Please refer to the following drawings for further understanding of the present invention.
Embodiments of the present invention realize accurate OFDM downlink frame synchronization and highly precise and large range carrier frequency offset estimation.
A first embodiment of the present invention provides a training sequence by joint the structure characteristics of the two training sequences with central symmetry structure and repeated data blocks structure respectively, which can realize accurate timing synchronization and highly precise and large range carrier frequency offset estimation.
A second embodiment of the present invention provides a frame structure based on the training sequence provided by the first embodiment of the present invention.
A third embodiment of the present invention comprises an adaptive method for OFDM downlink joint frame synchronization and carrier frequence offset estimation, which can realize accurate OFDM downlink frame synchronization and highly precise and large range carrier frequency offset estimation.
A fourth embodiment of the present invention comprises an adaptive communication system for OFDM downlink joint frame synchronization and carrier frequence offset estimation, which can realize accurate OFDM downlink frame synchronization and highly precise and large range carrier frequency offset estimation.
According to the first embodiment of the present invention, a training sequence generating method is provided. The training sequence includes a first training symbol and a second training symbol with equal-length, but without a cyclic prefix (CP), which is characterized in that, generating the first training symbol randomly according to the method for generating normal data symbols; subdividing the generated first training symbol logically into M sub-blocks with equal-length, wherein the structure characteristic M is a natural number larger than or equal to 1 and less than or equal to N; and copying the M sub-blocks in an reverse order to form the second training symbol, which together with the first training symbol constitute the training sequence.
According to the second embodiment of the present invention, a first-type OFDM frame is provided, wherein the first-type OFDM frame includes two first-type training sequences and data symbols, wherein the structure characteristic M of the first-type training sequence is equal to N.
According to the second embodiment of the present invention, a second-type OFDM frame is provided, wherein the second-type OFDM frame includes a first-type training sequence, a second-type training sequence and data symbols, wherein the structure characteristic M of the first-type training sequence is equal to N, the structure characteristic M of the second-type training sequence is a natural number larger than or equal to 1 and less than N.
According to the third embodiment of the present invention, a communication method is provided, which uses the first-type OFDM frame and the second-type OFDM frame, the first and second-type OFDM frames utilizing the training sequence generated by the method described above, the first-type OFDM frame including two first-type training sequences and data symbols, wherein the structure characteristic M of the first-type training sequence is equal to N, and the second-type OFDM frame includes a first-type training sequence, a second-type training sequence and data symbols, wherein the structure characteristic M of the first-type training sequence is equal to N, the structure characteristic M of the second-type training sequence is a natural number larger than or equal to 1 and less than N, the communication method includes the following operations,
According to the third embodiment of the present invention, a communication method is provided, which uses the first-type OFDM frame and the second-type OFDM frame, the first and second-type OFDM frames utilizing the training sequence generated by the method described above, the first-type OFDM frame including two first-type training sequences and data symbols, wherein the structure characteristic M of the first-type training sequence is equal to N, and the second-type OFDM frame includes a first-type training sequence, a second-type training sequence and data symbols, wherein the structure characteristic M of the first-type training sequence is equal to N, the structure characteristic M of the second-type training sequence is a natural number larger than or equal to 1 and less than N, wherein the communication method includes the following operations:
According to the fourth embodiment of the present invention, a communication system is provided, which is configured to use the first-type OFDM frame and the second-type OFDM frame, the first-type and the second-type OFDM frame utilizing the training sequence generated by the method described above, the first-type OFDM frame including two first-type training sequences and data symbols, wherein the structure characteristic M of the first-type training sequence is equal to N, and the second-type OFDM frame includes a first-type training sequence, a second-type training sequence and data symbols, wherein the structure characteristic M of the first-type training sequence is equal to N, the structure characteristic M of the second-type training sequence is a natural number larger than or equal to 1 and less than N, the communication system including a base station including a transmitter, which communicates with the mobile terminal through the wireless channel based on the first-type and second-type OFDM frames, wherein the first frame transmitted by the base station is the first-type OFDM frame and the subsequent frames are all the second-type OFDM frames; and a mobile terminal including a receiver, which performs the initial acquisition and adaptive tracking sequentially in order to perform timing synchronization and carrier frequency offset estimation for each frame according to the received OFDM frames, i.e. the first-type OFDM frame or the second-type OFDM frames.
