CN103200142A - Two-state simplified method of non-recursive shaped offset quadrature phase shift keying (SOQPSK)-TG signal - Google Patents

Abstract

The invention discloses a two-state simplified method of a non-recursive partial response shaped offset quadrature phase shift keying (SOQPSK)-TG signal. The method mainly solves the problem that in the prior art, a non-recursive SOQPSK-TG signal receiver is high in achievement complexity. The method includes the achievement steps: carrying out combination on a non-recursive SOQPSK-TG four-state grid chart through a state combination rule to obtain a two-state grid chart; then, generating a branch standard signal set of the two-state grid chart; and carrying out demodulation on a receiving signal in the receiver by using a MAX-LOG-MAP algorithm or a Viterbi algorithm through the branch standard signal set and the two-state grid chart, and recovering an initial input data sequence. The method achieves the demodulation on the non-recursive SOQPSK-TG signal through low algorithm complexity, and is applied to a system requiring the low achievement complexity.

Description

The two condition method for simplifying of onrecurrent SOQPSK-TG signal

Technical field

The invention belongs to wireless communication technology field, particularly adopt a kind of communication system of non-recursive part response shaping offset quadrature phase-shift-keying SOQPSK-TG signal, can be used for fields such as military communication, satellite communication, telemetry communication.

Background technology

In fields such as military communication, satellite communication and telemetry communication, its working channel all belongs to power and the dual limited channel of bandwidth, and has big carrier deviation.Shaping offset quadrature phase-shift-keying SOQPSK signal is because having constant envelope, and characteristics such as power validity height in the communications field, are one of the most frequently used modulation systems.The SOQPSK signal is a kind of continuous phase modulated signal, according to the correlation length difference of phase place, can be divided into two kinds: a kind of is total regression SOQPSK signal, and as the SOQPSK-MIL signal, another kind is partial response signal, as the SOQPSK-TG signal.Wherein the SOQPSK-MIL signal is a kind of signal of recommending in the American army mark, and the SOQPSK-TG signal is the signal of recommending in the IRIG106-04 aerial remote reconnaissance standard.By document " The Capacity of SOQPSK-TG " (IEEE Military Communications Conference.Baltimore, MD, USA:IEEE, 2011:555-560.) as can be known, compare with the SOQPSK-MIL signal, the SOQPSK-TG signal has good power spectrum characteristic, but its receiver implementation complexity height, in practical engineering application, be subjected to certain limitation.Therefore, the simplification status method of research SOQPSK-TG signal receiver is SOQPSK-TG signal key in application in practice.

At document " Reduced-complexity approach to iterative detection of SOQPSK " (IEEE Transactions on Communications, 2007,55 (7): the characteristic of having analyzed the SOQPSK-TG signal 54-62.), the modulation of SOQPSK-TG signal is decomposed into precoding and continuous phase signal modulation two parts, and according to the form of precoding, the SOQPSK-TG signal is divided into recurrence SOQPSK-TG signal and onrecurrent SOQPSK-TG signal.Analyze in the literature, no matter recurrence SOQPSK-TG or onrecurrent SOQPSK-TG signal, its phase association length L is equal to 8, at the receiving terminal maximum likelihood receiver optimum as if employing, then the receiver grid needs 512 phase states, computational complexity is high, almost can't realize in practice.

