Transmission and receiving method in a wireless communication system

Abstract

The present patent application relates to a transmission method and corresponding receiving methods. The transmission method comprises: selecting K number of data streams for transmission, wherein each data stream comprises one or more data messages; and independently for each data stream k=0, . . . , K1: encoding and modulating data messages of the kth data stream using a Trellis Coded Modulation, TCM, scheme TCM.sub.k, interleaving the encoded and modulated data messages of the k-th data stream by using a stream specific interleaver .sub.k, scrambling the interleaved data messages of the k-th data stream by using a stream specific scrambling sequence; combining all scrambled data messages of the K number of data streams into at least one transmission signal s(l); and transmitting the at least one transmission signal s(l) over a radio channel of the wireless communication system.

Claims

1. A transmission method in a wireless communication system, comprising: selecting, by the wireless communication system, K data streams for transmission from a set of Z>K data streams based on a channel quality of a radio channel, wherein each data stream comprises one or more data messages; for k-th data stream, wherein k=0, . . . , K1: encoding and modulating, by the wireless communication system, the one or more data messages of the k-th data stream using a Trellis Coded Modulation (TCM) scheme, TCM.sub.k, to generate one or more processed data messages; interleaving, by the wireless communication system, the one or more processed data messages of the k-th data stream using a stream specific interleaver .sub.k; and scrambling, by the wireless communication system, the interleaved one or more processed data messages of the k-th data stream using a stream specific scrambling sequence, wherein coefficients of the scrambling sequence for K=2 maximize a minimum Euclidean distance of the symbols of a signal constellation subject to symbol power constraint, wherein the coefficients of the scrambling sequence for K{3,4,5}, are uniformly phase spaced in [0, $\frac{}{})$ where is a constant dependent on the TCM.sub.k, and wherein the coefficients of the scrambling sequence for K6, are coefficients of a complex pseudo-random sequence; combining, by the wireless communication system, scrambled data messages of the K data streams into at least one transmission signal s(l); and transmitting, by the wireless communication system, the at least one transmission signal s(l) over the radio channel of the wireless communication system.

2. The method according to claim 1, wherein the K data streams experience substantially a same channel quality.

3. The method according to claim 1, wherein encoding parameters of the TCM.sub.k are dependent on the K data streams.

4. The method according to claim 3, wherein the TCM.sub.k encoding has four states for K3 and two states for K4.

5. The method according to claim 1, wherein the interleavers .sub.k use any of a set of randomly generated permutations, a set of permutations generated from circular shifts of a single randomly generated permutation, or a set of permutations generated according to a rule of quadratic polynomial permutation.

6. The method according to claim 1, wherein is equal to one for BPSK, PAM, and ASK, wherein is equal to two for QPSK and QAM, wherein is equal to four for 8PSK, and wherein is equal to eight for 16PSK.

7. The method according to claim 1, wherein K2.

8. The method according to claim 1, wherein the combining is achieved by summing the scrambled data messages of the K data streams to form the at least one transmission signal s(l).

9. The method according to claim 1, wherein the at least one transmission signal s(l) is transmitted by a single transmission device.

10. The method according to claim 1, wherein the at least one transmission signal s(l) is transmitted by a plurality of independent transmission devices.

11. The method according to claim 1, wherein each data message comprises a finite length sequence of information symbols or comprises a block of information symbols.

12. The method according to claim 1, wherein the K data streams uses same TCM.sub.k encoding parameters.

13. The method according to claim 1, wherein a modulation scheme of the TCM.sub.k is dependent on the K data streams.

14. The method according to claim 1, wherein the K data streams uses a same TCM.sub.k modulation scheme.

15. A transmission device for processing and transmitting communication signals in a wireless communication system, comprising: a non-transitory memory storage comprising instructions; and one or more processors in communication with the memory, wherein the one or more processors execute the instructions to: select K data streams for transmission from a set of Z>K data streams based on a channel quality of a radio channel, wherein each data stream comprises one or more data messages; for k-th data stream, wherein k=0, . . . , K1: encode and modulate the one or more data messages of the k-th data stream using a Trellis Coded Modulation (TCM) scheme, TCM.sub.k, to generate one or more processed data messages; interleave the one or more processed data messages of the k-th data stream using a stream specific interleaver .sub.k; and scramble the interleaved one or more processed data messages of the k-th data stream using a stream specific scrambling sequence, wherein coefficients of the scrambling sequence for K=2 maximize a minimum Euclidean distance of the symbols of a signal constellation subject to symbol power constraint, wherein the coefficients of the scrambling sequence for K{3,4,5}, are uniformly phase spaced in [0, $\left[0,\frac{}{}\right)$ wherein is a constant dependent on the TCM.sub.k, and wherein the coefficient of the scrambling sequence for K6, are coefficients of a complex pseudo-random sequence; combine scrambled data messages of the K data streams into at least one transmission signal s(l); and transmit the at least one transmission signal s(l) over the radio channel of the wireless communication system.

