KR20180078824A - Wavelet based non-orthogonal multiple access downlink transceiver - Google Patents

Abstract

The present invention relates to a wavelet-based non-orthogonal multiple access downlink transceiver. According to the present invention, the wavelet-based non-orthogonal multiple access downlink transceiver includes: a signal allocating and converting unit which dynamically allocates signals to be transmitted to a user terminal and converts a binary bitstream signal to be transmitted in series into being in parallel; a constellation mapping unit which respectively maps input signals paralleled by the signal allocating and converting unit with signal points of a two-dimensional signal constellation; a fractional transmission power allocation (FTPA) unit which performs a coding for overlapping a plurality of user terminals in a power domain by receiving an output signal from the signal mapping unit and allocates corresponding power according to channel gains; a synthesis filter-bank unit which receives an output signal from a serial/parallel converting unit and reproduces a complete signal from each filter; and an analysis filter-bank unit which receives a signal from the outside and a signal transmitted from the synthesis filter-bank unit and analyzes components of each subchannel through each of the filters from the received signal. The present invention like the above needs a relatively smaller bandwidth than an existing OFDM spectral shaping method and is able to lower PAPR.

Description

[0001] Wavelet based non-orthogonal multiple access downlink transceiver [

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a multiple access transceiver and, more particularly, to an orthogonal frequency division multiplexing (OFDM) system by introducing a wavelet-based spectral shaping scheme into a non-orthogonal multiple access transceiver for fifth generation or later wireless mobile communication. Orthogonal frequency division multiplexing) spectral shaping scheme and a wavelet-based non-orthogonal multiple access downlink transceiver that can reduce the peak-to-average power ratio (PAPR).

4G-based communication systems are deployed in major regions around the world. However, along with the diversity of services, the growth of mobile communication traffic is also increasing day by day, which has led to the study of 5G systems. Therefore, industrial and academic research for designing a next generation (5G) mobile network has been attracting attention in order to meet the demand for huge traffic of a mobile communication network. Compared to previous generation mobile communications, 5G systems need to support high data rates, wider coverage, lower latency, improved reliability and better energy efficiency. Flexibility and scalability are key design goals of a planned 5G system to address the needs of various services.

One of the key requirements for designing future radio access (FRA) technology towards 2020 is to improve spectral efficiency by at least 3 times over conventional LTE (Long Term Evolution). Therefore, to support the exponential growth of traffic, the multiple access (MA) technology for 5G is one of the most important research projects. The MA scheme for 5G is classified as orthogonal MA and non - orthogonal MA based on resource allocation among different users. In current mobile networks, such as LTE and LTE-Advanced (LTE-A), the MA technology is used for orthogonal frequency division multiple access (OFDMA) for downlink or single carrier frequency division multiplex And is adopted in the form of single carrier frequency division multiple access (SC-FDMA). In order to accommodate future traffic demand, non-orthogonal multiple access (NOMA) with successive interference cancellation (SIC) is attracting attention as a candidate for 5G MA. In previous generations, users were multiplexed using time and frequency domains, but NOMA uses a power or code domain to serve other users. NOMA is known to significantly improve system performance by allocating different power levels to different users according to channel conditions. In order to demultiplex different users, SIC is often used at the receiver side to cope with multiuser interference.

Another promising technology that helps 5G systems achieve better spectral suppression is the design of new waveforms. Waveform selection plays an important role in system complexity, transceiver design, and link performance. Moreover, careful selection of waveforms is inevitable to avoid inter-symbol interference (ISI), one of the challenging problems of distributed channels. At 3.9 / 4G, OFDM is adopted to provide different advantages, namely robustness to multipath, low complexity, and compatibility with multiple input multiple output (MIMO) techniques. OFDM can achieve higher data rates instead of increasing system throughput, but OFDM still has many limitations. In OFDM, poor response of an IFFT / FFT filterbank causes spectral leakage, and a low side lobe of 13dB causes more ICI.

To date, all researchers are interested in OFDM-based NOMA systems. However, due to spectral efficiency degradation due to the inclusion of spectral leakage, PAPR, and cyclic prefix (CP), OFDM is not a viable option for waveform shaping to meet the spectrum requirements for 5G systems. Therefore, to improve capacity and SE (spectral efficiency), a new waveform shaping technique is required for the power domain NOMA transceiver.