Document " Decision Feedback Detector for SOQPSK " (IEEE Transactions on Communications, 2009,57 (8): 2359-2368.) recurrence and onrecurrent SOQPSK-TG signal have been proposed the method for simplifying of four states, and proposed to simplify the method for two condition at recurrence SOQPSK-TG signal.The document is pointed out: at recurrence and onrecurrent SOQPSK-TG signal, the four state reduction methods of blocking based on pulse are reduced to four states with the receiver grid chart of SOQPSK-TG signal, adopt the Viterbi algorithm, carry out state transitions on four state trellis bases, recover information sequence.But when recovering information sequence with the grid chart of this four state, computational complexity is still higher.At recurrence SOQPSK-TG signal on the basis of recurrence four state reduction grid charts, by introducing the merging of decision-feedback and trellis state, simplify obtaining the two condition grid chart, in this receiver, adopt the Viterbi algorithm to carry out trellis state and shift, recover information sequence.In carrying out the grid transfer process, need pass through feedback branch, by the survival branch road information that obtains in the previous moment grid, generate expression formula by tributary signal, calculate the branch road information that respectively shifts in the current time grid, obtain the survival branch road information in the current time grid, calculate thus that again next respectively shifts branch road information in the grid constantly, by that analogy, carry out trellis state and shift, recover whole information sequence at last.Though adopted the two condition grid chart in this receiver, in the trellis state transfer process, the branch road information in a certain moment grid all needs to recomputate, and has increased computational complexity so again, makes overall implementation complexity also unreduced.Simultaneously, the two condition grid chart that the document is carried only is only applicable to the receiver of recurrence SOQPSK-TG signal, and is not suitable for the receiver of onrecurrent SOQPSK-TG signal.

Summary of the invention

The objective of the invention is to the deficiency at above-mentioned prior art, propose a kind of two condition method for simplifying of onrecurrent SOQPSK-TG signal, to reduce the implementation complexity of receiver.

Realize that technical thought of the present invention is: on onrecurrent four state trellis bases, merge by rational state, obtain onrecurrent two condition grid chart, in receiver, adopt the two condition grid chart to carry out demodulation as the state transitions grid, thereby the recovery information sequence, implementation step is as follows:

(1) merge rule by following state onrecurrent SOQPSK- TG signal 00,01,10,11 4 state trellis figure merged, obtain T ∈ I, Q} road 0,1 two condition grid chart constantly:

To the I road constantly, 00 state in four states, 10 states are merged into 0 state, 01 state, 11 states are merged into 1 state;

To the Q road constantly, 00 state in four states, 01 state are merged into 0 state, 10 states, 11 states are merged into 1 state;

(2) merge rule according to the state in the step (1), obtain T road { 00,01,10, the 11}, and two four state variables are designated as numbering l of two four state variable S ∈ in corresponding four state trellis of initial state m ' in the two condition grid chart constantly _i(T, m '), i ∈ 0,1}:

Four states, 00 state, 10 states to initial state m '=0 correspondence in the T=I road moment two condition grid chart are numbered, and namely 00 state is designated as numbering l ₀(I, 0), 10 states are designated as numbering l ₁(I, 0);

Four states, 01 state, 11 states to initial state m '=1 correspondence in the T=I road moment two condition grid chart are numbered, and namely 01 state is designated as numbering l ₀(I, 1), 11 states are designated as numbering l ₁(I, 1);

Four states, 00 state, 01 state to initial state m '=0 correspondence in the T=Q road moment two condition grid chart are numbered, and namely 00 state is designated as numbering l ₀(Q, 0), 10 states are designated as numbering l ₁(Q, 0);

Four states, 10 states, 11 states to initial state m '=1 correspondence in the T=Q road moment two condition grid chart are numbered, and namely 01 state is designated as numbering l ₀(Q, 1), 11 states are designated as numbering l ₁(Q, 1).

(3) according to the numbering l in the step (2) _i(T, m ') determines four corresponding state variable S, according to the one-to-one relationship of four state variable S and phase state θ, determines the value of θ;

(4) utilize numbering l in the step (2) _i(T, m ') determines that { 0,1} by onrecurrent precoding formula, obtains the output symbol α on the corresponding branch road for four corresponding state variable S and the input bit d ∈ on the corresponding branch road;

(5) utilize the value of output symbol α and phase state θ, obtain the T road constantly, initial state is that m ', done state are m, numbering l _i(T, m ') corresponding branch road standard signal

(6) with all branch road standard signals

Store, obtain the branch road standard signal collection of two condition grid;

(7) utilize the branch road standard signal collection of two condition grid and storage at receiving terminal, carry out demodulation to received signal, obtain the original input data sequence.

