CN100574143C - Array antenna communication device - Google Patents

Abstract

The invention provides a kind of communicator that adopts a plurality of antennas to control the dual-mode antenna directional diagram.Carry out under the occasion of computing at the weight vectors that adopts the RLS algorithm to two-way vector modulator, shorten the RLS convergence of algorithm time.In each antenna (unit antenna) 12, be provided with transmitting-receiving shared two-way vector modulator 22 and not by shared non-common circuit portion 24.Non-shared circuit part 24 is adjusted according to transmitting system and receiving system, makes the phase characteristic amplitude response become the constant times of determining between antenna.Adaptive processing section 30 adopts following weight vectors that two-way vector modulator 22 is controlled, and this weight vectors is by the modified R LS algorithm computation after the part of RLS algorithm was improved is gone out in the past.By adopting modified R LS algorithm, compare with RLS algorithm in the past to be reduced to and satisfy the operation times of specifying the condition of convergence.

Description

Array antenna communication device

Technical field

The present invention relates to a kind of communicator that adopts a plurality of antennas to control the dual-mode antenna directional diagram.

Background technology

The communicator that possesses adaptive array antenna is for well-known, this adaptive array antenna is by synthetic to suitably carried out addition by each received signal of following a plurality of antennas, and the direction of arrival that is formed on useful wave has wave beam and has the reception diagram of zero-signal (null) and receive useful signal selectively at the direction of arrival of disturbing wave, and above-mentioned a plurality of antennas spatially make it the configuration of being separated by.Carrying out under the occasion that sends, be preferably formed in the transmitting antenna directional diagram that useful direction has wave beam and have zero-signal in the jamming station direction by this communicator.In view of the above, can turn to useful direction to transmitted power selectively because of forming wave beam in useful direction, and because of making zero-signal for the jamming station, not be subjected to interference from this communicator towards the jamming station direction, the jamming station that group with this communicator and useful is communicated independently is difficult for exerting an influence, and is effective in this sense.

At this, relevant possess the communicator of adaptive array antenna in the past, illustrated with reference to Fig. 5.Also have, relevant transmitting-receiving frequency is identical, carry out the situation of the adaptive array antenna that transmitting-receiving communication and employing spatially be separated by 4 antennas 52 to be disposed herein as an example in the time-division mode, illustrated.

At first, the processing in butt joint time receiving is illustrated.By the received signal of antenna 52 via the anti T-R device 54 that transmission/reception is changed (in Fig. 5, expressing the connection status when receiving), after amplifying by low noise amplifier (LNA) 56, be input in the frequency mixer 58, at this by converting thereof into IF-FRE (IF) with intersecting from the local frequency of local transmitter 60.Then, signal becomes near the receive frequency frequency signal by IF filter 62, is amplified by IF amplifier 64, be input in the frequency mixer 66, this with mix from the local frequency of local transmitter 68, convert thereof into baseband signal.Then, signal is differentiated to necessary bandwidth by low pass filter 70, is converted thereof into digital signal by A-D converter (A/D) 72.Received signal in 4 antennas 52 is converted into baseband signal separately like this.These signals are imported in recipient's handling part 74, and (coefficient: w1～w4) after this carries out add operation and handled as received signal to carry out characteristic weighting at this according to amplitude and phase place.A though not only useful signal but also comprise the jamming station signal in the signal received by antenna, but, receive only useful signal and can remove the jamming station signal from received signal by in adaptive array handling part 76, suitably determining weight coefficient (reception weighted value) according to contrast signal and received signal.Relevant this processing has detailed record in non-patent literature 1.

Below, the processing during to relevant the transmission is illustrated.Transmit leg handling part 78 is divided into 4 with the transmission signal of being imported, and to carrying out weighting separately.At this, the weighted value during as transmission is used the weighted value when receiving sometimes.This is to come owing to utilizing the opposite property that sends signal and received signal to form the such design of following transmitting antenna directional diagram, and above-mentioned transmitting antenna directional diagram has the wave beam identical with reception diagram, zero-signal.So each signal that quilt is cut apart after the also weighting is input in the frequency mixer 84 via digital analog converter (D/A) 80 and low pass filter 82, converts thereof into the IF frequency at this by the mixing that is produced by local frequency.Then, signal is filtered by IF filter 86, after being amplified by IF amplifier 88, is input in the frequency mixer 90, converts thereof into the RF frequency at this by the mixing that is produced by local frequency.Subsequently, signal is sent by antenna 52 via transmit power amplifier (PA) 92 and anti T-R device 54.

Yet the weighting directional diagram is identical when making weighting directional diagram when sending and reception in above-mentioned conventional art.Though this is to make according to the opposite property of signal on the space after the antenna 52, with regard to radio section since send signal process sending part (TX) and received signal the acceptance division (RX) of process different, thereby property is false on the contrary.Therefore, as above-mentioned conventional art, even transmit leg handling part 78 adopt with in recipient's handling part 74 employed identical weighted values, can not obtain transmission directive property identical when receiving.That is to say, because transmission signal phase rotation amount and amplitude variations amount through TX one side are different with the phase of received signal rotation amount and the amplitude variations amount of process RX one side, thereby the signal amplitude when sending signal and reaching antenna through TX and phase place amplitude and the phase place when being received with received signal is different, therefore, for for carrying out identical weighting in the transmitting-receiving process, send with receiving course in antenna pattern, beam direction and zero-signal direction all be different.

