CN1271806C - Transmitting set and receiving set of multiple antenna communication system under time division duplexing mode, and method - Google Patents

Abstract

The present invention provides a high-speed data system of multiple antennas under a time division duplex mode and an implementation method thereof. The high-speed data system comprises a transmitting device and a receiving device, wherein the transmitting device comprises a demultiplexing device, a band spread and modulation scrambling device, a burst data stream forming device, a merging device and a transmitting antenna, wherein the demultiplexing device is used for splitting data streams of a burst structure into at least two layers of data; the receiving device comprises a plurality of receiving antennas of which the number is equivalent to the layer number of the data at least, a channel estimation device, a multidiameter combiner, a signal detection device, a disturbing zero adjustment device, a matched filtering and deciding device, a disturbing reconstruction device, a judgment device and a multipath combiner. By the present invention, a time division duplexing CDMA system and a multiple-input multiple-output (MIMO) system can be combined to increase a transmission rate.

Description

Many antenna datas communication system emitter under the TDD mode and receiving system and method

Technical field

The present invention relates to mobile communication, relate in particular to the high-speed data communication system of the many antennas under the TDD mode, emitter and receiving system and implementation method.

The present invention is specifically related to utilize in the transmitter a plurality of antennas in a plurality of antennas and receiver to carry out the wireless communication system of high speed data transfer, i.e. multiple-input and multiple-output (MIMO) system.And the time-division dual-mode uses a plurality of transmitting antennas and a plurality of reception antenna to carry out the method for high speed data transfer down.

Background technology

Along with the abundant development of wireless communication system, speech business can not have been satisfied the requirement of people to high-speed data service.The data service that web page browsing, multi-medium data transmission and other types are provided is a free-revving engine of development wireless communication system and service.Particularly, based on the 3-G (Generation Three mobile communication system) of code division multiple access (CDMA) system.

Though proposed the suggestion that the existing Radio Resource (such as code channel, time slot etc.) of multiple utilization provides high speed data transfer, it only exchanges the method for data capacity for voice capacity.Along with the appearance of the technology of MIMO, a kind of method of utilizing a plurality of transmitting antennas, a plurality of reception antenna to carry out high speed data transfer is suggested, and becomes a kind of trend of following wireless communication technology development.This technology is applicable to the wireless environment of multiple scattering body, and feasible signal from each transmitting antenna is highly incoherent in each reception antenna, and utilizes this irrelevance that the data of a plurality of antenna transmission are distinguished and detected at receiver end.Wherein the most famous is exactly so-called V-BLAST structure and corresponding V-BLAST detector.Verified, have the mimo system of M transmitting antenna and P reception antenna, almost can be under the situation of P 〉=M so that channel capacity is brought up to original M doubly.

Fig. 1 shows the schematic diagram of the typical V-BLAST system of prior art.Wherein, a single data flow is decomposed into M sub data flow by multichannel, and each sub data flow is encoded into symbol, and passes to transmitter separately.M transmitter is operated in the same frequency range, adopts synchronization timing to send between the transmitter.Wherein, each sub data flow is called as one deck (Layer), the Space Time Coding structure of layering so BLAST is otherwise known as.System adopts burst structure to send, and space channel remains unchanged in the time that each burst structure continues, and changes between different burst structure.Receiver end uses P independently receiver, and is operated in the same frequency range.Each receiver receives the signal from all M transmitting antenna.Traditional V-BLAST system postulation space channel is a flat fading, and space channel can be represented with the gain matrix H that a dimension is P * M so, its element h _JiFor transmitter i to the channel impulse response between the receiver j, and require M≤P.In the transport process of each burst structure, ignore channel variation in time, and suppose that channel is accurately estimated, such as in each burst structure, using training sequence to carry out channel estimating.

Suppose a=(a ₁a ₂Λ a _M) ^TBe the symbolic vector of emission, r=(r ₁r ₂Λ r _P) ^TBe the symbolic vector that receives, wherein symbol () ^TThe transposition of expression vector can be expressed as relation between the receiving and transmitting signal so:

r＝Ha+n

Wherein n is one and has independent equally distributed additive white Gaussian noise vector.

Typical V-BLAST detection method is to use interference that zero (InterferenceNulling), orderly interference eliminated are detected many antenna datas.Fig. 2 is that use ZF (ZF) criterion of prior art disturbs the V-BLAST to zero to detect schematic diagram.

The MIMO technology also is suggested (with reference to Chinese patent 00810923.0) with the scheme that combines of cdma system, but it only is directed to abundant relatively Frequency Division Duplexing (FDD) (FDD) cdma system (such as WCDMA and CDMA2000) of spread spectrum code resources, uses same spread sign indicating number or different spreading code to carry out high speed data transfer on different antennae.

Under existing technical background, the many antennas that utilize under the TDD mode of moulding do not carry out scheme, the especially 3-G (Generation Three mobile communication system) of high speed data transfer.

Though utilize wireless resource allocation such as code channel, time slot can reach the purpose that improves transmission rate, it just to the reallocation of existing resource, does not increase extra resource can utilize altogether, so the raising of its transmission rate is limited.Utilize multi-antenna technology, can be so that above-mentioned resource become the growth (relevant with the antenna number) of multiple, thus make that the heap(ed) capacity of system is provided, and can support more number of users.

But TDD mode and mode of frequency division duplexing have bigger different, and such as code source, detection method etc., the MIMO technology under the existing mode of frequency division duplexing can not be applied directly to TDD mode.

Summary of the invention

The objective of the invention is shortcoming, a kind of high-speed data transmission of many antennas and the device and implementation method of reception of realizing under TDD mode is provided at above-mentioned prior art.The present invention utilizes in the transmitter that a plurality of antennas carry out high speed data transfer in a plurality of antennas and receiver, is a kind of multiple-input and multiple-output (MIMO) system.

The invention provides the high-speed data transmission device of the many antennas under a kind of TDD mode, comprising:

The multichannel decomposer, the data flow that is used for distributing to all burst structure of time slot is divided at least two layer data, and described every layer data comprises at least one height stream;

Spread spectrum corresponding with the number of plies of described data and modulation scrambling apparatus are respectively applied for each son stream spread spectrum and modulation scrambling with described at least two layer data;

The bursty traffic corresponding with the number of plies of described data forms device, is used for inserting the substream of data that training sequence forms burst structure at each son stream of described spread spectrum and modulation scrambling;

The merging device corresponding with the number of plies of described data is used for the substream of data that described each bursty traffic forms all burst structure that device forms merged into and transmits;

The signal transmitting antenna corresponding with the number of plies of described data is used to launch transmitting of described modulation.