On the other hand, in Japanese Patent Application Laid-Open No. 10-2014-0104315 (Patent Document 1), "OFDM communication system and method thereof for reducing peak-to-average power ratio" is disclosed, and OFDM communication The system includes an inverse fast Fourier transform unit performing an inverse fast Fourier transform on data signals to be transmitted to a receiving apparatus; A controller for performing clipping on signals output from the inverse fast Fourier transform unit with a predetermined threshold amplitude value and for comparing the clipped signal with a pre-clipping signal to determine a size of each of the subcarriers; And a subcarrier allocator for transmitting the data signal to a receiver using the determined plurality of subcarriers.

In the case of Patent Document 1 as described above, the threshold value corresponding to each of the different subcarriers is calculated using the clipping noise, and the size of the subcarriers is limited based on the calculated threshold value, Although it may be possible to effectively reduce the peak-to-average power ratio (PAPR) increase while minimizing the interference to the noise, this is basically a communication system based on the OFDM, which causes spectrum leakage or interference .

Japanese Patent Application Laid-Open No. 10-2014-0104315 (Aug. 28, 2014)

The present invention has been made in view of the above problems, and it is an object of the present invention to provide an orthogonal frequency division multiplexing (OFDM) system by introducing a wavelet-based spectrum shaping method into a non-orthogonal multiple access transceiver for fifth generation or later wireless mobile communication. Orthogonal frequency division multiplexing) spectral shaping method and a wavelet-based non-orthogonal multiple access downlink transceiver that can reduce a peak-to-average power ratio (PAPR).

According to an aspect of the present invention, there is provided a wavelet-based non-orthogonal multiple access downlink transceiver,

A signal assignment and conversion unit for dynamically allocating signals to be transmitted to the user terminal and converting the binary bit string signals to be transmitted in series into parallel signals;

A constellation mapping unit that maps input signals parallelized by the signal allocation and conversion unit to signal points of a two-dimensional signal constellation, respectively;

A fractional transmit power allocation (FTPA) scheme that receives an output signal from the signal mapping unit to perform superposition coding of a plurality of user terminals in a power domain, and allocates a corresponding power according to a channel gain; part;

A serial / parallel converter for receiving a signal transmitted through the partial transmission power allocation (FTPA) and parallelizing the binary bit string signal to be transmitted serially;

A synthesis filter-bank unit for receiving an output signal from the serial-to-parallel conversion unit and reproducing an entire signal from each filter;

An analysis filter-bank unit for receiving signals transmitted from the synthesis filter-bank unit and an external signal, and analyzing the components of each subchannel through the respective filters from the received signals;

A first parallel / serial converter for receiving an output signal (parallel signal) from the analysis filter-bank unit and converting the parallel signal into a serial signal;

A successive interference cancellation (SIC) unit for receiving an output signal from the parallel / serial converter and separating signals of a plurality of user terminals and sequentially recovering the original signal by eliminating interference;

A constellation de-mapping unit for receiving an output signal from the sequential interference cancellation (SIC) unit and extracting an original binary bit signal from the two-dimensional signal constellation; And

And a second parallel / serial converter for receiving an output signal from the constellation de-mapping unit and converting the parallel signals into serial signals once again.

Herein, the partial transmission power allocation (FTPA) allocates a low power to a user terminal having a high channel gain and a high power to a user terminal having a low channel gain when allocating a corresponding power according to a channel gain .

In addition, a signal transmitted from the synthesis filter-bank unit and an external signal are received at the front end of the analysis filter-bank unit to alleviate the distortion of the amplitude and phase of the received signal or to remove the distortion included in the signal pulse, And an equalizer that equalizes the waveform so as to be close to an ideal waveform.

According to the present invention, by introducing a wavelet-based spectrum shaping method into a non-orthogonal multiple access transceiver for fifth generation or later wireless mobile communication, orthogonal frequency division multiplexing (OFDM) spectral shaping A relatively small bandwidth is required and a peak-to-average power ratio (PAPR) can be lowered.

