CN101150515B - A broadband regeneration multi-hop relay communication method based on dual time slot interweaving and iteration - Google Patents

Abstract

The invention belongs to the field of wireless communications technology, specifically, a method for broadband OFDM regenerative multi-hop relay communications based on double-time-slot interleaved iteration. According to the invention, all relay nodes of a system have baseband processing modules and different baseband pseudo-random interleavers through which different interleaving patterns are generated and distributed to each multi-hop relay node to eliminate mutual interference of multi-hop signals with interleaved iterative interference cancellation algorithm and to divide a relay flow into two time-slot stages for running by turns with the application of a distributing method of time resource normalization. The method of the invention can realize higher system capacity than traditional TDMA and FDMA relay networks, avoid defect of signal self-coupling of OFDM relay nodes, the elimination operations on coupling interferences and signal mutual interferences among multi-hop nodes. Therefore, traditional omnidirectional antennas can be used, thus reducing complexity of the system.

Description

A kind of broadband regeneration multi-hop relay communication method based on dual time slot interweaving and iteration

Technical field

The invention belongs to wireless communication technology field, be specifically related to a kind of wireless multi-hop relay communication method based on dual time slot interweaving and iteration.

Background technology

Multi-hop relay (Multi-hop Relay) is a kind of crucial wireless communication technology.Along with the fast development of radio multimedium business, the user is more and more higher to the requirement of data communication capacity and transmission quality.Yet, because stop in the complex wireless environments, the influence of factor such as shade, formed many communication dead angle.These will make the user be difficult to two-forty and the high-quality communication service that obtains to continue.International numerous manufacturers are developing the multi-hop relay communication equipment, and formulate the corresponding techniques standard, try hard to exploitation by novel relay station, build up this to reduce communication network cloth, increase reliability of data transmission, and promote network coverage ability and operating benefit.

Junction network is made up of sending node, via node and receiving node usually.Via node receives the signal from sending node, and after carrying out certain processing, transmits to receiving node.The pattern of via node processing signals generally has two kinds, promptly non-regenerative relaying (AF, Amplify-and-Forward) with regenerative repeating (DF, Decode-and-Forward).

So-called non-regenerative relaying is exactly to the received signal, sends after the radio-frequency head that carries out certain power amplifies again, and claims analog junction again.This pattern has advantages such as signal processing flow is simple, equipment cost is lower; But each analog junction node has also amplified the noise of system in enhancing signal intensity, the drawback of having brought noise to add up thus.

So-called regenerative repeating is deciphered at the signal of via node spontaneous emission in the future node exactly, recovers fully, produces distortion hardly, and then with the signal of certain power to next receiving node forwarding recompile, claims digital junction again.Regenerative relay system has been eliminated the noise accumulation in the transmission course, can realize remote, big capacity, high-quality digital communication; And its have link circuit self-adapting performance flexibly, stronger security performance, and can be with sky the time treatment technology combine so that good space diversity effect to be provided.

The technology that relay system is handled multi-hop communication mutual interference problem has two kinds usually: time division multiple access (TDMA, TimeDivision Multiple Access) mode and frequency division multiple access (FDMA, Frequency Division Multiple Access) mode.The time division multiple access trunking scheme will be divided into continuous transmission time slot the time exactly, and be assigned to each stage of multi-hop transmission, collide interference problem mutually with this nothing that realizes that data transmit between adjacent two or more via node.This method channel allocation mechanism is simple ripe, do not need too much system control information, but it is had relatively high expectations synchronously to system time.The frequency division multiple access trunking scheme by entire spectrum being divided into separate frequency band, is assigned to each stage of multi-hop relay exactly, reaches the purpose of avoiding mutual interference.This method is loose to the management expectancy of time synchronized, but to the designing requirement strictness of band filter.From the angle of information theory power system capacity, two kinds of trunking schemes of this time-division multiple access and frequency division multiple access are equivalent, can be referred to as the quadrature relaying.

Special ripple (Turbo) iterative detection is a basic theory branch of message area, suitable development and application have been obtained for many years, professor C.Berrou as French Britain university has proposed Turbo Code channel decoding, the Turbo channel equalization of scholar's research such as M.Sandell of Sydney University, Australia and R.Yan, the branch multiple access multiuser detection that interweaves that Li professor Ping of City University of Hong Kong proposes, or the like.In recent years, in a lot of emerging communication technical fields (as multi-hop relay communication), how further developing the Turbo detection technique also is the direction that numerous scientific research institutions make great efforts.

Present multi-hop relay research and product are confined in the narrow-band communication system more.Design for the broadband multihop relay system also is short of very much, and is especially then at the early-stage for the research of broadband orthogonal frequency division multiplexing (OFDM, Orthogonal Frequency DivisionMultiplexing) multihop relay system.

Summary of the invention

The object of the invention is the multihop relay system at OFDM, proposes a kind of maximizing system traversal capacity that has, and the signal that can avoid via node overcomes the broadband regeneration multi-hop relay communication method of the mutual interference problem of multi-hop transmission from being coupled drawback.