16. The transmission device according to claim 15, wherein encoding parameters of the TCM.sub.k are dependent on the K data streams.

17. The transmission device according to claim 15, wherein the interleavers .sub.k use any of a set of randomly generated permutations, a set of permutations generated from circular shifts of a single randomly generated permutation, or a set of permutations generated according to a rule of quadratic polynomial permutation.

18. The transmission device according to claim 15, wherein coefficients of the scrambling sequence are dependent on the K data streams.

19. The transmission device according to claim 15, wherein a modulation scheme of the TCM.sub.k is dependent on the K data streams.

20. The transmission device according to claim 15, wherein the K data streams uses a same TCM.sub.k modulation scheme.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) The appended drawings are intended to clarify and explain different embodiments of the present patent application in which:

(2) FIG. 1 shows the normalized spectral efficiency of the TCMA scheme;

(3) FIG. 2 illustrates a set of TCM transmitters according to prior art;

(4) FIG. 3 illustrates a TCM encoder-modulator having a 4 state encoder and QPSK modulation;

(5) FIG. 4 illustrates a TCMA receiver according to prior art;

(6) FIG. 5 shows a flow chart illustrating a prior art iterative detection and decoding method;

(7) FIG. 6 shows aggregate spectral efficiency of TCMA with 1 to 7 number of streams;

(8) FIG. 7 illustrates and overview of a general transmission and receiving scheme according to the present patent application;

(9) FIG. 8 illustrates a TCM encoder-modulator having 2 encoding states and a QPSK modulator;

(10) FIG. 9 illustrates the minimum Euclidean distance of a super-constellation according to the present patent application;

(11) FIG. 10 illustrates an embodiment of the present patent application applied in the downlink;

(12) FIG. 11 illustrates an embodiment of the present patent application applied in the uplink;

(13) FIG. 12 shows aggregate SE performance for the present patent application compared to the SE of a prior art TCMA system;

(14) FIG. 13 shows aggregate SE performance for the present patent application compared to a turbo-coded LTE system;

(15) FIG. 14 shows the super-constellation of a five stream ETCMA system where each stream is modulated using QPSK modulation;

(16) FIG. 15 illustrates a transmission device according to an embodiment of the present patent application;

(17) FIG. 16 illustrates an alternative transmission device according to an embodiment of the present patent application;

(18) FIG. 17 illustrates a first or a second receiving device according to an embodiment of the present patent application; and

(19) FIG. 18 illustrates an alternative first or second receiving device according to an embodiment of the present patent application.

DETAILED DESCRIPTION

(20) The present patent application relates to a method for the concurrent transmission of multiple independent data streams, possibly intended for different users, by efficiently sharing the same physical time-frequency-space radio resources, such as Resource Elements (REs). The present patent application also relates to corresponding receiving methods.

(21) The present patent application applies both to the downlink and to the uplink of cellular wireless communication systems, operating either in the FDD mode or in the TDD mode. However, the patent application is not limited to cellular systems and can be applied in any suitable wireless communication system, such as general OFDM and single carrier systems. A general scheme of present system is illustrated in FIG. 7 in which a transmit device (left) of the present patent application transmits communication signal s(l) to a receiver device (right) of the present patent application which receives signal r(l) over the radio channel of a wireless communication system.

(22) Transmission of data/information according to the present patent application comprises the following steps: selecting K number of data streams for transmission, wherein each data stream comprises one or more data messages; and independently for each data stream k=0, . . . , K1: encoding and modulating data messages of the kth data stream using a Trellis Coded Modulation, TCM, scheme TCM.sub.k, interleaving the encoded and modulated data messages of the k-th data stream by using a stream specific interleaver .sub.k, scrambling the interleaved data messages of the k-th data stream by using a stream specific scrambling sequence; combining all scrambled data messages of the K number of data streams into at least one transmission signal s(l); and transmitting the at least one transmission signal s(l) over a radio channel of the wireless communication system.

(23) According to an embodiment of the patent application, the data streams are selected for transmission according to the channel quality, such as signal-to-noise ratio (SNR), signal-to-interference plus noise ratio (SINR), or any other channel quality indicator. Accordingly, data streams with similar SNR or SINR values are grouped together and transmitted concurrently on the same REs. In the wireless communication system, there are Z data streams that must be transmitted to users. So, the K number of data streams is selected from a large set of Z>K data streams based on the channel quality of the radio channel. By this selection the data streams can be chosen such that the different data streams experience substantially the same channel quality, i.e., similar channel quality. This is particularly important when all TCM schemes are configured according to the same parameters, i.e., same convolutional encoder and same modulation scheme. In such case. similar error protection levels can be provided to all streams only if the experienced channel quality is similar. The number of data streams K should be two or more.

