Disclosure of Invention
The present invention aims to solve the problems in the prior art, and provides a frame structure for wireless communication and a signal processing method thereof, which ensure the validity of channel state information acquired through pilot frequency, fully utilize channel resources, and reduce the overhead of transmitting pilot frequency and synchronization signals.
In order to solve the technical problems, the invention adopts the technical scheme that:
a frame structure for wireless communication, characterized in that: the frame structure is an uplink frame and comprises a data code element sequence, a pilot frequency code element sequence and a synchronization code element sequence, wherein the data code element sequence is divided into two sections with the lengths of L1 and L2, the pilot frequency code element sequence with the length of LP is inserted between the two sections of data code element sequences, the first section of the synchronization code element sequence with the length of LS1 is inserted between the first section of the data code element sequence and the pilot frequency sequence, the first LS2 code elements of the pilot frequency code element sequence are also the second section of the synchronization code element sequence at the same time, and the first section and the second section of the synchronization code element sequence are combined to form a complete synchronization code element sequence.
A frame structure signal processing method for the wireless communication is characterized by mainly comprising the following steps:
step one, generating an uplink signal to be sent: input data is firstly subjected to channel coding to obtain a coding sequence;
step two, carrying out XOR on the coding sequence and the scrambling sequence to obtain a scrambling sequence;
step three, interleaving the scrambling sequence to obtain an interleaved sequence;
mapping the interleaving sequence into a data code element sequence;
step five, combining the data code element sequence, the pilot frequency code element sequence and the synchronous code element sequence according to the frame structure to form an uplink frame;
and sixthly, performing code element up-sampling on the code element sequence of the uplink frame to obtain an uplink signal to be sent.
Preferably, the channel coding in the step one adopts any one or a combination of several of convolutional code, turbo code, LDPC code and polarization code.
Preferably, in step three, the sequence elements of the interleaved sequence take values of 0 or 1, and the mth element of the interleaved input sequence is denoted as xmL-1, where L is the length of the sequence and the mth element of the output sequence is numbered ymThe interleaving is expressed as: y ismod(am,L)=xmWhere mod (am, L) represents the remainder for am/L, a and L are parameters of the formula mod (am, L), a is a constant, and the greatest common divisor of a and L is 1.
Preferably, in the fourth step, the interleaving sequence is mapped to the data symbol sequence
And adjusting mapping, wherein the specific process is as follows: 0 and 1 at odd positions of the sequence are mapped to +1 and-1, and 0 and 1 at even positions of the sequence are mapped to + i and-i.
Preferably, in step six, the specific process of symbol upsampling is as follows: the K times up-sampling is performed on the symbol first, and then filtering is performed by using a filter.
Preferably, the filter used for the symbol upsampling is a square root raised cosine roll-off filter and/or a sine filter.
Preferably, the pilot symbol sequence is filtered to obtain a single frequency signal or a direct current signal, and if the single frequency signal is obtained, the pilot frequency is deviated from the signal center frequency; if a direct current signal, the pilot frequency is equal to the signal center frequency.