Advantages of embodiments of the present invention includes, but are not limited to: realizing joint frame synchronization and carrier frequency offset estimation; the precision of frame synchronization is far higher than that of the traditional algorithms; the acquisition range of the carrier frequency offset is large and can reach at most a half of the whole transmission bandwidth; the parameter M can be adjusted adaptively with the change of the wireless channel; the precision of carrier frequency offset estimation is higher than that of the traditional algorithms; and the calculation complexity is reduced while the estimation precision is enhanced.
Embodiments of the present invention aim to realize accurate timing synchronization and highly precise and large range carrier frequency offset estimation, by joint the structure characteristics of the two training sequences with central symmetry structure and repeat data blocks structure respectively. Embodiments of the present invention provide a new training sequence and an adaptive communication system and communication method for OFDM downlink joint frame synchronization and carrier frequence offset estimation on the basis of the new training sequence. The joint frame synchronization and carrier frequence offset estimation can be realized in the communication system; the precision of frame synchronization is far higher than that of the traditional algorithms; the acquisition range of the carrier frequency offset is large and can reach at most a half of the whole transmission bandwidth; the parameter M can be adjusted adaptively with the change of the wireless channel; the precision of carrier frequency offset estimation is higher than that of the traditional algorithms; the calculation complexity is reduced while the estimation precision is enhanced.
The communication system and method of embodiments of the present invention are both realized on the basis of the training sequence provided herein. Before illustrating the communication system, the structure characteristic of the training sequence will be described first.
Next the structure characteristic of the training sequence will be illustrated with reference to examples. For example, when the first training symbol is {1, 2, 3, 4} and M is 2, the first training symbol is subdivided into two sub-blocks, i.e. {[1, 2], [3, 4]}. Then the sub-blocks are copied in reverse order to form the second training symbol {[3, 4], [1, 2]}. The first and the second training symbols together constitute the training sequence of one embodiment of the present invention {1, 2, 3, 4, 3, 4, 1, 2}.
An example of the normal form of the current training sequence is {x(0), x(1), . . . , x(N−1), x(0), x(1), . . . , x(N−1)}. Here {1, 2, 3, 4, 1, 2, 3, 4} is taken as an example. Suppose the correlation distance of sample “1” is 4 (i.e., the distance between the two sample “1”), the sum of the square of the correlation distance group of the current training sequence is 4×42=64. On the other hand, the correlation distance group of the above training sequence {1, 2, 3, 4, 3, 4, 1, 2} is 2×62+2×22=80. Since when the training sequence is adopted in downlink synchronization, the precision is proportional to the sum of the square of the correlation distance group. Therefore, the training sequence of the present invention can realize highly precise downlink synchronization as 80>64.
Based on the above, the training sequence can realize accurate timing synchronization and highly precise and large range carrier frequency offset estimation, by joint the structure characteristics of the two training sequences with central symmetry structure and a repeat data blocks structure respectively.
Based on the above-mentioned training sequence, an OFDM frame (the first-type OFDM frame F1 and the second-type OFDM frame F2) as shown in
The structure characteristic of the first-type training sequence S1 of the OFDM frame is M=N, i.e., the data of the second training symbol in the first-type training sequence S1 is the reverse repetition of the samples in the first training symbol and thus the central symmetry training sequence is created. Since the central symmetry training sequence can realize timing synchronization with high precision, it is used for the initial acquisition.
The structure characteristic M of the second-type training sequence S2 is the optimal value (1≦M<N) determined by the maximum multipath channel delay obtained in the initial acquisition. Since the central symmetry training sequence has low precision in the carrier frequency offset estimation in the multipath channel environment while block training sequence with repeated data-block has high precision in the carrier frequency offset estimation in the multipath channel environment, the second-type training sequence S2 can be used to finish the rest carrier frequency offset estimation in the adaptive tracking after the initial acquisition.
Next the communication system and method using the training sequence and OFDM frame of the present invention will be illustrated in detail.
As shown in
The first frame transmitted by the mobile terminal 1 in the base station 2 is the first-type OFDM frame F1 and the subsequent frames are all the second-type OFDM frames F2.
In the downlink transmission in the OFDM system, actually, when a mobile terminal 1 begins to access the communication system, the initial acquisition is implemented for every frame to perform timing synchronization and carrier frequency offset estimation. After the initial acquisition, the mobile terminal 1 is required to implement carrier frequency offset tracking in order to finish the carrier frequency offset estimation and to realize accurate timing synchronization and highly precise and large range carrier frequency offset estimation.