The broadband regeneration multi-hop relay communication method that the present invention proposes comprises the dual time slot interweaving and iteration regeneration multi-hop relay communication method of employing at wideband OFDM system; And the method for corresponding time resource distribution.In dual time slot interweaving and iteration multi-hop relay communication means and the research work in the past of time resource distribution method thereof, also do not occur.Experimental results show that this method can significantly improve the throughput of system, and can avoid of the mutual interference of the signal of OFDM relay station from coupling and each stage of relaying.

The system model of multi-hop wireless relay communication method of the present invention as shown in Figure 1.In this system model, communication node is divided into source node, via node and destination node.Each via node all has baseband processing module, thereby all has a baseband signal decoding disposal ability, and each node has different base band pseudo random interleavers, its interlacing pattern design can be adopted methods such as the diagonal angle interweaves, block interleaving, also can select the Turbo code inner interleaver method for designing in the 3G (Third Generation) Moblie agreement for use.

The working mechanism of dual time slot interweaving and iteration multi-hop relay communication means is to produce different interlacing patterns by pseudo random interleaver, is dispensed to each multi-hop relay node.The mutual interference that each node utilizes the interleaving iterative interference cancellation techniques to carry out between the multi-hop signal is eliminated.Need emphasis to be pointed out that, different with traditional F DMA system, because the ofdm system filter for receiver will receive whole frequency band, when multi-hop relay, be easy to generate serious coupling interference phenomenon certainly.In order to overcome the coupling interference problem certainly of trunking traffic node, the whole trunking traffic flow process of the present invention's design is divided into two time slot stages carries out in turn, and scheme is referring to Fig. 1, Fig. 3.

All odd number relaying stage transmission data of (1) first time slot

In first time slot, all odd number relaying stages 1,3 ..., (T-1) communication.Be specially: in the 1st relaying stage, source node data interweaves through interleaver, sends to second via node; At this moment, all the OFDM frequency sub-carrier is used, shown in Fig. 3 (a).The useful signal of the second via node reception sources node and the interference signal of other working nodes.Then, the second via node receiving element will carry out base band signal process such as iteration interference eliminated and decoding.

The baseband processing module of t via node as shown in Figure 2 in the multi-hop wireless relay system.This module is formed with the emission processing unit by receiving processing unit.The radio-frequency module that receives processing unit receives wireless signal, and carries out intermediate frequency process, fast fourier transform (FFT, Fast Fourier Transform), symbol de-maps, iteration interference eliminated, decoding judgement, and is sent to the medium Access Layer and stores; And the emission processing unit will be encoded, interweaved, sign map, and anti-fast fourier transform (IFFT, Inverse Fast Fourier Transformation), be emitted to next via node via intermediate frequency/RF processing unit at last.Corresponding multi-hop relay interleaving iterative interference cancellation algorithm is seen appendix 1.

For other the 3rd, the 5th ... in the odd number relaying stages such as (T-1), will transmit in the same way; And, use all OFDM frequency sub-carrier during the via node transmission.

All even number relaying stage transmission data of (2) second time slots

In second transmission time slot, all even number relaying stage work.For example, in the second relaying stage, digital coding that the second via node transmitter unit is stored the medium Access Layer again and transmission.And the like, jump the relaying stage through last T, the target receiving node will obtain the data of source node.The concrete signal processing method is identical with the first time slot workflow.

Two time slot multi-hop relay communication meanss of this system's transmitting terminal can adopt traditional omnidirectional antenna, have avoided numerous and diverse in being coupled interference elimination treatment, have simplified the equipment of system, and can reduce the investment of whole system.This is one of innovative point of the present invention.

Corresponding to above-mentioned regeneration multi-hop trunking traffic pattern, the present invention proposes the time resource normalization distribution method of two time slot stages employings, specifically describe as follows:

The power system capacity of dual time slot interweaving and iteration multi-hop relay communication can be represented by a minimum value of jumping the normalization time throughput of relaying, wherein:

The throughput of system of (1) first time slot:

${\mathrm{\&Gamma;}}_{\mathrm{thp}}^{\left(1\right)}={\mathrm{<figure-callout\; id="1"\; label="T"\; filenames="S2007101704454E00021.png"\; state="\{\{state\}\}">T</figure-callout>}}_1\min ({\mathrm{<figure-callout\; id="1"\; label="C"\; filenames="S2007101704454E00021.png"\; state="\{\{state\}\}">C</figure-callout>}}_1,{\mathrm{<figure-callout\; id="3"\; label="C"\; filenames="S2007101704454E00021.png"\; state="\{\{state\}\}">C</figure-callout>}}_3,\&CenterDot;\&CenterDot;\&CenterDot;C_{t-1}\&CenterDot;\&CenterDot;\&CenterDot;C_{T-1})---\left(1\right)$

Here: Γ _Thp ⁽¹⁾Be the throughput of system of first time slot, T ₁It is the normalization transmission time of first time slot;