Preferably, the pilot symbol sequence is:
after K times up-sampling and the sine filter, exactly one complex sine wave, i.e. a single frequency signal, with a sine wave period of 4K samples, is generated in the middle of the output waveform, assuming a sample interval TsSecond, then the pilot frequency is 1/(4 kT) higher than the signal center frequencys)Hz。
Preferably, the synchronization symbol sequence is a symbol sequence with good correlation, that is, the autocorrelation function of the synchronization symbol sequence has a sharp peak and very low side lobes, and the synchronization symbol sequence with a length of 50 is as follows:
{+1,+i,-1,+i,-1,+i,-1,+i,-1,-i,+1,+i,+1,+i,-1,-i,+1,+i,+1,+i,+1,+i,-1,-i,+1,-i,-1,+i,-1,+i,-1,+i,-1,+i,+1,-i,-1,-i,-1,+i,+1,-i,-1,-i,-1,-i,-1,+i,+1,-i}
preferably, the uplink signal transmission method includes:
the terminal determines the uplink transmission starting time by receiving the downlink synchronous signal, randomly selects the signal center frequency and transmits the uplink signal to be transmitted at the uplink transmission starting time;
preferably, the method for receiving the uplink signal is as follows:
the base station receives an uplink signal in the transmission time range of an uplink frame through multiple antennas, extracts a pilot signal from the uplink signal, detects a sending terminal in the pilot signal, and acquires the signal center frequency of the sending terminal according to the pilot signal, wherein the specific acquisition method comprises the following steps:
the base station performs J-point discrete Fourier transform on pilot signals received by all the antennas, wherein J is the length of the discrete Fourier transform, and the result is recorded as a vector S
i,S
iRepresenting the Fourier transform result of the ith antenna branch, taking the absolute value of the Fourier transform result received by each antenna, then squaring, and then adding the results of all the antenna branches and recording as a vector
Wherein, ABS
2(S
i) Denotes a general formula S
iTaking absolute values of all elements in the solution and then squaring; then, according to the threshold value set by the system, the position of the peak value in E is judged and recorded as n
1,n
2…,n
KK is the number of detected peak values, each peak value corresponds to a terminal for transmitting an uplink signal, and then the pilot frequency of each transmitting terminal is obtained according to the following formula;
f
sobtaining the center frequency of the uplink signal of each terminal as f for the sampling rate of the system
1-Δf,f
2-Δf,....,f
K- Δ f, where Δ f is the deviation of the signal center frequency from the pilot frequency;
the method for acquiring the channel state information of the terminal by the base station comprises the following steps:
according to the position n of the peak value in E corresponding to the k terminal
kK is 1,2, …, K is a natural number, K is the total number of detected terminals, and the fourier transform results for all antenna branches: s
1,S
2,…S
MTaking the n-th
kElements and are combined into a vector h
kI.e. the channel state information of terminal k to each antenna of the base station,
h
kthe dimensionality is as follows: m rows, 1 column, M is the number of base station receiving antennas, (S)
i)
nkRepresents a vector S
iN of (2)
kAn element;
in the process of carrying out beam forming receiving on a terminal by a base station, carrying out beam forming by adopting a conjugate, zero forcing or minimum mean square error method according to the channel response of the terminal, thereby realizing the MIMO receiving of a plurality of terminals.
The invention has the beneficial effects that:
the receiving end of the communication system estimates channel state information using the pilot symbol sequence and demodulates the data symbols using the channel estimate. But the channel conditions are in many cases time-varying, the channel estimates are less accurate for data symbols that are further away from the pilot symbol sequence. Placing the pilot symbol sequence in the middle of the data symbol sequence helps to shorten the distance of the data symbols from the pilot symbol sequence, thereby improving demodulation.
The receiving end of the communication system acquires timing synchronization using the synchronization symbol sequence and demodulates the data symbols using the synchronization estimate. But the timing synchronization is time-varying and the synchronization estimate is less accurate for data symbols that are further away from the synchronization symbol sequence. Placing the synchronization symbol sequence in the middle of the data symbol sequence helps to shorten the distance of the data symbol from the synchronization symbol sequence, thereby improving demodulation.
The synchronous code element sequence and the pilot code element sequence share a part of code elements, so that the total number of the code elements in the data frame can be reduced, the length of the data frame is shortened, and the utilization efficiency of channel resources is improved.
The pilot frequency code element sequence is filtered to obtain a single frequency signal or a direct current signal, meanwhile, a receiving end adopts Fourier transform to detect a terminal sending an uplink signal, and extracts the signal center frequency and channel state information of the terminal.
The terminal randomly selects the signal center frequency to transmit, so that the probability of pilot frequency collision when a plurality of terminals transmit simultaneously is reduced, and detection of all terminals transmitting uplink signals is facilitated.