Alternatively, described spread spectrum comprises the spread spectrum device identical with the sub-fluxion of described every number of plies, the sub-device of modulation and the sub-device of scrambling with the modulation scrambling apparatus, is used for respectively with each son stream spread spectrum, modulation and scrambling.

Preferably, described bursty traffic formation device comprises: the basic training sequences generator is used to produce basic training sequences; Layer concentrator marker is used for by training sequence label layer information; Inserter is used for inserting identical training sequence at each son stream of described every layer data; Wherein, the training sequence difference of different layers, and by described basic training sequences skew generation.

The present invention also provides the receiving system of the high-speed data of the many antennas under a kind of TDD mode, comprising:

At least equal a plurality of reception antennas of the number of plies of described data, be used to receive the signal of the transmission antennas transmit of the described number of plies that equals described data;

Channel estimating apparatus is used for a plurality of signals of the described reception of channel estimating, obtains the impulse response on the different paths;

The multipath combiner is used for carrying out according to the impulse response of described acquisition the merging of multipath data, and making it equivalence is the reception data of a flat channel, produces multilayer and receives data, and produce corresponding with it equivalent channel impulse response matrix;

Signal supervisory instrument is used to detect every layer and receives data, obtains the estimated value of all layer data;

Interference is used for the equivalent channel impulse response matrix according to described acquisition to zero device, and the every layer of interference that described multilayer is received data be to zero, suppressing the interference of other layer data, and obtains the data of optimum layer by the method for ordering;

Matched filtering and judgment device are used for the described optimum layer data of matched filtering and judgement, obtain the estimated value of this layer data, and wherein channel code can be used to distinguish different users, also can combine the different user of common differentiation with layer sequence number;

The disturbance reconstruction device is used for rebuilding the pairing reception data of this layer according to the estimated value of this layer data of described acquisition;

Judgment means is used for the information according to described received signal, has judged whether to obtain the estimated value of all layer data in the time slot;

The multichannel combiner is used to merge all detected layer data, to obtain the estimated value of all layer data in the time slot.

Alternatively, the reception data of multipath combiner output satisfy predetermined arrangement, make the channel impulse response matrix of equivalence satisfy the characteristic of piece diagonal matrix, wherein each the height piece on the diagonal is corresponding to the channel impulse response matrix of a chip, and all sub-pieces are all identical; To disturb to zero the time, only the equivalent channel impulse response to a chip disturbs zero, and the result is applied on all chips in the same time slot that receives.

Preferably, described interference also comprises interference blanking unit to zero device, is used for receiving the influence that data are eliminated the data of the described layer that has obtained estimated value in multilayer, so that carry out the detection of next layer data.

The present invention provides the receive-transmit system of the high-speed data of the many antennas under a kind of TDD mode again, comprising: aforementioned emitter and receiving system.

The present invention provides the launching technique of the high-speed data of the many antennas under a kind of TDD mode again, comprises step:

The data flow of decomposing all burst structure in the time slot is at least one layer data, and described at least one layer data comprises at least one height stream of decomposition;

Identify every layer data;

Described each the son stream of spread spectrum, modulation and scrambling;

In described each son stream, add training sequence respectively and form burst structure;

A plurality of son streams of every layer data are all superposeed formation multilayer emission data by a plurality of transmission antennas transmit.

Alternatively, the step of the every layer data of described sign comprises: the basic training sequences different to the data-bias of different layers, and using the deviant presentation layer, and/or the relation of layer and scrambler, and/or carry channelization code information.

The method of reseptance of the high-speed data of the many antennas under a kind of TDD mode provided by the invention comprises step:

Antenna with the layer that is no less than the emission data receives the signal of described emission, produces a plurality of received signals;

A plurality of signals of the described reception of channel estimating obtain the impulse response on the different paths;

Carry out the merging of multipath data according to the impulse response of described acquisition, making it equivalence is the reception data of a flat channel, produces multilayer and receives data, and produce corresponding with it equivalent channel impulse response matrix;

According to the equivalent channel impulse response matrix of described acquisition, the every layer of interference that described multilayer is received data be to zero, suppressing the interference of other layer data, and obtains the data of optimum layer by the method for ordering;

The data of matched filtering and the described optimum layer of judgement obtain the estimated value of this layer data, and wherein channel code can be used to distinguish different users, also can combine the different user of common differentiation with layer sequence number;

To the signal disturbance reconstruction of described reception, be used for rebuilding the pairing reception data of this layer according to the estimated value of this layer data of described acquisition;

Repeat above-mentioned steps, the estimated value of all layer data in obtaining a time slot;

Multichannel merges all detected layer data, to obtain the estimated value of all layer data in the time slot.

Alternatively, the reception data that multipath merges output satisfy predetermined arrangement, make the channel impulse response matrix of equivalence satisfy the characteristic of piece diagonal matrix, wherein each the height piece on the diagonal is corresponding to the channel impulse response matrix of a chip, and all sub-pieces are all identical; With described interference to zero step in, only the equivalent channel impulse response to a chip disturbs zero, and the result is applied on all chips in the same time slot that receives.

The method that transmits and receives of the high-speed data of the many antennas under a kind of TDD mode of the present invention comprises aforementioned step of transmitting and receiving step.

The present invention on the basis of existing technology, the uncorrelated channel of the multichannel of utilizing a plurality of dual-mode antenna to provide, transmit different data streams to improve the transmission rate of entire system, and in each incoherent channel, continue to use different spreading codes to form the CDMA code channel of a plurality of quadratures, make it to combine with existing C DMA system.Briefly, in the cdma system that promptly basic principle of V-BLAST system is applied to.Traditional V-BLAST system is a non-cdma system that is applied in the flat channel environment, and method and apparatus provided by the invention further can be applied in the frequency-selective channel environment, and and existing C DMA system combine, make it to have more the possibility of practical application.

The present invention utilizes the distinctive frame structure of time division duplex cdma system to carry out the transmission of data.Specifically, the data flow under the TDD mode is that time slot of a time slot " burst (burst) " formula sends, and this continuous send mode with Frequency Division Duplexing (FDD) is different.One of them burst structure is made of front and back two parts of the data of transmission and training sequence (midamble) placed in the middle.In traditional time division duplex cdma system, training sequence is mainly used in channel estimating and transmits spreading code information.The present invention further utilizes training sequence transfer layer serial number information, and provides corresponding exercise sequence distributing method, channelization code allocation method and scrambling code distribution method, and making the time division duplex cdma system combine with mimo system becomes possibility.

Can utilize channel code and the different user of layer sequence number data separation in the transmission plan of the present invention, thereby provide the multi-user to support.