$C_{t-1}=\frac{1}{N_c}E\left\{\underset{n=1}{\overset{N_c}{\mathrm{\&Sigma;}}}\frac{{\left|H_n^{t-1}\right|}^2}{\underset{i<t-1}{\mathrm{\&Sigma;}}{\left(\frac{d_{t-1}}{\underset{m=i}{\overset{t-1}{\mathrm{\&Sigma;}}}{d}_m}\right)}^{\mathrm{\&alpha;}}{\left|H_n^i\right|}^{2}w\left({\mathrm{SNR}}_n^i\right)+\underset{i>t-1}{\mathrm{\&Sigma;}}{\left(\frac{d_{t-1}}{\underset{m=t}{\overset{i-1}{\mathrm{\&Sigma;}}}{d}_m}\right)}^{\mathrm{\&alpha;}}{\left|H_n^i\right|}^{2}w\left({\mathrm{SNR}}_n^i\right)+{\mathrm{\&sigma;}}^2}\right\}---\left(2\right)$

Here: N _cBe ofdm system number of sub carrier wave, d _T-1Be the transmission range that (t-1) jumps relaying, H _n ^T-1Be the frequency domain channel decline value that (t-1) jumps n subcarrier of relaying, SNR _n ⁱBe the iteration signal to noise ratio that i jumps n subcarrier of relaying, E{x} is the mathematic expectaion about variable x.

The throughput of system of (2) second time slots:

${\mathrm{\&Gamma;}}_{\mathrm{thp}}^{\left(2\right)}={\mathrm{<figure-callout\; id="2"\; label="T"\; filenames="S2007101704454E00031.png"\; state="\{\{state\}\}">T</figure-callout>}}_2\min ({\mathrm{<figure-callout\; id="2"\; label="C"\; filenames="S2007101704454E00031.png"\; state="\{\{state\}\}">C</figure-callout>}}_2,{\mathrm{<figure-callout\; id="4"\; label="C"\; filenames="S2007101704454E00031.png"\; state="\{\{state\}\}">C</figure-callout>}}_4,\&CenterDot;\&CenterDot;\&CenterDot;C_t\&CenterDot;\&CenterDot;\&CenterDot;C_T)---\left(3\right)$

Here: Γ _Thp ⁽²⁾Be the throughput of system of second time slot, T ₂It is the normalization transmission time of second time slot;

$C_t=\frac{1}{N_c}E\left\{\underset{n=1}{\overset{N_c}{\mathrm{\&Sigma;}}}\frac{{\left|H_n^t\right|}^2}{\underset{i<t}{\mathrm{\&Sigma;}}{\left(\frac{d_t}{\underset{m=i}{\overset{t}{\mathrm{\&Sigma;}}}{d}_m}\right)}^{\mathrm{\&alpha;}}{\left|H_n^i\right|}^{2}w\left({\mathrm{SNR}}_n^i\right)+\underset{i>t}{\mathrm{\&Sigma;}}{\left(\frac{d_t}{\underset{m=t+1}{\overset{i-1}{\mathrm{\&Sigma;}}}{d}_m}\right)}^{\mathrm{\&alpha;}}{\left|H_n^i\right|}^{2}w\left({\mathrm{SNR}}_n^i\right)+{\mathrm{\&sigma;}}^2}\right\}---\left(4\right)$

(3) simultaneously, for making dual time slot interweaving and iteration multi-hop relay communication entire system capacity optimum, slot sytem normalization throughput equates in the time of must making two,

C _multihop＝T ₁min(C ₁，C ₃，…C _t-1…C _T-1)＝T ₂min(C ₂，C ₄，…C _t…C _T) (5)

Wherein, T ₁+ T ₂=1 is normalization time, C _MultihopTraversal capacity for multihop relay system.

Order $C_{\min }^{\left(1\right)}=\min ({\mathrm{<figure-callout\; id="1"\; label="C"\; filenames="S2007101704454E00021.png"\; state="\{\{state\}\}">C</figure-callout>}}_1,{\mathrm{<figure-callout\; id="3"\; label="C"\; filenames="S2007101704454E00021.png"\; state="\{\{state\}\}">C</figure-callout>}}_3,\&CenterDot;\&CenterDot;\&CenterDot;C_{t-1}\&CenterDot;\&CenterDot;\&CenterDot;C_{T-1}),$ $C_{\min }^{\left(2\right)}=\min ({\mathrm{<figure-callout\; id="2"\; label="C"\; filenames="S2007101704454E00031.png"\; state="\{\{state\}\}">C</figure-callout>}}_2,{\mathrm{<figure-callout\; id="4"\; label="C"\; filenames="S2007101704454E00031.png"\; state="\{\{state\}\}">C</figure-callout>}}_4,\&CenterDot;\&CenterDot;\&CenterDot;C_t\&CenterDot;\&CenterDot;\&CenterDot;C_T);$ And C _Min ⁽²⁾, C _Min ⁽²⁾Two numerical value just can obtain when network design makes up, and thus, can obtain two time slots resource normalization distribution methods:

$\left\{\begin{array}{c}{\mathrm{<figure-callout\; id="1"\; label="T"\; filenames="S2007101704454E00021.png"\; state="\{\{state\}\}">T</figure-callout>}}_1=\frac{C_{\min }^{\left(2\right)}}{C_{\min }^{\left(1\right)}+C_{\min }^{\left(2\right)}}\\ {\mathrm{<figure-callout\; id="2"\; label="T"\; filenames="S2007101704454E00031.png"\; state="\{\{state\}\}">T</figure-callout>}}_2=\frac{C_{\min }^{\left(1\right)}}{C_{\min }^{\left(1\right)}+C_{\min }^{\left(2\right)}}\end{array}\right.---\left(6\right)$

Two time slots resource normalization distribution methods are another innovative points of the present invention.

