CN101667893B - Virtual multi-input multi-output relay transmission method based on space-time block coding - Google Patents

Abstract

The invention relates to a virtual multi-input multi-output relay transmission method based on space-time block coding. The chain between each relay station and each base station carries out transmission in a virtual MIMO manner of space-time block coding, and the chain between user equipment and a relay station carries out transmission in a maximum specific value combination manner. Each relay station transmitter in an uplink chain adopts the transmission manner of space-time block coding of two transmission antennas and uses the same time-frequency resource to communicate with a base station receiver under the base station dispatch. In the relay station receiver of the uplink chain, estimation and detection of the transmission modulating symbols on a user terminal are finished, and in the time slot of the subsequent relay station and the base station chain, the detected modulating symbols transmitted by the user terminal are transferred to the receiver of the base station. The base station receiver in the uplink chain finishes separation of user signals by adopting a multi-user space-time combined equalization algorithm. The invention increases the network coverage area, the uplink transmission volume and the frequency band use ratio and maintains low-peak average ratio of the transmitter in the uplink chain of the relay station.

Description

Virtual multiple-input multiple-output relay transmission method based on block space-time block coding

Technical Field

The invention relates to an uplink relay transmission method of an LTE communication system. In particular to a virtual multiple input multiple output relay transmission method based on block space-time block coding, which can fully utilize the diversity gain provided by multiple antennas between a relay station and a base station and obviously provide the reliability of user link transmission in edge cells or network blind areas in a cellular mobile communication system.

Background

The LTE system adopts a single carrier frequency division multiplexing (SC-FDMA) multiple access scheme for an uplink and an Orthogonal Frequency Division Multiplexing (OFDMA) transmission scheme for a downlink, and currently, 3GPP has completed LTE Rel8 system standardization. However, there is a large gap between the LTE Rel8 system and the ITU IMT-Advanced technical requirements, which mainly results in low cell edge user throughput and spectrum efficiency. In order to meet and exceed the requirements of technical indexes of the IMT-Advanced system, the 3GPP further starts an LTE-Advanced research plan, and the plan aims to optimize an uplink/downlink transmission scheme of the LTE system on the basis of a technical framework determined by LTE Rel8 so as to meet the requirements of the IMT-Advanced system.

The downlink of the LTE-Advanced system needs to solve the core problem: the throughput of the interference limited user in the middle of the cell/the edge of the cell is improved, and the main technical means are adopted: for non-interference limited users in the center of the cell, the single-cell MU-MIMO technology is adopted to improve the frequency spectrum efficiency of a downlink; for the users with limited cell edge interference, multi-cell coordinated multi-point transmission (CoMP) is adopted to improve the throughput of the users at the cell edge. Uplink transmission needs to solve the core problem: the method overcomes the network coverage blind area caused by the limited transmitting power and shadow fading of the user terminal, improves the transmission reliability and the spectrum efficiency of the users at the edge of the cell, and adopts the main technical means: and (5) relaying the transmission. Relevant studies have shown that: in a cellular mobile communication system, network coverage can be increased and link transmission reliability can be improved through relay transmission, but because a relay station needs to use extra time/frequency resources to complete relay transmission (in a time division relay communication system, the relay station needs to use extra time slot resources; in a frequency division relay communication system, the relay station needs to use extra frequency resources), the utilization rate of a system channel is reduced. Therefore, how to improve the link reliability of cell edge users and network coverage blind areas in LTE-Advanced uplink transmission and ensure that the channel utilization rate of the system is not reduced is a problem to be solved urgently.

Fig. 1 shows a schematic diagram of a DFT-S-OFDM based relay transmission system, which consists of three different types of network devices: a base station (eNodeB), a Relay (Relay), and a User Equipment (UE). The downlink of the system works in a mode of combining MIMO with OFDM, and the uplink works in a mode of DFT-S-OFDM. In uplink transmission, a user terminal located in the coverage range of a base station directly communicates with the base station in a DFT-S-OFDM mode; and the user terminal located in the network blind area or the cell edge keeps communication with the base station through the relay station.

The communication process between the user terminal and the base station through the relay station is as follows: assuming that before the relay communication starts, the user terminal applies for the time-frequency resource used by the relay communication (the time-frequency resource used by the user terminal for the communication with the relay station (abbreviated as user time-frequency resource), and the time-frequency resource used by the relay station for forwarding the user terminal information to the base station (abbreviated as relay time-frequency resource)), the user terminal uses the user time-frequency resource allocated by the system to transmit the user information to the relay station in a DFT-S-OFDM manner, the relay station receives the user terminal transmission signal in the user time-frequency resource designated by the system, and obtains the bit information transmitted by the user after FFT, frequency domain equalization, IDFT and demodulation, and then re-modulates, DFT, maps, IFFT and inserts the cyclic prefix to obtain the relay forwarding signal, and uses the system-allocated relay time-frequency resource to forward to the base station receiver, the base station receives the signal forwarded by the relay station in the relay, and recovers the user terminal transmission information.

Compared with a non-relay mode uplink transmission scheme, the relay transmission scheme based on DFT-S-OFDM has the following advantages: the method can improve the network coverage and overcome the network communication blind area, but in a relay communication system, the scheme has the defect that the links from the user terminal to the relay station occupy additional time-frequency resources, thereby reducing the system throughput and the frequency spectrum utilization rate.

The DFT-S-OFDM-based single-antenna relay transmission scheme has the defects of low link transmission capacity of a user terminal and a relay station and low link reliability of the relay station and a base station, and has the defects that the diversity gain provided by a multi-antenna system of the link from the relay station to the base station cannot be utilized, so that the link transmission reliability of the user terminal in a marginal cell or a network blind area cannot be fully ensured, and the like. And the DFT-S-OFDM-based single antenna relay transmission scheme has a disadvantage of increasing the detection complexity of the base station receiver when the number of relay stations is large.

Disclosure of Invention

The technical problem to be solved by the invention is as follows: in an LTE uplink, a virtual multi-input multi-output relay transmission method based on block space-time block coding can be provided, which can fully utilize diversity gain provided by multiple antennas between a relay station and a base station and remarkably improve the reliability of user link transmission in a marginal cell or a network blind area in a cellular mobile communication system.

The technical scheme adopted by the invention is as follows: in a relay transmission system composed of user equipment, relay stations and a base station, a link between each relay station and the base station adopts a virtual MIMO mode of block space time block coding for transmission, and a link between the user equipment and the relay stations adopts a maximum ratio combining mode for transmission.

In the uplink, each relay station transmitter adopts a transmission mode of space-time block coding of two transmitting antenna blocks, and uses the same time-frequency resource to realize communication with a base station receiver under the scheduling of the base station.

In the relay station receiver in the uplink, the estimation and detection of the modulation symbols transmitted by the user terminal are completed, and the detected modulation symbols transmitted by the user terminal are forwarded to the receiver of the base station in the time slot of the subsequent relay station and base station link.

In the base station receiver in the uplink, the separation of user signals is completed by adopting a multi-user space-time joint equalization algorithm.

The transmission of the block space-time block coding is realized by the following steps:

the first step is as follows: modulating an input bit sequence;

the second step is as follows: performing N-point fast Fourier transform on the output of the modulator;

the third step: sending the output of the fast Fourier transform into a space-time block encoder for encoding;

the fourth step: and respectively carrying out the same processing on the two paths of signals after the coding processing, and transmitting the processed signals of each path through the corresponding antenna of the path.

The same process described in the fourth step includes the following processes:

1) respectively mapping two paths of data signals to L frequency domain subcarriers, wherein L is more than N, and N is an integer more than or equal to 1;

2) performing inverse fast Fourier transform of L points on the output of the mapper;

3) inserting cyclic prefix, D/A converting;

4) sending the signals into an intermediate frequency and radio frequency unit;

5) and transmitting via the antenna.

The signal processing of the base station receiver in the uplink comprises the following steps:

1) the reception paths corresponding to the respective reception antennas perform the same signal processing on the received signals:

a. processing a radio frequency signal from an antenna by a radio frequency and intermediate frequency unit, and sampling to form a digital baseband signal;

b. removing a cyclic prefix from the digital baseband signal;

c. performing L-point Fourier transform;

d. performing demapping processing, namely taking out data signals from the L subcarriers;

2) collecting the signals processed by each receiving channel and carrying out space-time joint equalization processing;

3) and carrying out the same signal processing on each user signal separated by the space-time joint equalizer by taking each channel as a unit, wherein the same signal processing comprises the steps of carrying out N-point inverse discrete Fourier transform, demodulation and decoding to obtain information transmitted by each user terminal.