Thereby, in possessing the communicator of this adaptive array antenna, for 4 systems separately for be necessary to carry out suitable adjustment, so that it is identical with the amount of phase rotation of received signal in the acceptance division (RX) to send the amount of phase rotation of signal in the sending part (TX), and, make the amplitude variations amount that sends signal in the sending part (TX) become the amplitude variations amount of received signal in the acceptance division (RX) and shared fixedly multiplying power between each antenna.

Under this occasion, usually adjust (recipient's calibration) for whole acceptance divisions (RX) of 4 systems, so that amplitude variations amount and the amount of phase rotation become fixed value mutually, and the whole sending parts (TX) for 4 systems are adjusted (transmit leg calibration), so that amplitude variations amount and the amount of phase rotation become fixed value mutually.This adjustment be by in recipient's handling part 74 to set amplitude phasing portion 94 of each system (each acceptance division) and in transmit leg handling part 78 to set amplitude phasing portion 96 of each system (each sending part), carry out.This calibration (correction) is particularly as being set forth in patent documentation 1 and the patent documentation 2, be amplitude and the phase place when in each system, being transformed into recipient or transmit leg and coming the sequentially determining received signal and amplitude and phase place when sending signal through sending part (TX) through acceptance division (RX), and carried out.

Patent documentation 1

Specially permit communique No. 3332911

Patent documentation 2

Special table 2003-501971 communique

Patent documentation 3

The spy opens the 2001-53663 communique

Non-patent literature 1

The good work of letter between chrysanthemum, " adopting the Adaptive Signal Processing of array antenna ", front page, Co., Ltd.'s science and technology is published in November, 1998

, according to the method for being set forth in patent documentation 1 or the patent documentation 2,, before finishing, calibration needs such problem of time so exist because send and amplitude and phase place when receiving sequentially determining to pass through about the order separately of a plurality of systems converts to.Moreover, new generation amplitude variations and phase place rotation in the implementation process of calibration, existence is difficult to the accuracy highland and implements the such problem of calibration.And, it is extremely difficult in the calibration implementation process sending part and acceptance division being remained the situation that does not have characteristic variations generally speaking, is continued such calibration that is starved of time as the sort of often the calibration in the lump when the utilization that its countermeasure has also that a lot of occasions have to implement to be set forth in the patent documentation 3.

In addition, owing to from useful, the signal level in jamming station according to producing than cataclysm to useful, the distance in jamming station etc., thereby in acceptance division, be provided with automatic gain adjustment device (AGC) usually, but because this automatic gain adjustment device is set, so the variation according to incoming level produces difference to amplitude variations amount and the amount of phase rotation each other at acceptance division sometimes, in fact the calibration correction value that also has a lot of occasions specially to make can not effectively be utilized.

Moreover, occur in recipient's amplitude phasing portion that certain is unusual, become the state that can not carry out normal correction, consequent error is appended in the weighted value of recipient's handling part under this occasion.And, because the weighted value of appending after this error is further used in transmit leg, so there are transmitting antenna directional diagram and the reception diagram obviously different such problems that become that make.

Summary of the invention

Array antenna communication device involved in the present invention adopts the adaptive array antenna that contains a plurality of unit antennas, it is characterized by: possess

RF transmitting system circuit is arranged in above-mentioned each unit antenna, and comprises transmit power amplifier at least;

RF receiving system circuit is arranged in parallel with above-mentioned RF transmitting system circuit in above-mentioned each unit antenna, and comprises low noise amplifier at least;

Two-way vector modulator is connected with above-mentioned RF transmitting system circuit and RF receiving system circuit jointly at the opposite side of unit antenna;

Divide to fit in portion, be connected with a plurality of two-way vector modulators;

Receiving and transmitting part (TRX) is connected with the above-mentioned branch portion of fitting in;

Adaptive processing section is controlled above-mentioned two-way vector modulator, and a plurality of unit antennas are used as adaptive array antenna;

Acceptance division (RX), be arranged at above-mentioned each RF receiving system circuit and above-mentioned adaptive processing section between,

For above-mentioned RF transmitting system circuit corresponding and RF receiving system circuit with the constituent parts antenna, the mutual difference of the difference that amplitude variations amount when making signal through above-mentioned RF transmitting system circuit corresponding with the constituent parts antenna and RF receiving system circuit is mutual and the amount of phase rotation between the constituent parts antenna about equally

Adaptive processing section is controlled above-mentioned two-way vector modulator according to satisfying the complex weighted vector W (m+1) that specifies the condition of convergence, and this appointment condition of convergence is to adopt the modified R LS algorithm computation according to following formula (11)～(15) to go out,

W(m+1)＝W(m)+γ′{κR′ _xx ^-1(m)X(m+1)}e*(m+1)

…(11)

e(m+1)＝r(m+1)-W ^H(m)X(m+1) …(12)

$r^{\′}=\frac{1}{\mathrm{\α}+X^H(m+1)\left\{{{{\mathrm{\κR}}^{\′}}_{\mathrm{xx}}}^{-1}\left(m\right)X(m+1)\right\}}\·\·\·\left(13\right)$

R ' _Xx ^-1(0)=δ ^-1I when m=0 (wherein) ... (14)

