in relation to the time offset of OQAM and also a pulse shaping. Furthermore, the subsets of W-H codes have to be appropriately selected (see Sections 3 and 4). The baseband equivalent of the transmitted signal can be written as

s(t) = X X xm,ngm,n(t) With xm,n = X a^u. (8)

neZ m=0

As the channel is distortion-free, the received signal is y(t) = s(t) and the demodulated symbols are obtained as follows:

ym) no = (y> gmo,no) .

Then, the despreading operation gives us the despread data for any code, for example, for u0, we get

2N-1 2N-1 2N-1

zn0,uo = X cp,uo ypno = X cp,uoX X xm,n(gm,n> gp,n„). p=0 p=o neZ m=o

Replacing xm,n and (gm,n,gp,n0) by their expression given in (8) and (4), respectively, we get

2N-1 2b-1 2N-1

\uo X cp,uo X X

p=o n = -2b+1 m=o


y V c d(c) + .Jprno) \

^ cm,uun+no,u\um-p,n-no ^ JYm,n+no J .

Then, splitting the summation over n in two parts, with n equal or not to 0, (11) can be rewritten as

U-1 2N-1

zn0,u0 = X dn0'),u X CP,Uo cp,u u=0 p=0

U-1 2b-1

u=o n=-2b+1 n=o

2N -1 2N-1

Y V d(c) I y y c c (pno) /. / . dn+no,u 1 cp,uo cm,u/m,n+no

p = o m = o

The W-H codes being orthogonal, that is,

c p , u c

1 if u = uo,

p,uo cp,u = 1 n .c ,

'o if u=uo,

we finally obtain:

dno,uo +J

U-1 2b — 1 /2N-12N-1

Y V d(c) y y c v(p,no)

/. / . dn+no,ul cp,uo cm,uYm,n+no

u=o n = -2b+1 V p=o m=o

The aim now is to show that, when U < M/2, for an appropriate choice of the U codes we can get z^U = d,^,^. Let us first examine the single user case.

Figure 3: Complex OQAM-CDMA transmission scheme.


As the channel is assumed to be distortion-free ifthere is only one single user, the demodulated and despread signal is the one obtained in (14) setting, for one user u0, U = 1. Then by splitting the summations over m and p in two parts, one for m = p and the other one for m=p, we get

Zntuo = dinOuo + j I X dn+nouo (si(n) + S2(n)) I, (15)

si(n) = X n p,uon p,uo(-1)pnjnAg(n,0)},

p=0 2N-1 p-1

S2(n) = X X n p,uo nm,uo (- 1)mnjm+n-pAg (n, m - p)j

p= o m = o

+j{ (-1)pnjp+n-mAg (n, p-m)}] •

For W-H codes, we have, Vn : Cp,u0Cp,u0 = 1/2N and the prototype filter g being real-valued, Ag(n,0), see (3), is also real-valued. Then, it is straightforward to show that for every n, s1(n) = 0.

Let us now look at

2N—1 p — 1

S2(n) = X X Cp,u0 Cm,u0 [l{ (—1)mT+"—pAg (n, m — p)}

p= o m = o

+j{ (-1)pn jp+n-mAg (n, p-m)j] •

With g being a real function, then Ag(n, m) = A*(n, —m), thus the imaginary terms in (17) are such that

Sj = l{( — 1)m"jm+"—pAg (n, m — p)j

+ l\( — 1)pnjp+n—mAg (n, p — m)\

mn g (18) = ^ ( — 1)mnjm+n—pAg (n, m — p)

+ ( — 1)pnjp+n—mA*g(n, m — p)].

It can be easily seen that for n even, SI = 0, while for n odd, the result depends upon the parity of m and p being given by

2(-l)mnj"+1R{ jm-pAg (n, m-p)}

if m and p have the same parity otherwise.

Then, the computation of (15) can be restricted to the terms obtained for odd values of n with p and m being of identical parity. After some computations, setting v = m — p, it can be shown that

= dSuo+2

V d(n) i2"+iy R

A dno+2n+1,uo J

n=-b n=0

r 2N-1-2v -)

^Ag(2n +1,2v) X (-1)knk+2v,uonk,uo\ k=0 j

With a first property ofthe W-H codes shown in Appendix A:

X (-1)knk+2v,uo nk,uo = 0 for v = 0, •••, N ;

k = 0, ••• ,2N - 1 - 2v,

(20) becomes

Vn0, U0' z«0,u0 = dn0,u0 •

This last equality is the result of a straightforward derivation of the demodulated and despread signal. It leads us to a property that a priori could not be easily intuitively apprehended. Nevertheless, we can attempt to justify it a posteriori. Let us notice, firstly, that if instead of complex data, we transmit real data over a distortion-free channel, thanks to the real orthogonality of the OQAM modulation scheme, we exactly recover these real data by taking the real part in (20). Again using (20), it can then be seen that to cancel the imaginary part, the interference, the condition (21) on the W-H CDMA codes is essential. Therefore, in the one user case, using the system linearity, we can transmit complex data and recover them perfectly at the receiver.