The multi-user space-time joint equalization processing is completed by the following formula:

two-user linear zero-forcing matrix is introduced by utilizing Alamouti-like characteristic of channel matrix

$\mathrm{\Φ}=\left[\begin{array}{cc}{I}_{2N}& -{\mathrm{\Λ}}_{\mathrm{1,2}}{\mathrm{\Λ}}_{\mathrm{2,2}}^{-1}\\ -{\mathrm{\Λ}}_{\mathrm{2,1}}{\mathrm{\Λ}}_{\mathrm{1,1}}^{-1}& I_{2N}\end{array}\right]$

And constructing a modified received signal vector:

$\tilde{Y}=\mathrm{\&Phi;}\left[\begin{array}{c}{Y}_D^1\\ {\mathrm{<figure-callout\; id="2"\; label="Y\; D"\; filenames="H2009100707149E00021.png,H2009100707149E00022.png"\; state="\{\{state\}\}">Y</figure-callout>}}_D^{}\end{array}\right]=\left[\begin{array}{cc}\mathrm{\&Sigma;}& 0\\ 0& \mathrm{\&Delta;}\end{array}\right]\left[\begin{array}{c}Z\\ C\end{array}\right]+\left[\begin{array}{c}{\tilde{N}}_D^1\\ {\tilde{N}}_{\mathrm{<figure-callout\; id="2"\; label="D"\; filenames="H2009100707149E00021.png,H2009100707149E00022.png"\; state="\{\{state\}\}">D</figure-callout>}}^2\end{array}\right]$

wherein, $\mathrm{\&Sigma;}={\mathrm{\&Lambda;}}_{\mathrm{1,1}}-{\mathrm{\&Lambda;}}_{\mathrm{1,2}}{\mathrm{\&Lambda;}}_{\mathrm{2,2}}^{-1}{\mathrm{\&Lambda;}}_{\mathrm{2,1}},$ $\mathrm{\&Delta;}={\mathrm{\&Lambda;}}_{\mathrm{2,2}}-{\mathrm{\&Lambda;}}_{\mathrm{2,1}}{\mathrm{\&Lambda;}}_{\mathrm{1,1}}^{-1}{\mathrm{\&Lambda;}}_{\mathrm{1,2}},$ reuse the Alamouti-like characteristics of the matrix sigma and delta, if it is recorded $\tilde{Y}={\left[{\tilde{Y}}_1{\tilde{Y}}_2\right]}^T,$ Then

$\tilde{Y}=\mathrm{\&Phi;}\left[\begin{array}{c}{Y}_D^1\\ {\mathrm{<figure-callout\; id="2"\; label="Y\; D"\; filenames="H2009100707149E00021.png,H2009100707149E00022.png"\; state="\{\{state\}\}">Y</figure-callout>}}_D^{}\end{array}\right]=\left[\begin{array}{cc}\mathrm{\&Sigma;}& 0\\ 0& \mathrm{\&Delta;}\end{array}\right]\left[\begin{array}{c}Z\\ C\end{array}\right]+\left[\begin{array}{c}{\tilde{N}}_D^1\\ {\tilde{N}}_{\mathrm{<figure-callout\; id="2"\; label="D"\; filenames="H2009100707149E00021.png,H2009100707149E00022.png"\; state="\{\{state\}\}">D</figure-callout>}}^2\end{array}\right]$

The treatment is carried out in the following way:

${\widetilde{\stackrel{~}{Y}}}_1={\mathrm{\&Sigma;}}^H{\tilde{Y}}_1={\mathrm{\&Sigma;}}^H\mathrm{\&Sigma;Z}+{\widetilde{\stackrel{~}{N}}}_{\mathrm{<figure-callout\; id="1"\; label="D"\; filenames="H2009100707149E00021.png,H2009100707149E00022.png"\; state="\{\{state\}\}">D</figure-callout>}}^1$

${\widetilde{\stackrel{~}{Y}}}_2={\mathrm{\&Delta;}}^H{\tilde{Y}}_2={\mathrm{\&Delta;}}^H\mathrm{\&Delta;C}+{\widetilde{\stackrel{~}{N}}}_{\mathrm{<figure-callout\; id="2"\; label="D"\; filenames="H2009100707149E00021.png,H2009100707149E00022.png"\; state="\{\{state\}\}">D</figure-callout>}}^2$

and use ${\mathrm{\&Sigma;}}^H\mathrm{\&Sigma;}=\left(\begin{array}{cc}\mathrm{<figure-callout\; id="0"\; label="A"\; filenames="H2009100707149E00111.png"\; state="\{\{state\}\}">A</figure-callout>}& 0\\ 0& A\end{array}\right)$ And ${\mathrm{\&Delta;}}^H\mathrm{\&Delta;}=\left(\begin{array}{cc}\mathrm{<figure-callout\; id="0"\; label="B"\; filenames="H2009100707149E00111.png"\; state="\{\{state\}\}">B</figure-callout>}& 0\\ 0& B\end{array}\right)$ the method is a diagonal array and is further simplified as follows:

${\widetilde{\stackrel{~}{Y}}}_{\mathrm{1,1}}=\mathrm{AZ}\left(k\right)+{\widetilde{\stackrel{~}{N}}}_{\mathrm{<figure-callout\; id="1"\; label="D"\; filenames="H2009100707149E00021.png,H2009100707149E00022.png"\; state="\{\{state\}\}">\; <figure-callout\; id="1"\; label="D"\; filenames="H2009100707149E00021.png,H2009100707149E00022.png"\; state="\{\{state\}\}">D</figure-callout>\; </figure-callout>}}^{\mathrm{1,1}};$ ${\widetilde{\stackrel{~}{Y}}}_{\mathrm{1,2}}=\mathrm{AZ}(k+1)+{\widetilde{\stackrel{~}{N}}}_{\mathrm{<figure-callout\; id="1"\; label="D"\; filenames="H2009100707149E00021.png,H2009100707149E00022.png"\; state="\{\{state\}\}">\; <figure-callout\; id="2"\; label="D"\; filenames="H2009100707149E00021.png,H2009100707149E00022.png"\; state="\{\{state\}\}">D</figure-callout>\; </figure-callout>}}^{\mathrm{1,2}}$

${\widetilde{\stackrel{~}{Y}}}_{\mathrm{2,1}}=\mathrm{BC}\left(k\right)+{\widetilde{\stackrel{~}{N}}}_{\mathrm{<figure-callout\; id="2"\; label="D"\; filenames="H2009100707149E00021.png,H2009100707149E00022.png"\; state="\{\{state\}\}">\; <figure-callout\; id="1"\; label="D"\; filenames="H2009100707149E00021.png,H2009100707149E00022.png"\; state="\{\{state\}\}">D</figure-callout>\; </figure-callout>}}^{\mathrm{2,1}};$ ${\widetilde{\stackrel{~}{Y}}}_{\mathrm{2,2}}=\mathrm{BC}(k+1)+{\widetilde{\stackrel{~}{N}}}_{\mathrm{<figure-callout\; id="2"\; label="D"\; filenames="H2009100707149E00021.png,H2009100707149E00022.png"\; state="\{\{state\}\}">\; <figure-callout\; id="2"\; label="D"\; filenames="H2009100707149E00021.png,H2009100707149E00022.png"\; state="\{\{state\}\}">D</figure-callout>\; </figure-callout>}}^{\mathrm{2,2}}$