The present invention compared with prior art has the following advantages:

First, the present invention has simplified the grid chart in the onrecurrent SOQPSK-TG receiver, existing onrecurrent SOQPSK-TG four state trellis figure are obtained the two condition grid chart by the state merging, therefore, can reduce the implementation complexity of receiver greatly and be easy to hardware to realize.

Second, owing at first all branch road standard signals of grid are generated and stored in advance at receiving terminal, S makes in the trellis state transfer process, can be according to the survivor path information in a last moment, concentrate at grid branch road standard signal and to choose next corresponding branch road standard signal in grid constantly, and need not recomputate each branch road standard signal, reduced the computational complexity of receiver.

Description of drawings

When being four states of existing onrecurrent SOQPSK-TG signal, Fig. 1 becomes grid chart;

Fig. 2 is general flow chart of the present invention;

Fig. 3 is for recovering the sub-process figure of original data sequence among the present invention;

Fig. 4 is the two condition grid chart among the present invention;

Fig. 5 is the former phase impulse function waveform figure of existing SOQPSK-TG signal;

Fig. 6 for the phase impulse function after with existing method Fig. 5 being blocked at [0T _b] at interval oscillogram;

Fig. 7 is for using the present invention and existing method to the errored bit comparison diagram of onrecurrent SOQPSK-TG signal demodulation.

Embodiment

The present invention will be further described below in conjunction with accompanying drawing.

With reference to accompanying drawing 2, specific implementation step of the present invention is as follows:

Step 1 obtains the two condition grid chart with four state trellis figure by the state merging

To T ∈ shown in Figure 1 I, Q} road 00,01,10,11 4 state trellis figure constantly merge rule according to state and carry out following division, and obtain T ∈ I, Q} road 0,1 two condition grid chart constantly, as shown in Figure 4:

To the I road constantly, 00 state in four states, 10 states are merged into 0 state, 01 state, 11 states are merged into 1 state;

To the Q road constantly, 00 state in four states, 01 state are merged into 0 state, 10 states, 11 states are merged into 1 state.

Step 2 merges rule according to the state in the step 1, obtains T road { 00,01,10, the 11}, and two state variables are designated as numbering l of two four state variable S ∈ in corresponding four state trellis of initial state m ' in the two condition grid chart constantly _i(T, m '), i ∈ 0,1}:

Four states, 00 state, 10 states to initial state m '=0 correspondence in the T=I road moment two condition grid chart are numbered, and namely 00 state is designated as numbering l ₀(I, 0), 10 states are designated as numbering l ₁(I, 0);

Four states, 01 state, 11 states to initial state m '=1 correspondence in the T=I road moment two condition grid chart are numbered, and namely 01 state is designated as numbering l ₀(I, 1), 11 states are designated as numbering l ₁(I, 1);

Four states, 00 state, 01 state to initial state m '=0 correspondence in the T=Q road moment two condition grid chart are numbered, and namely 00 state is designated as numbering l ₀(Q, 0), 10 states are designated as numbering l ₁(Q, 0);

Four states, 10 states, 11 states to initial state m '=1 correspondence in the T=Q road moment two condition grid chart are numbered, and namely 01 state is designated as numbering l ₀(Q, 1), 11 states are designated as numbering l ₁(Q, 1).