Detailed Description
The invention will now be illustrated by way of example with reference to the accompanying drawings,
as shown in fig. 1, a schematic diagram of a frame structure of the present invention is shown, the frame structure is an uplink frame, and the frame structure is composed of a data symbol sequence, a pilot symbol sequence and a synchronization symbol sequence, the data symbol sequence is split into two segments with lengths of L1 and L2, the pilot symbol sequence with length of LP is inserted between the two segments of the data symbol sequence, a first segment of the synchronization symbol sequence with length of LS1 is inserted between the first segment of the data symbol sequence and the pilot sequence, the first LS2 symbols of the pilot symbol sequence are also the second segment of the synchronization symbol sequence at the same time, and the first segment and the second segment of the synchronization symbol sequence are combined together to form a complete synchronization symbol sequence.
As shown in fig. 2, a method for processing the frame structure signal mainly includes the following steps:
the method for generating the uplink signal to be transmitted comprises the following steps:
input data is firstly subjected to channel coding to obtain a coding sequence, and the value of a sequence element is 0 or 1. Typical channel coding is convolutional code, turbo code, LDPC code, and polar code, among others.
And carrying out XOR on the coding sequence and the scrambling sequence to obtain a scrambling sequence, wherein the value of the sequence element is 0 or 1.
And interleaving the scrambling sequence to obtain an interleaved sequence, wherein the value of the sequence element is 0 or 1. The mth element of the interleaved input sequence is denoted xmWhere L is the length of the sequence and the sequence number of the mth element of the output sequence is ymThe interleaving is expressed as: y ismod(am,L)=xmWhere mod (am, L) represents the remainder for am/L, a and L are parameters of the formula mod (am, L), a is a constant, and the greatest common divisor of a and L is 1.
The interleaved sequence is mapped to a sequence of data symbols, a process also referred to as digital modulation.
The process of modulation mapping is as follows: 0 and 1 at odd positions of the sequence are mapped to +1 and-1, and 0 and 1 at even positions of the sequence are mapped to + i and-i.
As shown in fig. 1, the data symbol sequence is split into two segments, a first segment having a length of L1 and a second segment having a length of L2.
A pilot symbol sequence is inserted between the two segments, the pilot symbol sequence being of length LP.
A first segment of a sequence of synchronization symbols is inserted between the first segment of the sequence of data symbols and the pilot sequence, the first segment of the sequence of synchronization symbols having a length LS 1.
The first LS2 symbols of the pilot symbol sequence are also the second segment of the synchronization symbol sequence, and the first segment and the second segment of the synchronization symbol sequence are combined together to form the complete synchronization symbol sequence. That is, the synchronization symbol sequence and the pilot symbol sequence share LS2 symbols.
And each code element is replaced by a code element waveform, namely the code element is up-sampled and filtered by a code element waveform filter to obtain an uplink signal to be sent. The specific process includes up-sampling the code element by K times and filtering by a filter. Typical filters include square root raised cosine roll-off filters and sine filters. The impulse response of the sine filter is g (K) sin (0.5 π K/K), where K denotes the filter sample number and 1 ≦ K ≦ 2K-1.
The pilot code element sequence is filtered to obtain a single frequency signal, namely the pilot frequency has deviation with the signal center frequency; or a direct current signal, i.e. the pilot frequency is equal to the signal center frequency. An important example is: the pilot symbol sequence is { -1, + i, +1, -i, -1, + i, +1, -i. -, -1, + i, +1, -i }, and after K times up-sampling and the sine filter, in the middle part of the output waveform, exactly one complex sine wave, i.e., a single frequency signal, whose sine wave period is 4K samples, and the interval of the samples is TsSecond, then the pilot frequency is 1/(4 kT) higher than the signal center frequencys) Hz. Another important example is: the pilot symbol sequence is { +1, +1, +1, +1, +1., +1, +1, +1, +1}, and is filtered and sampled to obtain a dc signal, i.e., the pilot frequency is equal to the signal center frequency.