1 is a diagram schematically illustrating a system configuration of a wavelet-based non-orthogonal multiple access downlink transceiver according to an embodiment of the present invention.
2 is a graph showing a comparison of PAPR reduction performance between C-NOMA and W-NOMA.
3 is a diagram illustrating various types of PAPR performance comparisons for a W-NOMA based system.
Figure 4 is a diagram showing simulated PSD estimates of C-NOMA and W-NOMA.
5 is a diagram showing SE comparison of individual users for C-NOMA and W-NOMA.
Figure 6 is a diagram showing the total SE achieved by the C-NOMA and W-NOMA based systems.

The terms and words used in the present specification and claims should not be construed as limited to ordinary or dictionary terms and the inventor can properly define the concept of the term to describe its invention in the best way Should be construed in accordance with the principles and meanings and concepts consistent with the technical idea of the present invention.

Throughout the specification, when an element is referred to as " comprising ", it means that it can include other elements as well, without excluding other elements unless specifically stated otherwise. Also, the terms " part, " " module, " and " device "Lt; / RTI >

Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings.

1 is a diagram schematically illustrating a system configuration of a wavelet-based non-orthogonal multiple access downlink transceiver according to an embodiment of the present invention.

1, a wavelet-based non-orthogonal multiple access downlink transceiver 100 according to the present invention includes a signal allocation and conversion unit 101, a constellation mapping unit 102, a fractional transmit the first parallel-to-serial conversion unit 103, the serial-to-parallel conversion unit 104, the synthesis filter-bank unit 105, the analysis filter-bank unit 108, A de-interleaver 109, a sequential interference eliminator 110, a constellation de-mapping unit 111, and a second parallel / serial converter 112.

The signal assignment and conversion unit 101 dynamically allocates signals to be transmitted to a user terminal (for example, a smart phone), and converts the binary bit string signals to be transmitted serially into parallel signals.

The constellation mapping unit 102 maps input signals parallelized by the signal allocation and conversion unit 110 to signal points of a two-dimensional signal constellation.

The partial transmission power allocation (FTPA) unit 103 receives an output signal from the signal mapping unit 102 and performs superposition coding on a plurality of user terminals in a power domain. The partial transmission power allocation (FTPA) . Here, the partial transmission power allocation (FTPA) unit 103 assigns a corresponding power according to the channel gain, for example, allocates a low power to a user terminal having a high channel gain, A high power can be assigned to the terminal.

The serial / parallel conversion unit 104 receives the signal transmitted through the partial transmission power allocation (FTPA) unit 103 and parallelizes the binary bit string signal to be transmitted serially.

A synthesis filter-bank unit 105 receives an output signal from the serial-to-parallel conversion unit 104 and reproduces a complete signal from each filter.

The analysis filter-bank unit 108 receives the signal transmitted from the synthesis filter-bank unit 105 and an external signal, and analyzes the components of each subchannel through the respective filters from the received signal. Here, the signal transmitted from the synthesis filter-bank unit 105 and a signal from the outside (for example, a transmission signal from another user terminal) are received at the front end of the analysis filter-bank unit 108, The equalizer 107 may further include an equalizer 107 that mitigates distortion of the amplitude and phase of the signal or removes distortion included in the signal pulse to equalize the signal waveform to an ideal waveform.

The first parallel / serial conversion unit 109 receives the output signal (parallel signal) from the analysis filter-bank unit 108 and converts it into a serial signal.

The sequential interference cancellation (SIC) unit 110 receives the output signal from the parallel / serial converter 109, separates signals of a plurality of user terminals, and sequentially removes interference to recover the original signal.

The constellation de-mapping unit 111 receives the output signal from the sequential interference cancellation (SIC) unit 110 and extracts the original binary bit signal from the two-dimensional signal constellation.

The second parallel / serial converter 112 receives the output signal from the constellation de-mapping unit 111 and converts the parallel signals into serial signals once again.

1, reference numeral 106 denotes a radio channel through which a signal is transmitted. The dotted line box area on the upper side represents the transmitter and the dotted line box area on the lower side represents the receiver.