(24) Furthermore, one message for each data stream is processed using a stream specific processing scheme. Each TCM encoder-modulator pair may consist of different trellis codes or modulation schemes. However, for simplification it is assumed that all data streams use the same TCM in the following description. Regarding the data messages of the streams, it is assumed according to an embodiment of the patent application that each data message comprises a finite length sequence of information symbols or a block of information symbols which is well understood.

(25) According to another embodiment of the patent application the TCM encoder and modulation parameters are chosen according to Table 3 which implies that the encoder parameters and/or the modulation scheme of the TCM.sub.k for the different data streams are all equal but is dependent on the total number of selected data streams K according to the more general embodiment. Hence, according to an embodiment of the patent application the different data streams use the same encoders-modulators, i.e., TCM having the same encoding parameters and modulation schemes.

(26) When the number of streams is three or less, a TCM scheme with four-state convolutional encoders (as shown in FIG. 3) is used. When more than three streams are transmitted, a TCM scheme with two-state convolutional encoders (shown in FIG. 8) is used in the TCM scheme. The modulation scheme is QPSK in all cases but can be any other suitable modulation scheme. It has been realised by the inventors that at least the modulation schemes in the group comprising: PSK, QAM, PAM, ASK and APSK works well, but the skilled person realises that other modulation schemes can be used together with the present patent application.

(27) TABLE-US-00003 TABLE 3 Choice of the TCM parameters as a function of the number of streams K Number of Type of convolutional streams K encoder Modulation K 3 Four-state encoders QPSK K 4 Two-state encoders QPSK

(28) Moreover, the interleavers .sub.k used in the present transmission method have to be stream specific in order to aid the present (iterative) receiver to separate the stream signals in the received radio communication signal. Indeed, interleavers can be considered as the signatures of the different data streams so stream specific means that it is unique for each data stream. The receiver is thus facilitated in recognizing and separating the stream information when the interleavers are different for the different data streams. Instead, when the data streams use equal interleavers, the system performance is significantly degraded.

(29) With reference to FIG. 7, the TCM decoder of the k-th stream computes soft information to be fed back to the MSD through the stream specific interleaver .sub.k. This soft information, after being processed by the MSD, is fed back to the next (i.e., the (k+1)-th) TCM decoder through the inverse stream specific interleaver .sub.k+1.sup.1. The resulting permutation applied to such soft information corresponds to the equivalent permutation .sub.k,k+1.sup.(EQ)=.sub.k+1.sup.1.sub.k, i.e., the composition of the two permutations .sub.k and .sub.k+1.sup.1.

(30) The inventors have also realised that in order to avoid decreased performance, all interleavers .sub.j,k.sup.(EQ) should avoid mapping sequences of consecutive input symbols into sequences of consecutive output symbols. This way, the correlation between the j-th TCM decoder and the k-th TCM decoder input is reduced, hence performance is improved.

(31) Therefore, according to an embodiment a set of randomly generated permutations is used for interleaving. Experimental results show that randomly generated permutations perform well with respect to this characteristic, thus featuring good de-correlation properties.

(32) According to another embodiment a set of permutations generated from circular shifts of a single randomly generated permutation is used for interleaving the data messages. The set of permutations {.sub.k}.sub.k=0.sup.K-1 can be generated starting from a single permutation .sub.0 and using circular shifts of the permutation for generating the whole set of K permutations. We define the input circular shift of a permutation as follows: given a permutation of size N defined as :n.fwdarw..sub.n, n=0, . . . , N1, its input circular shift of P positions is the permutation .sup.(P,I): (n+P)mod N.fwdarw..sub.n, n=0, . . . , N1. Similarly, the output circular shift of a permutation is defined as .sup.(P,O): n.fwdarw.(P+.sub.n)mod N, n=0, . . . , N1. Starting from a randomly generated permutation .sub.0, we define the set of stream-specific permutations as .sub.k=.sub.0.sup.(kQ,I) (or alternatively .sub.k=.sub.0.sup.(kQ,O)) where Q is a suitably chosen integer. Any choice of Q larger than the constraint length of the constituent TCM encoder yields good results. This embodiment is convenient since only one permutation has to be stored in the memory of the transmitter and receiver, while all other stream-specific permutations can be easily generated starting from the only one permutation.

(33) According to yet another embodiment of the patent application a set of permutations generated according to the Quadratic Polynomial Permutation (QPP) rule, also used in the 3GPP LTE standard, is used for interleaving. QPP interleavers are of particular interest since they exhibit the contention-free property, a nice characteristic that permits parallelized high-speed implementation of decoding algorithms.

(34) QPP permutations are functions of three parameters, namely: the permutation size N and the two coefficients f.sub.1, f.sub.2 of the quadratic polynomial used to generate the permutation:
.sub.n=(f.sub.1n+f.sub.2n.sup.2)mod N,n=0, . . . ,N1.(5)

(35) For each size N, two sets of possible values for f.sub.1 and for f.sub.2 are defined. Values for f.sub.1 and for f.sub.2 can be independently chosen from the two sets. QPP interleavers for the present patent application have been generated choosing a fixed value for f.sub.2 and different values for f.sub.1 in order to obtain different permutations.