wherein, ${\widetilde{\stackrel{~}{Y}}}_1={\left[{\widetilde{\stackrel{~}{Y}}}_{\mathrm{1,1}}^T{,\widetilde{\stackrel{~}{Y}}}_{\mathrm{1,2}}^T\right]}^T,$ ${\widetilde{\stackrel{~}{Y}}}_2={[{\widetilde{\stackrel{~}{Y}}}_{\mathrm{2,1}}^T,{\widetilde{\stackrel{~}{T}}}_{\mathrm{2,2}}^T]}^T,$ ${\widetilde{\stackrel{~}{N}}}_{\mathrm{<figure-callout\; id="1"\; label="D"\; filenames="H2009100707149E00021.png,H2009100707149E00022.png"\; state="\{\{state\}\}">D</figure-callout>}}^1={[{\widetilde{\stackrel{~}{N}}}_{\mathrm{<figure-callout\; id="1"\; label="D"\; filenames="H2009100707149E00021.png,H2009100707149E00022.png"\; state="\{\{state\}\}">D</figure-callout>}}^{\mathrm{1,1}T},{\widetilde{\stackrel{~}{N}}}_{\mathrm{<figure-callout\; id="1"\; label="D"\; filenames="H2009100707149E00021.png,H2009100707149E00022.png"\; state="\{\{state\}\}">D</figure-callout>}}^{\mathrm{1,2}T}]}^T,$ ${\widetilde{\stackrel{~}{N}}}_{\mathrm{<figure-callout\; id="2"\; label="D"\; filenames="H2009100707149E00021.png,H2009100707149E00022.png"\; state="\{\{state\}\}">D</figure-callout>}}^2={[{\widetilde{\stackrel{~}{N}}}_{\mathrm{<figure-callout\; id="2"\; label="D"\; filenames="H2009100707149E00021.png,H2009100707149E00022.png"\; state="\{\{state\}\}">D</figure-callout>}}^{\mathrm{2,1}T},{\widetilde{\stackrel{~}{N}}}_{\mathrm{<figure-callout\; id="2"\; label="D"\; filenames="H2009100707149E00021.png,H2009100707149E00022.png"\; state="\{\{state\}\}">D</figure-callout>}}^{\mathrm{2,2}T}]}^T$ then, by using the diagonal matrix characteristics of A and B, the estimation of { Z (k), Z (k +1) } and { C (k), C (k +1) } can be conveniently obtained as follows:

$\tilde{Z}\left(k\right)=A^{-1}{\widetilde{\stackrel{~}{Y}}}_{\mathrm{1,1}};$ $\tilde{Z}(k+1)=A^{-1}{\widetilde{\stackrel{~}{Y}}}_{\mathrm{1,2}}$

$\tilde{C}\left(k\right)=A^{-1}{\widetilde{\stackrel{~}{Y}}}_{\mathrm{2,1}};$ $\tilde{C}(k+1)=A^{-1}{\widetilde{\stackrel{~}{Y}}}_{\mathrm{2,2}}$

and

carrying out IDFT operation on N points to obtain

And

finally are respectively paired

Andthe maximum likelihood detection is carried out on each component to obtain the estimated values of the relay forwarding symbol vectors { z (k), z (k +1) } and { c (k), c (k +1) }

And

the virtual MIMO relay transmission method based on the block space-time block coding can fully utilize the diversity gain provided by multiple antennas between the relay station and the base station, obviously provide the reliability of user link transmission in edge cells or network blind areas in a cellular mobile communication system, and further increase the network coverage; in addition, between the relay station and the base station link, the virtual MIMO technology is utilized to obviously improve the uplink transmission capacity and the frequency band utilization rate and overcome the defect of reduced frequency spectrum efficiency of the traditional relay scheme; the block space time block coding based transmission scheme also maintains the excellent characteristics of low peak-to-average ratio of the uplink transmitter of the relay station. The method of the invention can be applied to LTE-Advanced and 4G broadband mobile communication systems.

Drawings

FIG. 1 is a diagram of single antenna relay transmission based on DFT-S-OFDM;

fig. 2 is a schematic diagram of LTE-based TDD uplink virtual MIMO relay transmission;

fig. 3 is a schematic diagram of a ue and relay station link transmission based on maximal ratio combining;

FIG. 4 is a schematic diagram of link transmission between a B-STBC-based virtual MIMO relay station and a base station;

fig. 5 is a relay station transmitter flow diagram based on block space-time block coding;

fig. 6 is an uplink base station virtual MIMO receiver flow diagram;

fig. 7-1 is a flow diagram of a first embodiment of LTE-based TDD uplink virtual MIMO relay transmission;

fig. 7-2 is a flowchart of a second embodiment of LTE-based TDD uplink virtual MIMO relay transmission;

fig. 7-3 is a flowchart of a third embodiment of LTE-based TDD uplink virtual MIMO relay transmission;

fig. 7-4 are flow diagrams of a fourth embodiment of LTE-based TDD uplink virtual MIMO relay transmission;

fig. 7-5 are flowcharts of a fifth embodiment of LTE TDD uplink virtual MIMO based relay transmission;

fig. 7-6 are flowcharts of a fifth embodiment of LTE TDD uplink virtual MIMO based relay transmission;

fig. 8-1 is a BER performance comparison (QPSK modulation) for the virtual MIMO relay transmission method;

fig. 8-2 is a virtual MIMO relay transmission scheme BER performance comparison (16QAM modulation).

Detailed Description

The following describes the virtual mimo relay transmission method based on block space-time block coding in detail with reference to the accompanying drawings.

As shown in fig. 2, a base station (eNodeB) is located in the center of a cell, several fixed Relay stations (relays) are deployed in the cell, there is no direct transmission link between a network blind area and a cell edge user terminal and the base station, and the Relay stations provide Relay transmission service for a cell edge and a user terminal (UE) in the network blind area in a time division manner. In the time division relay communication system, a base station time slot is divided into two parts, one part of the time slot is allocated to a link from a user terminal to a relay station for transmission, and the other part of the time slot is allocated to a link from the relay station to the base station or a link from the user terminal to the base station for use.

In the uplink of an LTE TDD (long term evolution planning based on time division duplex) system, considering that the volume of a user terminal is limited, it is assumed that the user terminal is installed with only a single antenna, a fixed deployment relay station is installed with a plurality of receiving/transmitting antennas, and a base station is also equipped with a plurality of receiving antennas. In addition, it is further assumed that the base station completely knows the channel information from each relay station to the base station by using the channel estimation, and the relay station can also obtain the channel information of its serving ue through the channel estimation.

The invention provides a virtual multiple-input multiple-output relay transmission method based on block space-time block coding, belonging to a detection-forwarding scheme (DF). In the first stage of link transmission from the user terminal to the relay station, the relay station receiver adopts a maximum ratio combining algorithm to receive information transmitted by the user terminal; in the second stage of link transmission from the relay station to the base station, each relay station adopts block space time block coding (B-STBC) transmission, and a plurality of relay stations work in a virtual MIMO mode under the scheduling of the base station.

The invention relates to a virtual MIMO relay transmission method based on block space-time block coding, in particular to a relay transmission system composed of user equipment A, relay stations B and a base station C, wherein a link between each relay station B and the base station C adopts a virtual MIMO mode of block space-time block coding for transmission, and a link between the user equipment A and the relay stations B adopts a maximum ratio combining mode for transmission.

Fig. 3 shows a method for link transmission between a ue and a relay station based on maximal ratio combining. Each user terminal uses a standard DFT-S-OFDM transmitter, and the working principle is as follows: the user terminal source output bit sequence is sent to a modulator after interleaving and channel coding, and the k modulation symbol grouping of the user terminal is expressed as x (k) ═ x (k, 0), x (k, 1),.. multidot.x (k, i),. multidot.x (k, N-1)]^TWherein X (k, i) represents the ith modulation symbol in the kth modulation symbol group of the user terminal, the length of the modulation symbol group N is the same as the number of sub-channels allocated to the user terminal by the base station, and X (k) is represented as X (k) ([ X (k, 0), X (k, 1),.. multidot.,. X (k, j),. multidot.,. X (k, N-1) after N-point DFT preprocessing]^TX (k, j) and X (k, i) are discrete fourier transform relationships:

$X(k,j)=\underset{i=0}{\overset{N-1}{\mathrm{\&Sigma;}}}x(k,i)\&CenterDot;e^{-\mathrm{<figure-callout\; id="2"\; label="j"\; filenames="H2009100707149E00021.png,H2009100707149E00022.png"\; state="\{\{state\}\}">j</figure-callout>}2\mathrm{\&pi;ij}/N},j=0,...,N-1---\left(1\right)$

x (k) is mapped by a mapper to N consecutive subchannel { l | l ═ 0., N-1} transmissions assigned by the base station; the output signal of the mapper is transformed by L-point IFFT, then inserted into the cyclic prefix, and sent to the intermediate frequency and radio frequency unit after D/A conversion, and finally the radio frequency signal is sent to a single transmitting antenna.