The mobile terminal 1 performs adaptive tracking for every newly arrived frame. If the channel characteristic changes slowly, the initial acquisition can be performed every a few frames in order to readjust the timing offset and frequency offset of the terminal user. If the channel characteristic changes fast, the performing frequency of the initial acquisition can be higher. In the fast speed mobile system, the mobile terminal 1 can perform the initial acquisition and adaptive tracking for every frame.
However, no matter the channel characteristic changes slowly or fast, for the unification of communication specifications, the terminal 1 performs the initial acquisition and adaptive tracking for all the OFDM frames from the base station 2 in order to implement accurate timing synchronization and carrier frequency offset estimation. The detailed process is as the following:
It can be seen from the above that the OFDM system of the present invention utilizes the first-type training sequence S1 (the structure characteristic is M=N, i.e., the central symmetry training sequence) to implement the initial acquisition and to realize accurate timing synchronization and the initial carrier frequency offset estimation and large range carrier frequency offset acquisition; the adaptive tracking is performed by utilizing the second second-type training sequence of the first frame from the base station or utilizing the second-type training sequence S2 (the training sequence with repeat data blocks and M is adjusted adaptively according to the change of the wireless channel characteristic) of the frames following the first frame, wherein according to the optimal value M for the next frame, which is obtained from the maximum multipath channel delay in the initial acquisition and fed back to the base station, the highly precise carrier frequency offset estimation in the multipath channel environment is further performed and the carrier frequency offset compensation can be realized.
It should be noted that the OFDM frame of the present invention can only include the second-type training sequence S2 and data symbols, while in the present invention, only the first-type OFDM frame F1 and the second-type OFDM frame F2 with two training sequences and data symbols are taken into consideration in the communication system.
As shown in
The data streams are first input into the data modulating section 30 in the transmitter 3 for data modulation and then the bit streams are mapped into the specific constellation map in order to obtain the modulation symbols constituting the training sequence or the data symbols.
The control unit 31 is used to control the first or the second-type OFDM frames generated by the training sequence generating section 32 and the data symbol generating section 33.
The training sequence generating section 32 includes an M value determining unit 321, a serial/parallel conversion unit 322, a frequency domain first training symbol generating unit 323, an IFFT unit 324, a logic subdividing unit 325 and a second training symbol generating unit 326.
The M value determining unit 321 first determines the structure characteristic M value (1≦M≦N) of the training sequence. For example, when the first-type OFDM frame F1 is generated under the control of the control unit 31, the M value determining unit 321 specifies that the structure characteristic of the two first-type training sequences S1 included in the first-type OFDM frame F1 is M=N; when the second-type OFDM frame F2 is generated, the M value determining unit 321 specifies that the structure characteristic of the first-type training sequence S1 is M=N and specifies the structure characteristic of the second-type training sequence S2 according to the structure characteristic optimal M fed back by the mobile terminal.
The serial/parallel conversion unit 322 converts the modulated symbols from the data modulating section 30 into parallel data. The frequency domain first training symbol generating unit 323 generates the frequency domain first training symbol with the method for generating normal data symbols according to the output of the serial/parallel conversion unit 322. Then, the IFFT unit 324 obtains the time domain first training symbol by implementing IFFT on the frequency domain first training symbol.
The logic subdividing unit 325 subdivides logically the first training symbol generated by the IFFT unit 324 into M sub-blocks with equal-length according to the M value determined by the M value determining unit 321 (1≦M≦N) and M is a natural number). The second training symbol generating unit 326 copies the M sub-blocks in reverse order to form the second training symbol. Thus the first training symbol and the second training symbol together constitute the first-type training sequence S1 or the second-type training sequence S2 of the present invention.
The modulated symbol of the data modulating section 30 is input into the data symbol generating section 33 under the control of the control unit 31, which generates the data symbols of the OFDM frames (F1 or F2) of the present invention.
The data symbol generating section 33 includes a serial/parallel conversion unit 331 and an IFFT unit 332. The serial/parallel conversion unit 331 converts the modulated symbols from the data modulating section 30 into parallel data. The IFFT unit 332 obtains the data symbols by implementing IFFT on the parallel data from the serial/parallel conversion unit 331.
The training sequence and data symbols generated by the training sequence generating section 32 and the data symbol generating section 33 can form the OFDM frame in the format of the first-type OFDM frame F1 or the second-type OFDM frame F2. To generate OFDM frame under the control of the control unit 31, the method for generating the data frame in prior art can be adopted, such that the buffer temporally stores the training sequence and the data symbols respectively to form an OFDM frame, and then the frame is transmitted to the mobile terminal 1 through the wireless channel; or the buffer whose capacity is the length of the training sequence is used and when the buffer is full, the control unit 31 is triggered to control the generating of the data symbols and creates the OFDM frame through the bus, and then the frame is transmitted to the mobile terminal 1 through the wireless channel.