Advantage of the present invention

(1) by the multi-hop signal mutual interference technology for eliminating based on interleaving iterative, every jumping stage of junction network can take whole OFDM subcarrier in frequency domain resources, can realize than traditional TDMA and the higher power system capacity of FDMA junction network.

(2) can avoid the signal of OFDM via node to eliminate processing, can adopt traditional omnidirectional antenna thus, thereby reduce the complexity of system from be coupled drawback and numerous and diverse coupled interference.

(3), be easy to realize because multi-hop interference source number is limited, and the operand of dual time slot interweaving and iteration multi-hop signal mutual interference removing method is less.

(4) in order to reach the minimum design object of error rate of system, traditional OFDM relay system must carry out numerous and diverse multistage subcarrier allocation.By comparison, dual time slot interweaving and iteration multi-hop relay Network Design is that the time-frequency domain resources distribution is provided convenience, and it will make the operand of allocation algorithm greatly reduce.

(5) system possesses the flexible net framework.

Description of drawings

Fig. 1 is the wireless multi-hop relay network system.

Fig. 2 is the signal processing flow block diagram of t node.

Fig. 3 is a wideband OFDM dual time slot interweaving and iteration multihop relay system communication mechanism diagram.Wherein, (a) being two time slot multi-hop relay transmission mechanisms, (b) is the Resource Allocation Formula based on the dual time slot interweaving and iteration interference eliminated.

Fig. 4 is power system capacity simulation analysis conclusion one of the present invention (comparing with traditional direct communication modular system capacity).

Fig. 5 is power system capacity simulation analysis conclusion two of the present invention (comparing with the power system capacity of traditional TDMA/FDMA multi-hop relay pattern).

Embodiment

To the broadband regeneration multi-hop relay communication method that the present invention proposes, carry out simulation analysis, concrete steps are as follows:

1, test data is the random data of Gaussian distributed.

2, produce one group of pseudo random interleaving pattern, and be assigned in each node interleaver of multihop relay system.

3, relay system is according to the distance in every jumping relaying stage, the type of channel repeated encoding, and according to two time slots resource normalization distribution methods among the present invention, sets the time of every slot transmission, referring to (1) formula-(6) formula.

4, the via node of system divides two time slots to carry out work in turn:

The emission of (1) first time slot is handled

In first time slot, the transmitting power that the medium Access Layer of source node and odd number working node is certain to the data setting that will transmit, and with data passes to the physical layer transmitter module; In the physical layer transmitter module, data will at first be carried out duplication code coding (Repetition Coding), carry out Bit Interleave according to interlacing pattern again, carry out the symbol constellation modulation then; The modulation constellation symbol carries out inverse-Fourier transform, and frequency domain data is transformed in the time domain processing domain, then carries out the digital filtering of intermediate frequency; According to the corresponding power emission factor, data transmission is gone out again by omnidirectional antenna;

(2) first time slots receive to be handled

In first working time slot, the receiver module of even number node part will receive decoding, and handling process is: the radio-frequency module of receiver receives wireless signal, and this received signal comprises interference signal and useful signal; Receiver module carries out corresponding intermediate frequency process to received signal, then, carries out fast fourier transform, time-domain signal is transformed into frequency domain handles, and follows symbol de-maps, obtains the soft information of signal; Subsequently, according to the interlacing pattern that is distributed, carry out the iteration interference elimination treatment of data, concrete algorithm such as appendix 1.

Simultaneously, will decipher the gained data and be input to even number node medium Access Layer, carry out verification and storage, in order to next slot transmission;

The emission of (3) second time slots is handled

In the second time slot emission process, the medium Access Layer transmitting power that the data setting of being stored is certain of even number node, and with data passes to the physical layer transmitter module; Carry out duplication code coding, Bit Interleave, symbol constellation modulation, IFFT conversion again; Digital IF Processing is gone out data according to repeat transmitted power again by omni-directional antenna transmission then;

(4) second time slots receive to be handled

In the second time slot receiving course, the receiver module of odd number node part will receive decoding, and handling process is: the radio-frequency module received signal of receiver, and carry out corresponding intermediate frequency process to received signal; Then, carry out the FFT conversion, time-domain signal is transformed into frequency domain, and carry out symbol de-maps, obtain the soft information of signal; Subsequently, carry out the iteration interference elimination treatment of data according to the interlacing pattern that is distributed, specific algorithm is with reference to appendix 1.

5, Fig. 4, Fig. 5 are last simulation result, wherein:

Fig. 4 shows, compares with traditional direct communication method, and the dual time slot interweaving and iteration multi-hop communication method that the present invention proposes can significantly improve power system capacity.For example: when the working point was 10dB, the double bounce relaying can provide the capacity gain of 2.6Bps/Sec/Hz, and four jump the capacity gain that relaying then can provide 4.6Bps/Sec/Hz.