In the relay station receiver, the radio frequency signal from the antenna is processed by the radio frequency and intermediate frequency unit, and is sent to the baseband digital signal processing unit after A/D conversion. The digital base signal firstly removes the cyclic prefix, then carries out L-point FFT conversion, a demapper extracts a receiving signal from { L | L ═ 0.., N-1} sub-channel, and the kth sub-channel receiving signal Y of the s-th receiving antenna of the relay station receiver at the kth moment_R ^s(k, l) is represented by:

$Y_R^s(k,l)=H_{\mathrm{SR}}^{s,1}(k,l)X(k,l)+N_R^s(k,l),l=0,...,N-1,s=0,...,N_{\mathrm{rr}}---\left(2\right)$

wherein H_SR ^s，t(k, l) represents the frequency response of the ith sub-channel from the tth transmitting antenna of the user terminal to the ith receiving antenna of the relay station at the kth time, X (k, l) represents the complex signal transmitted by the ith sub-channel at the kth time of the user terminal, and N_R ^s(k, l) represents that the complex white Gaussian noise is received by the ith sub-channel of the s-th receiving antenna of the relay station at the kth moment, N_rrRepresenting the number of antennas of the relay station receiver. Relay station N_rrMaximum ratio combination is carried out on the signals received by the same sub-channels of the antennas, and an estimated value of X (k, l) is obtained through zero-forcing equalization:

$\hat{X}(k,l)=\frac{\underset{s=0}{\overset{N_{\mathrm{rr}}-1}{\mathrm{\&Sigma;}}}({\left(H_{\mathrm{SR}}^{s,1}(k,l)\right)}^{*}\&times;Y_R^s(k,l\left)\right)}{\underset{s=0}{\overset{N_{\mathrm{rr}}-1}{\mathrm{\&Sigma;}}}{\left|H_{\mathrm{SR}}^{s,1}(k,l)\right|}^2},l=0,...,N-1---\left(3\right)$

further on the signal vector $\hat{X}\left(k\right)=[\hat{X}(k,0),\hat{X}(k,1),...,\hat{X}(k,N-1)]$ Carrying out N-point IDFT processing to obtain $\hat{x}\left(k\right)=[\hat{x}(k,0),\hat{x}(k,1),...,\hat{x}(k,N-1)],$ Last pair of

And each component is subjected to maximum likelihood detection to obtain an estimated value of a modulation symbol transmitted by the user terminal, and the relay station transmitter transmits a detected signal to the base station receiver in a subsequent relay station-base station link time slot.

In the uplink, each relay station B transmitter adopts a transmission mode of space-time block coding of two transmitting antenna blocks, and uses the same time-frequency resource to realize communication with a base station C receiver under the scheduling of the base station C.

The transmitter transmission based on block space-time block coding is as shown in fig. 5, and is realized by the following steps:

the first step is as follows: modulating an input bit sequence;

the second step is as follows: performing N-point Fast Fourier Transform (FFT) on the output of the modulator;

the third step: the output of the fast Fourier transform is sent to a block space-time block coding device B-STBC for coding processing;

the fourth step: and respectively carrying out the same processing on the two paths of signals after the coding processing, and transmitting the processed signals of each path through the corresponding antenna of the path.

The same process described in the fourth step includes the following processes:

1) respectively mapping two paths of data signals to L frequency domain subcarriers, wherein L is more than N, and N is an integer more than or equal to 1;

2) performing Inverse Fast Fourier Transform (IFFT) of L points on the output of the mapper;

3) inserting cyclic prefix, D/A converting;

4) sending the signals into an intermediate frequency and radio frequency unit;

5) and transmitting via the antenna.

Fig. 4 shows a link transmission method based on a B-STBC virtual MIMO relay station and a base station. Relay station 1, 2, … N_rtBoth adopt two-antenna block space-time block coding (B-STBC) transmission scheme, and use the same/frequency resource to form 2N under the scheduling of the base station_rt×N_dThe virtual MIMO system of (1). In an actual communication system, considering that the detection complexity of a base station receiver is extremely high when the number of relay nodes based on virtual MIMO is large, a virtual MIMO system formed by two relay stations is generally considered, and since the operating principle of each relay station transmitter is the same, the block space time block coding process will be described below by taking only the relay station 1 as an example.

As shown in fig. 5, the modulation symbol packet to be transmitted by the 1 st relay station at the kth time is represented as z (k) ═ z (k, 0), z (k, 1),. ·, z (k, N-1)]^TZ (k) is represented by Z (k) ═ Z (k, 0), Z (k, 1), Z (k, N-1) after N-point DFT treatment]^TTwo continuous groups { Z (k), Z (k +1) } of the relay station 1 are simultaneously sent to a block space time block coder (B-STBC) for coding processing, and the block space time block coder outputs signal vectors { Z (k), -Z^*(k +1) } into the first launching branch, { Z (k +1), Z^*(k) Sending the data to a second emission branch; mapping two branch signals to N continuous sub-channels { l | l ═ 0., N-1} transmission allocated by a base station through a mapper; the output signal of the mapper is transformed by L-point IFFT, then inserted into the cyclic prefix, and sent to the intermediate frequency and radio frequency unit after D/A conversion, and finally the transmission signal is sent to two antennas for transmission.

The transmission process of the relay station 2 is completely the same as that of the relay station 1, the modulation symbol vector to be transmitted is marked as C (k), the signal vector after DFT preprocessing is C (k), two continuous groups of the relay station 2 are { C (k), C (k +1) }, and the signal vector is sent to the 1 st transmitting branch circuit after block space time group coding is { C (k), -C^*(k +1) }, the signal vector sent to the 2 nd transmitting branch is { C (k +1), C^*(k)}。

In the receiver of the relay station B in the uplink, the estimation and detection of the modulation symbol transmitted by the user terminal a are completed, and the detected modulation symbol transmitted by the user terminal a is forwarded to the receiver of the base station C in the subsequent time slot of the link between the relay station B and the base station C.

The signal processing of the base station C receiver in the uplink is shown in fig. 6, and includes the following steps:

1) the reception paths corresponding to the respective reception antennas perform the same signal processing on the received signals:

a. processing a radio frequency signal from an antenna by a radio frequency and intermediate frequency unit, and sampling to form a digital baseband signal;

b. removing a cyclic prefix from the digital baseband signal;

c. performing L-point Fourier transform;

d. performing demapping processing, namely taking out data signals from the L subcarriers;

2) collecting the signals processed by each receiving channel and carrying out space-time joint equalization processing;

3) and carrying out the same signal processing on each user signal separated by the space-time joint equalizer by taking each channel as a unit, wherein the same signal processing comprises the steps of carrying out N-point inverse discrete Fourier transform, demodulation and decoding to obtain information transmitted by each user terminal.