As shown in
The receiver 4 adds the initial carrier frequency offset result obtained by the initial acquisition section 5 and the adaptive tracking result obtained by the adaptive tracking section 6 to realize carrier frequency offset compensation.
The receiver 4 performs the initial acquisition and adaptive tracking for every newly arrived second-type OFDM frame F2 until the end of communication.
Specifically, for every first transmitted first-type OFDM frame F1, the receiver 4 utilizes the first first-type training sequence S1 to implement the joint frame synchronization and carrier frequency offset acquisition and utilizes the second first-type training sequence S1 to implement the adaptive tracking, in order to achieve the carrier frequency offset estimation, wherein the receiver 4 can obtain the maximum multipath channel delay, which will be fed back to the base station to form the second-type OFDM frame F2.
For each OFDM frame F2 subsequently transmitted by the base station, the structure characteristic of the second-type training sequence S2 is determined by the receiver 4 according to the optimal M fed back to the base station based on the previous OFDM frame (the first-type OFDM frame F1 or the second-type OFDM frame F2). The initial acquisition section 5 in the receiver 4 performs the joint frame synchronization and carrier frequency offset acquisition again for every first second-type OFDM frame F2, then the adaptive tracking section 6 performs the carrier frequency offset tracking, in order to achieve the carrier frequency offset estimation.
As shown in
The joint frame synchronization and carrier frequency offset acquisition unit 51 combines the frame synchronization and carrier frequency offset acquisition for the data sequence r(k) received through the wireless channel in order to perform the timing synchronization and the initial carrier frequency offset estimation by using the timing metric Mθ(ε) specific to the first-type training sequence S1 (with the structure characteristic M=N)
wherein the timing metric Mθ(ε) is a function of the timing offset θ and frequency offset ε, N is the length of the training symbol of the training sequence, r(k) is the data sequence received by the mobile terminal and r*(k+θ) is the conjugation of the data sequence r*(k+θ).
The accurate timing offset θ and frequency offset ε can guarantee the timing metric Mθ, (ε) reaches the local peak, i.e., the joint the frame synchronization and carrier frequency offset acquisition of the present invention is to adjust the θ and ε simultaneously to get the local peak of the timing metric Mθ(ε). The process to adjust θ and ε is as the following.
wherein {circumflex over (ε)} represents the frequency offset precompensation value and
represents the frequency offset precompensation operation.
When Mθ(ε) reaches the local peak, i.e., the beginning position of the training sequence and the carrier frequency offset of the system are found, the initial carrier frequency offset is {circumflex over (ε)}A and timing offset is {circumflex over (θ)}.
In multipath environment, every local peak of Mθ(ε) means a path training sequence is detected at the receiver. According to the local peaks of a plurality of timing metrics obtained by the joint frame synchronization and carrier frequency offset acquisition unit 51, the multipath tap detecting unit 52 can calculate the maximum multipath channel delay (for example L samples) of the wireless channel as shown in
The optimal M determining unit 53 calculates the currently optimal M according to the maximum multipath channel delay from the multipath tap detecting unit 52. When the prerequisites that the accuracy of the carrier frequency offset estimation of the training sequence of the present invention is higher than the traditional algorithm (Moose algorithm) is met, the optimal M's range is
the carrier frequency offset estimation accuracy is the highest, wherein └x┘ represents the maximum integer less than or equal to x. And the structure characteristic optimal value M determined by the optimal M determining unit
When the currently optimal M is obtained, the feedback unit 54 feeds back the optimal M to the base station and the initial acquisition section 5 realizes the joint frame synchronization and carrier frequency offset estimation.
It should be noted that for the first-type OFDM frame transmitted by the base station 2 in the communication system of the present invention, the initial carrier frequency offset obtained by the mobile terminal 1 is represented as {circumflex over (ε)}A.
For every OFDM frame transmitted by the base station, since the initial carrier frequency offset acquisition accuracy is not high enough, the adaptive tracking section 6 further performs the carrier frequency offset tracking after the initial acquisition, i.e., the adaptive tracking. The adaptive tracking section 6 performs the carrier frequency offset tracking according to the second first-type training sequence S1 of the first-type OFDM frame F1 or according to the second-type training sequence of the second-type OFDM frame F2.