Fig. 5 shows, compares with FDMA quadrature relay communication method with traditional TDMA, and the dual time slot interweaving and iteration multi-hop communication method that the present invention proposes can obviously improve power system capacity; This result has proved advantage of the present invention.For example, jump in the relay system four, as when the working point is low signal-to-noise ratio 0dB, method of the present invention can obtain the capacity gain of 3.3Bps/Sec/Hz.

Appendix 1

The interleaving iterative interference cancellation algorithm is as follows in the multihop relay system:

(1) initialization

If $L_{\mathrm{REP}}\left(X_j^{t-1}\right)=0,\&ForAll;j\&Element;(1,\&CenterDot;\&CenterDot;\&CenterDot;,J)---(A-1)$

(2) main flow process

$E\left(X_j^{t-1}\right)\&DoubleLeftArrow;\tanh (L_{\mathrm{REP}}\left(X_j^{t-1}\right)/2)\&ForAll;j\&Element;(1,\&CenterDot;\&CenterDot;\&CenterDot;J)---(A-2)$

$\mathrm{Var}\left(X_j^{t-1}\right)\&DoubleLeftArrow;1-E{\left(X_j^{t-1}\right)}^2$

j∈(1，…，J) (A-3)

$E\left({\mathrm{\&zeta;}}_j^{t-1}\right)\&DoubleLeftArrow;\underset{i<t-1}{\mathrm{\&Sigma;}}\sqrt{{\left(\frac{d_{t-1}}{\underset{m=i}{\overset{t-1}{\mathrm{\&Sigma;}}}{d}_m}\right)}^{\mathrm{\&alpha;}}}{H}_j^{i}E\left(X_j^i\right)+\underset{i>t-1}{\mathrm{\&Sigma;}}\sqrt{{\left(\frac{d_{t-1}}{\underset{m=t}{\overset{i-1}{\mathrm{\&Sigma;}}}{d}_m}\right)}^{\mathrm{\&alpha;}}}{H}_j^{i}E\left(X_j^i\right)---(A-4)$

i∈(1，3，…，(t-1)…，(T-1))

j∈(1，…J)

Wherein, α is the path attenuation factor.

$\mathrm{Var}\left({\mathrm{\&zeta;}}_j^{t-1}\right)\&DoubleLeftArrow;\underset{i<t-1}{\mathrm{\&Sigma;}}{\left(\frac{d_{t-1}}{\underset{m=i}{\overset{t-1}{\mathrm{\&Sigma;}}}{d}_m}\right)}^{\mathrm{\&alpha;}}{\left|H_j^i\right|}^2\mathrm{Var}\left(X_j^i\right)+\underset{i>t-1}{\mathrm{\&Sigma;}}{\left(\frac{d_{t-1}}{\underset{m=t}{\overset{i-1}{\mathrm{\&Sigma;}}}{d}_m}\right)}^{\mathrm{\&alpha;}}{\left|H_j^i\right|}^2\mathrm{Var}\left(X_j^i\right)+{\mathrm{\&sigma;}}^2---(A-5)$

i∈(1，3，…，(t-1)…，(T-1))

j∈(1，…J)

$L_{\mathrm{ESE}}\left(X_j^{t-1}\right)=\frac{{2H}_j^{t-1}}{\mathrm{Var}\left({\mathrm{\&zeta;}}_j^{t-1}\right)}(r_j^{t-1}-E\left({\mathrm{\&zeta;}}_j^{t-1}\right)),---(A-6)$

Here:

$r_j^{t-1}=H_j^{t-1}{X}_j^{t-1}+(\underset{i=1}{\overset{t-3}{\mathrm{\&Sigma;}}}\sqrt{{\left(\frac{d_{t-1}}{\underset{m=i}{\overset{t-1}{\mathrm{\&Sigma;}}}{d}_m}\right)}^{\mathrm{\&alpha;}}}{H}_j^i{X}_j^i+\underset{i=t+1}{\overset{T-1}{\mathrm{\&Sigma;}}}\sqrt{{\left(\frac{d_{t-1}}{\underset{m=t}{\overset{i-1}{\mathrm{\&Sigma;}}}{d}_m}\right)}^{\mathrm{\&alpha;}}}{H}_j^i{X}_j^i)+{\mathrm{\&sigma;}}^2---(A-7a)$

${\mathrm{\&zeta;}}_j^{t-1}\stackrel{\mathrm{\&Delta;}}{=}(\underset{i=1}{\overset{t-3}{\mathrm{\&Sigma;}}}\sqrt{{\left(\frac{d_{t-1}}{\underset{m=i}{\overset{t-1}{\mathrm{\&Sigma;}}}{d}_m}\right)}^{\mathrm{\&alpha;}}}{H}_j^i{X}_j^i+\underset{i=t+1}{\overset{T-1}{\mathrm{\&Sigma;}}}\sqrt{{\left(\frac{d_{t-1}}{\underset{m=t}{\overset{i-1}{\mathrm{\&Sigma;}}}{d}_m}\right)}^{\mathrm{\&alpha;}}}{H}_j^i{X}_j^i)+{\mathrm{\&sigma;}}^2---(A-7b)$

Then, with L _ESE(X _j ^T-1) substitution posterior probability (APP:A Posteriori Probability) channel decoder, decipher.(A-2) formula of again decode results being brought into is carried out next iteration.