In order to generalize the relation (22) to a multiuser case, we propose in this section a selection mode for the subsets of W-H codes. Then, the generalization can be carried out step by step, considering firstly a two-user OQAM-CDMA system and secondly a U-user system with U < M/2.

4.1. Selection of the U codes

For a Walsh-Hadamard matrice of size M = 2N = 2n, there are two subsets of column indices, Sn and Sn, with cardinal equal to M/2 making a partition of all the index set. We propose a recurrent rule of construction for these two subsets that can guarantee the absence of interference between users. For n0 = 1, each subset is initialized setting S1 = {0} and

s2 = {1}.

Let us now assume that, for a given integer n = n0, the two subsets contain the following list ofindices:

Sn° = {^1,1, il,2,2*1,3, ¿1,2"0-1}, s' = {¿2,1, 2*2,2,2*2,3, •••, ¿2,2"0-1} •

These subsets are afterwards used to build new subsets of identical size such that

Sr = {22,1 +2n0,22,2 +2n0, i2,3 +2n0,..., ¿2,2*0-1 +2n0 },

Sn = {¿1,1 + 2n°, ¿1,2 + 2n0, ¿1,3 + 2n0,..., ¿12n0-1+2n0}.

Then, we get the subsets of higher size, n = n0 + 1, as follows:

S'0+1 = Sr u S'0,

S2n»+1 = S'0 u S'0 •

4.2. Case of two users in the same subset (U = 2)

In the second step of our proof, we want to show now that, again for W-H codes such that M = 2N = 2n, if two users u0 and u1 take their codes into the same subset, for example, all in SI or all in Sn, there in no interference between these 2 users, Zn,u0 = dn,u0 and Znu = dnu.

Let us show at first that for u0 and u1 e SI (resp., Sn), Zn^u, = d^m. Indeed, setting U = 2 in (14), for two given users u0 and u1 e SI (resp., Sn), we get

= d(c) =d

20 ,U0 + J (

2b-1 /2N-12N-1

Y >) V V c c v(p,n0} I

/ . dn+n0,U0 I cp,U0 cm,U0 ym,n+n0 I

= —2b+1 \ p=0 m=0 '

n=-2b+1 \ n=0

/2N-1 2N-1

y d(c) I Y V c c y(p,'0)

/ . dn+n0,ui 1 cp,u0 cm,u1 Ym,n+n0

n=-2b+1 \ n=0

V p=0 m=0

As it has been shown for one user that z20,u0 = dn^, based on (20), we can deduce that at the right-hand side the second term is zero.

Then, by splitting again the summation over m in two parts, one for m = p and the second one for m= p, we get

= d(c) + j ^n0,U0 " n0,U0 J

X dn+n0,U1 (w(n) + Tg(n, U0, U1)

n = -2b+1 \ n=0

with w(n) containing the terms obtained for p = m and Tg(n, u0, u1), the ones for m=p leading to

w(n) = X Cp,u0CPMI3{(-1)pnj"Ag(n,0)},

Tg (n, u0, u^

2N-1 p-1

= X X (Cp,u0 Cm,u !{(-1)mn jm+n-pAg (n, m - p)}

p=0 m=0

+ Cp,Ul Cm,u01{(-1)pn jp+n-mAg (n, p-m)}),


Ag(n, 0) being real-valued, it is obvious that w(n) = 0 for n even. For n odd, it is shown in Appendix B that

X (-1)pC(p%1 Cph = 0 for u0, u1 e sn (resp., Sn2), (29)

which leads again to w(n) = 0. Thus for every n, w(n) = 0.

The expression of Tg can be rewritten introducing a new variable u = p - m and using the fact that Ag(n, -u) = A* (n, u), thus we obtain

Tg(n, u0, u\)

2N-1 r2N-1-u

= X ? X ( 1)mnJn-UCm+u,u

u=1 v. m=0

cm,u1 Ag (n, u)

m=0 2N — 1 —u

2N — 1 —u -j

+ £ (-1)mn+unCm+u,u1 Cm,u0 j"+uAg (n, u) f •

For n even (n = 2k), we get Tg (2k, u0, uj

2N-1 r2N -1-u

= X (-1)k^j X Cm+u,u0 Cm,u1 j-uAg(n, u)

u=1 m=0 (31)



+ X cm+u,u1 cm,u0 J Ag (n, u) j" •

In Appendix C, it is shown that for s > 0, and for any WH matrix of order n, that is, a size M= 2n, the corresponding codes Cm,u0 are such that


y Cn c(21

/. Cm,u0 Cm+s,ui

= X ttU, for u0, u1 e Sn (resp^, S^^

2 — 1- s

Then as Tg (2k, u0, u1) is the imaginary part of the sum of two conjugate quantities, we have

Tg (2k, u0, ui) = 0.

The same lines of arguments can be applied to show that if n is odd (n = 2k + 1), we get Tg(2k + 1, u0, u1) = 0.

