CN105429922B - The channel estimation method based on Comb Pilot for DDO-OFDM systems - Google Patents

Abstract

The invention discloses a comb-shaped pilot-based channel estimation method for a DDO-OFDM system. The present invention comprises the following steps: Step 1, obtain the pilot frequency position; Step 2, insert the pilot frequency subcarrier in the pilot frequency position in the transmission signal of the transmitting end; Step 3, extract the received signal at the same pilot frequency position at the receiving end Step 4: Estimate the average frequency response of the pilot sub-carriers using the LS estimation averaging method; Step 5: Estimate the frequency response of the data sub-carriers in the channel using the linear interpolation method. Under the same pilot expenditure, the algorithm of the present invention reduces the bit error rate, improves the estimation accuracy, and improves the signal-to-noise ratio gain; the algorithm bit error rate of the present invention is less affected by the pilot expenditure, that is, a smaller pilot is used The cost can also ensure a low bit error rate and improve the frequency band utilization; the algorithm of the invention has low complexity and strong practicability.

Description

Comb Pilot Based Channel Estimation Method for DDO-OFDM System

technical field

The invention belongs to the technical field of optical communication, in particular to a comb pilot-based channel estimation method used in a DDO-OFDM system.

Background technique

With the development of communication technology and the continuous improvement of communication requirements, optical communication shows two obvious development trends: the data rate of single-channel transmission is greatly increased, approaching 100Gb/s; the network must have fast dynamic adjustment capabilities . However, when the data rate reaches 100Gb/s, the traditional optical fiber segment compensation becomes expensive and time-consuming, and it is difficult to accurately realize the compensation for system dispersion. However, due to its good calculation characteristics, OFDM can perform complex operations in the frequency domain. The fiber dispersion can be easily compensated. Specific references: SHIEHW, BAOH, TANGY. Coherent Optical OFDM: Theory and De-sign [J]. Optics Express 2008, 16:41-859.

In 2005, N.E.Jolley and T.M.Tang et al first proposed the application of OFDM technology in the optical fiber transmission system at the OFC2005 conference, and experimentally proved that 10Gb/s signals can be transmitted for 1km in multimode optical fibers. Since then, people have started Research on Optical Orthogonal Frequency Division Multiplexing (O-OFDM). References: Jolley N E, Kee H, Pickaed P, etal. Generation and propagation of a 1550nm 10Gbit/s optical orthogonal frequency division multiplexed signal over 1000m of multimode fiber using adirectly modulated DFB. In: Optical Fiber Communication. Technical Conference, 2000 OFC/NFOEC. Harlow, UK, 2005, 6-11.

Orthogonal Frequency Division Multiplexing (OFDM) is a multi-carrier modulation technology. Its basic principle is to use digital signal processing to perform inverse Fourier transform (IFFT) to generate a set of orthogonal subcarriers for low-rate Parallel transmission of digital signals to complete the transmission task of high-speed digital signals. The most prominent advantage of this process is that it improves the spectrum utilization rate of the system and reduces the complexity of calculation at the same time.

Compared with traditional optical communication, O-OFDM has shown many outstanding advantages such as high spectral efficiency, anti-dispersion and nonlinear effects, and these will be very important for the next generation of high-speed communication systems. Many current research results also indicate the potential application value of O-OFDM in future large-capacity, long-distance optical communication systems. Most importantly, O-OFDM has important application value in the future dynamic switching optical network due to its ability of self-adaptive single-tap frequency domain equalization.

The channel needs to be estimated in the direct detection of the O-OFDM system, and the accuracy of the channel estimation will directly affect the performance of the whole system. Channel parameter estimation is a key technology in the realization of wireless communication systems. Whether the detailed channel information can be obtained, so that the transmitted signal can be correctly demodulated at the receiving end is an important index to measure the performance of an optical communication system.

According to the previous literature research, the current channel estimation methods are mainly based on the average method based on block pilots and the blind estimation algorithm, but these two algorithms have certain deficiencies. The averaging method based on block pilots has a large frequency cost and low spectrum utilization; the blind channel method has defects such as low precision, high complexity, and long statistical time, so it is difficult to play a role in practical applications. Therefore, it is necessary to study a new algorithm that overcomes the defects of current channel estimation algorithms.