(36) Furthermore, scrambling codes are used to further separate data streams in the signal domain so that the discrimination of the data streams is facilitated by using stream specific scrambling sequences. Each encoded, modulated and interleaved stream is scrambled and combined with the other streams before transmission according to the present patent application. This is a very important aspect of the present patent application.

(37) The scrambling consists in multiplying the sequence of interleaved modulation symbols s.sub.k(l) with a stream-specific sequence of complex coefficients c.sub.k=(c.sub.k(l)), where index k=0, . . . , K1 refers to the stream, and index l=0, . . . , L1 refers to the symbol interval (L is the block length). The signal transmitted by the present system at symbol interval l is

(38) $\begin{array}{cc}s\left(l\right)=\underset{k=0}{\overset{K-1}{.Math.}}{c}_k\left(l\right)s_k\left(l\right)& \left(6\right)\end{array}$

(39) and it belongs to the superposed-constellation x(l) which contains all possible values of s(l). This means that the combining of the different data streams is an addition of the different scrambled streams so as to form the transmission signal s(l) for transmission in the wireless communication system.

(40) To optimize the SE of the present scheme through scrambling, different solutions are possible. According to an embodiment the coefficients c.sub.k=(c.sub.k(l)) are made independent of the time index l, therefore c.sub.k(l)=c.sub.k, l=0, . . . , L1. Moreover, the values of coefficients c.sub.k are chosen to maximize the minimum Euclidean distance between symbols of the superposed constellation, or super-constellation, for short. The minimum Euclidean distance of the super-constellation is defined as

(41) $\begin{array}{cc}{D}_{E.\min }\left(l\right)=\underset{lj}{\min }{D}_E\left(x_i\left(l\right),x_j\left(l\right)\right)& \left(7\right)\end{array}$
where symbols x.sub.i(l), x.sub.j(l) are any distinct pair of symbols belonging to the super-constellation x(l). The scrambling sequences are optimized in order to achieve the maximum D.sub.E.min(l) for all l=0, . . . , L1.

(42) As an example, we consider the case of two streams using QPSK signal constellations. The first stream (k=0) is scrambled using the sequence c.sub.0(l)=1 and the second stream (k=1) is scrambled using the sequence c.sub.1(l)=exp (j). In FIG. 9 D.sub.E.min of the super-constellation is shown as a function of . Two solutions maximize D.sub.E.min:=/6 and =/3.

(43) In general, for any number of streams K and for any TCM constellation, it is possible to find a set of optimal scrambling sequences c.sub.k.sup.opt(l), k=0, . . . , K1 that maximizes D.sub.E.min which means that the coefficients of the scrambling sequence are dependent on the number of selected streams K according to an embodiment of the patent application.

(44) Accordingly, for K{3,4,5} it has been observed that the best performance is obtained choosing the phases as uniformly spaced in a given interval [0,/), where is a constant dependent on the TCM constellation as

(45) $\begin{array}{cc}{c}_k\left(l\right)=\exp \left(j\frac{k}{K}\right),k=0,\mathrm{.Math.},K-1.& \left(8\right)\end{array}$

(46) Therefore, according to an embodiment of the patent application =1 for BPSK, PAM, ASK, =2 for QPSK and QAM, =4 for 8 PSK, and =8 for 16 PSK. This solution does not result in a super-constellation with maximum D.sub.E.min. However, simulations show that its performance is the best for K=3, 4, 5.

(47) For the case K6, any suitable complex set of sequences with random-like properties can be used for scrambling. In particular, Zadoff-Chu (ZC) sequences, a class of complex sequences with low cross-correlation properties, are well suited in this respect. We have hence

(48) $\begin{array}{cc}{c}_k\left(l\right)=\exp \left(-\frac{j{r}_{k}l\left(l+1\right)}{L}\right),l=0,\mathrm{.Math.},L-1& \left(9\right)\end{array}$
where r.sub.k is a stream-specific value in this case.

(49) The choice of the scrambling sequence type is made as a function of the number of streams K, as summarized in the embodiment of Table 4. These parameters are the result of a search performed through simulation. Better results could be obtained by further extending the search.

(50) TABLE-US-00004 TABLE 4 Example of scrambling sequence as a function of the number of streams K Number of Scrambling sequence streams K family K = 2 maximum D.sub.E.min K = 3, . . . , 5 Uniformly spaced in [0, /) K 6 Zadoff-Chu

(51) The high performance of the present receivers is achieved thanks to the new iterative detection and decoding algorithm that operates on soft information related to modulation symbols s.sub.k(l) instead of soft information related to coded bits d.sub.k.sup.(i)(l) as done in prior art. In this way, the loss of information incurred when converting symbol-wise soft information to bit-wise soft information is avoided. For simplicity, the scrambling sequences and the channel state information are assumed to be perfectly known at the receiver.