The synchronization symbol sequence is a symbol sequence with good correlation, i.e. the autocorrelation function has a sharp peak and very low side lobes. An example is as follows: the length 50 synchronization symbol sequence is as follows:
{+1,+i,-1,+i,-1,+i,-1,+i,-1,-i,+1,+i,+1,+i,-1,-i,+1,+i,+1,+i,+1,+i,-1,-i,+1,-i,-1,+i,-1,+i,-1,+i,-1,+i,+1,-i,-1,-i,-1,+i,+1,-i,-1,-i,-1,-i,-1,+i,+1,-i}
after 8 times of upsampling and the sine filter, the autocorrelation function has good autocorrelation, as shown in fig. 3.
Comparing the synchronization symbol sequence and the pilot symbol sequence, it can be seen that the last 4 symbols of the synchronization symbol sequence are identical to the first 4 symbols of the pilot symbol sequence, and thus, the synchronization symbol sequence and the pilot symbol sequence are overlapped by 4 symbols.
The uplink signal sending method comprises the following steps:
the terminal determines the uplink transmission starting time by receiving the downlink synchronous signal, randomly selects the signal center frequency, and transmits the uplink signal to be transmitted at the uplink transmission starting time. In this way, the transmission times of the plurality of terminals are the same, that is, the start time and the end time of the pilots of the plurality of terminals are the same, but the signal center frequencies of the plurality of terminals are random and not necessarily the same, that is, the pilot frequencies of the plurality of terminals are random and not necessarily the same.
The uplink signal receiving method comprises the following steps:
the base station receives signals through multiple antennas within a transmission time range of an uplink frame. And extracts a pilot signal therefrom, where the pilot signal is the sum of the pilot signals of the plurality of terminals whose pilot frequencies are random and thus not necessarily the same.
The base station detects a sending terminal in a pilot signal, and acquires a signal center frequency of the sending terminal according to the pilot signal, wherein the specific acquisition method comprises the following steps: the base station performs J-point discrete Fourier transform on pilot signals received by all the antennas, wherein J is the length of the discrete Fourier transform, and the result is recorded as a vector S
i,S
iRepresenting the Fourier transform result of the ith antenna branch, taking the absolute value of the Fourier transform result received by each antenna, then squaring, and then adding the results of all the antenna branches and recording as a vector
Wherein, ABS
2(S
i) Denotes a general formula S
iTaking absolute values of all elements in the solution and then squaring; then, according to the threshold value set by the system, the position of the peak value in E is judged and recorded as n
1,n
2…,n
KK is the number of detected peak values, each peak value corresponds to a terminal for transmitting an uplink signal, and then the pilot frequency of each transmitting terminal is obtained according to the following formula;
f
sfor the system sampling rate, obtain each terminalThe center frequency of the uplink signal at the end is denoted as f
1-Δf,f
2-Δf,....,f
K- Δ f. Where Δ f is the deviation of the signal center frequency from the pilot frequency.
The method for acquiring the channel state information of the terminal by the base station comprises the following steps:
according to the position n of the peak value in E corresponding to the k terminal
kK is 1,2, …, K is a natural number, and the fourier transform results for all antenna branches: s
1,S
2,…S
MTaking the n-th
kElements and are combined into a vector h
kI.e. the channel state information of terminal k to each antenna of the base station,
h
kthe dimensionality is as follows: m rows, 1 column, M is the number of base station receiving antennas, (S)
i)
nkRepresents a vector S
iN of (2)
kAnd (4) each element.
In the process of carrying out beam forming receiving on a terminal by a base station, carrying out beam forming by adopting a conjugate, zero forcing or minimum mean square error method according to the channel response of the terminal, thereby realizing the MIMO receiving of a plurality of terminals.