Contents of the invention

In order to overcome the shortcomings of traditional channel estimation methods such as low frequency spectrum utilization and low precision, the present invention proposes a novel comb-shaped pilot-based channel estimation method for DDO-OFDM systems. The invention can reduce pilot expenses, improve frequency band utilization rate, reduce bit error rate, and increase signal-to-noise ratio gain. The previous simulation results show that the algorithm has good reliability and stability.

The technical scheme adopted by the present invention to solve its technical problems: a kind of channel estimation method based on the comb-shaped pilot frequency suitable for DDO-OFDM system, first use the LS (least square least square method) channel estimation algorithm average method to estimate the pilot frequency The frequency response of the subcarrier, and then estimate the frequency response of the data subcarrier by linear interpolation on the basis of the frequency response of the pilot subcarrier.

The specific implementation steps are as follows:

Step 1. Calculate the pilot position.

Step 2. Inserting pilot subcarriers at pilot positions in the transmitted signal at the transmitting end.

Step 3: Extracting pilot subcarriers at the same pilot position from the received signal at the receiving end.

Step 4, using the LS estimation averaging method to estimate the average frequency response of the pilot subcarriers;

Step 5. Estimate the frequency response of the data subcarriers in the channel by using the linear interpolation method.

The solution of the pilot position in the step 1 is specifically as follows:

(1-1) Calculate the number p of pilot frequency, its formula is:

Wherein, N is the actual number of effective carriers, and L is the pilot interval. ceil(X) means take the smallest integer greater than or equal to the specified expression X.

(1-2) Calculate the position of the pilot subcarrier.

Among them, i _n represents the position of the nth pilot on the frequency axis.

The pilot subcarrier vector inserted in the transmitted signal in step 2 is:

X _p ＝[X(i ₁ ,:);X(i ₂ ,:);...;X(i _n ,:);...;X(i _p ,:)] _p×M

Wherein, X(i _n ,:) is the nth pilot subcarrier inserted in the transmitted signal.

The step (3) extracts the pilot subcarrier vector in the received signal as:

Y _p ＝[Y(i ₁ ,:); Y(i ₂ ,:);...; Y(i _n ,:);...; Y(i _p ,:)] _p×M

Wherein, Y(i _n ,:) is the nth pilot subcarrier extracted from the received data.

Described step (4) utilizes LS estimation average method to estimate the average frequency response of pilot frequency subcarrier, specifically as follows:

(4-1) Utilize the LS estimation method to estimate the frequency response of the pilot subcarrier, its formula is:

in, is the frequency response of the estimated nth pilot subcarrier.

(4-2) Average the frequency response at the pilot position on the time axis to obtain the average frequency response of the pilot subcarriers. The obtained average frequency response vector of the pilot subcarrier is:

in, is the average frequency response of the nth pilot subcarrier.

Described step (5) utilizes the linear interpolation method to estimate the frequency response of the data subcarrier in the channel, specifically as follows:

(5-1) Perform linear interpolation on the frequency axis for the average frequency response of the pilot subcarriers to estimate the frequency response of the data subcarriers in the channel. The estimated data subcarrier frequency response is:

(5-2) Extend the frequency response of the data subcarrier on the time axis to obtain the required channel frequency response. The extended channel frequency response on the time axis is:

Wherein, M is the number of OFDM symbols.

Compared with the prior art, the present invention has the following advantages and beneficial effects:

1. Under the same pilot frequency expenditure, the algorithm of the present invention reduces the bit error rate, improves the estimation accuracy, and improves the signal-to-noise ratio gain.

2. The bit error rate of the algorithm of the present invention is less affected by the pilot expense, and can ensure a low bit error rate even with a smaller pilot expense, thereby improving the frequency band utilization rate.

3. The algorithm of the present invention has low complexity and strong practicability.

Description of drawings

Fig. 1 is a system block diagram of O-OFDM.