${{R^{\′}}_{\mathrm{xx}}}^{-1}\left(m\right)$

$=\frac{1}{\mathrm{\α}}{{R^{\′}}_{\mathrm{xx}}}^{-1}(m-1)-\frac{\left\{{{{\mathrm{\κR}}^{\′}}_{\mathrm{xx}}}^{-1}(m-1)X\left(m\right)\right\}\left\{{\mathrm{\κX}}^H\left(m\right){{R^{\′}}_{\mathrm{xx}}}^{-1}(m-1)\right\}}{{\mathrm{\α}}^2+{\mathrm{\αX}}^H\left(m\right)\left\{{{{\mathrm{\κR}}^{\′}}_{\mathrm{xx}}}^{-1}(m-1)X\left(m\right)\right\}}$

(m=1 wherein, 2,3 ... the time) ... (15)

Here,

W(m)＝R _xx ^-1(m)r _xr(m)

$R_{\mathrm{xx}}\left(m\right)=\underset{i=1}{\overset{m}{\mathrm{\Σ}}}{\mathrm{\α}}^{m-i}X\left(i\right)X^H\left(i\right)$

$r_{\mathrm{xr}}\left(m\right)=\underset{i=1}{\overset{m}{\mathrm{\Σ}}}{\mathrm{\α}}^{m-i}X\left(i\right)r^{*}\left(i\right)$

X (i): plural input vector, r (i): contrast signal, i: the positive integer corresponding, H: complex-conjugate transpose, *: complex conjugate, α: satisfy the real number of 0＜α≤1, κ: satisfy the real number of 0＜κ＜1, δ: positive constant, I: unit ranks with sample time.

And array antenna communication device of the present invention adopts the adaptive array antenna that contains a plurality of unit antennas, it is characterized by: possess

RF transmitting system circuit is arranged in above-mentioned each unit antenna, and comprises transmit power amplifier at least;

RF receiving system circuit is arranged in parallel with above-mentioned RF transmitting system circuit in above-mentioned each unit antenna, and comprises low noise amplifier at least;

Two-way vector modulator is connected with above-mentioned RF transmitting system circuit and RF receiving system circuit jointly at the opposite side of unit antenna;

Divide to fit in portion, be connected with a plurality of two-way vector modulators;

Receiving and transmitting part (TRX) is connected with the above-mentioned branch portion of fitting in;

Adaptive processing section is controlled above-mentioned two-way vector modulator, and a plurality of unit antennas are used as adaptive array antenna;

Acceptance division (RX), be arranged at above-mentioned each RF receiving system circuit and above-mentioned adaptive processing section between,

For above-mentioned RF transmitting system circuit corresponding and RF receiving system circuit with the constituent parts antenna, the mutual difference of the difference that amplitude variations amount when making signal through above-mentioned RF transmitting system circuit corresponding with the constituent parts antenna and RF receiving system circuit is mutual and the amount of phase rotation between the constituent parts antenna about equally

Adaptive processing section is controlled above-mentioned two-way vector modulator according to satisfying the complex weighted vector W (m+1) that specifies the condition of convergence, and this appointment condition of convergence is to adopt the modified R LS algorithm computation according to following formula (21)～(25) to go out,

W(m+1)＝W(m)+κγR″ _xx ^-1(m)X(m+1)e*(m+1)

…(21)

e(m+1)＝r(m+1)-W ^H(m)X(m+1) …(22)

$r=\frac{1}{\mathrm{\α}+X^H(m+1){R_{\mathrm{xx}}}^{-1}\left(m\right)X(m+1)}\·\·\·\left(23\right)$

R " _Xx ^-1(0)=δ ^-1I when m=0 (wherein) ... (24)

${{R^{,,}}_{\mathrm{xx}}}^{-1}\left(m\right)$

$=\frac{1}{\mathrm{\α}}{{R^{,,}}_{\mathrm{xx}}}^{-1}(m-1)-\mathrm{\κ}\frac{{{R^{,,}}_{\mathrm{xx}}}^{-1}(m-1)X\left(m\right)X^H\left(m\right){{R^{,,}}_{\mathrm{xx}}}^{-1}(m-1)}{{\mathrm{\α}}^2+{\mathrm{\αX}}^H\left(m\right){{R^{,,}}_{\mathrm{xx}}}^{-1}(m-1)X\left(m\right)}$

(m=1 wherein, 2,3 ... the time) ... (25) here,

W(m)＝R _xx ^-1(m)r _xr(m)

$R_{\mathrm{xx}}\left(m\right)=\underset{i=1}{\overset{m}{\mathrm{\Σ}}}{\mathrm{\α}}^{m-i}X\left(i\right)X^H\left(i\right)$

$r_{\mathrm{xr}}\left(m\right)=\underset{i=1}{\overset{m}{\mathrm{\Σ}}}{\mathrm{\α}}^{m-i}X\left(i\right)r^{*}\left(i\right)$

X (i): plural input vector, r (i): contrast signal, i: the positive integer corresponding, H: complex-conjugate transpose, *: complex conjugate, α: satisfy the real number of 0＜α≤1, κ: satisfy the real number of 0＜κ＜1, δ: positive constant, I: unit ranks with sample time.

With regard to the related array communications device of the invention described above, it is desirable to, for above-mentioned RF transmitting system circuit corresponding with the constituent parts antenna and RF receiving system circuit, the time of delay when making signal through above-mentioned RF transmitting system circuit corresponding with the constituent parts antenna and RF receiving system circuit about equally.