Give in the output detection method in the present invention, utilized the distinctive frame structure of time division duplex cdma system to carry out Data Detection equally.The existence of training sequence placed in the middle in burst structure, what impel that the time division duplex cdma system uses is the piece detection mode of a time slot of a time slot, this continuous detecting mode with the Frequency Division Duplexing (FDD) cdma system is different.Because the training sequence that the different chips (chip) in piece use is identical, promptly its channel estimation value is identical, can be applied to all chips in the piece to the V-BLAST detector for a chip simultaneously so, thereby simplify testing process greatly.In patent of the present invention, the CDMA detector is included in the V-BLAST detector, after promptly one deck Data Detection finishes, carries out the matched filtering of CDMA sign indicating number, recovers the data of each code channel, and then carries out the interference cancellation process in traditional V-BLAST detector.

Transmitter architecture that the present invention provides and receiver structure schematic diagram are under the principle that guarantees above-mentioned transmission plan and detection method, and design in conjunction with the characteristic of former time division duplex cdma system.Therefore use transceiver arrangement that this patent provides and can realize combining of time division duplex cdma system and mimo system, reach the purpose of raising transmission rate in conjunction with transmission plan of the present invention, detection method etc.

Description of drawings

Fig. 1 shows the schematic diagram of the typical V-BLAST system of prior art;

Fig. 2 is that use ZF (ZF) criterion of prior art disturbs the V-BLAST to zero to detect schematic diagram;

Fig. 3 provides a plurality of transmitting antennas of having of embodiments of the invention and the transmitter architecture schematic diagram of high speed data transfer is provided, and the number of transmitting antenna is M;

Fig. 4 shows the deviant allocative decision of TD-SCDMA system in the embodiments of the invention 1, and wherein, minimum offset values is 16, and transmitting antenna is 2;

Fig. 5 shows the deviant allocative decision of TD-SCDMA system in the embodiments of the invention 2, and wherein, minimum offset values is 16, and transmitting antenna is 4;

Fig. 6 shows the receiver structure schematic diagram of embodiments of the invention;

Figure 7 shows that the structural representation of channel impulse response matrix H of the present invention;

Fig. 8 has provided and utilized the present invention is 4 at the dual-mode antenna number average, the relation curve of the error rate under the frequency-selective channel situation (BER) and signal to noise ratio (Eb/No);

Fig. 9 has provided and utilized the present invention is 8 at the dual-mode antenna number average, the relation curve of the error rate under the frequency-selective channel situation (BER) and signal to noise ratio (Eb/No).

Embodiment

Implement and understand the present invention for the ease of persons skilled in the art, describe method and apparatus of the present invention with reference to accompanying drawing by embodiment respectively below.Though will be that example is specifically described the present invention with the TD-SCDMA system in the narration below, persons skilled in the art know that the present invention can use other TDD cdma systems or other similar wireless communication systems, comprises wireless fixed communications.

Transmitter architecture

Fig. 3 provides a plurality of transmitting antennas of having of embodiments of the invention and the transmitter architecture schematic diagram of high speed data transfer is provided, and the number of transmitting antenna is M.Situation for encoding in the schematic diagram that Fig. 3 provides also is to describe at this not coding situation in the narration below, should be understood that the present invention is equally applicable to the coding situation.Persons skilled in the art know that coding can carry out before multichannel is decomposed, also can carry out after multichannel is decomposed, and perhaps carry out simultaneously.

In the schematic diagram that Fig. 3 provides, the corresponding relation that M group sends between data and the M transmit antennas is fixing, this do not get rid of this corresponding relation can be on different time-gap randomly changing or periodically-varied.Under the situation of coding, when especially same son stream was encoded on different time-gap, the change of this corresponding relation can utilize coding gain to improve the reliability of link.

According to transmitter architecture shown in Figure 3, the workflow of whole transmitter terminal is as follows:

(1) distribute to the data flow s of all burst structure in the time slot, at first be divided into MN son stream in " multichannel decomposition " unit, wherein every N son stream is called a layer data corresponding to a transmit antennas, has the M layer data in the structure so shown in Figure 3;

(2) the sub-flow point of each in every layer data does not carry out spread spectrum, modulation and scrambling, and the principle of its spread spectrum, modulation and scrambling is identical with former TDD cdma system, and concrete channel code, scrambling code distribution method will specifically be described with reference to following transmission plan;

(3) the sub-flow point of each in every layer data does not add training sequence formation burst structure, and its burst structure is identical with former TDD cdma system, and concrete exercise sequence distributing method can be with reference to the description of the transmission plan of mask body down;

(4) N burst structure of of N in every layer data stream formation, being superimposed to form before sending sends data d _i(i=1 Λ M), and from the antenna of correspondence, send.

Transmission plan

For the tdd mode cdma system that has used the MIMO technology, there are three kinds of resources to be utilized: time slot, code channel and antenna.One of them user's data sequence can take a time slot and send, and also can take a plurality of time slots and carry out the data transmission.User data s in time slot at first is broken down into M sub data flow according to the antenna number, and each height stream is called as a layer data.The total data of this one deck all sends by a corresponding transmitting antenna, and before sending, each layer data is divided into the N group once more, and is assigned on N the CDMA code channel.Then, the data on each code channel are carried out spread spectrum, modulation and scrambling respectively, and add training sequence formation burst structure.At last, the code channel of the N in each layer data merges, and goes out by corresponding antenna transmission.

Data on the different layers can be used transmission of same spread sign indicating number or the transmission of different spreading code.Spreading code is made of jointly channel code and scrambler, wherein refer to that with the sign indicating number transmission different layers upward uses identical channel code and identical scrambler, different sign indicating number transmission comprise multiple situation, it can be channel code difference but scrambler is identical, also can be identical but scrambler difference, perhaps their combination of channel code.Among the present invention, utilize channel code to distinguish different users jointly with antenna, each user can distribute one or more channel code different with other users according to the difference of business, if two users' channel code is identical, can further rely on it to distinguish from different antennae (layer).This transmission plan makes maximum system capacity improve original M doubly, and can support a maximum MQ number of users with respect to former TD-SCDMA system so, and wherein Q is the number of channel code.But be to use layer to carry out the user data area timesharing, this user itself must know which layer is the data of oneself come from, and the order of layer, thereby can select, recover to belong to this user's data in numerous layers data.When using different scramblers on the different antennae, can replace layer to carry out user's differentiation with scrambler.What adopt in an embodiment of the present invention is exactly that different layers uses different this defeated modes of scrambler.

In order to obtain channel impulse response at receiver end from different transmit antennas, every layer need to be distributed different training sequences, and though promptly with all data of one deck its from which user, take which code channel and all use same training sequence to send.The training sequence that different layers uses is produced by skew by same basic training sequences (basic midamble code).Here it should be noted that, different antennae is distributed different training sequences, cause up when carrying out high speed data transfer, maximum number of user is except top restriction, also be subjected to the restriction of terminal antenna number P, training sequence length L _ M and minimum training sequence deviation value Delta, be that maximum number of user is: L_M//(PDelta), the downlink user number is then irrelevant therewith.