(52) The present receiver method comprises the steps of: receiving the at least one transmission signal transmitted by a method according to the present patent application; computing a joint probability for the modulation symbols of all K streams conditioned to the received signal r(l); and for each data stream k=0, . . . , K1: a) computing soft information related to the modulation symbols for the kth data stream from the joint probability, b) feeding the soft information related to the modulation symbols through a stream specific deinterleaver .sub.k.sup.1 to a TCM.sub.k decoder for the k-th data stream, c) decoding the deinterleaved soft information so as to obtain soft information related to information symbols of data messages for the k-th data stream and updated soft information related to the modulation symbols for the k-th data stream, d) feeding the updated soft information related to the modulation symbols for the k-th data stream through a stream specific interleaver .sub.k, and e) updating the joint probability with the updated soft information related to the modulation symbols for the k-th data stream; repeating steps a)-e) an arbitrary number of iterations for all K streams, and in the final iteration: using the soft information related to information symbols for the k-th data stream so as to obtain decoded data messages for the k-th data stream, and outputting the decoded messages for the k-th data stream.

(53) With reference to FIG. 7, the TCM decoder of the k-th stream computes soft information to be fed back to the MSD through the stream specific interleaver .sub.k. This soft information, after being processed by the MSD, is fed back to the next (i.e., the (k+1)-th) TCM decoder through the inverse stream specific interleaver .sub.k+1.sup.1.

(54) The Enhanced MSD (EMSD) of the present patent application performs the computation of the joint probability distribution P(s(l)=x.sub.i(l)|r(l))=P(s.sub.0(l)=m.sub.j.sub.0, . . . , s.sub.K-1(l)=m.sub.j.sub.K-1|r(l)) for all points x.sub.i(l) of the super-constellation x(l) obtained as

(55) $\begin{array}{cc}{x}_i\left(l\right)=\underset{k=0}{\overset{K-1}{.Math.}}{c}_k\left(l\right)m_{\mathrm{jk}}& \left(10\right)\end{array}$
and iteratively updates the joint probability by iterating with the present Enhanced TCM (ETCM) decoders. For each stream k, the EMSD computes the soft information related to the modulation symbols for the k-th data stream p.sub.k,j(l)=P(s.sub.k(l)=m.sub.j|r(l)) that the l-th modulation symbol of stream k equals the j-th symbol m.sub.j of the TCM.sub.k constellation. Then, the EMSD feeds these probabilities to the ETCM decoder through the deinterleaver .sub.k.sup.1.

(56) The ETCM decoder receives in input the probability distributions p.sub.k,j(l), executes a trellis-based decoding algorithm (e.g, Viterbi, BCJR or SISO) that applies the TCM code constraints and computes updated probabilities q.sub.k,j(l), which are fed back to the EMSD through the interleaver .sub.k. The EMSD uses the updated probabilities q.sub.k,j(l) to update the joint probability distribution P(s(l)|r(l)) thus gaining improved knowledge of the transmitted symbol s(l). Then, the EMSD proceeds to processing the next data stream, and accordingly the processing for the different data streams is performed sequentially according to an embodiment of the patent application.

(57) More in detail, the EMSD performs the following steps: Receiving updated probabilities q.sub.k,j(l) of soft information related to modulation symbols from at least one of the ETCM decoders; Re-computing the joint probability distribution of modulation symbol P(s(l)=x.sub.i(l)|r(l)) taking into account the updated probabilities:

(58) $\begin{array}{cc}P\left(s\left(l\right)=x_i\left(l\right).Math.r\left(l\right)\right)=\underset{k=0}{\overset{K-1}{.Math.}}\underset{j\text{:}{x}_k=m_j}{\overset{}{.Math.}}{q}_{k,j}\left(l\right),l=0,\mathrm{.Math.},L-1& \left(11\right)\end{array}$ For each data stream k=0, . . . , K1, re-computing the soft information related to the modulation symbols for the k-th data stream p.sub.k,j(l) to be provided to the k-th ETCM decoder at the next iteration:

(59) $\begin{array}{cc}\begin{array}{c}{p}_{k,j}\left(l\right)=P\left(s_k=m_j.Math.r\left(l\right)\right)\\ =\underset{x\text{:}{x}_k=m_j}{\overset{}{.Math.}}P\left(s\left(l\right)=x_i\left(l\right).Math.r\left(l\right)\right),\end{array}l=0,\mathrm{.Math.},L-1& \left(12\right)\end{array}$ Feeding the probability distributions p.sub.k,j(l) to each ETCM decoder through their associated deinterleavers.