Step 3 is according to the numbering l in the step 2 _i(T, m ') determines four corresponding state variable S, according to the one-to-one relationship of four state variable S and phase state θ, determines that the value of θ is:

$\mathrm{\θ}=\left\{\begin{array}{cc}0,& S=00\\ 3\mathrm{\π}/2,& S=01\\ \mathrm{\π}/2,& S=10\\ \mathrm{\π},& S=11\end{array}\right..$

Step 4 is utilized the numbering l in the step 2 _i(T, m ') definite four corresponding state variable S and the input bit d ∈ on the corresponding branch road 0,1} by onrecurrent precoding formula, obtains the output symbol α on the corresponding branch road:

α＝(-1) ^k(2d ₂-1)(d-d ₁)

Wherein k is determined by moment T:

$k=\left\{\begin{array}{cc}1& T=I\\ 0& T=Q\end{array}\right.$

d ₁, d ₂With the corresponding relation of four state variable S and moment T be:

$S=\left\{\begin{array}{cc}({\mathrm{<figure-callout\; id="1"\; label="d"\; filenames="HDA00002955625000011.png,HDA00002955625000031.png"\; state="\{\{state\}\}">d</figure-callout>}}_1,d_2)& T=I\\ ({\mathrm{<figure-callout\; id="2"\; label="d"\; filenames="HDA00002955625000042.png,HDA00002955625000052.png"\; state="\{\{state\}\}">d</figure-callout>}}_2,d_1)& T=Q\end{array}\right.;$

Described numbering l _iCorresponding relation such as the table 1 of (T, m ') and four state variable S, phase state θ, input bit d, output symbol α, branch road done state m:

Step 5 according to the value of output symbol α and phase state θ, obtains the T road moment, initial state m ', done state m, numbering l _i(T, m ') corresponding branch road standard signal

${\mathrm{Rstd}}_i^T(m^{\&prime;},m)=\exp \left(j(2\mathrm{\&pi;h\&alpha;}{q}_{\mathrm{PT}}\left(t\right)+\mathrm{\&theta;})\right),0\&le;t\&le;T_b$

T wherein _bIn the is-symbol cycle, the present embodiment value is 1/2000s, and exp () asks exponent arithmetic, and h is modulation index, and value is that 1/2, L is phase restriction length, and value is 8, q in the present embodiment _PT(t) be with former phase impulse function q _TG(t) block, and will the time become part and only be kept at interval [0, T _b] interior phase impulse function:

$q_{\mathrm{PT}}\left(t\right)=\left\{\begin{array}{cc}0,& t<0\\ q_{\mathrm{TG}}(t+\frac{(L-1)T_b}{2})& 0\&le;t\&le;{\mathrm{<figure-callout\; id="1"\; label="T\; b"\; filenames="HDA00002955625000011.png,HDA00002955625000031.png"\; state="\{\{state\}\}">T</figure-callout>}}_b\\ 1/2,& t>T_b\end{array}\right.,$

In the formula, q _TG(t) be former phase impulse function:

$q_{\mathrm{TG}}\left(t\right)=\left\{\begin{array}{cc}0,& t<0\\ {\&Integral;}_0^{\mathrm{<figure-callout\; id="0"\; label="t"\; filenames="HDA00002955625000011.png,HDA00002955625000041.png"\; state="\{\{state\}\}">t</figure-callout>}}0.3116\&times;\frac{\cos (0.4375\mathrm{\&pi;\&tau;}/T_b)}{1-4{(0.4375\mathrm{\&tau;}/T_b)}^2}\&times;\frac{\sin (0.625\mathrm{\&pi;\&tau;}/T_b)}{0.625\mathrm{\&pi;\&tau;}/T_b}\&times;w\left(\mathrm{\&tau;}\right)\mathrm{d\&tau;},& 0\&le;t{<\mathrm{<figure-callout\; id="1"\; label="T\; b"\; filenames="HDA00002955625000011.png,HDA00002955625000031.png"\; state="\{\{state\}\}">T</figure-callout>}}_b\\ 1/2,& t\&GreaterEqual;{\mathrm{LT}}_b\end{array}\right.$

In the formula, w (τ) is window function:

$w\left(\mathrm{\&tau;}\right)=\left\{\begin{array}{cc}1,& 0\&le;\left|\frac{\mathrm{\&tau;}}{{2T}_b}\right|\&le;1.5\\ \frac{1}{2}+\frac{1}{2}\cos \left(2\mathrm{\&pi;}(\frac{\mathrm{\&tau;}}{{2T}_b}-1.5)\right)& 1\&CenterDot;5\&le;\left|\frac{\mathrm{\&tau;}}{{2T}_b}\right|\&le;2\\ 0,& 2<\left|\frac{\mathrm{\&tau;}}{{2T}_b}\right|\end{array}\right..$

This former phase impulse function q _TG(t) waveform only is kept at [0, T at interval as shown in Figure 5 after Fig. 5 blocked _b] in the phase impulse function waveform as shown in Figure 6.

Step 6 is with all branch road standard signals Store, obtain the branch road standard signal collection of two condition grid.

By the table 1 in the step 4 as can be known, phase state θ and output symbol α have 16 kinds of corresponding relations, determine a kind of branch road standard signal according to the corresponding relation of a kind of θ and α

Obtain 16 kinds of branch road standard signals Each branch road standard signal Expression formula as shown in table 2:

Step 7 is utilized the branch road standard signal collection of two condition grid and storage at receiving terminal, carries out demodulation to received signal, obtains the original input data sequence.

With reference to Fig. 3, being implemented as follows of this step:

7a) signal that arrives receiver is carried out down-conversion, extraction, filtering, obtain baseband complex signal y _k, with baseband complex signal y _kIn before (L-1) * samplenum/2 sampling point give up, remain sampling point and be designated as y _k', receiving symbol add up to N, wherein L is phase restriction length, the present embodiment value is that 8, samplenum is that each symbol sample is counted, the present embodiment value is that 8, N is the sum of receiving symbol, the present embodiment value is 200;

7b) initialization setting:

The initial moment of initialization is the n=0 moment, is determined the value of T by n and T road relation constantly:

Initialization n=0 is the survival status number l of initial state m ' correspondence constantly _n(m ') ∈ { 0,1} and path metric M _n(m '): l ₀(0)=0, l ₀(1)=1, M ₀(0)=100, M ₀(1)=0;

7c) at n constantly, intercepting y _kThe sampled point of ' the n symbol interval obtains the signal y of current time _n

7d) utilize the n survival status number l of initial state m ' correspondence constantly _n(m '), n be the T road moment of correspondence constantly, and initial state m ' and done state m concentrate the standard signal that takes out the correspondence branch road from initial state m ' to done state m from standard signal

7e) utilize standard signal

And reception data y _n, obtain corresponding each branch road branched measurement value from initial state m ' to done state m by the conjugate multiplication method

${\widetilde{\mathrm{\&gamma;}}}_n(m^{\&prime;},m)=\mathrm{Re}(\mathrm{\&Sigma;}{y}_n\&times;\mathrm{conj}\left({\mathrm{Rstd}}_{l_n\left(m^{\&prime;}\right)}^T(m^{\&prime;},m)\right)),$

Wherein, conjugation is asked in conj () expression, and real part is got in Re () expression;

7f) utilize n each the branch road branch metric of the moment that obtains

And the path metric M of initial state m ' correspondence _n(m ') obtains the n+1 survival status number l of initial state m correspondence constantly _N+1(m) and path metric M _N+1(m):

7f1) with branch metric

Be added to the n path metric M of initial state m ' correspondence constantly _nOn (m '), n each done state m of the moment is only kept the path of a paths metric maximum as its survivor path, namely $M_{n+1}\left(m\right)=\underset{m^{\&prime;}}{\max }(M_n\left(m^{\&prime;}\right)+{\widetilde{\mathrm{\&gamma;}}}_n(m^{\&prime;},m)),$ Give up other path;