A kind of so in addition training sequence method of salary distribution is different with default (default) mode with general (common) mode of former TD-SCDMA system.These two kinds of methods of salary distribution are intended to terminal when receiving, and can obtain other users' spreading code information by training sequence deviation value relevant information, thereby carry out Multiuser Detection.The general mode in its Central Plains is meant that when base station down sends all code channels use identical training sequence, and informs that with the deviant between this training sequence and the basic training sequences how many spreading codes are terminal have in use; Default mode is the one-to-one relationship of setting up between a kind of training sequence deviation value and the spreading code, and each code channel uses the training sequence of different deviants, and terminal obtains all users' spreading code by this one-to-one relationship after carrying out channel estimating.

So after supporting MIMO, originally two kinds of methods of salary distribution certainly will will change, this is because training sequence must be corresponding with transmitting antenna (layer), and training sequence also needs the serial number information of attendant layers, inform that promptly the current detected layer of receiver is in position in all layers at transmitter terminal, so that multichannel can correctly recover initial data when merging.In addition, receiver also needs to know scrambling code information and channelization code information.For layer sequence number information, can utilize the size of every layer of training sequence deviation value to characterize so, promptly deviant arrange from small to large represent ground floor, the second layer respectively, up to the M layer, vice versa.When different layers used the different scrambling codes transmission, the information of scrambler also can make in this way and transmit, and promptly ground floor is corresponding to first scrambler, and the second layer is corresponding to second scrambler, by that analogy; If exist different layers to use the situation of same scrambler, and all layers have used a plurality of scramblers again, so also can make the relation of setting up in this way between layer and the scrambler, as long as receiver end is also known this corresponding relation.

Under the principle that provides in the above, transmission for channelization code information, can adopt following two kinds of schemes: different layers can use any training sequence deviation value in first kind of scheme, can characterize a layer information as long as guarantee the size of different layers deviant, deviant is not informed any channelization code information of terminal, terminal can use single user to detect, and also can know channelization code information by methods such as code channel activating detections, carries out Multiuser Detection; Another allocative decision is that different layers uses different training sequence deviation values, but select the difference of deviant by all layers, can attach extraneous information, inform that how many channel codes are terminal have in use, and require channel code to distribute according to a definite sequence, come the front such as same group of the inside sequence number and be assigned with earlier, as long as terminal knows that having how many channel codes to be used can know that promptly which channel code is used like this.

Provide two kinds of deviant allocative decisions under 2 transmit antennas and the 4 transmit antennas situations below respectively.With the TD-SCDMA system is example, and its maximum channel number of codes is 16, and training sequence length L _ M is 144, and supposes that training sequence that each layer uses and the deviant between the basic training sequences are as the nonnegative integer of minimum offset values Delta doubly.At transmitting antenna M=2, a solution under the Delta=16 situation as shown in Figure 4 so.This a kind of method of salary distribution wherein only can be represented 8 kinds of different deviants respectively with 8 positions of a box the inside as can be seen, and two antennas are placed on the different positions and just can determine different corresponding relations so, promptly $C_8^2=28$ Kind of corresponding relation can be chosen wherein 16 kinds of relations between antenna deviation value and the code channel number the most wantonly, and such compound mode has C ₂₈ ¹⁶Kind.In the form text wherein shown in Figure 4, a kind of code channel numerical value of line display situation, totally 16 row represent that respectively 16 kinds of code channels take situation; Preceding 8 row in the delegation, represent the ratio between 8 kinds of different deviants and the minimum offset values, if this cell has " X " symbol, represent that this ratio is selected, the training sequence deviation value of this layer data is promptly determined so, and last tabulation in delegation's form is shown in preceding 8 row corresponding code channel number under " X " symbol distribution situation.That Fig. 5 provides is transmitting antenna M=4, a solution under the Delta=16 situation.According to top analysis, can draw total $C_8^4=70$ Kind of corresponding relation is available, the basically identical that provides of the meaning of form shown in Figure 5 and Fig. 4 wherein, and what only Fig. 5 provided is one group of combination more clocklike, promptly position 0 represents 2 ⁰=1, position 1 expression 2 ¹=2, position 3 expressions 2 ³=8, position 4 expressions 2 ⁴=16, position 5,6,7 represents 0 respectively, so number of channelization codes be four positions representative of choosing value and.According to this design, in actual use,, need preserve this form simultaneously at transmitter terminal and receiver end for solution shown in Figure 4, transmitting terminal carries out the distribution of training sequence according to this form and code channel number number; Then need deviant at receiver end, the code channel number that obtains taking by the method for tabling look-up according to training sequence.And solution shown in Figure 5, then need not preserve the data of whole form at receiver end, only need the corresponding relation between location aware and the numerical value to get final product, receiver can be from the deviant of training sequence, directly calculate the code channel number that takies according to the numerical value corresponding relation, thus the memory space of the receiver end of effectively saving and the cost of tabling look-up and being paid.

Need only by analyzing as can be seen $C_{L\_M/\mathrm{Delta}}^M\≥16,$ Promptly can provide the solution under M (M 〉=2) the root antenna situation.For the TD-SCDMA system, under the Delta=16 situation, the number of antennas that can support is 6 at most so; If Delta=8 can support that so maximum number of antennas is 15; If Delta can be littler, perhaps cancelling deviant is the restriction doubly of Delta nonnegative integer, can support more antennas so.

Receiver structure

Corresponding to transmitter architecture shown in Figure 3 or similar transmitter architecture, Fig. 6 has provided the respective receiver structural representation, and wherein the reception antenna number is P, and requires P 〉=M.Data are at first carried out channel estimating on the many reception antennas, obtain the impulse response on the different paths, carry out the merging of multipath data then, and the reception data that make it equivalence and be a flat channel detect.Rake combiner when the merging of multipath data can be used sky (Space-Time Rake Combiner) also can use selectivity to merge device (Selective Combiner).The following mode of The data after multipath merges is carried out layering and is detected, at first disturb the interference that zero (Interference Nulling) is suppressed other layer data, select to disturb one deck to carry out yard matched filtering and judgement to wanted to interfering signal ratio (SINR) maximum after zero or error minimum, obtain the estimated value of this layer data, carry out disturbance reconstruction then, and in received signal, cancel the influence of this layer data, carry out the detection of next layer data.After detecting fully and finish, all layer data carry out the merging of multichannel data.Wherein by disturbing zero, calculating unit such as optimal detection order and interference eliminated and constitute typical V-BLAST detector.