With regard to the related array communications device of the invention described above, it is desirable to, at above-mentioned RF transmitting system circuit corresponding with the constituent parts antenna and at least one side among the RF receiving system circuit adjustment part is set, this adjustment part is used for adjusting above-mentioned amplitude variations amount, the amount of phase rotation and among time of delay at least one.

It is desirable to regard to the related array communications device of the invention described above, is a kind of time-division bidirectional communication apparatus that same communication counterpart is carried out transceiver with same frequency.

Description of drawings

Fig. 1 is the block diagram of the related array antenna communication device major part structure of expression embodiment of the present invention.

Fig. 2 represents is the convergence number of times that has between TRX and each RX in the related array antenna communication device of embodiment of the present invention under the occasion of phase error when going out weight vectors by the second modified R LS algorithm computation.

Fig. 3 represents is the convergence number of times that has between TRX and each RX in the related array antenna communication device of embodiment of the present invention under the occasion of phase error when going out weight vectors by the first modified R LS algorithm computation.

Fig. 4 represents is the convergence number of times that has between TRX and each RX in the related array antenna communication device of embodiment of the present invention under the occasion of amplitude error when going out weight vectors by the first modified R LS algorithm computation.

Fig. 5 is the block diagram of array antenna communication device in the past.

Embodiment

Below, relevant preferred forms of the present invention is illustrated with reference to Fig. 1.Fig. 1 is the block diagram of the related example of communicator 10 major parts of expression present embodiment.Also have, the example when in the present embodiment relevant 4 antennas of employing (unit antenna) 12 being constituted adaptive array antenna is illustrated.

Be input to signal in each antenna 12 under anti T-R device 14,20 and state that the recipient is connected,, pass through anti T-R device 20 again, be imported in the two-way vector modulator 22 by band pass filter (BPF) 16 and low noise amplifier (LAN) 18.Herein, be provided with between anti T-R device 14 and anti T-R device 20 that independently circuit is (just respectively separately according to transmitting system and receiving system, RF transmitting system circuit and RF receiving system circuit), in the present embodiment this part is called non-common circuit portion 24.Then, a plurality of system signals after the weighting fit in portion 26 by branch and carry out add operation separately in two-way vector modulator 22, are received (received signal) through receiving and transmitting part (TRX) 28.The part of this received signal is imported in the adaptive processing section (APU) 30.

In addition, the signal of being exported from low noise amplifier (LAN) 18 passes through the acceptance division (RX) 32 that is arranged at each system, is imported in the adaptive processing section 30.

Adaptive processing section 30 is according to the contrast signal of being imported and from the signal of TRX28, obtain following weighted value (weighted value in each two-way vector modulator 22), be set in each two-way vector modulator 22, this weighted value is used for separating from disturbing wave, noise etc. takes out the useful wave signal.In view of the above, can be formed on the antenna pattern that useful direction has wave beam and have zero-signal in the jamming station direction.In addition, can also improve signal SN ratio from useful.

On the other hand, baseband transmit signals fits in portion 26 by branch and distributes to each system through receiving and transmitting part (TRX) 28.The signal that is distributed is imported into respectively in the two-way vector modulator 22, process anti T-R device 20 and adjuster are (though play a role mainly as phase regulator, but also can comprise function as the amplitude adjuster) 34, carry out power amplification by transmit power amplifier (PA) 36, this after anti T-R device 14 exported by antenna 12.Also have, anti T-R device 14,20 all is connected with transmit leg during transmission.

Yet, in this structure on the RF section in the transmission/receiving course signal path different be the circuit structure part between transmitting-receiving conversion transducer 14 and the anti T-R device 20, just non-common circuit portion 24.; in non-common circuit portion 24, be provided with adjuster 34 in the present embodiment; about each system is constituted (perhaps adjust) according to transmit path (RF transmitting system circuit) and RX path (RF receiving system circuit), make the mutual difference of the mutual difference of amplitude variations amount and the amount of phase rotation become roughly the same value at separately 12 on each antenna (unit antenna).In addition, in the present embodiment to the weighted value of each system be in the transmission/receiving course the weighted value of shared two-way vector modulator 22.That is to say, the communicator 10 related according to present embodiment, owing to adopt adjuster 34 to make the signal characteristic variation in transmit path/RX path become constant times according to each system, thereby can adopt shared weighted value in the transmission/receiving course at each system, the transmitting antenna directional diagram is formed as identical directional diagram (directional diagram that just, has same beam, zero-signal) with reception diagram.

Moreover, with regard to the related communicator 10 of this enforcement, be preferably in and the difference of vibration that detects between following two kinds of signals and phase difference be set and in the receiving system its device of proofreading and correct (adaptive processing section 30 just is equivalent to this device in the present embodiment), above-mentioned signal one is through two-way vector modulator 22, divides and to fit in portion 26 and TRX28 and signal in the input adaptive handling part 30, and the 2nd, distribute and the acceptance division (RX) 32 of each system of process and signal the input adaptive handling part from the leading portion (antenna 12 1 sides) of two-way vector modulator 22.Though do not interfere with communication itself, but in order to improve the convergence of adaptive control, and can seek the constringent further raising of adaptive control by this device, and then seek the further improvement of SN ratio for the output of set acceptance division (RX) 32 to adaptive processing section 30.Also have, though when receiving relevant each system of being used for this device obtains the signal of weighting before synthetic independently of one another, and the signal from weighting after synthetic (just from signal that TRX28 exported) is obtained useful signal selectively, the convergence of weighting control is improved, but in order to carry out this control, need be in two-way vector modulator 22 when the weighting phase-shift phase being made as zero the phase difference of TRX28 output signal, in advance will through the signal phase difference behind the RX32 be made as at least 0 ° ± below 90 °.