The adaptive tracking section 6 includes a tracking unit 61 and a carrier frequency offset compensation unit 62. The tracking unit 61 performs the carrier frequency offset tracking with the estimator {circumflex over (ε)}T according to the data sequence r(k) received through the wireless channel. The estimator {circumflex over (ε)}T can realize the function as shown in formula (3).
wherein P is an index with the range from 1 to M,
the M is the structure characteristic optimal M obtained by the mobile terminal in the initial carrier frequency offset acquisition and fed back to the base station.
Since the front L samples of every sub-block of the training sequence received by the mobile terminal 1 are influenced by the multipath channel and thus the interference between sub-blocks exists, they are not used to implement carrier frequency offset tracking. After the carrier frequency offset tracking is achieved by the estimator with formula (3), i.e., after obtaining the {circumflex over (ε)}T value, the receiver 4 calculates the estimated overall carrier frequency offset by adding the initial carrier frequency offset {circumflex over (ε)}A with the adaptive tracking result {circumflex over (ε)}T obtained by the initial acquisition section 5, and feeds back the sum to the carrier frequency offset compensation unit 62 for carrier frequency offset compensation, so that the carrier frequency offset tracking is completed.
2k+1 carrier frequency precompensation values {circumflex over (ε)}, i.e., (−kΔε, −(k−1)Δε, . . . −Δε, 0, Δε, . . . , kΔε) are required for the initial acquisition to implement the carrier frequency precompensation on the stored data sequence. Every carrier frequency precompensation value is used to compensate a data sequence buffered, wherein 1<kΔε<DFTlength/4.
The joint frame synchronization and carrier frequency offset acquisition unit 51 utilizes the carrier frequency offset precompensation operation such as
to calculate the value of the timing metric Mθ(ε) of every sequence after precompensated sample by sample (from the 0th to the (G−1)th sample), and to record the position with the maximum Γ, wherein 1 represents an index of the buffered data in the buffer and its range is 1˜G.
When a carrier frequency offset value for precompensation is closest to the actual carrier frequency offset value, the position where Mθ(ε) reaches the maximum value is highly probable the beginning position of the training sequence. In all the 2k+1 precompensation sequences, if the position where there is the maximum Mθ(ε) value is found, the beginning position ({circumflex over (θ)}) of the training sequence and the carrier frequency offset acquisition result (εacqλ) are also found. Formula (1) and (2) show that the largest range of the carrier frequency offset acquisition of the present invention is (−N/4, N/4), i.e. the largest acquisition range can reach a half of the whole transmission bandwidth, so the carrier frequency offset precompensation value in
wherein N is the DFTlength/4, Δε>0 is the carrier frequency offset precompensation interval. The smaller Δε is, the higher the carrier frequency offset acquisition accuracy is and the calculation complexity during the acquisition process increases accordingly. Δε is normally taken as 0.1 in the present invention. Formula (3) shows the carrier frequency offset tracking range of the present algorithm is
should be met. It can be seen from the above that, compared with the current algorithms, the present invention reduces the calculation complexity while enhancing the estimation precision.
Table 1 is the environment of the two wireless channels (Scenario I and Scenario II) used in performance analysis of the present invention.
The structure of the mobile terminal 1 and the base station 2 is not limited to the above description.
One embodiment of a communication method of the present invention according to the modified embodiment is the same with that of the above embodiments mostly and only the difference will be illustrated here.
In step b), the mobile terminal transmits the maximum multipath channel delay determined by the initial acquisition results to the base station, not the structure characteristic optimal M.
In step c), the base station determines the structure characteristic optimal M and generates the second-type OFDM frame according to the maximum multipath channel delay for the previous frame,
As same in step b), in step d) the mobile terminal transmits the maximum multipath channel delay determined by the initial acquisition result to the base station, not the structure characteristic optimal M.
According to the modified embodiment of the present invention, the base station 2 can calculates the structure characteristic optimal M quickly according to the maximum multipath channel delay.
The above discussion proves the present invention can realize joint frame synchronization and carrier frequency offset estimation; the precision of frame synchronization is far higher than that of the traditional algorithms; the acquisition range of the carrier frequency offset is large and can reach at most a half of the whole transmission bandwidth; the parameter M can be adjusted adaptively with the change of the wireless channel; the precision of carrier frequency offset estimation is higher than that of the traditional algorithms; the calculation complexity is reduced while the estimation precision is enhanced.