The computation for n odd uses the following properties of Walsh-Hadamard codes:


I (-1Г


= - X (-l)mc(nui Cm+2s,Uo for Mo, Mi G S", (resp., S") ,


Y (_1)mc(") r(2)

/ . V -V cm,u0 cm+2s+i,i

26-1 2N-12N-l

X(u0, u) = X I I

" = -2b+1 p=0 m=0

It has been shown for one user, with u = u0 (see (15) in Section 3), and afterwards for 2 users, with u0 and u1 e SI (resp., S") (see (26) in Section 4.2), that X(u0, u) = 0. Therefore, if the U codes are all chosen in e SI (resp., S" ), we get

w J-c) — л(с)

v"0, u0, Z"0,u0 — d"0,u0.

So, in this last and more general case the result can be a posteriori justified using the same lines of arguments we developed previously for the one- and two-user case.


= X ( — 1)m4t cm+2s+1,u0 for «0,u1 e S" (resp.,S^.

The proof of these properties, not reported here to avoid another lengthy mathematical derivation, is quite similar to the one used to get by recurrence the result presented in Appendix B. Finally, as

Tg(", u0, щ) = 0 for u0, u1 g S" (resp., S"), (35)

we get

V"o, u0,


- d") = d"0,rn.

As in the one-user case, this last equality is the result of a straightforward derivation of the demodulated and despread signal and it could not be so easily intuitively apprehended. However in this case, based on our previous study of OQAM-CDMA systems for real-data transmission [7], it was clear that to cancel the imaginary part, some specific conditions on W-H codes were required. Indeed looking at [7, Figures 4 and 5], it is clear that whatever the orthogonal pulse shape g(t) being used, the imaginary part is zero only for some pairs of codes. What we show here is that these pairs of W-H codes can be grouped in two subsets, forming a partition of the set of all codes (see Section 4.1), where they satisfy the essential relations (29), (32), (34). So using again the system linearity, we can transmit complex data and recover them perfectly at the receiver.

4.3. Case of U users in the same subset (U < M/2)

Now let us consider the case U < M/2 where the U codes are all chosen either in S" or in Si?. Setting U < M/2 in (14) for U given users e SI (resp., S2), we get

z"0,u0 = d"0,u0 + iX X(u0, u) ,


In MC-CDMA, and CP taking apart, the transmitted data are complex and the full load is obtained when using all the codes of the W-H matrix (U = M). When considering the full diversity, that is, one spread symbol transmitted over all modulated carriers, the maximum spectral efficiency (full load) is obtained for M complex data symbols transmitted at every T0 symbol duration.

In OQAM-CDMA with real-data symbol transmission, the full load is again obtained when U = M. Therefore, we obtain the maximum spectral efficiency when M realdata symbols are transmitted at every T0/2 symbol duration, which is equivalent to the transmission of M complex-data symbols at T0.

In the proposed OQAM-CDMA scheme with complex data symbol transmission, the system guarantees a complex orthogonality up to a number of users U = M/2 which corresponds to the maximum load. As for the complex OQAM-CDMA system, M/2 complex-data symbols are transmitted at every T0/2 symbol duration, this scenario is equivalent to the one where M complex-data symbols are transmitted every T0 duration.

So, these 3 scenarios lead to the same spectral efficiency, without taking into account the CP, but consider different number of spreading codes to reach this spectral efficiency. Since the number of spreading codes used directly impacts on the multiple access interference (MAI) [13, 14], a first analysis ofthe different systems shows that using less spreading codes may lead to better performance results. Indeed, if U increases, the MAI term also increases. As an illustration of the reduction of the MAI, we can notice that when there is only one user in the OQAM-CDMA complex transmission scheme, that is, no MAI, the same spectral efficiency is obtained either in MC-CDMA or in OQAM-CDMA real-transmission schemes, with the use of 2 W-H spreading codes, with a nonzero MAI term. So, the OQAM-CDMA with complex symbol transmission should outperform the two other systems as it uses twice less spreading codes to achieve the same spectral efficiency. Some simulation results will also confirm this analysis in the following section.


This section gives the main parameters used in simulations and provides an evaluation of the 3-transmission schemes: MC-CDMA, OQAM-CDMA with real symbols transmission, and the new proposed OQAM-CDMA with complex-symbol transmission. This evaluation leads to a fair comparison between the 3 systems either in terms of BER or in percentage of load.

6.1. System parameters

The static propagation channel is modelled by a 3-tap delay profile having the following characteristics.

(i) Delay (ps): 0 0.2527 0.32.

(ii) Powers (in dB): -0 - 3 - 2.2204.

The other main parameters of the uncoded system are the following.

(i) Carrier frequency: fc = 1000 MHz.

(ii) FFT size = 32.

(iii) Sampling frequency = 10 MHz.

(iv) Symbol duration, to (to): 1.6 ps (3.2 ps).

(v) Cyclic prefix = 0.5 ps for MC-CDMA.

(vi) Walsh-Hadamard spreading codes of length 32.

(vii) One-tap MMSE (minimum mean squared error) equalization.

(viii) For OFDM/OQAM, either the IOTA prototype filtering of length 128, b = 4, or TFL prototype of length 32, b = 1, is implemented.