Yet, this is in and comprises preamble, single sign indicating number institute's signal of hypothesis (or setting) in advance during implements communication such as base station IDs, travelling carriage ID in addition in the received signal, and the contrast signal when part that can it is known under this occasion is handled as adaptive array is used.Under the occasion that can utilize contrast signal, and the well-known recurrence least square theory of error (RLS[=Recursive Least-Squares] algorithm) be comparatively effective.In the RLS algorithm, have to carry out at a high speed and control and the fast such feature of weighting weight convergence to the beam shaping of useful wave sense and to the zero-signal of disturbing wave sense.Specific description is because of there being detailed record in non-patent literature 1, and omits detailed description herein.

; the occasion of said structure; just the signal from the constituent parts antenna is weighted and synthetic after with carry out frequency translation by R-T unit (TRX) and by acceptance division (RX) output of constituent parts antenna carried out under the occasion of frequency translation respectively in the lump; if between TRX and each RX, amplitude and phase place are had error, then give and adopt the RLS convergence to bring bigger influence.That is to say that because the RLS algorithm minimizes the recurrence mean square error of contrast signal and synthesized received signal at a high speed, thereby if above-mentioned error is big, then evaluated error becomes big, reach satisfied according to the RLS algorithm and specify the time of the condition of convergence to increase.

Therefore, adopt the following first or second modified R LS algorithm after RLS algorithm in the past made amendment in the present embodiment, the convergence when between TRX and each RX amplitude error or phase error being arranged is greatly improved.Below, relevant this algorithm is illustrated.

At first, relevant RLS algorithm in the past is illustrated.If will be made as plural input vector X (i) [i: the positive integer corresponding with sample time] by the signal that a plurality of unit antenna is imported, then complex weighted vector W (m) [m: the parameter of expression convergence computing step] becomes,

Formula 5

W(m)＝R _xx ^-1(m)r _xr(m)

$R_{\mathrm{xx}}\left(m\right)=\underset{i=1}{\overset{m}{\mathrm{\Σ}}}{\mathrm{\α}}^{m-i}X\left(i\right)X^H\left(i\right)$

$r_{\mathrm{xr}}\left(m\right)=\underset{i=1}{\overset{m}{\mathrm{\Σ}}}{\mathrm{\α}}^{m-i}X\left(i\right)r^{*}\left(i\right)$

In formula, r (i): be contrast signal, H represents the complex conjugate transposition, and * represents complex conjugate.In addition, α is the real number that satisfies 0＜α≤1.At this moment, W (m+1) can be expressed as with W (m)

Formula 6

W(m+1)＝W(m)+γR _xx ^-1(m)X(m+1)e*(m+1) …(1)

e(m+1)＝r(m+1)-W ^H(m)X(m+1) …(2)

$r=\frac{1}{\mathrm{\α}+X^H(m+1){R_{\mathrm{xx}}}^{-1}\left(m\right)X(m+1)}\·\·\·\left(3\right)$

Herein, R _Xx ^-1(m) adopt following recurrence formula to obtain.

Formula 7

R _Xx ^-1(0)=δ ^-1I when m=0 (wherein) ... (4)

${R_{\mathrm{xx}}}^{-1}\left(m\right)$

$=\frac{1}{\mathrm{\α}}{R_{\mathrm{xx}}}^{-1}(m-1)-\frac{{R_{\mathrm{xx}}}^{-1}(m-1)X\left(m\right)X^H\left(m\right){R_{\mathrm{xx}}}^{-1}(m-1)}{{\mathrm{\α}}^2+{\mathrm{\αX}}^H\left(m\right){R_{\mathrm{xx}}}^{-1}(m-1)X\left(m\right)}$

(m=1 wherein, 2,3 ... the time) ... (5)

In this formula, δ: positive constant, I are the unit ranks.Then, the relevant W (m that obtains by above-mentioned formula ₁+ 1) (but step m=m ₁The time), check whether to satisfy and specify the condition of convergence (for example, SIR is that 40dB is with first-class).When satisfying the condition of convergence, adopt the W (m of this moment ₁+ 1) carries out weighting.Under the occasion that does not satisfy the condition of convergence, transfer to following step m ₂(=m ₁+ 1).That is to say, will be by step m ₁W (the m that is obtained ₁+ 1) as next step m ₂W (the m of above-mentioned formula (1)～(5) ₂) substitution in addition, obtain W (m ₂+ 1) (=W (m ₁+ 2)).Like this, specify the condition of convergence, all step m is given increment until satisfying.Also have, the m when finally satisfying the condition of convergence under this occasion becomes the convergence number of times.Top formula (1)～(5) are RLS algorithms in the past.