In conjunction with receiver structure shown in Figure 6, the workflow of whole receiver end can be described below:

(1) the data r that receives of many reception antennas at first enters " channel estimating " unit, obtain each dual-mode antenna between channel estimation value, the method that its " channel estimating " unit adopts can be identical with former TDD cdma system;

(2) receive data r and channel estimation value and sent into " multipath combiner " unit simultaneously, carry out the multipath merging and make frequency-selective channel equivalence to be flat fading channel, concrete " multipath combiner " design can be with reference to following detection method;

(3) signal and the channel estimation value of sending from " multipath combiner " sent into " disturbing zero " unit simultaneously, and the value according to a preliminary estimate of a layer data is only exported in " disturbing zero " unit at every turn $\widehat{d_i}(i=1\mathrm{\ΛM}),$ Method of operation can be with reference to following detection method more specifically;

(4) value according to a preliminary estimate of a layer data of " interference is to zero " unit output $\widehat{d_i}(i=1\mathrm{\ΛM}),$ Be admitted to " sign indicating number matched filter ", use traditional sign indicating number matched filtering method to obtain the estimated value of each code channel in the layer data;

After the estimated value of all code channels obtains in (5) one layer data, be admitted to " disturbance reconstruction " unit together with channel estimation value, the processing basically identical of processing that the estimated value of one layer data experiences in " disturbance reconstruction " unit and a layer data process shown in Figure 3, and use channel estimation value virtual space channel to obtain the reconstruction signal of this layer data;

The reconstruction signal of (6) one layer data and channel estimation value are delivered to " disturbing zero " unit once more, " disturb " signal the unit at first cuts reconstruction from receive data r after to zero, carry out the operation in (3) and (4) link then, obtain the estimated value of next layer data;

(7) carry out the operation of (3-6) repeatedly, up to the estimated value that obtains all layer data;

(8) estimated value of all layer data is admitted to " multichannel combiner " and carries out multichannel and merge, and obtains the estimated value  of all data in the time slot.

The received signal model

Send because the TD-SCDMA system carries out data with burst structure (burst), promptly the data of the transmission in time slot are made of former and later two data divisions and training sequence (midamble) placed in the middle.Therefore need to receive a data part and training sequence placed in the middle at least and just can carry out the detection of data, this makes that also the Data Detection of TD-SCDMA is to carry out with the form of piece, and carries out continuous Data Detection unlike other Frequency Division Duplexing (FDD) cdma systems or V-BLAST system with the form of chip (chip) or symbol (symbol).

According to structure shown in Figure 3 and associated transport scheme, the signal d that goes out from every antenna transmission of transmitter as can be seen _i(i=1 ... M) be the stack of many code channels spread spectrum data, and the data difference on every antenna.Because carry out transfer of data with burst structure, and chip (chip) length of supposing user data in this burst structure is the L (length that does not promptly comprise training sequence, symbol shown in Fig. 3 and Fig. 6 such as d and r etc. all do not comprise training sequence for convenience, training sequence is considered to follow them and transmits together, and enter corresponding channel estimating unit, do not enter detecting unit), sending burst d so can be expressed as follows:

$d={\left[\begin{array}{ccccccccccccc}{d}_1^{\left(1\right)}& d_1^{\left(2\right)}& \mathrm{\Λ}& d_1^{\left(M\right)}& d_2^{\left(1\right)}& d_2^{\left(2\right)}& \mathrm{\Λ}& d_2^{\left(M\right)}& \mathrm{\Λ}& d_L^{\left(1\right)}& d_L^{\left(2\right)}& \mathrm{\Λ}& d_L^{\left(M\right)}\end{array}\right]}^T---\left(1\right)$

D wherein _l ^(m)Represent the 1st chip on the m root antenna.Subscript ^TThe matrix transpose operation of representing matrix or vector.

Consider that space channel is the situation of frequency-selective channel, all can produce the multidiameter delay expansion at each chip of receiver end.Suppose the time delay extension width that receiver can be distinguished, promptly the window of channel estimating is long is W, and the received signal sequence r that receives of receiver can be expressed as so:

$r={\left[\begin{array}{ccccccccccccc}{r}_1^{\left(1\right)}& r_1^{\left(2\right)}& \mathrm{\Λ}& r_1^{\left(P\right)}& r_2^{\left(1\right)}& r_2^{\left(2\right)}& \mathrm{\Λ}& r_2^{\left(P\right)}& \mathrm{\Λ}& r_{L+W-1}^{\left(1\right)}& r_{L+W-1}^{\left(2\right)}& \mathrm{\Λ}& r_{L+W-1}^{\left(P\right)}\end{array}\right]}^T---\left(2\right)$

R wherein _l ^(p)Not only comprise the signal of this corresponding with it chip, comprise time delay spread signal equally from adjacent chip.

For a chip, its transmission signal and received signal can be represented with following relation, are example with first chip:

$\left[\begin{array}{c}{r}_1^{\′\left(1\right)}\\ r_1^{\′\left(2\right)}\\ M\\ r_1^{\′\left(P\right)}\\ r_2^{\′\left(1\right)}\\ r_2^{\′\left(2\right)}\\ M\\ r_2^{\′\left(P\right)}\\ M\\ r_W^{\′\left(1\right)}\\ r_W^{\′\left(2\right)}\\ M\\ r_W^{\′\left(P\right)}\end{array}\right]=\left[\begin{array}{cccc}{h}_1^{\left(\mathrm{1,1}\right)}& h_1^{\left(\mathrm{1,2}\right)}& \mathrm{\Λ}& h_1^{(1,M)}\\ h_1^{\left(\mathrm{2,1}\right)}& h_1^{\left(\mathrm{2,2}\right)}& \mathrm{\Λ}& h_1^{(2,M)}\\ M& M& O& M\\ h_1^{(P,1)}& h_1^{(P,2)}& \mathrm{\Λ}& h_1^{(P,M)}\\ h_2^{\left(\mathrm{1,1}\right)}& h_2^{\left(\mathrm{1,2}\right)}& \mathrm{\Λ}& h_2^{(1,M)}\\ h_2^{\left(\mathrm{2,1}\right)}& h_2^{\left(\mathrm{2,2}\right)}& \mathrm{\Λ}& h_2^{(2,M)}\\ M& M& O& M\\ h_2^{(P,1)}& h_2^{(P,2)}& \mathrm{\Λ}& h_2^{(P,M)}\\ M& M& M& M\\ h_W^{\left(\mathrm{1,1}\right)}& h_W^{\left(\mathrm{1,2}\right)}& \mathrm{\Λ}& h_W^{(1,M)}\\ h_W^{\left(\mathrm{2,1}\right)}& h_W^{\left(\mathrm{2,2}\right)}& \mathrm{\Λ}& h_W^{(2,M)}\\ M& M& O& M\\ h_W^{(P,1)}& h_W^{(P,2)}& \mathrm{\Λ}& h_W^{(P,M)}\end{array}\right]\left[\begin{array}{c}{d}_1^{\left(1\right)}\\ d_1^{\left(2\right)}\\ M\\ d_1^{\left(M\right)}\end{array}\right]+\left[\begin{array}{c}{n}_1^{\′\left(1\right)}\\ n_1^{\′\left(2\right)}\\ M\\ n_1^{\′\left(P\right)}\\ n_2^{\′\left(1\right)}\\ n_2^{\′\left(2\right)}\\ M\\ n_2^{\′\left(P\right)}\\ M\\ n_W^{\′\left(1\right)}\\ n_W^{\′\left(2\right)}\\ M\\ n_W^{\′\left(P\right)}\end{array}\right]---\left(3\right)$