When considering the MC-CDMA technique, the performance results are given by taking into account the loss in power (10 log10(T0/(T0+CP) = 0.63 dB) induces by the cyclic prefix insertion. For the OQAM-CDMA with complex symbol transmission, the W-H codes are issued from the first subset Sf. In MC-CDMA and OQAM-CDMA with real symbol transmission, when S1 is not sufficient to achieve the targeted spectral efficiency (use of more than M/2 codes), then the spreading codes from the S52 subset are selected.

6.2. Simulation results

Figure 4 shows the performance results obtained at 1/16 of the maximum system spectral efficiency. To achieve this spectral efficiency, the MC-CDMA and real OQAM-CDMA techniques use 2 W-H spreading codes, whereas the complex OQAM-CDMA system uses only one W-H code. It is shown that the OQAM-CDMA with real symbols outperforms the MC-CDMA technique of the gain induced by the absence of the CP insertion. When comparing both OQAM-CDMA schemes, we can note that the complex-symbol transmission system provides around a 2 dB gain at BER= 10-2 compared to the real-symbol transmission. This gain shows that using only one W-H code (complex-symbol transmission) instead of 2 (real-symbol transmission) allows to reduce the MAI term and so to obtain better performance.

Figure 5 shows the performance results obtained at the maximum system spectral efficiency. To achieve this spectral

-0- OQAM-CDMA complex (IOTA4) -e- OQAM-CDMA real (IOTA4) - MC-CDMA

Figure 4: BER performance results of the 3-transmission schemes at 1/16 of the maximum spectral efficiency.

-e- OQAM-CDMA real (IOTA4) - MC-CDMA

-0- OQAM-CDMA complex (IOTA4)

Figure 5: BER performance results of the 3-transmission schemes at the maximum spectral efficiency.

efficiency, the MC-CDMA and the real-OQAM-CDMA techniques use the 32 W-H spreading codes, whereas the complex OQAM-CDMA system uses 16 W-H codes corresponding to the whole S1 subset. It is shown that both OQAM-CDMA systems have the same performance results and outperform the MC-CDMA system as no CP is required. In that case, we note that the MAI term does not provide any gain in favour of the OQAM-CDMA with complex symbol transmission.

In Figure 5, it can also be noted that the OQAM and OFDM curves merge around Eb/N0 = 13 dB. Indeed, in this context, where we assume a perfect channel knowledge and a one-tap MMSE equalization, the OQAM system, which has

0 10 20 30 40 50 60 70 80 90 100 Percentage of the system load

-■*- OQAM-CDMA complex (TFL1) -e- OQAM-CDMA real (TFL1) —*— OQAM-CDMA complex (IOTA4) -e- OQAM-CDMA real (IOTA4) - MC-CDMA

Figure 6: BER performance result regarding to the system's load for the 3-transmission schemes.

no guard interval, suffers from ISI in the presence of a time dispersive channel. So for Ey/N0 beyond 13 dB the curves cross. This phenomenon, named intrinsic interference, is explained in details in [15, 16]. As also shown in [17], if the delay spread is not too long, less than 1/8 of t0, a one tap equalization maybe enough. For larger delay spread, as is the case here (being 20% of t0), a more complex equalization procedure should be used.

To quantify the impact of the MAI term, we have plotted in Figure 6, the performance with regard to the system load at fixed Ey/N0 (a fixed Ey/N0 ratio leads to a lower BER in OQAM systems compared to MC-CDMA since the CP is taken into account). This Ey/N0 is the same for the 3 systems and is equal to 10 dB that corresponds approximatively to a BER= 10-2 at full load. For OQAM-CDMA systems, we have either considered the IOTA prototype function or the TFL one. Figure 6 shows that OQAM-CDMA systems give better performance results than the MC-CDMA whatever the load. This gain is always provided by the no-CP insertion. When comparing OQAM-CDMA systems, Figure 6 shows that until 35% of the load, the OQAM-CDMA with complex symbol transmission outperforms the OQAM-CDMA with realsymbol transmission. These results illustrate the impact of the MAI term on the performance results, showing the advantage of using the OQAM-CDMA complex-symbol transmission. Now, if we compare the results with regard to the prototype function, we can comment that the TFL prototype provides better performance results than the IOTA one in real-symbol transmission and for 2 W-H codes. For the complex OQAM-CDMA transmission, both prototypes have almost the same performance. Note also that, for complexity implementation, TFL prototype is more suitable than the IOTA one since its length is 4 times less.


In this paper, we have proposed an OQAM-CDMA system with complex-data symbol transmission, which allows a reduction of the MAI term while keeping the same spectral efficiency as in MC-CDMA (CP taking apart) or OQAM-CDMA with real-data symbol transmission. We have proved that the transmission of complex symbols in OQAM-CDMA requires a judicious selection of the W-H spreading codes to guarantee the complex orthogonality, that is, in theory limited to the real field in OQAM. The performance results obtained in the considered system have shown that OQAM-CDMA with complex symbol transmission outperforms the MC-CDMA technique whatever the system load thanks to the no-CP insertion and to the lower number of spreading codes used. Compared to OQAM-CDMA with real-symbol transmission, owing to the reduction of the MAI term, our proposed technique gives better performance results up to 35% of the system load. The choice of the prototype function in OQAM-CDMA has no major impact on the performance results in our studied system. However, the TFL prototype function has an advantage with regard to the implementation.