In the present embodiment, adopt above-mentioned formula (1)～(5) are changed to following RLS algorithm (the first modified R LS algorithm) behind formula (11)～(15) respectively.Be until the satisfied appointment condition of convergence m to be carried out increment and carry out computing with the something in common of the above-mentioned algorithm of RLS in the past.

Formula 8

W(m+1)＝W(m)+γ′{κR′ _xx ^-1(m)X(m+1)}e*(m+1)

…(11)

e(m+1)＝r(m+1)-W ^H(m)X(m+1) …(12)

$r^{\′}=\frac{1}{\mathrm{\α}+X^H(m+1)\left\{{{{\mathrm{\κR}}^{\′}}_{\mathrm{xx}}}^{-1}\left(m\right)X(m+1)\right\}}\·\·\·\left(13\right)$

R ' _Xx ^-1(0)=δ ^-1I when m=0 (wherein) ... (14)

${{R^{\′}}_{\mathrm{xx}}}^{-1}\left(m\right)$

$=\frac{1}{\mathrm{\α}}{{R^{\′}}_{\mathrm{xx}}}^{-1}(m-1)-\frac{\left\{{{{\mathrm{\κR}}^{\′}}_{\mathrm{xx}}}^{-1}(m-1)X\left(m\right)\right\}\left\{{\mathrm{\κX}}^H\left(m\right){{R^{\′}}_{\mathrm{xx}}}^{-1}(m-1)\right\}}{{\mathrm{\α}}^2+{\mathrm{\αX}}^H\left(m\right)\left\{{{{\mathrm{\κR}}^{\′}}_{\mathrm{xx}}}^{-1}(m-1)X\left(m\right)\right\}}$

(m=1 wherein, 2,3 ... the time) ... (15)

In formula, α (0＜α≤1) and κ (0＜κ＜1) are real numbers.

Perhaps, also can adopt above-mentioned formula (1)～(5) are changed to following RLS algorithm (the second modified R LS algorithm) behind (21)～(25) respectively.Be until the satisfied appointment condition of convergence m to be carried out increment and carry out computing with the something in common of the above-mentioned algorithm of RLS in the past and the first modified R LS algorithm.

Formula 9

W(m+1)＝W(m)+κγR″ _xx ^-1(m)X(m+1)e*(m+1)

…(21)

e(m+1)＝r(m+1)-W ^H(m)X(m+1) …(22)

$r=\frac{1}{\mathrm{\α}+X^H(m+1){R_{\mathrm{xx}}}^{-1}\left(m\right)X(m+1)}\·\·\·\left(23\right)$

R " _Xx ^-1(0)=δ ^-1I when m=0 (wherein) ... (24)

${{R^{,,}}_{\mathrm{xx}}}^{-1}\left(m\right)$

$=\frac{1}{\mathrm{\α}}{{R^{,,}}_{\mathrm{xx}}}^{-1}(m-1)-\mathrm{\κ}\frac{{{R^{,,}}_{\mathrm{xx}}}^{-1}(m-1)X\left(m\right)X^H\left(m\right){{R^{,,}}_{\mathrm{xx}}}^{-1}(m-1)}{{\mathrm{\α}}^2+{\mathrm{\αX}}^H\left(m\right){{R^{,,}}_{\mathrm{xx}}}^{-1}(m-1)X\left(m\right)}$

(m=1 wherein, 2,3 ... the time) ... (25)

In formula, α (0＜α≤1) and κ (0＜κ＜1) are real numbers.

Fig. 2 represents, the convergence number of times when comprising [phase error] between TRX and each RX in the second modified R LS algorithm.Transverse axis be when between TRX and each RX, giving phase error amount of phase error (°; Deg), the longitudinal axis is the convergence number of times when the condition of convergence being made as " power ratio (SIR) between useful wave signal and the disturbing wave signal is more than the 40dB ".In addition, also represent following method in order to compare, promptly among the algorithm of adaptive array as identical employing contrast signal with RLS to mean square error carry out minimized MMSE (: Minimum MeanSquare Error) one of method and generally be widely adopted according to LMS (: the method for algorithm Least MeanSquare).Also has employed constant when α is the weighting renewal.In this Fig. 2, κ=0.9, κ=0.7 and κ=0.5th, the result when adopting the second modified R LS algorithm.κ=1 o'clock, identical with in the past RLS algorithm.

Yet, do not have between TRX and each RX to learn when just error is 0 ° under the occasion of phase error that RLS algorithm (κ=1) was than the obvious convergence of LMS algorithm fast (the convergence number of times lacks) in the past.LMS needed 21 times renewal before convergence, (but α=0.2 o'clock) RLS algorithm is restrained by 1 time therewith relatively.

But,, learn that then in the past RLS algorithm and LMS algorithm become about equally convergence rate (convergence number of times), but the convergence of LMS algorithm aspect is fast for the phase error more than it if the phase error between TRX and each RX becomes 18 °.

To this, under the occasion of the second modified R LS algorithm (κ=0.9, κ=0.7 and κ=0.5), learn, then convergence rate is slow if compare with in the past RLS algorithm near error is 0 °, phase error reach 30 ° near then convergence rate change little and restrain than more quickly.Also have, under the big occasion (κ=0.9) of the value of κ among the above-mentioned second modified R LS algorithm, error hour convergence is fast, and then convergence rate is slack-off if error increases.On the other hand, under the little occasion of the value of κ (κ=0.5), though slow in the little regional convergence rate of error, not slack-off even error increases convergence rate with it yet.Thereby, under the occasion of using the second modified R LS algorithm, preferably select suitable κ according to the margin of error of being supposed between TRX and each RX.