R wherein _w ^{L (p)}Be w the chip signal that receives on the p root antenna, h _w ^{(p, m)}W from the m transmit antennas to p root reception antenna of expression receives the channel impulse response that chip lives through, n _w ^{L (p)}Represent on the p root antenna noise that receives in w the chip.

Think that in testing process space channel is a quasi-static channel opens, promptly in the time that a burst structure continues, remain unchanged, and use the training sequence in this burst structure to carry out channel estimating, draw the channel impulse response of all chips in the burst structure, channel matrix promptly in the formula, and use symbol H _cExpression, dimension is PW * M.It can also be seen that a chip through after the space channel from top formula, time delay expands to W chip, and L sends chip through space channel so, and expands to L+W-1 chip, i.e. received signal sequence r after the mutual superposition.So if represent the space channel of chip process in the burst structure with H, the relation that sends so between burst d and the received signal sequence r can be expressed as follows:

r＝Hd+n (4)

Wherein n is the noise vector that receives.The channel impulse response matrix H is a matrix in block form, and it is by ML H _cMatrix constitutes, and dimension is P (L+W-1) * ML.Its structure as shown in Figure 7, wherein except H _cElement beyond the matrix is 0.

Detection method

The present invention utilizes the multipath combiner that the frequency-selective channel equivalence is flat channel, and uses the V-BLAST detection method to carry out the detection of multi-layer data, and every layer data re-uses yard matched filtering and carries out Multiuser Detection.And utilize the block characteristic of channel impulse response matrix H to simplify testing process, the complexity and the number of plies that make multi-layer data detect are linear.

1. selectivity merges device

It is the situation that selectivity merges device that the multipath combiner at first is discussed now.According to formula (3), between transmitting antenna to a reception antenna, exist the W paths, it is exactly to select a minimum footpath of bars decline to replace all W paths among this W paths that selectivity merges device, and this footpath is called as optimum footpath.A kind of optimum footpath system of selection is that the transmission data on all antennas are regarded as an integral body, selects it to arrive the minimum path of average decline on a certain reception antenna, as the optimal path of this root reception antenna, promptly selects H _cThe path of delegation's correspondence of two norm maximums is as the optimum footpath in the matrix.The optimum footpath of supposing p root reception antenna correspondence is t ^(p), so:

$t^{\left(p\right)}=\arg \underset{t=1\mathrm{\ΛW}}{\max }{\left|\right|{\left(H_c\right)}_{p+(t-1)P}\left|\right|}^2,(p=1\mathrm{\ΛP})---\left(5\right)$

Wherein () _iThe i that represents this matrix is capable, ‖ ‖ ²2 norms of expression vector.

After the decision of optimum footpath, channel matrix is reconstructed into F _c, it is made of the corresponding channel impulse response in optimum footpath:

${\left(F_c\right)}_p={\left(H_c\right)}_{p+(i^{\left(p\right)}-1)P},(p=1\mathrm{\ΛP})---\left(6\right)$

And, be arranged in the matrix in block form form, and represent this matrix with F according to the mode of Fig. 7.Because F _cDimension be P * M, the dimension of F is PL * ML so, and matrix F is a block diagonal matrix.

After channel matrix was reconstructed, corresponding received signal sequence also needed to be reconstructed, if arrange the burst on the p root reception antenna as follows:

$r^{\left(p\right)}=\left[\begin{array}{c}{r}_1^{\left(p\right)}\\ r_2^{\left(p\right)}\\ M\\ r_{L+W-1}^{\left(p\right)}\end{array}\right]---\left(7\right)$

Received signal sequence x on the p root reception antenna after the reconstruct so ^(p)By r ^(p)T (p) row to the capable formation of t (p)+L-1.After received signal sequence reconstruct on all reception antennas finishes, can be according to following structure, the received signal sequence x that reconstruct is total:

$x={\left[\begin{array}{ccccccccccccc}{x}_1^{\left(1\right)}& x_1^{\left(2\right)}& \mathrm{\Λ}& x_1^{\left(P\right)}& x_2^{\left(1\right)}& x_2^{\left(2\right)}& \mathrm{\Λ}& x_2^{\left(P\right)}& \mathrm{\Λ}& x_L^{\left(1\right)}& x_L^{\left(2\right)}& \mathrm{\Λ}& x_L^{\left(P\right)}\end{array}\right]}^T---\left(8\right)$

Formula (4) just can be rewritten as following form like this:

$x=\left[\begin{array}{c}{x}_1\\ x_2\\ M\\ x_L\end{array}\right]=\left[\begin{array}{cccc}{\mathrm{<figure-callout\; id="0"\; label="F\; c"\; filenames="C200310103385.600331.png,C200310103385.600341.png"\; state="\{\{state\}\}">F</figure-callout>}}_c& 0& \mathrm{\&Lambda;}& 0\\ 0& F_c& \mathrm{\&Lambda;}& 0\\ M& \mathrm{<figure-callout\; id="0"\; label="M\; O\; M"\; filenames="C200310103385.600331.png,C200310103385.600341.png"\; state="\{\{state\}\}">M</figure-callout>}& O& M\\ 0& 0& \mathrm{\&Lambda;}& F_c\end{array}\right]\left[\begin{array}{c}{d}_1\\ d_2\\ M\\ d_L\end{array}\right]+\left[\begin{array}{c}{v}_1\\ v_2\\ M\\ v_L\end{array}\right]=\mathrm{Fd}+v---\left(9\right)$

D wherein _lThe 1st the chip signal sequence that expression sends, x _lThe 1st the chip signal sequence that expression receives, v _lThe 1st noise sequence that chip signal sequence the inside comprises that expression receives.