(60) Moreover, the patent application also relates to a second aspect of the receiving method which comprises the steps of: receiving a communication signal r(l), wherein the communication signal r(l) has been encoded and modulated according to a Trellis Coded Multiple Access, TCMA, scheme and comprises K number of data streams indexed k=0, . . . , K1, respectively, before transmission; computing a joint probability for the modulation symbols of all K streams conditional to the received signal r(l); and for each data stream k=0, . . . , K1: a) computing soft information related to the modulation symbols for the kth data stream from the joint probability, b) feeding the soft information related to the modulation symbols through a stream specific deinterleaver .sub.k.sup.1 to a TCM.sub.k decoder for the k-th data stream, c) decoding the deinterleaved soft information so as to obtain soft information related to information symbols of data messages for the k-th data stream and updated soft information related to the modulation symbols for the k-th data stream, d) feeding the updated soft information related to the modulation symbols for the k-th data stream through a stream specific interleaver .sub.k, and e) updating the joint probability with the updated soft information related to the modulation symbols for the k-th data stream; repeating steps a)-e) an arbitrarily number of iterations for all K streams, and in the final iteration: using the soft information related to information symbols for the k-th data stream so as to obtain decoded data messages for the k-th data stream, and outputting the decoded messages for the k-th data stream.

(61) Hence, it is realised that the second aspect of the receiving method can be used with the general Trellis Coded Multiple Access, TCMA, schemes. In fact, the TCMA scheme of prior art can be regarded as a special case of ETCMA in which only one stream per transmitter/device is transmitted and no scrambling is applied.

(62) Receiver Complexity Reduction

(63) The EMSD computes and updates the probability distribution P(s(l)=x.sub.i(l)|r(l)) over the whole super-constellation, i.e., for all x.sub.i(l)x(l). Since the size of x(l) grows exponentially with the number of streams K, the complexity of the receiver algorithm grows exponentially with K as well. Hence, the present receiver methods have a complexity which grows exponentially with the number of data streams K.

(64) It has been noted that the present system, when a large number of streams are transmitted, requires high SNR. In such condition, the received signal r(l) is, with high probability, close to the transmitted symbol s(l). Hence, the inventors have realised that it is unnecessary for the EMSD to compute the probability distribution P(s(l)=x.sub.i(l)|r(l)) over the whole super-constellation. Instead, the number of super-constellation points x.sub.i(l)x(l) over which the probability distribution is computed can be restricted to a smaller subset of symbols in the vicinity of the received symbol r(l). Therefore, the proposed approach for reducing the complexity of the ETCMA receiver consists in reducing the number of super-constellation points x.sub.i(l) of the probability distribution P(s(l)=x.sub.i(l)|r(l)) computed by the EMSD to a region around the received signal r(l).

(65) A first embodiment consists in defining the reduced set of super-constellation points x.sub.i(l) as consisting of the symbols x.sub.i(l) that fall within a distance R.sub.0(SNR) of the received symbol r(l). This concept is sketched in FIG. 14, where two possible received symbols are indicated by black dots and the circles with radius R.sub.0 and R.sub.1 delimit the number of super-constellation points x.sub.i(l) over which the MSD performs its computation hence the computational complexity is substantially reduced.

(66) The radius R.sub.0 is adaptively computed as a function of the channel quality (e.g., SNR or SINR) such that, for low SNR, it takes large values while, for high SNR, it takes small values. In other words the distance R.sub.0 increases with decreasing channel quality and decreases with increasing channel quality. This is essential for minimizing the chance that the transmitted symbol (l) is not among the signals x.sub.i(l) included in the reduced signal set. In such case, the iterative algorithm would never be able to deliver the correct decoded message. It is realised that other measures than SNR or SINR can be used for representing the channel quality.

(67) The first embodiment, although simple, leads to the following drawback: since the super-constellation does not feature a uniform density, the number of points in the reduced signal set is not constant. In particular, the super-constellation x(l) exhibits a high density close to its centre and a lower density far from the centre. A second embodiment, aimed at keeping almost constant the number of points in the reduced signal set, consists in setting the radius R.sub.0(SNR,|r(l)|) as a function of the channel quality and the amplitude of the received symbol |r(l)| so that when |r(l)| is small, R.sub.0 is small, and when |r(l)| is large, R.sub.0 is large (R.sub.1<R.sub.0 in FIG. 14). So, the distance R.sub.0 is further dependent on the amplitude of the received signal r(l) such that the distance R.sub.0 increases with increasing amplitude of the received signal r(l) and decreases with decreasing amplitude of the received signal r(l). Accordingly, for good performance the distance R.sub.0 may be computed as R.sub.0=e.sup.|r(l)|.sup.2.sup./(2K), where is a first scaling parameter applied to the standard deviation of the noise and is a second scaling parameter applied to the received signal amplitude |r(l)|, and is the standard deviation of the noise. These mentioned embodiments compensate for the variable density of the super-constellation, although still results in a reduced signal set containing a variable number of points.