On the other hand, Fig. 3 is illustrated in the convergence rate when comprising " phase error " in the first modified R LS algorithm between TRX and each RX.The record method of Fig. 3 is identical with Fig. 2.Learn that by this Fig. 3 even under the occasion that adopts the first modified R LS algorithm, compare the convergence number of times with RLS algorithm (κ=1) in the past still few.

In addition, Fig. 4 is illustrated in the convergence rate when comprising " amplitude error " in the first modified R LS algorithm between TRX and each RX.Transverse axis is the amplitude error amount (dB) when giving phase error between TRX and each RX, and the longitudinal axis is the convergence number of times when the condition of convergence being made as " power ratio (SIR) between useful wave signal and the disturbing wave signal is more than the 40dB ".Learn that by this Fig. 4 adopting under the occasion of the first modified R LS algorithm, even for " amplitude error ", to compare the convergence number of times still few with RLS algorithm (κ=1) in the past.

Also have, non-shared circuit part 24 best its constitute and further make in transmitting system and receiving system by equating (perhaps possessing adjustable structure important document) time of delay in the present embodiment.This is according to equating that in the frequency slope by phase place between time of delay (more detailed is the group delay time) equal circuit such group delay time definition makes.That is to say,, also can on other frequency, prevent to produce phase difference from fixed value even on certain characteristic frequency the phase difference between transmitting system and the receiving system is being made as fixingly.That is to say, can make the phase difference between transmit path and the RX path roughly the same, so relevant to utilize the sort of communicator of a plurality of frequencies be effective especially because this constitutes about more vast frequency band.

As top illustrated, according to the present invention, since in transmission/receiving course with the path of signal sharing as much as possible, thereby obtain following effect, promptly can realize adopting the self adaptation action of identical parameters in the transmitting-receiving process, the more easy and accuracy antenna pattern between more the highland reduces to send/receive poor, meanwhile by adopting the above-mentioned first or second modified R LS algorithm, even between TRX and each RX, produce amplitude and phase error or amplitude error, the calculation process time (reaching the convergence time of specifying the condition of convergence in the RLS algorithm) that also can shorten the RLS algorithm significantly compared with the past.

Claims (5)

1. an array antenna communication device adopts the adaptive array antenna that contains a plurality of unit antennas, it is characterized by: possess

RF transmitting system circuit is arranged in above-mentioned each unit antenna, and comprises transmit power amplifier at least;

RF receiving system circuit is arranged in parallel with above-mentioned RF transmitting system circuit in above-mentioned each unit antenna, and comprises low noise amplifier at least;

Two-way vector modulator is connected with above-mentioned RF transmitting system circuit and RF receiving system circuit jointly at the opposite side of unit antenna;

Divide to fit in portion, be connected with a plurality of two-way vector modulators;

Receiving and transmitting part (TRX) is connected with the above-mentioned branch portion of fitting in;

Adaptive processing section is controlled above-mentioned two-way vector modulator, and a plurality of unit antennas are used as adaptive array antenna;

Acceptance division (RX), be arranged at above-mentioned each RF receiving system circuit and above-mentioned adaptive processing section between,

For above-mentioned RF transmitting system circuit corresponding and RF receiving system circuit with the constituent parts antenna, the mutual difference of the difference that amplitude variations amount when making signal through above-mentioned RF transmitting system circuit corresponding with the constituent parts antenna and RF receiving system circuit is mutual and the amount of phase rotation between the constituent parts antenna about equally

Adaptive processing section is controlled above-mentioned two-way vector modulator according to satisfying the complex weighted vector W (m+1) that specifies the condition of convergence, and this appointment condition of convergence is to adopt the modified R LS algorithm computation according to following formula (11)～(15) to go out,

W(m+1)＝W(m)+γ′{κR′ _xx ^-1(m)X(m+1)}e*(m+1)

···(11)

e(m+1)＝r(m+1)-W ^H(m)X(m+1) ···(12)

${\mathrm{\γ}}^{\′}=\frac{1}{\mathrm{\α}+X^H(m+1)\left\{\mathrm{\κ}{{R^{\′}}_{\mathrm{xx}}}^{-1}\left(m\right)X(m+1)\right\}}---\left(13\right)$

R ' _Xx ^-1(0)=δ ^-1I when m=0 (wherein) (14)

${{R^{\′}}_{\mathrm{xx}}}^{-1}\left(m\right)$

$=\frac{1}{\mathrm{\α}}{{R^{\′}}_{\mathrm{xx}}}^{-1}(m-1)\frac{\left\{\mathrm{\κ}{{R^{\′}}_{\mathrm{xx}}}^{-1}(m-1)X\left(m\right)\right\}\left\{\mathrm{\κ}{X}^H\left(m\right){{R^{\′}}_{\mathrm{xx}}}^{-1}(m-1)\right\}}{{\mathrm{\α}}^2+\mathrm{\α}{X}^H\left(m\right)\left\{\mathrm{\κ}{{R^{\′}}_{\mathrm{xx}}}^{-1}(m-1)X\left(m\right)\right\}}$

(m=1 wherein, 2,3 ... the time) (15)