From formula (9) as can be seen, the signal of any one chip transmission all with typical V-BLAST structural similarity, so all can use the V-BLAST detection method to detect, but in the data that estimate arbitrary layer $\widehat{d_m}(m=1...M)$ Afterwards, need to increase the sign indicating number matched filter and carry out Multiuser Detection, reenter interference cancellation unit then.The estimated value  of this user's initial data finally can multichannel be merged in the output of several layers multi-user detector.And, since the piece diagonal structure of channel matrix F, the channel matrix F of all chips _cAll identical, so in fact only need carry out V-BLAST and detect correlation computations, and result of calculation is applied on other chips gets final product, thereby save amount of calculation greatly a chip.

2. Rake combiner when empty

Except use selectivity merging device carried out the multipath merging, the Rake combiner carried out the multipath merging in the time of can also selecting to use sky.The Rake combiner was promptly made related operation to channel H when what is called was empty on received signal sequence r, made the time delay spread signal that the received signal sequence is L+W-1 by a length become the equivalent flat channel received signal x that a length is L.Promptly multiply by the conjugate transpose of channel matrix H specifically respectively on formula (4) both sides:

x＝H ^Hr＝H ^HHd+H ^Hn (10)

Wherein () ^HThe conjugate transpose computing of representing matrix.Make R=H ^HH is the cross-correlation matrix of space channel, v=H ^HN is the noise sequence of equivalence, and formula (10) can be write as following form so:

x＝Rd+v (11)

Promptly similar with formula (9), but wherein the R matrix is a diagonally dominant matrix of piece, and be a piece diagonal matrix unlike the F matrix.In order to simplify calculating, ignore the element beyond the piece diagonal in the R matrix here, and the new matrix that order generates is F, by the block matrix character of H matrix, can obtain so:

$F=\left[\begin{array}{cccc}{\mathrm{<figure-callout\; id="0"\; label="F\; c"\; filenames="C200310103385.600331.png,C200310103385.600341.png"\; state="\{\{state\}\}">F</figure-callout>}}_c& 0& \mathrm{\&Lambda;}& 0\\ 0& F_c& \mathrm{\&Lambda;}& 0\\ M& \mathrm{<figure-callout\; id="0"\; label="M\; O\; M"\; filenames="C200310103385.600331.png,C200310103385.600341.png"\; state="\{\{state\}\}">M</figure-callout>}& O& M\\ 0& 0& \mathrm{\&Lambda;}& F_c\end{array}\right]---\left(12\right)$

F _c＝H _c ^HH _c (13)

Use the R matrix in the F matrix replacement formula (11), the detection of the data that the same procedure that just can use " selectivity merging device " part to introduce is simplified.

Fig. 8 and Fig. 9 have further provided corresponding simulation result, and what channel model adopted is link level simulation model 7.0 versions that space channel model tissue (SCM AHG) is issued, and what select for use is the vehicle-mounted model of case II wherein, and that translational speed is chosen is 40km/h.What wherein used the base station is the uniform straight line array row, and antenna distance is 4 wavelength, and what terminal was used also is the uniform straight line array row, and antenna distance is a half-wavelength.

It is M=P=4 that Fig. 8 has provided antenna number, the relation curve of the error rate under the frequency-selective channel situation (BER) and signal to noise ratio (Eb/No).This curve is the simulation result under single user's downlink transfer and 8 CDMA code channels of the every CU situation.Wherein, SC represents that selectivity merges, and Rake merged when STRC represented sky, and ZF-V-BLAST represents to use the V-BLAST detector of Zero Forcing, and MMSE-V-BLAST represents to use the V-BLAST detector of minimum mean square error criterion.

It is M=P=8 that Fig. 9 has provided antenna number, the relation curve of the error rate under the frequency-selective channel situation (BER) and signal to noise ratio (Eb/No).This curve is the simulation result under single user's downlink transfer and 8 CDMA code channels of the every CU situation.Wherein, SC represents that selectivity merges, and Rake merged when STRC represented sky, and ZF-V-BLAST represents to use the V-BLAST detector of Zero Forcing, and MMSE-V-BLAST represents to use the V-BLAST detector of minimum mean square error criterion.

As can be seen, use device of the present invention and method, can realize combining of time division duplex cdma system and mimo system, improve message transmission rate.

Though described the present invention by embodiment, those of ordinary skills know, the present invention has many distortion and variation and do not break away from spirit of the present invention, wish that appended claim comprises these distortion and variation and do not break away from spirit of the present invention.

Claims (18)

1. the high-speed data transmission device of the many antennas under the TDD mode comprises:

The multichannel decomposer, the data flow that is used for distributing to all burst structure of time slot is divided at least two layer data, and described every layer data comprises at least one height stream;

Spread spectrum corresponding with the number of plies of described data and modulation scrambling apparatus are respectively applied for each son stream spread spectrum and modulation scrambling with described at least two layer data;

The bursty traffic corresponding with the number of plies of described data forms device, is used for inserting the substream of data that training sequence forms burst structure at each son stream of described spread spectrum and modulation scrambling;

The merging device corresponding with the number of plies of described data is used for the substream of data that described each bursty traffic forms all burst structure that device forms merged into and transmits;

The signal transmitting antenna corresponding with the number of plies of described data is used to launch transmitting of described modulation.

2. emitter as claimed in claim 1, wherein, described spread spectrum comprises spread spectrum device, modulation sub-device and the scrambling sub-device identical with the son stream number of described every layer data with the modulation scrambling apparatus, is used for respectively with each son stream spread spectrum, modulation and scrambling.

3. emitter as claimed in claim 2, wherein, the sub-device of described spread spectrum uses different channel codes for each the son stream with layer, and the sub-flow point of different layers does not use different channel codes, or uses identical channelizing code character to carry out multiplexing; And/or the sub-device of described scrambling uses identical scrambler or different scramblers to the identical scrambler of same layer data use to the different layers data.

4. emitter as claimed in claim 1, wherein, described bursty traffic forms device and comprises: the basic training sequences generator is used to produce basic training sequences; Layer concentrator marker is used for by training sequence label layer information; Inserter is used for inserting identical training sequence at each son stream of described every layer data; Wherein, the training sequence difference of different layers, and by described basic training sequences skew generation.

5. emitter as claimed in claim 4, wherein, described layer concentrator marker comprises deviator, be used to be offset described basic training sequences and produce training sequence with deviant presentation layer by training sequence, wherein, the relation of all right presentation layer of deviant and scrambler, and can also be used to carry channelization code information.