Hereinafter, a wavelet-based non-orthogonal multiple access downlink transceiver 100 according to the present invention having the above configuration will be described in further detail.

In the present invention, a power-domain non-orthogonal multiple access (WDM) transceiver is used as a waveform shaping method to improve SE (spectral efficiency) and reduce OOB (out of band) radiation and PAPR (peak-to- (wavelet OFDM). In the present invention, a system including NOMA and FFT-OFDM in a waveform shaping method is referred to as a conventional NOMA (C-NOMA), and a system including NOMA and WOFDM in a waveform shaping method is referred to as a wavelet NOMA (W-NOMA) do. In a simulated system, users are multiplexed in a non-orthogonal manner in the power domain, whereas two different waveform shaping techniques are considered for data transmission. Link-level performance of both systems is analyzed and compared for PAPR reduction, OOB radiation, SE and computational complexity (CC). The simulation results show that W-NOMA outperforms C-NOMA in terms of reduction, OOB radiation and SE.

Filter-bank multicarrier (FBMC) technology began in the mid-1960s. Various implementations of FBMC such as cosine modulated filter bank (CMFB), cosine modulated multitone (CMT), staggered multitone (SMT) and filtered multi-tone (FMT) are considered candidates for future radio access. CMT uses the same synthesis and analysis filter bank as the Discrete Wavelet Multitone (DWMT) based wavelet filter bank.

In the case of a transceiver according to the present invention, a wavelet filter bank such as a CMT is implemented with an octave interval or a two-channel tree structure. In the tree structure, a low pass filter (LPF) and a high pass filter (HPF) are used to decompose the input data into two subbands and then downsample to 2. The lowpass data is repeated through the same filtering process. This structure is also referred to as QMF (Quadrature Mirror Filterbank) due to the complementary relationship between a low pass filter (LPF) and a high pass filter (HPF). Therefore, PR-QMF using cosine modulation is introduced to achieve a low-complexity design process in the structure of the transceiver of the present invention. Filter banks are implemented with various types of multiple access (MA) techniques to affect multipath effects on data transmission.

Having a single transmit antenna, i.e., N _t = 1T _x in the wavelet-based non-orthogonal multiple access downlink transceiver 100 according to the present invention, for simplicity, a base station (BS), such as in the aforementioned Figure 1 single-input multiple- Output (SIMO) system, the number of receive antennas (N) at the user equipment (UE) is the antennas (N _r = NR _x ) for the total users. It is assumed that the transceiver system of the present invention has two users, a user near the BS as a near user (NEU) and a user remote from the BS as a remote user (FEU). Also, it is considered that the two users are paired with one another, which is a key feature of the power domain NOMA. Power allocation with two or more users in the NOMA system is still under investigation by several research groups. Although two users are considered in the present invention, the analysis in the transceiver system of the present invention can be applied to two or more users. W-NOMA is similar to C-NOMA. The only difference is in the way the signal waveform is transformed by replacing the DFT filter bank with a wavelet filter bank.

At the transmitter input data bits to the user equipment (UE ₁ ~UE _T) it is loaded into an adaptive manner. The signal-to-noise ratio (SNR) for all sub-channels is calculated, and the bits are loaded based on these signal-to-noise ratios. More bits are packed into subchannels with a high signal-to-noise ratio, and subchannels with a low signal-to-noise ratio are filled with fewer bits. Rate adaptive water filling algorithm is used to calculate the signal-to-noise ratio of all subchannels. The bits are mapped to symbols using M-QAM. At this point, each data stream can be represented by the following equation:

는 q번째 부반송파 내의 u번째 심볼이다. Here, Q denotes the number of subcarriers, and u denotes an index of a symbol existing in any of the qth subcarriers.

Is the u-th symbol in the q-th subcarrier.

The transmit power p is assigned to all users according to the channel gain using fractional transmit power allocation (FTPA). That is, the lower power is assigned to the user with the higher channel gain, and vice versa.

In the case of single input multiple output (SIMO), all i-th data streams with different assigned power are now multiplexed using superposition coding (SC). It allows multiplexing of all users in a non-orthogonal manner since a larger capacity can be achieved compared to orthogonal multiplexing. As a result, the superposition coded (coded) signal S is given by the following equation.