For convenience of description, the base station receiver uses two pairs of receiving antennas, and the transmission scheme can be conveniently popularized to the condition that the number of the receiving antennas is 1/4/6/8. In the base station receiver, the radio frequency signal from the antenna is processed by the radio frequency and intermediate frequency units and forms a digital baseband signal after sampling, the digital baseband signal is subjected to L-point FFT (fast Fourier transform) after removing a cyclic prefix, and the 1 st subchannel receiving signal Y of the 1 st receiving antenna at the kth moment of the base station receiver_D ¹(k, l) is represented by:

$Y_D^1(k,l)={\mathrm{<figure-callout\; id="1"\; label="H\; RD"\; filenames="H2009100707149E00021.png,H2009100707149E00022.png"\; state="\{\{state\}\}">H</figure-callout>}}_{\mathrm{RD}}^{}(k,l)Z(k,l)+{\mathrm{<figure-callout\; id="1"\; label="H\; RD"\; filenames="H2009100707149E00021.png,H2009100707149E00022.png"\; state="\{\{state\}\}">H</figure-callout>}}_{\mathrm{RD}}^{}(k,l)Z(k+1,l)+{\mathrm{<figure-callout\; id="1"\; label="G\; RD"\; filenames="H2009100707149E00021.png,H2009100707149E00022.png"\; state="\{\{state\}\}">G</figure-callout>}}_{\mathrm{RD}}^{}(k,l)C(k,l)+{\mathrm{<figure-callout\; id="1"\; label="G\; RD"\; filenames="H2009100707149E00021.png,H2009100707149E00022.png"\; state="\{\{state\}\}">G</figure-callout>}}_{\mathrm{RD}}^{}(k,l)C(k+1,l)+N_D^1(k,l)---\left(4\right)$

wherein H_RD ^m，n(k, l) and G_RD ^m，n(k, l) respectively represent the frequency response of the l sub-channel from the nth transmitting antenna of the relay station 1 and the relay station 2 to the mth receiving antenna of the base station at the kth time, N_D ^m(k, l) represents that complex white Gaussian noise is input into the ith subchannel of the mth receiving antenna of the base station at the time of k. Further, assuming that the frequency response of each sub-channel from the relay station to the base station at the k-th and k + 1-th time remains constant, the l-th sub-channel receiving signal Y of the 1-th receiving antenna at the k + 1-th time of the base station can be obtained_D ¹The first subchannel receiving signal Y of the 2 nd receiving antenna at the (k +1, l) time and the k +1 time_D ²(k, l) and Y_D ²(k+1，l)：

$Y_D^1(k+1,l)=-{\mathrm{<figure-callout\; id="1"\; label="H\; RD"\; filenames="H2009100707149E00021.png,H2009100707149E00022.png"\; state="\{\{state\}\}">H</figure-callout>}}_{\mathrm{RD}}^{}(k,l)Z^{*}(k+1,l)+{\mathrm{<figure-callout\; id="1"\; label="H\; RD"\; filenames="H2009100707149E00021.png,H2009100707149E00022.png"\; state="\{\{state\}\}">H</figure-callout>}}_{\mathrm{RD}}^{}(k,l)Z^{*}(k,l)-{\mathrm{<figure-callout\; id="1"\; label="G\; RD"\; filenames="H2009100707149E00021.png,H2009100707149E00022.png"\; state="\{\{state\}\}">G</figure-callout>}}_{\mathrm{RD}}^{}(k,l)C^{*}(k+1,l)+{\mathrm{<figure-callout\; id="1"\; label="G\; RD"\; filenames="H2009100707149E00021.png,H2009100707149E00022.png"\; state="\{\{state\}\}">G</figure-callout>}}_{\mathrm{RD}}^{}(k,l)C^{*}(k,l)+N_D^1(k+1,l)---\left(5\right)$

$Y_D^2(k,l)={\mathrm{<figure-callout\; id="2"\; label="H\; RD"\; filenames="H2009100707149E00021.png,H2009100707149E00022.png"\; state="\{\{state\}\}">H</figure-callout>}}_{\mathrm{RD}}^{}(k,l)Z(k,l)+{\mathrm{<figure-callout\; id="2"\; label="H\; RD"\; filenames="H2009100707149E00021.png,H2009100707149E00022.png"\; state="\{\{state\}\}">H</figure-callout>}}_{\mathrm{RD}}^{}(k,l)Z(k+1,l)+{\mathrm{<figure-callout\; id="2"\; label="G\; RD"\; filenames="H2009100707149E00021.png,H2009100707149E00022.png"\; state="\{\{state\}\}">G</figure-callout>}}_{\mathrm{RD}}^{}(k,l)C(k,l)+{\mathrm{<figure-callout\; id="2"\; label="G\; RD"\; filenames="H2009100707149E00021.png,H2009100707149E00022.png"\; state="\{\{state\}\}">G</figure-callout>}}_{\mathrm{RD}}^{}(k,l)C(k+1,l)+N_D^2(k,l)---\left(6\right)$

$Y_D^2(k+1,l)=-{\mathrm{<figure-callout\; id="2"\; label="H\; RD"\; filenames="H2009100707149E00021.png,H2009100707149E00022.png"\; state="\{\{state\}\}">H</figure-callout>}}_{\mathrm{RD}}^{}(k,l)Z^{*}(k+1,l)+{\mathrm{<figure-callout\; id="2"\; label="H\; RD"\; filenames="H2009100707149E00021.png,H2009100707149E00022.png"\; state="\{\{state\}\}">H</figure-callout>}}_{\mathrm{RD}}^{}(k,l)Z^{*}(k,l)-{\mathrm{<figure-callout\; id="2"\; label="G\; RD"\; filenames="H2009100707149E00021.png,H2009100707149E00022.png"\; state="\{\{state\}\}">G</figure-callout>}}_{\mathrm{RD}}^{}(k,l)C^{*}(k+1,l)+{\mathrm{<figure-callout\; id="2"\; label="G\; RD"\; filenames="H2009100707149E00021.png,H2009100707149E00022.png"\; state="\{\{state\}\}">G</figure-callout>}}_{\mathrm{RD}}^{}(k,l)C^{*}(k,l)+N_D^2(k+1,l)---\left(7\right)$

Equations (4), (5), (6) and (7) are further expressed in matrix form as:

$\left[\begin{array}{c}{Y}_D^1\left(k\right)\\ Y_D^1{(k+1)}^{*}\\ Y_D^2\left(k\right)\\ Y_D^2{(k+1)}^{*}\end{array}\right]=\left[\begin{array}{cccc}{\mathrm{<figure-callout\; id="1"\; label="H\; RD"\; filenames="H2009100707149E00021.png,H2009100707149E00022.png"\; state="\{\{state\}\}">H</figure-callout>}}_{\mathrm{RD}}^{}& {\mathrm{<figure-callout\; id="1"\; label="H\; RD"\; filenames="H2009100707149E00021.png,H2009100707149E00022.png"\; state="\{\{state\}\}">H</figure-callout>}}_{\mathrm{RD}}^{}& {\mathrm{<figure-callout\; id="1"\; label="G\; RD"\; filenames="H2009100707149E00021.png,H2009100707149E00022.png"\; state="\{\{state\}\}">G</figure-callout>}}_{\mathrm{RD}}^{}& {\mathrm{<figure-callout\; id="1"\; label="G\; RD"\; filenames="H2009100707149E00021.png,H2009100707149E00022.png"\; state="\{\{state\}\}">G</figure-callout>}}_{\mathrm{RD}}^{}\\ {\mathrm{<figure-callout\; id="1"\; label="H\; RD"\; filenames="H2009100707149E00021.png,H2009100707149E00022.png"\; state="\{\{state\}\}">\; <figure-callout\; id="2"\; label="H\; RD"\; filenames="H2009100707149E00021.png,H2009100707149E00022.png"\; state="\{\{state\}\}">H</figure-callout>\; </figure-callout>}}_{\mathrm{RD}}^{}& -{\mathrm{<figure-callout\; id="1"\; label="H\; RD"\; filenames="H2009100707149E00021.png,H2009100707149E00022.png"\; state="\{\{state\}\}">\; <figure-callout\; id="1"\; label="H\; RD"\; filenames="H2009100707149E00021.png,H2009100707149E00022.png"\; state="\{\{state\}\}">H</figure-callout>\; </figure-callout>}}_{\mathrm{RD}}^{}& {\mathrm{<figure-callout\; id="1"\; label="G\; RD"\; filenames="H2009100707149E00021.png,H2009100707149E00022.png"\; state="\{\{state\}\}">\; <figure-callout\; id="2"\; label="G\; RD"\; filenames="H2009100707149E00021.png,H2009100707149E00022.png"\; state="\{\{state\}\}">G</figure-callout>\; </figure-callout>}}_{\mathrm{RD}}^{}& -{\mathrm{<figure-callout\; id="1"\; label="G\; RD"\; filenames="H2009100707149E00021.png,H2009100707149E00022.png"\; state="\{\{state\}\}">\; <figure-callout\; id="1"\; label="G\; RD"\; filenames="H2009100707149E00021.png,H2009100707149E00022.png"\; state="\{\{state\}\}">G</figure-callout>\; </figure-callout>}}_{\mathrm{RD}}^{}\\ {\mathrm{<figure-callout\; id="2"\; label="H\; RD"\; filenames="H2009100707149E00021.png,H2009100707149E00022.png"\; state="\{\{state\}\}">H</figure-callout>}}_{\mathrm{RD}}^{}& {\mathrm{<figure-callout\; id="2"\; label="H\; RD"\; filenames="H2009100707149E00021.png,H2009100707149E00022.png"\; state="\{\{state\}\}">H</figure-callout>}}_{\mathrm{RD}}^{}& {\mathrm{<figure-callout\; id="2"\; label="G\; RD"\; filenames="H2009100707149E00021.png,H2009100707149E00022.png"\; state="\{\{state\}\}">G</figure-callout>}}_{\mathrm{RD}}^{}& {\mathrm{<figure-callout\; id="2"\; label="G\; RD"\; filenames="H2009100707149E00021.png,H2009100707149E00022.png"\; state="\{\{state\}\}">G</figure-callout>}}_{\mathrm{RD}}^{}\\ {\mathrm{<figure-callout\; id="2"\; label="H\; RD"\; filenames="H2009100707149E00021.png,H2009100707149E00022.png"\; state="\{\{state\}\}">\; <figure-callout\; id="2"\; label="H\; RD"\; filenames="H2009100707149E00021.png,H2009100707149E00022.png"\; state="\{\{state\}\}">H</figure-callout>\; </figure-callout>}}_{\mathrm{RD}}^{}& -{\mathrm{<figure-callout\; id="2"\; label="H\; RD"\; filenames="H2009100707149E00021.png,H2009100707149E00022.png"\; state="\{\{state\}\}">\; <figure-callout\; id="1"\; label="H\; RD"\; filenames="H2009100707149E00021.png,H2009100707149E00022.png"\; state="\{\{state\}\}">H</figure-callout>\; </figure-callout>}}_{\mathrm{RD}}^{}& {\mathrm{<figure-callout\; id="2"\; label="G\; RD"\; filenames="H2009100707149E00021.png,H2009100707149E00022.png"\; state="\{\{state\}\}">\; <figure-callout\; id="2"\; label="G\; RD"\; filenames="H2009100707149E00021.png,H2009100707149E00022.png"\; state="\{\{state\}\}">G</figure-callout>\; </figure-callout>}}_{\mathrm{RD}}^{}& -{\mathrm{<figure-callout\; id="2"\; label="G\; RD"\; filenames="H2009100707149E00021.png,H2009100707149E00022.png"\; state="\{\{state\}\}">\; <figure-callout\; id="1"\; label="G\; RD"\; filenames="H2009100707149E00021.png,H2009100707149E00022.png"\; state="\{\{state\}\}">G</figure-callout>\; </figure-callout>}}_{\mathrm{RD}}^{}\end{array}\right]\left[\begin{array}{c}Z\left(k\right)\\ Z(k+1)\\ C\left(k\right)\\ C(k+1)\end{array}\right]+\left[\begin{array}{c}{N}_D^1\left(k\right)\\ N_D^1{(k+1)}^{*}\\ N_D^2\left(k\right)\\ N_D^2{(k+1)}^{*}\end{array}\right]---\left(8\right)$