In future work, we will investigate the potential utilization of more than M/2 codes in OQAM-CDMA system transmitting complex-data symbols in order to increase the system spectral efficiency and exceed the theoretical MC-CDMA and OQAM-CDMA transmitting real-data symbol systems.



We denote by A(n) = [c^, C,n),..., cM"] the W-H matrix of order n of size M X M with M= 2n, and with a kth column given by ckn) = [c0nk, c(nk,..., Mik]T for k = 0,1,...,M - 1.

In this appendix, we show that for any positive integer n and any code of index k, that is, for the kth column of A(n), we have

2"-1-2 p

(-ir4"VS = 0 for p = 0,1, ...,2"-1,

m = 0,1,... ,2" - 1 - 2p.

The proofis carried out in 2 steps. Step 1. We first show by recurrence that

for n = 1,2,..., oo;

A") A") _ A") A")

L2m,kL2p,k = c2m+1,kc2p+1,k

m = 0,1,... ,2("-1) - 1; p = 0,1, ...,2("-1)

Case n = 1

The W-H matrix being such that

A«- -1- i1 1 A = —2 \1 "1

it can be easily checked that 4^4,0 = ci|i3ci|i3 and c'qJc'qJ c( jC ( j, which shows that the property is true for n = 1 .

Case n = n0

We assume that the property is true for n = n0,

r("o) r(«o) _ r(«o) r(«o) f p _ 0 1 2(no-1 ) _ 1 .

c2m,kc2p,k = c2m+1 ,kc2p+1 ,k> IOr p = 0' 1 ' ••• ,2 1 .

m = 0,1 , ••• ,2("0- 1) - 1 - 2p.

Case n = n0 + 1

Step 2. Let us simply notice that for a given m and p, we have

2n-1-2p V (_1)mr(n) c(n)

zl ( 1) cm+2p,kcm,k



.(n) r(n) _ (n) (n)

'2u+2p,kr2u,k r2p+2u+1,kr2u+1,k) •

Then, with the result proved in Step 1, we conclude that


(n) r(n)

m+2 p,krm,k

Ck (-1)m = 0-


Let us show that the property is therefore true for n = n0 +1. As

a-"=(i;;o; _ao>) , <">

we also have

(r(n) r(n)

(n+1) _ I \rk , Ck

if k < 2n0,

(ckn)2„, -rk-J if k> 2n0 - 1

Let us now only consider the case (k > 2;o - 1) knowing that the computation principle is similar for the case (k < 2;o -1). For m = 0,1,... ,2;o - 1 and p = 0,1,... ,2;o - 1, we have the following set of equalities:

c(;o+1> _ c(;o) . c(;o+1> _ c(;o)

C2m,k C2m,k-2n> C2p,k C2p,k-2n

c(;o+1> _ (;o> . C(;o+1> _ c(;o>

C2m+1,k C2m+1,k-2n> C2p+1,k C2p,k-2n

if 2m < 2;o - 1, 2p < 2;o - 1,


Keeping the same notations as in Appendix A and using the definitions of the subsets S;, S2, S;0, and S2° presented in Section 4.1, we can also show that

X (-1)pc(p2)uoCph = 0 for «0,u e s; (resp.,SI). (B.1)

The proof is established by recurrence on ;, starting by ; = 2.

Case ; = 2

The W-H matrix is given by

A(2) = -L V4

/11 1 1 \

1 -11 -1

11 -1 -1

V1 -1 -1 1)

r(n0+1) _ r(n0) . r(n0+1) _ _r(n>)

r2m,k = r2m,k-2n; r2p,k = r2p,k-2n

r(n0+1) _ (n0) . r(n0+1) _ _ (n0) (A 7)

r2m+1,k = r2m+1,k-2n; r2p+1,k = L2p,k-2" v ^

if 2m < 2n0 - 1,2p> 2n0 - 1,

(n0+1) _ Jn0)

~(n0+1) _ (n0)

= r2m,k-2n. r2p,k = r2p,k-2n

(;o+1> _ _ (20> . (;o+1) _ J;o>

C2m+1,k = C2m+1,k-2n > C2p+1,k = C2p,k-2n

if 2m > 220 - 1, 2p > 2 20 - 1.

As the case 2m > 2 20 - 1 and 2p < 2 20 - 1 can be directly derived, exchanging m and p, from the one where 2m < 2 20 -1 and 2p > 220 - 1, we always get

r(n0+1)r(n0+1) _ r(n0) r(n0)

r2m,k r2p,k = L2m,k-2"L2p,k-2",

r(n0+1) r(n0+1) __r(n0) r(n0)

r2m+1,kr2p+1,k = L2m+1,k-2"L2 p+1,k-2»•

Then using the recurrence assumption for ; = ;0 given in (A.4), we get

4^4^ = r(m!+1,kr2p)+1,k for n =1,2, •••,

n = 0,1, ••• ,2(n-1) - 1, p = 0,1, ••• ,2(n-1) - 1

It can be easily checked, testing the different possible cases when u0, u1 e Sj = {0,3}: u0 = u1 = 0 or 3, u0 = 0 and u1 = 3, similarly for u0, ui e S\ = {1,2} that the property (B.1) is true for n = 2.