(68) Another approach to reduce the receiver complexity is to limit the number of considered signal points, i.e., the joint probability is only computed for a limited and fixed number of signal points D.sub.0 closest to the received symbol r(l) with respect to the Euclidean distance. This approach leads to similar results as the previous approaches while featuring the advantage of a constant number of points included in the reduced signal set. In practice, the super-constellation symbols are ordered based on their Euclidean distance from the received symbol r(l) and then the D.sub.0 symbols with lowest Euclidean distance from r(l) are included in the restricted signal space.

(69) It should further be realised by the skilled person that the joint probability and/or the soft information used in the present receivers related to the modulation symbols may be expressed in any of: probability distributions, logarithms of probability distributions, likelihood ratios, or logarithms of likelihood ratios.

(70) Examples of Application Scenarios

(71) A first possible application of the present transmission/receiving scheme concerns the Downlink (DL) of wireless communication systems, e.g. cellular systems such as 3GPP LTE and LTE Advanced. With reference to FIG. 10 a single transmitter device sends multiple (at least two) data streams to at least one receiver device. The complex coefficients c.sub.k(l) are chosen according to the aforementioned solutions. For example, a maximum D.sub.E.min solution for the case of two streams with QPSK constellations would be the following:

(72) 0$\begin{array}{cc}{c}_0\left(l\right)=\exp \left(j\right);c_1\left(l\right)=\exp j\left(+\frac{}{6}\right),l=0,\mathrm{.Math.},L-1.& \left(13\right)\end{array}$
Such solution is optimal for any value of
For the case of K{3,4,5} streams, we would choose

(73) $\begin{array}{cc}{c}_k\left(l\right)=\exp j\left(+\frac{k}{2K}\right),k=0,\mathrm{.Math.},K-1,l=0,\mathrm{.Math.},L-1.& \left(14\right)\end{array}$

(74) A second possible application of the present transmission/receiving scheme concerns the Uplink (UL) of wireless communication systems, see FIG. 11. Here, multiple (at least one) devices transmit their modulated signals to a Radio Access Network (RAN) transmission point. All transmitted signals are thereafter combined at the receiver antenna as a result of the wireless channel linearity. Carrier signals from different devices are necessarily characterized by phase and frequency offsets, such that a phase control of the transmitted signals aimed at obtaining the desired combination at the receiver antenna is very complex to implement.

(75) Concerning the UL case, two cases have to be considered: the first case concerns data streams transmitted by the same communication device (e.g., a UE). In this case, the same approach used for DL scrambling to set the phase differences between streams can be applied. The second case concerns data streams transmitted by different devices. In this case, a simple and effective solution consists in randomizing the phases of the streams according to different device-specific randomization sequences. Randomization is performed by associating to each device a different pseudo-random scrambling sequence. Again, ZC sequences or any other pseudo random sequences are proposed in this purpose.

(76) The scrambling sequence used on the k-th data stream transmitted by the u-th device is c.sub.k,u(l)e.sub.u(l), where c.sub.k,u(l) is a stream-specific scrambling sequence unique within the device and e.sub.u(l) is a device-specific scrambling sequence defined as

(77) $\begin{array}{cc}{e}_u\left(l\right)=\exp j\left(\frac{r_{u}l\left(l+1\right)}{L}\right),l=0,\mathrm{.Math.},L-1,u=0,\mathrm{.Math.},U-1.& \left(15\right)\end{array}$

(78) FIG. 12 shows the achieved SE of the present scheme. The TCM encoders and modulators, interleavers and scrambling sequences are selected according to the solutions described in previous description. The observed improvement is significant: e.g., in the case of K=7 streams, the SNR gain is close to 8 dB which is very good.

(79) FIG. 13 shows a comparison between the SE of the present scheme and the SE of an LTE turbo-coded system. It is observed that the present scheme performs better than the LTE system for almost all SNR and SE values. Moreover, it is shown that the present system is capable of achieving spectral efficiencies as high as 7 bits/s/Hz, which is impossible to achieve in the LTE system.

(80) Moreover, as understood by the person skilled in the art, any method according to the present patent application may also be implemented in a computer program, having code means, which when run by processing means causes the processing means to execute the steps of the method. The computer program is included in a computer readable medium of a computer program product. The computer readable medium may comprises of essentially any memory, such as a ROM (Read-Only Memory), a PROM (Programmable Read-Only Memory), an EPROM (Erasable PROM), a Flash memory, an EEPROM (Electrically Erasable PROM), or a hard disk drive.

(81) Furthermore, the present method in a transmitter and a receiver can be implemented and executed in suitable transmitter and receiver communication devices, respectively. It is realized by the skilled person that the present transmitter and receiver devices may comprise the necessary communication capabilities in the form of e.g., functions, means, units, elements, etc., for executing the methods according to the patent application which means that the devices can be modified, mutatis mutandis, according to any method of the present patent application. Examples of other such means, units, elements and functions are: memory, encoders, decoders, mapping units, multipliers, interleavers, deinterleavers, modulators, demodulators, inputs, outputs, antennas, amplifiers, DSPs, MSDs, TCM encoder, TCM decoder, etc. which are suitably arranged together. Examples of transmitter and receiver devices are base stations (such as eNB), mobile devices (such as UEs), relay devices, remote radio heads, hyper transmitters-receiver, virtual transmitters-receiver, coordinated multi-point transmitters-receiver, etc.