For n done state m=0 constantly, the path metric value of supposing to obtain to done state m=0 from initial state m '=0 this moment is greater than the path metric value that obtains to done state m=0 from initial state m '=1, then only keep branch road from initial state m '=0 to done state m=0 as survivor path, obtain the n+1 path metric of initial state m=0 correspondence constantly by survivor path

7f2) keep the initial state m ' of n moment survivor path as the survival status number l of n+1 moment initial state m correspondence _N+1(m), be about to initial state m '=0 of n moment survivor path in the step (7f1), as the survival status number l of n+1 moment initial state m=0 correspondence _N+1(0)=0;

7g) make next n=n+1 constantly, by n in the step (7b) and T road relation constantly, determine the n value of T constantly; Turn back to step (7c), circulation successively is up to calculating all N code elements branch metric and path metric constantly;

7h) utilize all N code elements branch metric and path metric constantly, recover the original input data sequence by MAX-LOG-MAP algorithm or Viterbi algorithm.

What adopt in the embodiment of the invention is to recover the original input data sequence by the MAX-LOG-MAP algorithm.

Effect of the present invention can further specify by following emulation:

1. simulation system parameters setting

Matlab 7.6 simulation softwares are used in emulation of the present invention, and the simulation parameter setting is consistent with used parameter among the embodiment, symbol period T _b=1/2000s, correlation length L=8, modulation index h=1/2, sampling number samplenum=8, receiving symbol add up to N=200; Channel model adopts additive white Gaussian noise channel; Adopt the MAX-LOG-MAP algorithm to recover the original input data sequence at receiving terminal.

2. emulation content

With two condition method of the present invention and existing four status method onrecurrent SOQPSK-TG signal is carried out demodulation, the errored bit contrast that obtains as shown in Figure 7.As can be seen from Figure 7, though existing four status method are better than two condition method of the present invention to the errored bit performance of onrecurrent SOQPSK-TG signal demodulation to the errored bit performance of onrecurrent SOQPSK-TG signal demodulation, compare with existing four status method, two condition method of the present invention is 10 to the mistake of onrecurrent SOQPSK-TG signal demodulation than performance ^-2～10 ^-5The time, signal to noise ratio has the loss of 0.5～1dB.But two condition method of the present invention has lower computational complexity, is applied in the system of low system implementation complexity.

Claims (5)

1. the two condition method for simplifying of a non-recursive part response shaping offset quadrature phase-shift-keying SOQPSK-TG signal comprises the steps:

(1) merge rule by following state 00,01,10,11 4 state trellis figure of onrecurrent SOQPSK-TG signal merged, obtain T ∈ I, Q} road 0,1 two condition grid chart constantly:

To the I road constantly, 00 state in four states, 10 states are merged into 0 state, 01 state, 11 states are merged into 1 state;

To the Q road constantly, 00 state in four states, 01 state are merged into 0 state, 10 states, 11 states are merged into 1 state;

(2) merge rule according to the state in the step (1), obtain T road { 00,01,10, the 11}, and two four state variables are designated as numbering l of two four state variable S ∈ in corresponding four state trellis of initial state m ' in the two condition grid chart constantly _i(T, m '), i ∈ 0,1}:

Four states, 00 state, 10 states to initial state m '=0 correspondence in the T=I road moment two condition grid chart are numbered, and namely 00 state is designated as numbering l ₀(I, 0), 10 states are designated as numbering l ₁(I, 0);

Four states, 01 state, 11 states to initial state m '=1 correspondence in the T=I road moment two condition grid chart are numbered, and namely 01 state is designated as numbering l ₀(I, 1), 11 states are designated as numbering l ₁(I, 1);

Four states, 00 state, 01 state to initial state m '=0 correspondence in the T=Q road moment two condition grid chart are numbered, and namely 00 state is designated as numbering l ₀(Q, 0), 10 states are designated as numbering l ₁(Q, 0);