Case ; = ;0

We now assume the property (B.1) is true for n = n0. Otherwise, for the sets S;0 and S;0, see (23) and u0, u1,..., we have

2n0 -1

V ( 1)pr(n0) r(n0) _ 0

zL (-1) rp,u0 rp,u1 = ° p=0

Case n = n0 + 1

Let us show that (B.3) also holds true for n = n0 + 1. For n = n0 + 1, (B.3) can be rewritten as

2n0+1-1 V p (n

¿^ ( 1) rp,u0 rp,u p=0

p (n0+1) (n0+1) r p , u r p

2n0 -1

2n0 -1

p (n0+1) (n0+1) p (n0+1) (n0+1)

= zL (-1) rp,u0 rp,ui + (-1) rp+2n0 ,u0 rp+2n0 ,ui,

where the second summation results from a substitution of p by p - 2n0.

Let us consider the 3 possible cases

(1) u0, u1 e Sn° (resp., e S20). Then for p< 2% we get

r(?o+1) _r(?o). r(?o+1) _ r(?o).

cp+2n0,u0 — cp,uo. cp,uo — cp,Uo.

r(?o+1) _r(?o). r(?o+1) _ (no)

cp+2no,u1 — cp,u1. rp,u1 — rp,u1 •

Based on (B.4) and (B.3), we get

2?o+1 —1 2no-1

)pr(?o) c(no) _ 2 V (_1)pr(?o) c(no)

I (-1)p4nu04?u1 — 2 x (-I)p4nu04?u) — o^ (B.6) p—0 p—0

(2) u0,u1 e S^1 (resp., S^3).

For p < 220, based on the same principles of computation, we now obtain

r(?o+1) _ i?o) . (?o + 1) _ (no) .

cp+2n,uo rp,uo-2S ' rp,uo rp,uo—2? '

r(no+1) _ (no) . r(?o+1) _ r(no)

rp+2S,u1 rp,u1—2n? > rp>u1 tp,u1—2?^

Then using relation (B.4) and noting that by the substitution v0 = u0 - 220 and v1 = u1 - 220 v0, v1 e S2 (resp., e Sf), then taking the recurrence relation (B.3) into account, we get

2?o+1 —1

2?o -1

x (-1)pcpnu+1)cpu:1' = 2 2 (-1)= 0. (b.8)

p=0 p=0

(3) u0 e S^ (resp., S^) and u1 e S^ (resp., S"^) For p < 220, the recurrence relation between the columns of Hadamard matrices of successive order leads to

(?o+1) (?o). (no+1) _ (no).

rp+2n0,u0 — rp,uo> rp,uo — rp,uo>

(no + 1) p+2?o ,u1

r(no) . rp,u1-2"o>

(no+1) (no)

p,u1—2?o •

Then, we find


Using again the notations introduced in Appendix A and in Section 4, we are going to show that for 0 < s < 2n - 1, the W-H codes satisfy the following properties:


r(n) r(n) /. rm,u0 rm+s,u1 m—0

2n — 1 —s

— X rmur!m+s,uo for uo, u1 e S? (resp., Si?),

'.n — 1—s

V r(n) r(nl

/. rm,u0 rm+s,u1

— — V mir(mn+s,uo for uo e S?, u1 e S?

As in the previous appendices, we use a recurrence on n. Case n = 1

For the W-H matrix of order 1

Ad) — — ( 1 1 ,

A V2U —1'

it can be checked with the partition S1 = {0} and S2 = {1} that for s = 1, the relations (C.1) are (C.2) are true.

Case n = n0

Let us now assume that these relations are also true for n = n0, that is, for 0 < s < 2n° - 1, we have

2no — 1— s

r(no) r(no) /. rm,u0 rm+s,:

2n? — 1—s

— X rmnurlm+luo for uo, u1 e S?0(resp. S??0),

2?o — 1—s

r(no) r(no) /. rm,u0 rm+s,:

2? —1—s

— — X r^ir^,uo for uo e s?0, u1 e S??0

2n0+1-1 V t npr (n0+1)>0+1)

¿^ ( 1) cp,u0 cp,u1 p=0

2n0 -1 2n0 -1

- v ( 1)pc{n0) v ( 1)pc{n0)

= zL (-1) cpu cp,u1-2"° - zL (-1) cpu cp,u1-2"° = 0 p=0 p=0


To conclude, for any integer n, there are two subsets, SI and Sn, which give a partition of all the index set such that for u0, u1 e SI (resp., S2) the property (B.1) is satisfied.

Case n = n0 + 1

To show that these properties are also true for n = n0 + 1, the seven different cases have to be considered

(1) u0, u1 e S^1 (resp., S^1) and u0, u1 e Sl° (resp.,


(2) u0, u1 e S2^1 (resp., S^1) and u0, u1 e Sf1 (resp.,

(3) u0, u1 e sn0+1 (resp., e S20+1) and u e S20 (resp., S20) and u0 e S2 (resp., S2^).