Fig. 2 is a flowchart of the implementation method.

FIG. 3 is a schematic diagram of a data frame structure of O-OFDM at a transmitting end.

Fig. 4 is a comparison chart of bit error rate (ber)-sensitivity-to-noise ratio (SNR) in previous experiments based on the block pilot LS estimation average method and the algorithm of the present invention when the pilot cost is 1/512 in the embodiment.

Detailed ways

The present invention will be described in further detail below in conjunction with examples and accompanying drawings, but the embodiments of the present invention are not limited thereto.

The invention mainly relates to the direct detection channel estimation problem of the optical orthogonal frequency division multiplexing O-OFDM system. As shown in Figure 1, O-OFDM includes S/P, QAM mapping, data subcarrier modulation, inserting pilot subcarriers, IFFT, adding CP, adding M sequence, removing M sequence, removing CP, FFT, and extracting pilot subcarriers Carrier, channel estimation, data subcarrier demapping, QAM demapping, P/S and other steps. Among them, inserting pilot subcarriers, extracting pilot subcarriers and channel estimation belong to the channel estimation module.

The steps of the comb pilot-based LS channel estimation algorithm of the present invention will be described in detail below with reference to FIG. 2 and FIG. 3 .

(1) Assuming that there are N= ^2K carriers in total, M OFDM symbols in total, and the pilot interval is L, then there are p pilot subcarriers in total.

(2) As shown in Figure 3, a pilot subcarrier is inserted into the pilot position in the transmitted signal at the transmitter, and the position of the nth pilot subcarrier on the frequency axis is _n , and the nth pilot subcarrier The carrier is X(i _n ,:), and the pilot subcarrier vector is X _p .

X _p ＝[X(i ₁ ,:);X(i ₂ ,:);...;X(i _n ,:);...;X(i _p ,:)] _p×M

(3) At the receiving end, extract the pilot subcarriers at the same pilot position i _n for the received signal, record the extracted nth pilot subcarrier as Y(i _n ,:), and record the extracted pilot subcarriers The vector is Y _p . And use the least squares method to estimate the frequency response of the pilot subcarrier in the channel, record the frequency response of the nth pilot subcarrier as Denote the frequency response vector as

Y _p ＝[Y(i ₁ ,:); Y(i ₂ ,:);...; Y(i _n ,:);...; Y(i _p ,:)] _p×M

(4) Frequency response vector to pilot position Calculate the average value on the time axis (by row) to calculate the average frequency response of the pilot subcarrier, and record the average frequency response of the nth pilot subcarrier as Denote the average frequency response vector of the pilot subcarrier as H _mean .

(5) Perform linear interpolation on the frequency axis for the average frequency response H _mean of the pilot subcarriers, and estimate the frequency response H _linear of the data subcarriers in the channel.

(6) Extend on the time axis to obtain H, and H is the finally estimated channel frequency response.

Wherein, M is the number of OFDM symbols.

Example:

As shown in FIG. 4 , it is a comparison chart of the pre-experimental noise ratio-SNR between the LS estimation averaging method based on block pilots in this embodiment and the LS channel estimation method based on comb pilots in the present invention. It sets O-OFDM to transmit 1023 OFDM symbols, including 1024 carriers (the first line does not contain data, that is, the effective subcarriers start from the second to the 1024th, that is to say, the actual number of effective subcarriers is 1023 ), the pilot expense is 1/512. Utilize the method of the present invention as follows:

(1) The pilot cost is 1/512, that is, the pilot interval L is 511, then there are 3 pilot subcarriers, which are the 2nd, 513th, and 1024th subcarriers respectively, and the pilot vector is denoted as X ₃ .

X ₃ =[X(2,:);X(513,:);X(1024,:)] _3×M

The first subcarrier does not contain data, and the remaining subcarriers are all data subcarriers.

(2) Extract the pilot sub-carrier from the pilot position in the received data, and form the pilot vector as Y ₃ .