Four states, 10 states, 11 states to initial state m '=1 correspondence in the T=Q road moment two condition grid chart are numbered, and namely 01 state is designated as numbering l ₀(Q, 1), 11 states are designated as numbering l ₁(Q, 1);

(3) according to the numbering l in the step (2) _i(T, m ') determines four corresponding state variable S, according to the one-to-one relationship of four state variable S and phase state θ, determines the value of θ;

(4) utilize numbering l in the step (2) _i(T, m ') determines that { 0,1} by onrecurrent precoding formula, obtains the output symbol α on the corresponding branch road for four corresponding state variable S and the input bit d ∈ on the corresponding branch road;

(5) utilize the value of output symbol α and phase state θ, obtain the T road constantly, initial state is that m ', done state are m, numbering l _i(T, m ') corresponding branch road standard signal

(6) with all branch road standard signals

Store, obtain the branch road standard signal collection of two condition grid;

(7) utilize the branch road standard signal collection of two condition grid and storage at receiving terminal, carry out demodulation to received signal, obtain the original input data sequence.

2. the two condition method for simplifying of SOQPSK-TG signal according to claim 1, the one-to-one relationship according to four state variable S and phase state θ described in the step (3) wherein, determine to realize the value of θ by following formula:

$\mathrm{\&theta;}=\left\{\begin{array}{cc}0,& S=00\\ 3\mathrm{\&pi;}/2,& S=01\\ \mathrm{\&pi;}/2,& S=10\\ \mathrm{\&pi;},& S=11\end{array}\right..$

3. the two condition method for simplifying of SOQPSK-TG signal according to claim 1 wherein passes through onrecurrent precoding formula described in the step (4), obtains the output symbol α on the corresponding branch road, realizes by following formula:

α＝(-1) ^k(2d ₂-1)(d-d ₁)

Wherein, k is and moment T value corresponding that its value is by moment T decision, namely

$k=\left\{\begin{array}{cc}1& T=I\\ 0& T=Q\end{array}\right.$

Wherein, T refers to T ∈ (I, Q) the road moment;

d ₁, d ₂Be and four state variable S value corresponding that its value is with the corresponding relation of moment T and four state variable S:

$S=\left\{\begin{array}{cc}(d_1,d_2)& T=I\\ (d_2,d_1)& T=Q\end{array}\right..$

4. the two condition method for simplifying of SOQPSK-TG signal according to claim 1, the described value of utilizing output symbol α and phase state θ of step (5) wherein, obtain the T road constantly, initial state is that m ', done state are m, numbering l _i(T, m ') corresponding branch road standard signal

Realize by following formula:

${\mathrm{Rstd}}_i^T(m^{\&prime;},m)=\exp \left(j(2\mathrm{\&pi;h\&alpha;}{q}_{\mathrm{PT}}\left(t\right)+\mathrm{\&theta;})\right),0\&le;t\&le;T_b,$

Wherein, h is modulation index, and exp () is the index computing, T _bBe symbol period, q _PT(t) be with former phase impulse function q _TG(t) block, and will the time become part and only be kept at interval [0, T _b] interior phase impulse function:

$q_{\mathrm{PT}}\left(t\right)=\left\{\begin{array}{cc}0,& t<0\\ q_{\mathrm{TG}}(t+\frac{(L-1)T_b}{2})& 0\&le;t\&le;T_b\\ 1/2,& t>T_b\end{array}\right.,$

In the formula, L is phase restriction length, q _TG(t) be former phase impulse function:

$q_{\mathrm{TG}}\left(t\right)=\left\{\begin{array}{cc}0,& t<0\\ {\&Integral;}_0^{t}0.3116\&times;\frac{\cos (0.4375\mathrm{\&pi;\&tau;}/T_b)}{1-4{(0.4375\mathrm{\&tau;}/T_b)}^2}\&times;\frac{\sin (0.625\mathrm{\&pi;\&tau;}/T_b)}{0.625\mathrm{\&pi;\&tau;}/T_b}\&times;w\left(\mathrm{\&tau;}\right)\mathrm{d\&tau;},& 0\&le;t<T_b\\ 1/2,& t\&GreaterEqual;{\mathrm{LT}}_b\end{array}\right.$