6. receiving system that is used to receive the signal of emitter as claimed in claim 1 comprises:

At least equal a plurality of reception antennas of the number of plies of described data, be used to receive the signal of the transmission antennas transmit of the described number of plies that equals described data;

Channel estimating apparatus is used for a plurality of signals of the described reception of channel estimating, obtains the impulse response on the different paths;

The multipath combiner is used for carrying out according to the impulse response of described acquisition the merging of multipath data, and making it equivalence is the reception data of a flat channel, produces multilayer and receives data, and produce corresponding with it equivalent channel impulse response matrix;

Signal supervisory instrument is used to detect every layer and receives data, obtains the estimated value of all layer data;

Interference is used for the equivalent channel impulse response matrix according to described acquisition to zero device, and the every layer of interference that described multilayer is received data be to zero, suppressing the interference of other layer data, and obtains the data of optimum layer by the method for ordering;

Matched filtering and judgment device are used for the described optimum layer data of matched filtering and judgement, obtain the estimated value of this layer data, and wherein channel code can be used to distinguish different users, also can combine the different user of common differentiation with layer sequence number;

The disturbance reconstruction device is used for rebuilding the pairing reception data of this layer according to the estimated value of this layer data of described acquisition;

Judgment means is used for the information according to described received signal, has judged whether to obtain the estimated value of all layer data in the time slot;

The multichannel combiner is used to merge all detected layer data, to obtain the estimated value of all layer data in the time slot.

7. receiving system as claimed in claim 6, wherein, the reception data of multipath combiner output satisfy predetermined arrangement, make the channel impulse response matrix of equivalence satisfy the characteristic of piece diagonal matrix, wherein each the height piece on the diagonal is corresponding to the channel impulse response matrix of a chip, and all sub-pieces are all identical; To disturb to zero the time, only the equivalent channel impulse response to a chip disturbs zero, and the result is applied on all chips in the same time slot that receives.

8. receiving system as claimed in claim 6, wherein, described interference also comprises collator to zero device, and being used for selecting disturbing at least one the layer data to satisfying after zero in wanted to interfering signal ratio SINR maximum and the error minimal condition is the described data that obtain optimum layer.

9. receiving system as claimed in claim 6, wherein, described interference also comprises interference blanking unit to zero device, is used for receiving the influence that data are eliminated the data of the described layer that has obtained estimated value in multilayer, so that carry out the detection of next layer data.

10. receiving system as claimed in claim 6, wherein, Rake combiner and selectivity merged device when described multipath combiner comprised sky.

11. the launching technique of the high-speed data of the many antennas under the TDD mode comprises step:

The data flow of decomposing all burst structure in the time slot is at least one layer data, and described at least one layer data comprises at least one height stream of decomposition;

Identify every layer data;

Described each the son stream of spread spectrum, modulation and scrambling;

In described each son stream, add training sequence respectively and form burst structure;

A plurality of son streams of every layer data are all superposeed formation multilayer emission data by a plurality of transmission antennas transmit.

12. launching technique as claimed in claim 11, wherein, with a plurality of son streams of every layer data all the stack step that forms the emission data transmission also comprise: whether each son stream of judging this layer data all forms corresponding burst structure, if all do not form, continue to form, if all form, the burst structure that stack all forms.

13. launching technique as claimed in claim 11, wherein, the step of described each the son stream of described spread spectrum, modulation and scrambling comprises: with the different son stream of different channel code spread spectrums with layer, the sub-flow point of different layers does not use different channel codes, or uses identical channelizing code character to carry out multiplexing; And/or, the different layers data are used identical scrambler or different scramblers to the identical scrambler of same layer data use.

14. launching technique as claimed in claim 11, the step of the every layer data of described sign comprises: the basic training sequences different to the data-bias of different layers, and using the deviant presentation layer, and/or the relation of layer and scrambler, and/or carry channelization code information.

15. launching technique as claimed in claim 14, wherein, the training sequence deviation of different layers satisfies one of following condition at least with the relation of layer:

Deviant satisfies the predetermined mapped relation with layer, so that receiver end is known the order of layer by these mapping relations, thereby merges multi-layer data accurately;

At least distribute two different scramblers being applied to different layers, and when the selection that requires scrambler needs to select in a scrambler group of determining, the sequence number of deviant and scrambler satisfies the second predetermined mapping relations, so that receiver end is known the scrambling code information of different layers by these mapping relations;

Requiring channel code to need to select in a channelizing code character of determining and clocklike divide timing, the quantity of deviant and channel code satisfies the 3rd preset mapping relation, so that receiver end is known channelization code information by mapping relations.

16. a method of reseptance that is used to receive the signal of launching technique emission as claimed in claim 11 comprises step:

Antenna with the layer that is no less than the emission data receives the signal of described emission, produces a plurality of received signals;

A plurality of signals of the described reception of channel estimating obtain the impulse response on the different paths;

Carry out the merging of multipath data according to the impulse response of described acquisition, making it equivalence is the reception data of a flat channel, produces multilayer and receives data, and produce corresponding with it equivalent channel impulse response matrix;

According to the equivalent channel impulse response matrix of described acquisition, the every layer of interference that described multilayer is received data be to zero, suppressing the interference of other layer data, and obtains the data of optimum layer by the method for ordering;

The data of matched filtering and the described optimum layer of judgement obtain the estimated value of this layer data, and wherein channel code can be used to distinguish different users, also can combine the different user of common differentiation with layer sequence number;

To the signal disturbance reconstruction of described reception, be used for rebuilding the pairing reception data of this layer according to the estimated value of this layer data of described acquisition;

Repeat above-mentioned steps, the estimated value of all layer data in obtaining a time slot;

Multichannel merges all detected layer data, to obtain the estimated value of all layer data in the time slot.

17. method of reseptance as claimed in claim 16, wherein, the reception data that multipath merges output satisfy predetermined arrangement, make the channel impulse response matrix of equivalence satisfy the characteristic of piece diagonal matrix, wherein each the height piece on the diagonal is corresponding to the channel impulse response matrix of a chip, and all sub-pieces are all identical; With described interference to zero step in, only the equivalent channel impulse response to a chip disturbs zero, and the result is applied on all chips in the same time slot that receives.

18. method of reseptance as claimed in claim 16, wherein,

Described interference comprises zero step: selecting to disturb at least one the layer data to satisfying after zero in wanted to interfering signal ratio SINR maximum and the error minimal condition is the described data that obtain optimum layer;

Described interference is in zero the step, is that the equivalent channel impulse response with a chip disturbs Z-operation, and the result is applied to all chips in the same time slot that receives;

Described interference comprises zero step: use reconstruction signal to carry out interference cancellation operation, so that carry out the detection of next layer data;

Described step of carrying out the merging of multipath data comprises: the Rake combiner merges described multipath data and utilizes selectivity merging device to merge described multipath data and merge described multipath data with different multipath merging methods when utilizing sky, produce multilayer and receive data, and produce corresponding with it equivalent channel impulse response matrix.

Images