Here,

W(m)＝R _xx ^-1(m)r _xr(m)

$R_{\mathrm{xx}}\left(m\right)=\underset{i=1}{\overset{m}{\mathrm{\Σ}}}{\mathrm{\α}}^{m-i}X\left(i\right)X^H\left(i\right)$

$r_{\mathrm{xr}}\left(m\right)=\underset{i=1}{\overset{m}{\mathrm{\Σ}}}{\mathrm{\α}}^{m-i}X\left(i\right)r^{*}\left(i\right)$

X (i): plural input vector, r (i): contrast signal, i: the positive integer corresponding with sample time, H: complex-conjugate transpose, *: complex conjugate, α: the real number that satisfies 0＜α≤1, κ: the real number that satisfies 0＜κ＜1, δ: positive constant, I: unit ranks, m: the parameter of expression convergence computing step.

2. an array antenna communication device adopts the adaptive array antenna that contains a plurality of unit antennas, it is characterized by: possess

RF transmitting system circuit is arranged in above-mentioned each unit antenna, and comprises transmit power amplifier at least;

RF receiving system circuit is arranged in parallel with above-mentioned RF transmitting system circuit in above-mentioned each unit antenna, and comprises low noise amplifier at least;

Two-way vector modulator is connected with above-mentioned RF transmitting system circuit and RF receiving system circuit jointly at the opposite side of unit antenna;

Divide to fit in portion, be connected with a plurality of two-way vector modulators;

Receiving and transmitting part (TRX) is connected with the above-mentioned branch portion of fitting in;

Adaptive processing section is controlled above-mentioned two-way vector modulator, and a plurality of unit antennas are used as adaptive array antenna;

Acceptance division (RX), be arranged at above-mentioned each RF receiving system circuit and above-mentioned adaptive processing section between,

For above-mentioned RF transmitting system circuit corresponding and RF receiving system circuit with the constituent parts antenna, the mutual difference of the difference that amplitude variations amount when making signal through above-mentioned RF transmitting system circuit corresponding with the constituent parts antenna and RF receiving system circuit is mutual and the amount of phase rotation between the constituent parts antenna about equally

Adaptive processing section is controlled above-mentioned two-way vector modulator according to satisfying the complex weighted vector W (m+1) that specifies the condition of convergence, and this appointment condition of convergence is to adopt the modified R LS algorithm computation according to following formula (21)～(25) to go out,

W(m+1)＝W(m)+κγR″ _xx ^-1(m)X(m+1)e*(m+1)

···(21)

e(m+1)＝r(m+1)-W ^H(m)X(m+1) ···(22)

$\mathrm{\γ}=\frac{1}{\mathrm{\α}+X^H(m+1){R_{\mathrm{xx}}}^{-1}\left(m\right)X(m+1)}---\left(23\right)$

R " _Xx ^-1(0)=δ ^-1I when m=0 (wherein) (24)

${{R^{,,}}_{\mathrm{xx}}}^{-1}\left(m\right)$

$=\frac{1}{\mathrm{\α}}{{R^{,,}}_{\mathrm{xx}}}^{-1}(m-1)-\mathrm{\κ}\frac{{{R^{,,}}_{\mathrm{xx}}}^{-1}(m-1)X\left(m\right)X^H\left(m\right){{R^{,,}}_{\mathrm{xx}}}^{-1}(m-1)}{{\mathrm{\α}}^2+\mathrm{\α}{X}^H\left(m\right){{R^{,,}}_{\mathrm{xx}}}^{-1}(m-1)X\left(m\right)}$

(m=1 wherein, 2,3 ... the time) (25)

Here,

W(m)＝R _xx ^-1(m)r _xr(m)

$R_{\mathrm{xx}}\left(m\right)=\underset{i=1}{\overset{m}{\mathrm{\Σ}}}{\mathrm{\α}}^{m-i}X\left(i\right)X^H\left(i\right)$

$r_{\mathrm{xr}}\left(m\right)=\underset{i=1}{\overset{m}{\mathrm{\Σ}}}{\mathrm{\α}}^{m-i}X\left(i\right)r^{*}\left(i\right)$

X (i): plural input vector, r (i): contrast signal, i: the positive integer corresponding with sample time, H: complex-conjugate transpose, *: complex conjugate, α: the real number that satisfies 0＜α≤1, κ: the real number that satisfies 0＜κ＜1, δ: positive constant, I: unit ranks, m: the parameter of expression convergence computing step.

3. according to the array antenna communication device of claim 1 or 2 records, it is characterized by:

For above-mentioned RF transmitting system circuit corresponding with the constituent parts antenna and RF receiving system circuit, the time of delay when making signal through above-mentioned RF transmitting system circuit corresponding with the constituent parts antenna and RF receiving system circuit about equally.

4. according to the array antenna communication device of claim 3 record, it is characterized by:

At above-mentioned RF transmitting system circuit corresponding with the constituent parts antenna and at least one side among the RF receiving system circuit adjustment part is set, this adjustment part is used for adjusting above-mentioned amplitude variations amount, the amount of phase rotation and among time of delay at least one.

5. according to the array antenna communication device of claim 1 or 2 records, it is characterized by:

Described array antenna communication device is a kind of time-division bidirectional communication apparatus that same communication counterpart is carried out transceiver with identical frequency.

Images