Y ₃ ＝[Y(2,:); Y(513,:); Y(1024,:)] _3×M

(3) Use the LS channel estimation algorithm to estimate the frequency response of the pilot subcarriers in the channel

(4) Frequency response to pilot position Averaging on the time axis (by row) estimates the mean frequency response _Hmean of the pilot subcarriers.

(5) Perform linear interpolation on the frequency axis for the estimated average frequency response H _mean of the pilot subcarriers, and estimate the frequency response H _linear of the data subcarriers in the channel.

(6) Extend on the time axis to obtain H, and H is the finally estimated channel frequency response.

Comparing the two BER-SNR curves in Figure 4, it can be seen that when the BER is controlled at the order of 10 ^-3 , compared with the traditional channel estimation based on block pilots, this algorithm has a performance-to-noise ratio of about 5dB gain.

The channel estimation algorithm based on LS estimation and linear interpolation based on comb pilots described in the present invention has been introduced in detail above, and the description of the above examples is only used to help understand the algorithm and its core idea of the present invention rather than explaining it. Any other changes, modifications, combinations, and simplifications that do not deviate from the spirit and principles of the present invention should be equivalent replacements and fall within the protection scope of the present invention.

Claims (4)

1. be used for the channel estimation method based on comb pilot of DDO-OFDM system, it is characterized in that comprising the steps: Step 1, find out the pilot position; Step 2, inserting pilot frequency subcarriers at the pilot frequency position in the transmission signal of the transmitter; Step 3, extracting pilot subcarriers at the same pilot position for the received signal at the receiving end; Step 4, using the LS estimation averaging method to estimate the average frequency response of the pilot subcarriers; Step 5, using the linear interpolation method to estimate the frequency response of the data subcarrier in the channel; The solution of the pilot position in the step 1 is specifically as follows: (1-1) Calculate the number p of pilot frequency, its formula is: Among them, N is the number of actual effective carriers, and L is the pilot interval; ceil(X) means to take the smallest integer greater than or equal to the specified expression X; (1-2) Calculate the position of the pilot subcarrier; Among them, i _n represents the position of the nth pilot on the frequency axis; The pilot subcarrier vector inserted in the transmitted signal in step 2 is: X _p ＝[X(i ₁ ,:);X(i ₂ ,:);...;X(i _n ,:);...;X(i _p ,:)] _p×M Wherein, X(i _n ,:) is the nth pilot subcarrier inserted in the transmitted signal, and M is the number of OFDM symbols. 2. the channel estimation method based on the comb pilot for DDO-OFDM system according to claim 1, it is characterized in that described step 3 extracts pilot frequency subcarrier vector in received signal and is: Y _p ＝[Y(i ₁ ,:); Y(i ₂ ,:);...; Y(i _n ,:);...; Y(i _p ,:)] _p×M Wherein, Y(i _n ,:) is the nth pilot subcarrier extracted from the received signal. 3. the channel estimation method based on the comb pilot for DDO-OFDM system according to claim 2, it is characterized in that described step 4 utilizes LS estimation average method to estimate the average frequency response of pilot frequency subcarrier, specifically as follows: (4-1) Utilize the LS estimation method to estimate the frequency response of the pilot subcarrier, its formula is: in, is the frequency response of the estimated nth pilot subcarrier; / indicates that the element at the position corresponding to Y _p and X _p is divided; (4-2) Average the frequency response at the pilot position on the time axis to obtain the average frequency response of the pilot subcarrier; the average frequency response vector of the pilot subcarrier obtained is: in, is the average frequency response of the nth pilot subcarrier. 4. the channel estimation method based on the comb pilot for DDO-OFDM system according to claim 3, it is characterized in that described step 5 utilizes the linear interpolation method to estimate the frequency response of the data subcarrier in the channel, specifically as follows : (5-1) Perform linear interpolation on the frequency axis to the average frequency response of the pilot subcarriers to estimate the frequency response of the data subcarriers in the channel; the estimated frequency response of the data subcarriers is: (5-2) Extend the frequency response of the data subcarrier on the time axis to obtain the required channel frequency response; the channel frequency response after expansion on the time axis is: Wherein, M is the number of OFDM symbols.