(4) u0 e S20+1, u1 e S20+1 and u0 e S20, and u1 e S20.

(5) u0 e S;0+1, u1 e S;0+1 and u0 e S;0, and u1 e S;0.

(6) u0 e S20+1, u1 e S20+1 and u0 e S20, and u1 e Sn2°.

(7) u0 e S;0+1, u1 e S;o+1 and u0 e S20, and u1 e S;o.

As the proofs are quite similar for these different cases in this appendix, as a matter of example, we only provide the proofs for the first 2 cases.

(1) u0, u1 e S;0+1 (resp., Sn0+1) and u0, u1 e S;0 (resp., S;0). We split the interval of analysis for 0 < s < 220+1 - 1 into two subcases.

(2) u0, u1 e S;0 (resp., S2 ) and u0, u1 e S;0 (resp.,

To prove the properties for 0 < s < 220+1 - 1, we proceed as in the first case, splitting the analysis into two 2 intervals for s.

Case 0 < s < 220 - 1

Si, the left-hand side in relation (C.1), can again be split into three terms:

Subcase 0 < s < 220 - 1

Let Sl denote the left-hand side in relation (C.1), it can be split into three terms:

2n0 — 1 —s

2n0 —1

S y C(;o+1>C(;o+1> + V C(;o+1>C(;o+1>

SL = Cm,u0 Cm+s,u1 + Cm,u0 Cm+s,i

m=0 2n0+1—1—s

m=2n0 —s

+ Y C(;o+1>C(;o+1> + Z-, Cm,u0 Cm+s,u1

Making a substitution by m^m — 2 20 + s and m^m — 220 in the second and third summations, respectively, then using in the first and third summations the relation of recurrence between the columns of W-H matrices of consecutive order, we get


S y C(;o+1>C(;o+1>

SL Cm,u1 Cm+s,uo

2n0 — 1 —s

2n0 — s— 1

- 2 V c(;o> c(;oJ + y c(;o> C-

= 2 Z, Cm,u1 Cm+s,uo + Z Cm,uo Cm+s,u1

C(20> C(;0>

with s = 2;o - s.

Then using the recurrence hypothesis (C.4), we have

2"0+1 —1—s

2"0+1 —1—s

V C(20+1>C(;0+1^ V C(20+1>C(;0+1> (C8)

/. Cm,u1 Cm+s,uo = Cm,u0 Cm+s,u1 (C.8)

Subcase 2 20 < s

Then we can write

2n0+1 —1—s

Si = I

c(;o) c(;o>

C m Ur. C,

m,uo Cm+s-2n<° u


c(;o> c(;o> (C 9)

Cm,u0 Cm+s—2n0 ,u1


Sl - y C(20+1>C(;0+1> + V C(20+1>C(;0+

SL Cm,u0 Cm+s,u1 + Cm,u0 Cm+s,,



m=2n0 —s


+ y C(;o+1> C(;o+1>

+ Z., Cm,u0 Cm+s,u1.

Making a substitution by m^m — 2 20 + s and m^m — 220 in the second and third summations, respectively, then using, in the first and third summations, the relation of recurrence between the columns of W-H matrices of consecutive order, we get


Sr - Y C(;o+1>C(;o+1> SL = Cm,u0 Cm+s,u1

2n0 — 1— s 2n0 — s— 1

= 2 y c(;o> J;0> _ y C(20> c(;o>

^ Z-. Cm,V0 Cm+s,V1 Cm,V1 Cm+s;


with s = 2;o - s, v1 = u1 - 2;, v0 = u0 - 2;, and v0, v1 e Sn2" (resp., S;0). In the same manner, we get


Y C(;o+1>C(;o+1> /. Cm,u1 Cm+s,uo

2n0 — 1— s 2n0 — s— 1

_ 2 y c(;o> C(;0> _ y c(;o) c(;o>

2 Z., Cm,V1 Cm+s,V0 /_, Cm,Vo Cm+s,v1.

m=0 m=0


Si = I

c(;o+1>c(;o+1> Cm+s,u1


From the recurrence hypothesis (C.4), we have

2n0+1 —1—s 2n0+1 —1—s

y C(;o+1)C(;o+1> _ y C(;o+1>C(;o+1> (C 14)

/. Cm,u1 Cm+s,uo = Cm,u0 Cm+s,u1 (C.14)

m=0 m=0

Case 220 < s

2n0 — 1— s

y c(;o> C(;0> m,u

m,u0 m+s,u1 m=0

with again s = s - 220. From the recurrence hypothesis (C.4), we have

2n0+1 —1—s 2"0+1 —1—s

y C(;o+1>C(;o+1> _ y C(;o+1>C(;o+1> (C]0)

/. Cm,u1 Cm+s,uo Cm,u0 Cm+s,u1. (C.10)

m=0 m=0

2"0+1 —1—s

(;0) (;0) Cm,u0—2n0 Cm+s—2n0 ,u1 —2n0

2n0+1—2n0 —1—(s—2n0>

y c(;o> C(;0>

X . um,vo m+s—2n0 ,V1

2no — 1— s

v c(;o> c(;o>

Z_. m,V0 m+s,V1


with s = 2"0 - s, vi = ui - 2", v0 = u0 - 2" and v0, vi e S"0 (resp., Sn0). In the same manner, we get