In the formula, w (τ) is window function:

$w\left(\mathrm{\&tau;}\right)=\left\{\begin{array}{cc}1,& 0\&le;\left|\frac{\mathrm{\&tau;}}{{2T}_b}\right|\&le;1.5\\ \frac{1}{2}+\frac{1}{2}\cos \left(2\mathrm{\&pi;}(\frac{\mathrm{\&tau;}}{{2T}_b}-1.5)\right)& 1.5\&le;\left|\frac{\mathrm{\&tau;}}{{2T}_b}\right|\&le;2\\ 0,& 2<\left|\frac{\mathrm{\&tau;}}{{2T}_b}\right|\end{array}\right..$

5. the two condition method for simplifying of SOQPSK-TG signal according to claim 1, wherein utilize the branch road standard signal collection of two condition grid and storage at receiving terminal described in the step (7), carry out demodulation to received signal, obtain the original input data sequence, carry out as follows:

7a) signal that arrives receiver is carried out down-conversion, extraction, filtering, obtain baseband complex signal y _k, with baseband complex signal y _kIn before (L-1) * samplenum/2 sampling point give up, remain sampling point and be designated as y _k', L is correlation length, samplenum counts for each symbol sample, receiving symbol add up to N;

7b) initialization setting:

The initial moment of initialization is the n=0 moment, is determined the value of T by n and T road relation constantly:

Initialization n=0 is the survival status number l of initial state m ' correspondence constantly _n(m ') ∈ { 0,1 and path metric M _n(m '): l ₀(0)=0, l ₀(1)=1, M ₀(0)=100, M ₀(1)=0;

7c) at n constantly, intercepting y _kThe sampled point of ' the n symbol interval obtains the signal y of current time _n

7d) utilize the n survival status number l of initial state m ' correspondence constantly _n(m '), n be the T road moment of correspondence constantly, and initial state m ' and done state m concentrate the standard signal that takes out the correspondence branch road from initial state m ' to done state m from standard signal

7e) utilize standard signal And reception data y _n, obtain corresponding each branch road branched measurement value from initial state m ' to done state m by the conjugate multiplication method

${\widetilde{\mathrm{\&gamma;}}}_n(m^{\&prime;},m)=\mathrm{Re}(\mathrm{\&Sigma;}{y}_n\&times;\mathrm{conj}\left({\mathrm{Rstd}}_{l_n\left(m^{\&prime;}\right)}^T(m^{\&prime;},m)\right)),$

Wherein, conjugation is asked in conj () expression, and real part is got in Re () expression;

7f) utilize each the branch road branch metric that obtains

And the path metric M of initial state m ' correspondence _n(m ') obtains the n+1 survival status number l of initial state m correspondence constantly _N+1(m) and path metric value M _N+1(m):

7f1) with branch metric Be added to the n path metric M of initial state m ' correspondence constantly _nOn (m '), n each done state m of the moment is only kept the path of a paths metric maximum as its survivor path, namely $M_{n+1}\left(m\right)=\underset{m^{\&prime;}}{\max }(M_n\left(m^{\&prime;}\right)+{\widetilde{\mathrm{\&gamma;}}}_n(m^{\&prime;},m)),$ Give up other path;

7f2) keep the initial state m ' of n moment survivor path as n+1 survival status number l constantly _N+1(m);

7g) make next n=n+1 constantly, by n in the step (7b) and T road relation constantly, determine the n value of T constantly; Turn back to step (7c), circulation successively is up to calculating all N code elements branch metric and path metric constantly;

7h) utilize all N code elements branch metric and path metric constantly, recover the original input data sequence by MAX-LOG-MAP algorithm or Viterbi algorithm.

Images