Here, x _i denotes a transmission signal of the UE _i, ρ _i represents the power allocated to UE _i. T represents the total number of users in the system.

The superposition coded signal S can be expressed by the following equation.

Here, S _{q, u, i} are QAM-encoded U-th symbols for the i-th user of the q-th subcarrier among Q subcarriers in total. It then passes through the synthesis filter bank (SFB).

OFDM, a widely used multi-carrier technique, uses a rectangular window as a prototype filter with the same size as the Fourier transform. In WOFDM, however, wavelet filters belonging to different wavelet families are used as prototype filters in the form of inverse discrete wavelet transform (IDWT) and discrete wavelet transform (DWT) on the transmitter and receiver sides, respectively. After passing through the synthesis filter bank SFB, the multiplexed QAM encoded symbol S is transformed into a wavelet symbol S _idwt . To describe the input signal S to the wavelet system, a scaling and wavelet function is required along with a known set of scaling and wavelet coefficients.

The spectrum of the input signal S is decomposed into high and low pass components after passing through a high pass filter (HPF) d _n and a low pass filter (LPF) z _n . The output signal after IDWT can be expressed by the following equation.

은 n = 0 ... j-1 인 복소수 값 QAM 심볼이다. I는 색인 집합이며 보통 정수로 취해진다. J는 양의 정수이며 분해 수준을 나타낸다. 스케일링 함수 φ(m)와 웨이브렛 함수 ψ(m)는 다음과 같이 주어진다.here

Is a complex-valued QAM symbol with n = 0 ... j-1. I is an index set and is usually taken as an integer. J is a positive integer and indicates the decomposition level. The scaling function φ (m) and the wavelet function φ (m) are given by

의 관계를 갖는다.At this time,

.

S _idwt is transmitted over the Rayleigh fading channel. the reception signal at the rth reception antenna of the i-th user, that is, N _t The received signal y _ir considering a single transmitter with an antenna is given by

h _irv represents the channel coefficient between transmit antenna v and receive antenna r for user i. eta _ir represents an additive white Gaussian noise (AWGN) with mean zero and variance for the user and antenna. S _idwt represents a signal for user i having an assigned power p _i transmitted via antenna v. The ranges of the transmit and receive antennas are r = 1,2, ..., Nr and v = 1,2, ..., Nt, respectively. The input / output relationship here can be expressed as follows.

_{Here, S idwt (S idwt = [} S idwt1, ..., S idwtNt] ') is the size of the complex transmission signal vector having a _{(N t × 1), y} ir (= [y i1, ..., y _iNr] ') is large (complex received signal vector having a N _r × 1) _{and, η ir (= [η i1} , ..., η iNr]') is added noise signal with a size (N _r × 1) It is a vector

On the receiver side, undesirable channel effects are compensated for by using equalization. The equalized data is applied to the analysis filter bank z- _n , and d- _n for DWT is given as follows.

After forward conversion, the data stream is forwarded to a sequential interference cancellation (SIC) unit for decoding. SIC can recover weak signals as well as strong signals. First, a strong signal is decoded and then a strong signal is subtracted from the combined signal to decode the weak signal. That is, when receiving a signal combining a strong signal and a weak signal, the sequential interference cancellation (SIC) unit first decodes the strong signal, and subtracts the strong signal from the combined signal to decode the weak signal. The original data bits are recovered after demodulation.

On the other hand, OFDM-based multi-carrier systems face high PAPR due to the nature of the IFFT / FFT operation. The OFDM signal is a combination of narrowband signals of different tones. When these Q narrowband signals are constructively added, the instantaneous power of these peaks becomes higher than the average signal power. As a result, the peak power of the OFDM signal will be Q times the average power. Due to the large PAPR, the high power amplifier (HPA) operates in a nonlinear mode, which results in intermodulation distortion (IMD).

The PAPR of all signals can be defined as the ratio of peak instantaneous power to average power. Therefore, the PAPR of S _idwt can be expressed by the following equation.

Here, max _n {circle over ()} denotes the maximum value of the time index n among all the instantaneous values. E {{ ² }} represents the total average calculated during one symbol period.