2n0+i — i —s 2n0-i-s

Y r("0+i)r(n0+i) _ _ y r(n0) r(n0) (Ci6)

/. rm,ui rm+s,U0 rm,vi rm+s,v0 • (C.i6)

m=0 m=0

From the recurrence hypothesis (C.4), we have

2n0+i-i-s 2"0+i-i-s

Y r("0+i)r(n0+i) _ y r(n0+i)r(n0+i) (C17) /. rm,ui rm+s,U0 rm,u0 rm+s,ui • V^-1'/ m=0 m=0

To conclude, for any integer n, there are two subsets, SI and S", which give a partition of all the index set such that for u0, ui e Sn (resp . , S"), the property (C .i) is satisfied and for u0 e SI (resp., S") and ui e S" (resp., Sn), the property (C.2) is satisfied.


[1] N. Yee, J. P. Linnartz, and G. Fettweis, "Multi-carrier CDMA in indoor wireless radio networks," in Proceedings of the 4th IEEE International Symposium on Personal, Indoor and Mobile Communications (PIMRC '93), pp. i09—ii3, Yokohama, Japan, September i993.

[2] K. Fazel and L. Papke, "On the performance of convolutionally-coded CDMA/OFDM for mobile communication system," in Proceedings of the 4th IEEE International Symposium on Personal, Indoor and Mobile Communications (PIMRC '93), pp. 468-472, Yokohama, Japan, September i993.

[3] R. W. Chang, "Synthesis of band-limited orthogonal signals for multi-channel data transmission," Bell System Technical Journal, vol. 45, no. i0,pp. i775-i796, i966.

[4] B. R. Saltzberg, "Performance of an efficient parallel data transmission system," IEEE Transactions on Communication Technology, vol. i5, no. 6, pp. 805-8ii, i967.

[5] B. Hirosaki, "An orthogonally multiplexed QAM system using the discrete Fourier transform," IEEE Transactions on Communications, vol. 29, no. 7, pp. 982-989, i98i.

[6] B. Le Floch, M. Alard, and C. Berrou, "Coded orthogonal frequency division multiplex," Proceedings of the IEEE, vol. 83, no. 6, pp. 982-996, i995.

[7] C. Lele, P. Siohan, R. Legouable, and M. Bellanger, "OFDM/OQAM for spread spectrum transmission," in Proceedings of the 6th International Workshop on Multi-Carrier Spread Spectrum (MC-SS '07), pp. i57-i66, Herrsching, Germany, May 2007.

[8] J. Guo, "Performance of a wavelet-based OFDM-CDMA high speed power line communication systems," in Proceedings of the 7th International Power Engineering Conference (IPEC '05), p. 259, Singapore, November-December 2005.

[9] P. Siohan, C. Siclet, and N. Lacaille, "Analysis and design of OFDM/OQAM systems based on filterbank theory," IEEE Transactions on Signal Processing, vol. 50, no. 5, pp. ii70-ii83, 2002.

[10] H. Bolcskei, "Orthogonal frequency division multiplexing based on offset QAM," in Advances in Gabor Analysis, pp. 32i-352, Birkhauser, Boston, Mass, USA, 2003.

[11] D. Pinchon, P. Siohan, and C. Siclet, "Design techniques for orthogonal modulated filterbanks based on a compact representation," IEEE Transactions on Signal Processing, vol. 52, no. 6, pp. i682-i692, 2004.

[12] E. H. Dinan and B. Jabbari, "Spreading codes for direct sequence CDMA and wideband CDMA cellular networks," IEEE Communications Magazine, vol. 36, no. 9, pp. 48-54, 1998.

[13] M. Helard, R. Legouable, J.-F. Helard, and J.-Y. Baudais, "Mul-ticarrier CDMA techniques for future wideband wireless networks," Annals of Telecommunications, vol. 56, no. 5-6, pp. 260-274, 2001.

[14] S. Kaiser, Multi-carrier CDMA mobile radio systems—analysis and optimization of detection, decoding, and channel estimation, Ph.D. dissertation, University of München, Munchin, Germany, 1997.

[15] D. Lacroix and J.-P. Javaudin, "A new channel estimation method for OFDM/OQAM," in Proceedings of the 7th International OFDM-Workshop (InOWo '02), Hamburg, Germany, September 2002.

[16] J.-P. Javaudin, D. Lacroix, and A. Rouxel, "Pilot-aided channel estimation for OFDM/OQAM," in Proceedings of the 57th IEEE Semiannual Vehicular Technology Conference (VTC '03), vol. 3, pp. 1581-1585, Jeju, Korea, April 2003.

[17] C. Lele, J.-P. Javaudin, R. Legouable, A. Skrzypczak, and P. Siohan, "Channel estimation methods for preamble-based OFDM/OQAM modulations," in European Wireless, Paris, France, April 2007.