Symbol description:

J: j chip, J: the chip number of a hardwood, X _j ⁱ: function variable, jump the emission modulation symbol of j chip, L in order to represent i _REP(X): about the log-likelihood ratio of X, tanh (): hyperbolic tangent function, E (X): about the expectation of variable X, Var (X): about the variance of variable X, t: t jumps relaying, H _j ⁱ: i jumps the frequency domain channel decline value of j chip of relaying, d _t: t jumps the transmission range of relaying, σ ²: noise variance, α: the path attenuation factor, r _j ^t: t jumps the receiving symbol of j chip, ζ _j ^t: t jumps the deviation between j chip emission modulation symbol and the receiving symbol.

Claims (5)

1. broadband regeneration multi-hop relay communication method based on dual time slot interweaving and iteration, adopt multi-hop wireless relay system model, wherein, communication node is divided into source node, via node and destination node, each via node all has baseband processing module, and each node all has different base band pseudo random interleavers, and its interlacing pattern adopts that diagonal angle interweaves, block interleaving or Turbo code inside interweave method for designing; Produce different interlacing patterns by pseudo random interleaver, be dispensed to each via node in the multi-hop; Each node utilizes the mutual interference between interleaving iterative interference elimination method elimination multi-hop signal, and whole trunking traffic flow process is divided into two time slot stages, and is specific as follows:

All odd number relaying stage transmission data of (1) first time slot

In first time slot, all odd number relaying stages 1,3 ..., (T-1) communication is specially: in the 1st relaying stage, source node data interweaves through interleaver, sends to second via node; At this moment, all the OFDM frequency sub-carrier is used, the useful signal of the second via node reception sources node and the interference signal of other working nodes; Then, the second via node receiving element will carry out the base band signal process of iteration interference eliminated and channel decoding;

For other the 3rd, the 5th ... in the odd number relaying stages such as (T-1), will transmit in the same way; And, use all OFDM frequency sub-carrier during the via node transmission;

All even number relaying stage transmission data of (2) second time slots

In second transmission time slot, all even number relaying stage work, digital coding that the second via node transmitter unit is stored the medium Access Layer again and transmission; Jump the relaying stage through last T, the target receiving node will obtain the data of source node; The concrete signal processing method is identical with the first time slot workflow;

It is characterized in that the method for the time resource normalization distribution that two time slot stages adopt, its particular content is:

The power system capacity of dual time slot interweaving and iteration multi-hop relay communication represented by a minimum value of jumping the normalization time throughput of relaying, wherein:

The throughput of system of (1) first time slot:

${\mathrm{\&Gamma;}}_{\mathrm{thp}}^{\left(1\right)}=T_1\min (C_1,C_3,\&CenterDot;\&CenterDot;\&CenterDot;C_{t-1}\&CenterDot;\&CenterDot;\&CenterDot;C_{T-1}),---\left(1\right)$

Here: Γ _Thp ⁽¹⁾Be the throughput of system of first time slot, T ₁It is the normalization transmission time of first time slot; C _tIt is the power system capacity that t jumps;

$C_{t-1}=\underset{n=1}{\overset{N_c}{\mathrm{\&Sigma;}}}\frac{{\left|H_n^{t-1}\right|}^2}{\underset{i\&NotEqual;t-1}{\mathrm{\&Sigma;}}{\left(\frac{d_{t-1}}{\underset{n=i}{\overset{t-1}{\mathrm{\&Sigma;}}}{d}_n}\right)}^{\mathrm{\&alpha;}}{\left|H_n^i\right|}^2{w}_n^i\left({\mathrm{SNR}}_n^i\right)+\underset{i\&NotEqual;t-1}{\mathrm{\&Sigma;}}{\left(\frac{d_{t-1}}{\underset{n=t}{\overset{i-1}{\mathrm{\&Sigma;}}}{d}_n}\right)}^{\mathrm{\&alpha;}}{\left|H_n^i\right|}^2{w}_n^i\left({\mathrm{SNR}}_n^i\right)+{\mathrm{\&sigma;}}_n^2},---\left(2\right)$

Here: N _cBe ofdm system number of sub carrier wave, d _T-1Be the transmission range that (t-1) jumps relaying, H _n ^T-1Be the frequency domain channel decline value that (t-1) jumps n subcarrier of relaying, SNR _n ⁱBe the iteration signal to noise ratio that i jumps n subcarrier of relaying, E{x} is the mathematic expectaion about variable x, σ ²: the additive Gaussian noise variance, w ( ^*) be the relation function of signal to noise ratio and decoder output valve variance, it is relevant with decoder, can obtain by Monte Carlo simulation, and α is the path attenuation factor;

The throughput of system of (2) second time slots:

${\mathrm{\&Gamma;}}_{\mathrm{thp}}^{\left(2\right)}=T_2\min (C_2,C_4,\&CenterDot;\&CenterDot;\&CenterDot;C_t\&CenterDot;\&CenterDot;\&CenterDot;C_T),---\left(3\right)$