The PAPR of all MCM techniques depends on the shape of the waveform. Therefore, the PAPR can be reduced by selecting the appropriate waveform as shown in the equation below.

Here,? (M) is a scaling function and Q is the total number of subcarriers. However, the complementary cumulative distribution function (CCDF) defined by CCDF = 1-CDF is mainly used to describe the statistical properties of the PAPR. The CCDF of the IDWT frame shows how the PAPR of a particular frame exceeds a given threshold.

Meanwhile, simulation results will be described after applying OFDM and WOFDM waveform shaping to data received from users sharing resources in non-orthogonal mode. 2 shows a comparison of PAPR reduction performance between C-NOMA and W-NOMA.

From Fig. 2, it can be seen that with the application of the simplest wave (i. E. Haar) for WOFDM based NOMA, a gain of 2 dB is obtained without any PAPR reduction technique.

In CCDF of 10 ^-4, C-NOMA system for only have a PAPR of 14dB, W-NOMA system, its value is 12dB. Thus, it can be inferred that the pulse shape design is more likely to have a higher PAPR in a C-NOMA based system than in a W-NOMA based system.

In addition, W-NOMA-based systems provide flexibility in base function selection depending on the characteristics of the transmission scenario due to the availability of different wavelet families. On the other hand, a C-NOMA based communication system can use sines or cosines as a base function. The superior performance of FFT filterbank based communication systems compared to wavelet based communication systems in terms of PAPR has become evident in a variety of available tasks. Thus, by replacing OFDM with WOFDM, the transceiver system of the present invention can effectively reduce the PAPR problem.

Figure 3 shows a comparison of various types of PAPR performance for a W-NOMA based system. First, the behavior of W-NOMA is evaluated when a wavelet filter belonging to the same family but different in filter length / order (L) is used. Second, investigate the behavior of a system that uses the same length but other household wavelet filters.

We also investigate the effect of subcarriers (Q) on the PAPR performance of the W-NOMA system. Different shapes of CCDF are obtained for Daubechies 6 (db6), Daubechies 10 (db10), Daubechies 20 (db20), Cioflet 2 (coif 2) and Symlet (sym6). In this set of simulations, the W-NOMA system is realized by using two-stage wavelet decomposition (J).

From Figure 3 it can be seen that db6 with L = 12 and Q = 128 performs better than db6 with filters L = 12 and Q = 256. The tendency of PAPR reduction with respect to the number of subcarriers is that the increase in the number of subcarriers degrades the PAPR reduction. In practice, this increase in PAPR is directly related to the number of subcarriers that increase the maximum power of the cumulative signal. Therefore, there is a tradeoff between the number of subcarriers and the PAPR value. Comparing db6, db10 and db20 with Q = 256, it can be seen that an increase in L degrades the PAPR reduction gain.

Figure 4 shows the simulated PSD estimates of the two systems using the number of subcarriers fixed at Q = 256. From the simulation results, it can be deduced that the signal power is kept constant in the main band in both systems. However, the adjacent channel leakage rate (ACLR) of C-NOMA is higher than that of W-NOMA. At 2.5 Hz, the ACLR levels are -60 dB and -71 dB for conventional NOMA and wavelet NOMA, respectively.

Figures 5 and 6 show the spectral efficiencies of conventional and wavelet NOMA, respectively.

In FIG. 5, SE of two users of the two systems are compared in the presence of noise, interference of other users, and HPA distortion. It can be observed that U ₁ and U ₂ of WNOMA achieve better spectral gain than C-NOMA. Of WNOMA-based systems of 20dB SINDR U ₁ and U ₂ are each 3.421b / s / Hz and U ₁ and U ₂ in the other hand, CNOMA based system to achieve 1.617b / s / Hz is 2.119b / s / Hz, and Respectively. Figure 6 shows the overall theory and simulated SE of the two systems, WNOMA and CNOMA.

From the graph, the theoretical SE of NOMA at SINDR 20dB is 6.66 b / s / Hz, while the simulated SE of WNOMA and CNOMA can be analyzed to be 5.037 b / s / Hz and 3.575 b / s / Hz, respectively. This shows that the NOMA using WOFDM achieves 20% higher spectral efficiency than the conventional OFDM based NOMA under the same system parameters. This is because WOFDM-based communication systems are more robust and do not use CPs, which makes the spectrum more efficient compared to conventional transmission methods.