(82) Especially, the processors of the present devices may comprise, e.g., one or more instances of a Central Processing Unit (CPU), a processing unit, a processing circuit, a processor, an Application Specific Integrated Circuit (ASIC), a microprocessor, or other processing logic that may interpret and execute instructions. The expression processor may thus represent a processing circuitry comprising a plurality of processing circuits, such as, e.g., any, some or all of the ones mentioned above. The processing circuitry may further perform data processing functions for inputting, outputting, and processing of data comprising data buffering and device control functions, such as call processing control, user interface control, or the like.

(83) The present transmitter devices comprise at least one processor which is arranged to execute the steps of the transmission method according to the present patent application. This embodiment is illustrated in FIG. 15 in which the transmitter device comprises a processor arranged for executing the different steps of the present method.

(84) Alternatively, according to another embodiment of the patent application the present transmitter device comprises dedicated units for the performance of the method steps. This embodiment is illustrated in FIG. 16 in which the transmitter device comprises dedicated units, suitably connected to each other, for the corresponding method steps. The device according to this embodiment comprises a selecting unit, encoder units, interleaving units, scrambling units, a combining unit, and a transmitting unit.

(85) The present receiver devices comprise a processor which is arranged to execute the steps of the receiver method according to the present patent application. This embodiment is illustrated in FIG. 17 in which the receiver device comprises a processor arranged for executing the different steps of the present method.

(86) Alternatively, according to another embodiment of the patent application the present receiver device comprises dedicated units for the performance of the method steps. This embodiment is illustrated in FIG. 18 in which the receiver device comprises dedicated units for the corresponding method steps. The device according to this embodiment comprises a receiving unit, an EMSD, computing units, deinterleaver units, decoding units, interleaver units, and outputting units.

(87) The second receiver device of the present patent application can be arranged and implemented as the first receiver device except that the receiver signal in this case has been encoded and modulated according to a general Trellis Coded Multiple Access, TCMA, scheme.

(88) The present transmission and receiver devices can all be modified, mutatis mutandis, to correspond to the different embodiments of the present transmission and receiver devices.

(89) As an example relating to the transmitter device, according to an embodiment of the device, the K number of data streams is selected from a set of Z>K data streams based on the channel quality of said radio channel, such as signal-to-noise ratio.

(90) According to another embodiment of the device, the K number of data streams experience substantially the same channel quality.

(91) According to yet another embodiment of the device, the encoder parameters and/or the modulation scheme of the TCM.sub.k is dependent on the number of selected streams K.

(92) According to yet another embodiment of the device, the TCM.sub.k encoder has four states for K3 and two states for K4.

(93) According to yet another embodiment of the device, the modulation scheme is any in the group comprising: PSK, QAM, PAM, ASK and APSK.

(94) According to yet another embodiment of the device, the interleavers .sub.k use any of: a set of randomly generated permutations; a set of permutations generated from circular shifts of a single randomly generated permutation; or a set of permutations generated according to the rule of quadratic polynomial permutation.

(95) According to yet another embodiment of the device, the coefficients of the scrambling sequence are dependent on the number of selected streams K.

(96) According to yet another embodiment of the device, the coefficients of the scrambling sequence for:

(97) K=2, maximize the minimum Euclidean distance of the symbols of the signal constellation subject to symbol power constraint;

(98) K{3,4,5}, are uniformly phase spaced in [0,

(99) $\frac{}{})$
where is a constant dependent on the modulation scheme; and

(100) K6, are the coefficients of a complex pseudo-random sequence, such as a Zadoff-Chu sequence.

(101) According to yet another embodiment of the device, is equal to one for BPSK, PAM, and ASK; two for QPSK and QAM; four for 8 PSK; and eight for 16 PSK.

(102) According to yet another embodiment of the device, the number of selected streams is K2.

(103) According to yet another embodiment of the device, the combining is achieved by summing all the scrambled data messages of the K number of data streams to form the at least one transmission signal s(l).

(104) According to yet another embodiment of the device, each data message comprises a finite length sequence of information symbols or a block of information symbols.

(105) According to yet another embodiment of the device, the wireless communication system is an OFDM system or a single carrier system.

(106) According to yet another embodiment of the device, the K number of data streams uses the same TCM encoder parameters and/or modulation schemes.

(107) This principle also relates to the first and second receiver devices, of the present patent application, in relation to their respective corresponding methods.

(108) Finally, it should be understood that the present patent application is not limited to the embodiments described above, but also relates to and incorporates all embodiments within the scope of the appended independent claims.