Here: Γ _Thp ⁽²⁾Be the throughput of system of second time slot, T ₂It is the normalization transmission time of second time slot;

$C_t=\underset{n=1}{\overset{N_c}{\mathrm{\&Sigma;}}}\frac{{\left|H_t\right|}^2}{\underset{i\&NotEqual;t}{\mathrm{\&Sigma;}}{\left(\frac{d_t}{\underset{n=i}{\overset{t}{\mathrm{\&Sigma;}}}{d}_n}\right)}^{\mathrm{\&alpha;}}{\left|H_i\right|}^2{w}_i\left({\mathrm{SNR}}_i\right)+\underset{i\&NotEqual;t}{\mathrm{\&Sigma;}}{\left(\frac{d_t}{\underset{n=t+1}{\overset{i-1}{\mathrm{\&Sigma;}}}{d}_n}\right)}^{\mathrm{\&alpha;}}{\left|H_i\right|}^2{w}_i\left({\mathrm{SNR}}_i\right)+{\mathrm{\&sigma;}}^2},---\left(4\right)$

Slot sytem normalization throughput equates when (3) making two,

C _multihop＝T ₁min(C ₁，C ₃，…C _t-1…C _T-1)＝T ₂min(C ₂，C ₄，…C _t…C _T)，(5)

Wherein, T ₁+ T ₂=1 is normalization time, C _MultihopTraversal capacity for multihop relay system;

Order

C _Min ⁽¹⁾, C _Min ⁽²⁾Two numerical value just can obtain when network design makes up, and thus, obtain two time slots resource normalization distribution methods:

$\left\{\begin{array}{c}{T}_1=\frac{C_{\min }^{\left(2\right)}}{C_{\min }^{\left(1\right)}+C_{\min }^{\left(2\right)}}\\ T_2=\frac{C_{\min }^{\left(1\right)}}{C_{\min }^{\left(1\right)}+C_{\min }^{\left(2\right)}}\end{array}\right.---\left(6\right).$

2. method according to claim 1 is characterized in that the described interleaving iterative interference cancellation algorithm of utilizing, and concrete steps are as follows:

(1) initialization

If $L_{\mathrm{REP}}\left(X_j^{t-1}\right)=0,j\&Element;(1,\&CenterDot;\&CenterDot;\&CenterDot;,J),---(A-1)$

(2) main flow process

$E\left(X_j^{t-1}\right)\&DoubleLeftArrow;\tanh (L_{\mathrm{REP}}\left(X_j^{t-1}\right)/2),---(A-2)$

$\mathrm{Var}\left(X_j^{t-1}\right)\&DoubleLeftArrow;1-E{\left(X_j^{t-1}\right)}^2,---(A-3)$

$E\left({\mathrm{\&zeta;}}_j^{t-1}\right)\&DoubleLeftArrow;\underset{i<t-1}{\mathrm{\&Sigma;}}\sqrt{{\left(\frac{d_{t-1}}{\underset{m=i}{\overset{t-1}{\mathrm{\&Sigma;}}}{d}_m}\right)}^{\mathrm{\&alpha;}}}{H}_j^{i}E\left(X_j^i\right)+\underset{i>t-1}{\mathrm{\&Sigma;}}\sqrt{{\left(\frac{d_{t-1}}{\underset{m=t}{\overset{i-1}{\mathrm{\&Sigma;}}}{d}_m}\right)}^{\mathrm{\&alpha;}}}{H}_j^{i}E\left(X_j^i\right),---(A-4)$

Here, i ∈ (1,3 ..., (t-1) ..., (T-1)); J ∈ (1 ... J); α is the path attenuation factor;

$\mathrm{Var}\left({\mathrm{\&zeta;}}_j^{t-1}\right)\&DoubleLeftArrow;\underset{i<t-1}{\mathrm{\&Sigma;}}{\left(\frac{d_{t-1}}{\underset{m=i}{\overset{t-1}{\mathrm{\&Sigma;}}}{d}_m}\right)}^{\mathrm{\&alpha;}}{\left|H_j^i\right|}^2\mathrm{Var}\left(X_j^i\right)+\underset{i>t-1}{\mathrm{\&Sigma;}}{\left(\frac{d_{t-1}}{\underset{m=t}{\overset{i-1}{\mathrm{\&Sigma;}}}{d}_m}\right)}^{\mathrm{\&alpha;}}{\left|H_j^i\right|}^2\mathrm{Var}\left(X_j^i\right)+{\mathrm{\&sigma;}}^2,---(A-5)$

Here, i ∈ (1,3 ..., (t-1) ..., (T-1)), j ∈ (1 ... J);

$L_{\mathrm{ESE}}\left(X_j^{t-1}\right)=\frac{{2H}_j^{t-1}}{\mathrm{Var}\left({\mathrm{\&zeta;}}_j^{t-1}\right)}(r_j^{t-1}-E\left({\mathrm{\&zeta;}}_j^{t-1}\right)),---(A-6)$

Here:

$r_j^{t-1}=H_j^{t-1}{X}_j^{t-1}+(\underset{i=1}{\overset{t-3}{\mathrm{\&Sigma;}}}\sqrt{{\left(\frac{d_{t-1}}{\underset{m=i}{\overset{t-1}{\mathrm{\&Sigma;}}}{d}_m}\right)}^{\mathrm{\&alpha;}}}{H}_j^i{X}_j^i+\underset{i=t+1}{\overset{T-1}{\mathrm{\&Sigma;}}}\sqrt{{\left(\frac{d_{t-1}}{\underset{m=t}{\overset{i-1}{\mathrm{\&Sigma;}}}{d}_m}\right)}^{\mathrm{\&alpha;}}}{H}_j^i{X}_j^i)+{\mathrm{\&sigma;}}^2,---(A-7a)$

${\mathrm{\&zeta;}}_j^{t-1}\stackrel{\mathrm{\&Delta;}}{=}(\underset{i=1}{\overset{t-3}{\mathrm{\&Sigma;}}}\sqrt{{\left(\frac{d_{t-1}}{\underset{m=i}{\overset{t-1}{\mathrm{\&Sigma;}}}{d}_m}\right)}^{\mathrm{\&alpha;}}}{H}_j^i{X}_j^i+\underset{i=t+1}{\overset{T-1}{\mathrm{\&Sigma;}}}\sqrt{{\left(\frac{d_{t-1}}{\underset{m=t}{\overset{i-1}{\mathrm{\&Sigma;}}}{d}_m}\right)}^{\mathrm{\&alpha;}}}{H}_j^i{X}_j^i)+{\mathrm{\&sigma;}}^2,---(A-7b)$

Then, with L _ESE(X _j ^T-1) substitution posterior probability channel decoder, decipher; (A-2) formula of again decode results being brought into is carried out next iteration;

Symbol description: j: j chip, J: the chip number of a frame, X _j ⁱ: function variable, jump the emission modulation symbol of j chip, L in order to represent i _REP(X): about the log-likelihood ratio of X, tanh (): hyperbolic tangent function, E (X): about the expectation of variable X, Var (X): about the variance of variable X, t: t jumps relaying, H _j ⁱ: i jumps the frequency domain channel decline value of j chip of relaying, d _t: t jumps the transmission range of relaying, σ ²: additive Gaussian noise variance, α: the path attenuation factor, r _j ^t: t jumps the receiving symbol of j chip, ζ _j ^t: t jumps the deviation between j chip emission modulation symbol and the receiving symbol, L _ESE(X _j ^T-1) be the outside likelihood value that t-1 jumps j chip, α is the path attenuation factor.

3. method according to claim 1, it is characterized in that in the described multi-hop wireless relay system, the baseband processing module of via node is formed with the emission processing unit by receiving processing unit, the radio-frequency module that receives processing unit receives wireless signal, and carry out intermediate frequency process, fast fourier transform, symbol de-maps, iteration interference eliminated, decoding judgement, and be sent to the medium Access Layer and store; The emission processing unit will be encoded, interweaved, sign map, and anti-fast fourier transform, be emitted to next via node via intermediate frequency/RF processing unit at last.

4. method according to claim 1 is characterized in that adopting omnidirectional antenna.

5. method according to claim 1, it is characterized in that described trunking traffic flow process being divided into two time slot stages, to carry out the step of work in turn as follows:

The emission of (1) first time slot is handled

In first time slot, the transmitting power that the medium Access Layer of source node and odd number working node is certain to the data setting that will transmit, and with data passes to the physical layer transmitter module; In the physical layer transmitter module, data will at first be carried out the duplication code coding, carry out Bit Interleave according to interlacing pattern again, carry out the symbol constellation modulation then; The modulation constellation symbol carries out inverse-Fourier transform, and frequency domain data is transformed in the time domain processing domain, then carries out the digital filtering of intermediate frequency; According to the corresponding power emission factor, data transmission is gone out again by omnidirectional antenna;

(2) first time slots receive to be handled

In first working time slot, the receiver module of even number node part will receive decoding, and handling process is: the radio-frequency module of receiver receives wireless signal, and this received signal comprises interference signal and useful signal; Receiver module carries out corresponding intermediate frequency process to received signal, then, carries out fast fourier transform, time-domain signal is transformed into frequency domain handles, and follows symbol de-maps, obtains the soft information of signal; Subsequently, according to the interlacing pattern that is distributed, carry out the iteration interference elimination treatment of data; Simultaneously, will decipher the gained data and be input to even number node medium Access Layer, carry out verification and storage, in order to next slot transmission;

The emission of (3) second time slots is handled

In the second time slot emission process, the medium Access Layer transmitting power that the data setting of being stored is certain of even number node, and with data passes to the physical layer transmitter module; Carry out duplication code coding, Bit Interleave, symbol constellation modulation, IFFT conversion again; Digital IF Processing is gone out data according to repeat transmitted power again by omni-directional antenna transmission then;

(4) second time slots receive to be handled

In the second time slot receiving course, the receiver module of odd number node part will receive decoding, and handling process is: the radio-frequency module received signal of receiver, and carry out corresponding intermediate frequency process to received signal; Then, carry out the FFT conversion, time-domain signal is transformed into frequency domain, and carry out symbol de-maps, obtain the soft information of signal; Subsequently, carry out the iteration interference elimination treatment of data according to the interlacing pattern that is distributed.

Images