As described above, the wavelet-based non-orthogonal multiple access downlink transceiver according to the present invention introduces a wavelet-based spectral shaping scheme to a non-orthogonal multiple access transceiver for fifth generation or later wireless mobile communication, orthogonal frequency division multiplexing (OFDM), a relatively small bandwidth is required and a peak-to-average power ratio (PAPR) is also reduced.

While the present invention has been particularly shown and described with reference to exemplary embodiments thereof, it is to be understood that the invention is not limited to the disclosed exemplary embodiments, but many variations and modifications may be made without departing from the spirit and scope of the invention. Be clear to the technician. Accordingly, the true scope of protection of the present invention should be construed according to the following claims, and all technical ideas within the scope of the same should be construed as being included in the scope of the present invention.

100: (inventive) wavelet based non-orthogonal multiple access downlink transceiver
101: Signal allocation and conversion unit 102: Constellation mapping unit
103: partial transmission power allocation (FTPA) unit 104: serial /
105: synthesis filter-bank unit 106: radio channel for transmission
107: equalization section 108: analysis filter-bank section
109: first parallel / serial converter 110: sequential interference eliminator
111: constellation de-mapping unit 112: second parallel /

Claims (5)

A signal assignment and conversion unit for dynamically allocating signals to be transmitted to the user terminal and converting the binary bit string signals to be transmitted in series into parallel signals;
A constellation mapping unit that maps input signals parallelized by the signal allocation and conversion unit to signal points of a two-dimensional signal constellation, respectively;
A fractional transmit power allocation (FTPA) scheme that receives an output signal from the signal mapping unit to perform superposition coding of a plurality of user terminals in a power domain, and allocates a corresponding power according to a channel gain; part;
A serial / parallel converter for receiving a signal transmitted through the partial transmission power allocation (FTPA) and parallelizing the binary bit string signal to be transmitted serially;
A synthesis filter-bank unit for receiving an output signal from the serial-to-parallel conversion unit and reproducing an entire signal from each filter;
An analysis filter-bank unit for receiving signals transmitted from the synthesis filter-bank unit and an external signal, and analyzing the components of each subchannel through the respective filters from the received signals;
A first parallel / serial converter for receiving an output signal (parallel signal) from the analysis filter-bank unit and converting the parallel signal into a serial signal;
A successive interference cancellation (SIC) unit for receiving an output signal from the parallel / serial converter and separating signals of a plurality of user terminals and sequentially recovering the original signal by eliminating interference;
A constellation de-mapping unit for receiving an output signal from the sequential interference cancellation (SIC) unit and extracting an original binary bit signal from the two-dimensional signal constellation; And
And a second parallel-to-serial converter for receiving an output signal from the constellation de-mapping unit and converting the parallel signals into a serial signal once again.
The method according to claim 1,
Wherein the partial transmission power allocation (FTPA) allocates a low power to a user terminal having a high channel gain and a high power to a user terminal having a low channel gain in assigning corresponding power according to a channel gain, Wavelet - based non - orthogonal multiple access downlink transceiver.
The method according to claim 1,
A signal transmitted from the synthesis filter-bank unit and a signal from the outside are received at the front end of the analysis filter-bank unit to alleviate the distortion of the amplitude and phase of the reception signal or to remove the distortion included in the signal pulse, Based wavelet based non-orthogonal multiple access downlink transceiver further comprising an equalizer that is equalized to approximate an ideal waveform.
The method according to claim 1,
Wherein the superposition coded signal (S) by the partial transmit power allocation (FTPA) section is given by the following equation: < EMI ID = 17.1 >

Here, x _i denotes a transmission signal of the UE _i, ρ _i represents the power allocated to UE _i. T represents the total number of users in the system.
The method according to claim 1,
Wherein the sequential interference cancellation (SIC) unit decodes the strong signal and subtracts a strong signal from the combined signal to decode the weak signal when the strong signal is combined with the weak signal. Non - orthogonal multiple access downlink transceiver.

Images