wherein, T_D ^m(k) N subchannel received signal vectors representing the mth receive antenna at time k of the base station receiver, $H_{\mathrm{RD}}^{m,n}=\mathrm{diag}(H_{\mathrm{RD}}^{m,n}(k,1),H_{\mathrm{RD}}^{m,n}(k,2),...,H_{\mathrm{RD}}^{m,n}(k,l),...H_{\mathrm{RD}}^{m,n}(k,N))$ representing the frequency response matrix from the nth transmit antenna of the relay station 1 to the mth receive antenna of the base station at time k, $G_{\mathrm{RD}}^{m,n}=\mathrm{diag}(G_{\mathrm{RD}}^{m,n}(k,1),G_{\mathrm{RD}}^{m,n}(k,2),...,G_{\mathrm{RD}}^{m,n}(k,l),...G_{\mathrm{RD}}^{m,n}(k,N))$ representing the frequency response matrix from the nth transmit antenna of the relay station 2 to the mth receive antenna of the base station at time k, N_D ^m(k) And the complex white Gaussian noise vector represents the input of N sub-channels of the mth receiving antenna of the base station at the time k.

For the received signal model given in (8), the base station receiver can directly use the linear zero forcing or minimum mean square error detection algorithm to obtain the estimated values of the signal vectors transmitted by the two user terminals, but when the number of sub-channels allocated by the base station is large, the dimension of the channel transmission matrix in (8) is very large (4N × 4N), the matrix inversion method is directly used for signal detection, and the complexity is very high, and a low-complexity virtual MIMO detection algorithm is given below. (8) Further expressed in the form of a block matrix:

$\left[\begin{array}{c}{Y}_D^1\\ {\mathrm{<figure-callout\; id="2"\; label="Y\; D"\; filenames="H2009100707149E00021.png,H2009100707149E00022.png"\; state="\{\{state\}\}">Y</figure-callout>}}_D^{}\end{array}\right]=\left[\begin{array}{cc}{\mathrm{\&Lambda;}}_{\mathrm{1,1}}& {\mathrm{\&Lambda;}}_{\mathrm{1,2}}\\ {\mathrm{\&Lambda;}}_{\mathrm{2,1}}& {\mathrm{\&Lambda;}}_{\mathrm{2,2}}\end{array}\right]\left[\begin{array}{c}Z\\ C\end{array}\right]+\left[\begin{array}{c}{N}_D^1\\ N_D^2\end{array}\right]---\left(9\right)$

wherein, ${\mathrm{<figure-callout\; id="1"\; label="Y\; D"\; filenames="H2009100707149E00021.png,H2009100707149E00022.png"\; state="\{\{state\}\}">Y</figure-callout>}}_D^{}={\left[\begin{array}{cc}{Y}_D^1\left(k\right)& Y_D^1{(k+1)}^{*}\end{array}\right]}^T,$ ${\mathrm{<figure-callout\; id="2"\; label="Y\; D"\; filenames="H2009100707149E00021.png,H2009100707149E00022.png"\; state="\{\{state\}\}">Y</figure-callout>}}_D^{}={\left[\begin{array}{cc}{Y}_D^2\left(k\right)& Y_D^2{(k+1)}^{*}\end{array}\right]}^T,$ Z＝[Z(k)Z(k+1)]^T，C＝[C(k)C(k+1)]^T。

in the base station C receiver in the uplink, the separation of user signals is completed by adopting a multi-user space-time joint equalization algorithm.

The multi-user space-time joint equalization processing is completed by the following formula:

two-user linear zero-forcing matrix is introduced by utilizing Alamouti-like characteristic of channel matrix

$\mathrm{\&Phi;}=\left[\begin{array}{cc}{I}_{2N}& -{\mathrm{\&Lambda;}}_{\mathrm{1,2}}{\mathrm{\&Lambda;}}_{\mathrm{2,2}}^{-1}\\ -{\mathrm{\&Lambda;}}_{\mathrm{2,1}}{\mathrm{\&Lambda;}}_{\mathrm{1,1}}^{-1}& I_{2N}\end{array}\right]---\left(10\right)$

And constructing a modified received signal vector:

$\tilde{Y}=\mathrm{\&Phi;}\left[\begin{array}{c}{Y}_D^1\\ {\mathrm{<figure-callout\; id="2"\; label="Y\; D"\; filenames="H2009100707149E00021.png,H2009100707149E00022.png"\; state="\{\{state\}\}">Y</figure-callout>}}_D^{}\end{array}\right]=\left[\begin{array}{cc}\mathrm{\&Sigma;}& 0\\ 0& \mathrm{\&Delta;}\end{array}\right]\left[\begin{array}{c}Z\\ C\end{array}\right]+\left[\begin{array}{c}{\tilde{N}}_D^1\\ {\tilde{N}}_D^2\end{array}\right]---\left(11\right)$

wherein, $\mathrm{\&Sigma;}={\mathrm{\&Lambda;}}_{\mathrm{1,1}}-{\mathrm{\&Lambda;}}_{\mathrm{1,2}}{\mathrm{\&Lambda;}}_{\mathrm{2,2}}^{-1}{\mathrm{\&Lambda;}}_{\mathrm{2,1}},$ $\mathrm{\&Delta;}={\mathrm{\&Lambda;}}_{\mathrm{2,2}}-{\mathrm{\&Lambda;}}_{\mathrm{2,1}}{\mathrm{\&Lambda;}}_{\mathrm{1,1}}^{-1}{\mathrm{\&Lambda;}}_{\mathrm{1,2}},$ reuse the Alamouti-like characteristics of the matrix sigma and delta, if it is recorded $\tilde{Y}={\left[{\tilde{Y}}_1{\tilde{Y}}_2\right]}^T,$ Equation (11) is processed as follows:

${\widetilde{\stackrel{~}{Y}}}_1={\mathrm{\&Sigma;}}^H{\tilde{Y}}_1={\mathrm{\&Sigma;}}^H\mathrm{\&Sigma;Z}+{\widetilde{\stackrel{~}{N}}}_{\mathrm{<figure-callout\; id="1"\; label="D"\; filenames="H2009100707149E00021.png,H2009100707149E00022.png"\; state="\{\{state\}\}">D</figure-callout>}}^1$

(12)

${\widetilde{\stackrel{~}{Y}}}_2={\mathrm{\&Delta;}}^H{\tilde{Y}}_2={\mathrm{\&Delta;}}^H\mathrm{\&Delta;C}+{\widetilde{\stackrel{~}{N}}}_{\mathrm{<figure-callout\; id="2"\; label="D"\; filenames="H2009100707149E00021.png,H2009100707149E00022.png"\; state="\{\{state\}\}">D</figure-callout>}}^2$

and use ${\mathrm{\&Sigma;}}^H\mathrm{\&Sigma;}=\left(\begin{array}{cc}A& 0\\ 0& A\end{array}\right)$ And ${\mathrm{\&Delta;}}^H\mathrm{\&Delta;}=\left(\begin{array}{cc}\mathrm{<figure-callout\; id="0"\; label="B"\; filenames="H2009100707149E00111.png"\; state="\{\{state\}\}">B</figure-callout>}& 0\\ 0& B\end{array}\right)$ is a diagonal matrix. (12) Further simplifying as follows:

${\widetilde{\stackrel{~}{Y}}}_{\mathrm{1,1}}=\mathrm{AZ}\left(k\right)+{\widetilde{\stackrel{~}{N}}}_{\mathrm{<figure-callout\; id="1"\; label="D"\; filenames="H2009100707149E00021.png,H2009100707149E00022.png"\; state="\{\{state\}\}">\; <figure-callout\; id="1"\; label="D"\; filenames="H2009100707149E00021.png,H2009100707149E00022.png"\; state="\{\{state\}\}">D</figure-callout>\; </figure-callout>}}^{\mathrm{1,1}};$ ${\widetilde{\stackrel{~}{Y}}}_{\mathrm{1,2}}=\mathrm{AZ}(k+1)+{\widetilde{\stackrel{~}{N}}}_{\mathrm{<figure-callout\; id="1"\; label="D"\; filenames="H2009100707149E00021.png,H2009100707149E00022.png"\; state="\{\{state\}\}">\; <figure-callout\; id="2"\; label="D"\; filenames="H2009100707149E00021.png,H2009100707149E00022.png"\; state="\{\{state\}\}">D</figure-callout>\; </figure-callout>}}^{\mathrm{1,2}}$

(13)

${\widetilde{\stackrel{~}{Y}}}_{\mathrm{2,1}}=\mathrm{BC}\left(k\right)+{\widetilde{\stackrel{~}{N}}}_{\mathrm{<figure-callout\; id="2"\; label="D"\; filenames="H2009100707149E00021.png,H2009100707149E00022.png"\; state="\{\{state\}\}">\; <figure-callout\; id="1"\; label="D"\; filenames="H2009100707149E00021.png,H2009100707149E00022.png"\; state="\{\{state\}\}">D</figure-callout>\; </figure-callout>}}^{\mathrm{2,1}};$ ${\widetilde{\stackrel{~}{Y}}}_{\mathrm{2,2}}=\mathrm{BC}(k+1)+{\widetilde{\stackrel{~}{N}}}_{\mathrm{<figure-callout\; id="2"\; label="D"\; filenames="H2009100707149E00021.png,H2009100707149E00022.png"\; state="\{\{state\}\}">\; <figure-callout\; id="2"\; label="D"\; filenames="H2009100707149E00021.png,H2009100707149E00022.png"\; state="\{\{state\}\}">D</figure-callout>\; </figure-callout>}}^{\mathrm{2,2}}$

wherein, ${\widetilde{\stackrel{~}{Y}}}_1={[{\widetilde{\stackrel{~}{Y}}}_{\mathrm{1,1}}^T,{\widetilde{\stackrel{~}{Y}}}_{\mathrm{1,2}}^T]}^T,$ ${\widetilde{\stackrel{~}{Y}}}_2={[{\widetilde{\stackrel{~}{Y}}}_{\mathrm{2,1}}^T,{\widetilde{\stackrel{~}{Y}}}_{\mathrm{2,2}}^T]}^T,$ ${\widetilde{\stackrel{~}{N}}}_{\mathrm{<figure-callout\; id="1"\; label="D"\; filenames="H2009100707149E00021.png,H2009100707149E00022.png"\; state="\{\{state\}\}">D</figure-callout>}}^1={[{\widetilde{\stackrel{~}{N}}}_{\mathrm{<figure-callout\; id="1"\; label="D"\; filenames="H2009100707149E00021.png,H2009100707149E00022.png"\; state="\{\{state\}\}">D</figure-callout>}}^{\mathrm{1,1}T},{\widetilde{\stackrel{~}{N}}}_{\mathrm{<figure-callout\; id="1"\; label="D"\; filenames="H2009100707149E00021.png,H2009100707149E00022.png"\; state="\{\{state\}\}">D</figure-callout>}}^{\mathrm{1,2}T}]}^T,$ ${\widetilde{\stackrel{~}{N}}}_{\mathrm{<figure-callout\; id="2"\; label="D"\; filenames="H2009100707149E00021.png,H2009100707149E00022.png"\; state="\{\{state\}\}">D</figure-callout>}}^2={[{\widetilde{\stackrel{~}{N}}}_{\mathrm{<figure-callout\; id="2"\; label="D"\; filenames="H2009100707149E00021.png,H2009100707149E00022.png"\; state="\{\{state\}\}">D</figure-callout>}}^{\mathrm{2,1}T},{\widetilde{\stackrel{~}{N}}}_{\mathrm{<figure-callout\; id="2"\; label="D"\; filenames="H2009100707149E00021.png,H2009100707149E00022.png"\; state="\{\{state\}\}">D</figure-callout>}}^{\mathrm{2,2}T}]}^T$ then, by using the diagonal matrix characteristics of A and B, the estimation of { Z (k), Z (k +1) } and { C (k), C (k +1) } can be conveniently obtained as follows:

$\tilde{Z}\left(k\right)=A^{-1}{\widetilde{\stackrel{~}{Y}}}_{\mathrm{1,1}};$ $\tilde{Z}(k+1)=A^{-1}{\widetilde{\stackrel{~}{Y}}}_{\mathrm{1,2}}$

(14)

$\tilde{C}\left(k\right)=A^{-1}{\widetilde{\stackrel{~}{Y}}}_{\mathrm{2,1}};$ $\tilde{C}(k+1)=A^{-1}{\widetilde{\stackrel{~}{Y}}}_{\mathrm{2,2}}$

and

carrying out IDFT operation on N points to obtain

And

finally are respectively paired

And

the maximum likelihood detection is carried out on each component to obtain the estimated values of the relay forwarding symbol vectors { z (k), z (k +1) } and { c (k), c (k +1) }

And

FIG. 7-1 is a first embodiment of the present invention; the relay station 2 sends and receives the signal 1, the base station 2 receives the antenna, and the user terminal and the relay station link: 1x1 maximum ratio combining; relay station to base station link: 4x2 virtual MIMO transmission.

FIG. 7-2 is a second embodiment of the present invention; the relay station 2 sends 2 receives, the base station 2 receives the antenna, the user terminal and relay station link: 1x2 maximum ratio combining; relay station to base station link: 4x2 virtual MIMO transmission.

FIGS. 7-3 are third embodiments of the present invention; the relay station 2 sends and receives 4, the base station 2 receives the antenna, and the user terminal and relay station link: 1x4 maximum ratio combining; relay station to base station link: 4x2 virtual MIMO transmission.

FIGS. 7-4 are fourth embodiments of the present invention; the relay station 2 sends 1 and receives, the base station 4 receives the antenna, and the user terminal and relay station link: 1x1 maximum ratio combining; relay station to base station link: 4x4 virtual MIMO transmission.

FIGS. 7-5 are fifth embodiments of the present invention; the relay station 2 sends and receives, the base station 4 receives the antenna, the user terminal and relay station link: 1x2 maximum ratio combining; relay station to base station link: 4x4 virtual MIMO transmission.

FIGS. 7-6 are sixth embodiments of the present invention; the relay station 2 sends and receives 4, the base station 4 receives the antenna, the user terminal and relay station link: 1x4 maximum ratio combining; relay station to base station link: 4x4 virtual MIMO transmission.

Only some embodiments of the virtual mimo relay transmission method based on block space-time block coding according to the present invention have been listed above. The method provided by the invention can be conveniently popularized to the number of receiver antennas of the relay station of 1/2/4/8; the number of base station receiver antennas is 2/4/6/8, etc.

The effect achieved by adopting the virtual multiple-input multiple-output relay transmission method based on the block space-time block coding of the invention is given below.

1. Comparison of system spectrum utilization under different transmission schemes

Table 1 shows the spectral efficiency of non-relay transmission, conventional relay transmission, single antenna based virtual MIMO relay, and block space time block coding based virtual MIMO relay transmission schemes. When calculating the spectrum efficiency of a transmission scheme, assuming that the subchannel interval of a DFT-S-OFDM system is delta f, the transmission period of a DFT-S-OFDM symbol is T, the modulation constellation number Q of a user terminal and the sub-channel number N of a user terminal; meanwhile, under the scheduling of the base station, it is assumed that the UE1 located at the cell edge communicates with the relay station 1, and the UE2 located in the network blind area also communicates with the relay station 2.

Table 1: comparison of spectrum utilization for different transmission schemes

Table 1 comparison shows that: in the traditional time division relay transmission, the relay station needs to occupy extra time resources, so that the frequency spectrum efficiency of the system is reduced by one time compared with the direct transmission, and the virtual MIMO-based relay transmission adopts the spatial multiplexing technology to improve the frequency spectrum efficiency by one time, so that the reduction of the frequency spectrum efficiency caused by the traditional relay transmission is compensated, and finally, the integral frequency spectrum utilization rate of the system is kept the same as the direct transmission.

2. Complexity analysis of detection algorithms

Table 2: complexity comparison under different detection algorithms

The operation complexity of the direct use of the (8) type linear zero-forcing detection algorithm and the algorithm proposed by the present invention is shown in table 2. The comparison shows that the operation complexity of the detection algorithm provided by the invention is only half of the complexity of the direct zero-forcing algorithm.

3. Link transmission performance of block space-time block coding virtual MIMO relay scheme

Without loss of generality, the invention provides the bit error performance of a virtual MIMO relay transmission system based on block space-time block coding under the antenna architecture (figure 7-2) of the relay station 2 sending and receiving and the base station 2 receiving. Meanwhile, for convenience of comparison, the invention also provides bit error performance of a virtual MIMO relay transmission system based on VBLAST (vertical layered space-time code) under the structure of a relay station 1 sending and receiving and a base station 2 receiving antenna.

FIG. 8-1 shows relay-based under QPSK modulationThe method comprises the following steps that a station sends a first transmission and receives a second transmission, and a relay station sends the first transmission and receives the second transmission based on the bit error performance of a system under a virtual MIMO relay transmission scheme under a block space-time block coding structure, wherein: QPSK modulation, N120, L2048. The comparison of the curves shows that: at BER of 10^-4And compared with the former virtual MIMO relay transmission system, the B-STBC DFT-S-OFDM based relay transmission system has 3dB performance improvement. The B-STBC based virtual MIMO relay transmission scheme proposed herein is feasible.

Fig. 8-2 shows the bit error performance of the system under the virtual MIMO relay transmission scheme based on the space-time block coding structure under the condition of one-transmission and two-reception based on the relay station under the 16QAM modulation, where N is 120 and L is 2048. The conclusion is substantially in accordance with FIG. 8-1.

Claims (7)

1. In a relay transmission system composed of a user equipment (A), a relay station (B) and a base station (C), a link between each relay station (B) and the base station (C) adopts a block space time block coded virtual MIMO mode for transmission, and the block space time block coded virtual MIMO mode is as follows: in the uplink, each relay station transmitter adopts a transmission mode of space-time block coding of two transmitting antenna blocks, and uses the same time-frequency resource to realize communication with a base station (C) receiver under the scheduling of the base station (C), and the link between the user equipment (A) and the relay station (B) adopts a maximum ratio combining mode to carry out transmission.

2. The transmission method according to claim 1, wherein the relay station receiver (B2) in the uplink performs estimation and detection of the modulation symbols transmitted by the user equipment (a), and then forwards the detected modulation symbols transmitted by the user equipment (a) to the receiver of the base station (C) in the time slot of the relay station (B) and base station (C) link.

3. The method according to claim 1, wherein the separation of the user signals is performed in the base station (C) receiver in the uplink by using a multi-user space-time joint equalization algorithm.

4. The method according to claim 1, wherein the transmission of the block space time block coding is implemented by the following steps:

the first step is as follows: modulating an input bit sequence;

the second step is as follows: performing an N-point fast Fourier transform on the modulator output, wherein N is an integer power of 2;

the third step: sending the output of the fast Fourier transform into a space-time block encoder for encoding;

the fourth step: and respectively carrying out the same processing on the two paths of signals after the coding processing, and transmitting the processed signals of each path through the corresponding antenna of the path.

5. The method according to claim 4, wherein the same process in the fourth step comprises the following steps:

1) respectively mapping two paths of data signals to L frequency domain subcarriers, wherein L is more than N, and N is an integer power of 2;

2) performing inverse fast Fourier transform of L points on the output of the mapper;

3) inserting cyclic prefix, D/A converting;

4) sending the signals into an intermediate frequency and radio frequency unit;

5) and transmitting via the antenna.

6. The method according to claim 3, wherein the multi-user space-time joint equalization algorithm of the base station (C) in uplink comprises the following steps:

1) the reception paths corresponding to the respective reception antennas perform the same signal processing on the received signals:

a. processing a radio frequency signal from an antenna by a radio frequency and intermediate frequency unit, and sampling to form a digital baseband signal;

b. removing a cyclic prefix from the digital baseband signal;

c. performing an L-point Fourier transform, wherein L is an integer power of 2;

d. performing demapping processing, namely taking out data signals from the L subcarriers;

2) collecting the signals processed by each receiving channel and carrying out space-time joint equalization processing;

3) the user signals separated by the space-time joint equalizer are still processed by the same signal processing by taking each channel as a unit, the processing comprises N-point inverse discrete Fourier transform, demodulation and decoding to obtain the information transmitted by each user terminal, and N is an integer power of 2.

7. The method according to claim 3, wherein the multi-user space-time joint equalization algorithm is performed by the following formula:

utilizes the Alamouti-like characteristic of the channel matrix and records phi as a two-user linear zero-forcing matrix

Wherein I_2NConstructing a modified received signal vector for an identity matrix of 2N x 2N, Λ being the introduced intermediate variable:

wherein,

reuse the Alamouti-like characteristics of the matrix sigma and delta, if it is recorded

Then

Wherein Z and C are symbol sequence sets to be transmitted by each relay station after DFT processing,

for the noise sequence set on the transmission link where each relay station is located, it will

The treatment is carried out in the following way:

and use

And

the method is a diagonal array and is further simplified as follows:

wherein,

t represents the conjugate transpose operation of the matrix, and then the diagonal matrix characteristics of A and B are utilized to conveniently obtain the estimates of { Z (k), Z (k +1) } and { C (k), C (k +1) }:

and

carrying out IDFT operation on N points to obtain

And

finally are respectively paired

And

the maximum likelihood detection is carried out on each component to obtain the estimated values of the relay forwarding symbol vectors { z (k), z (k +1) } and { c (k), c (k +1) }

And

Images