阅读说明：本技术 基于omp信道估计的ofdm系统性能分析方法 (OFDM system performance analysis method based on OMP channel estimation ) 是由 乔钢 肖雨竹 强夕竹 于 2019-09-26 设计创作，主要内容包括：本发明提供的是一种基于OMP信道估计的OFDM系统性能分析方法。针对基于OMP信道估计的OFDM通信系统,有效信噪比ESNR需要成功解码后才能获取这一限制。本发明主要通过建立各个信噪比与系统参数关系的模型,通过系统参数如(导频数量、ICI、信道多径条数与输入信噪比等)获得有效信噪比与导频信噪比,并得到各个信噪比之间的关系。针对水声信道条件的复杂性带来的输入信噪比无法准确指导解码性能的问题,本发明利用已知典型水声通信系统和信道设定简化问题。本发明所公开的方法能够有效通过估计系统参数估计系统各个信噪比性能指标。(The invention provides an OFDM system performance analysis method based on OMP channel estimation. For an OFDM communication system based on OMP channel estimation, the effective snr ESNR needs to be successfully decoded before this limit can be obtained. The invention mainly obtains the effective signal-to-noise ratio and the pilot signal-to-noise ratio through establishing a model of the relation between each signal-to-noise ratio and system parameters (such as pilot frequency quantity, ICI, channel multipath number, input signal-to-noise ratio and the like) and obtains the relation between each signal-to-noise ratio. Aiming at the problem that the decoding performance cannot be accurately guided by the input signal-to-noise ratio caused by the complexity of the underwater acoustic channel condition, the invention utilizes the known typical underwater acoustic communication system and the channel setting simplification problem. The method disclosed by the invention can effectively estimate each signal-to-noise ratio performance index of the system by estimating the system parameters.)

1. An OFDM system performance analysis method based on OMP channel estimation is characterized in that:

(1) signal pre-processing

1) The received passband signal is subjected to down-conversion and low-pass filtering to obtain the average energy of Gaussian white noise in the time domain, which is equivalent to the average noise energy of the frequency domain subcarrier

2) Synchronizing, namely estimating Doppler frequency shift by using cyclic prefix, and then compensating for consistent and inconsistent Doppler;

3) estimating the interference between the average subcarriers by using the null subcarriers and the time domain average noise;

4) channel estimation is carried out by adopting an OMP algorithm to obtain the number N of pilot frequencies_PAnd channel parameters including: channel multipath number S, delay tau of the ith path_lAnd amplitude coefficient beta l]And channel energy

(2) Estimating system performance from system parameters

1) Using formulas

2) using formulas

2. The method of claim 1 for OFDM system performance analysis based on OMP channel estimation, wherein the average inter-subcarrier interference V_ICIBy inter-subcarrier interference V on the pilot_kObtaining:

V_ICI＝minV_k

B^-(D_I,k,K)＝{n|n＝(n+q)_K,-D_I≤q≤D_I,n≠k}

wherein the content of the first and second substances,representing the energy of the ith path of the channel, s representing the transmitted frequency domain vector, K representing the number of subcarriers contained in an OFDM symbol, alpha representing the residual Doppler factor, f_cRepresenting the carrier frequency, T is the duration of one OFDM symbol.

Technical Field

The invention relates to a method for analyzing the performance of an Orthogonal Frequency Division Multiplexing (OFDM) communication system.

Background

Orthogonal Frequency Division Multiplexing (OFDM) is widely used in wireless communications with its high spectrum utilization and low complexity system implementation. In recent years, with the increasing demand for high data rate and reliability performance of a underwater acoustic (UWA) communication system, OFDM has become one of the alternatives of the underwater acoustic communication system with its good resistance to long channel delay dispersion and multipath fading. However, in underwater acoustic communication, due to relative motion between a transmitter and a receiver and low sound velocity, doppler shift in the system is often caused, so that inter-carrier interference (ICI) is brought, and the performance of OFDM is seriously affected. It is therefore critical to accurately estimate system performance.

Common performance indicators for communication systems include MSE, BER, BLER of the equalized transmitted symbols. BER and BLER perform well but require a large amount of data to be accurately obtained. In a single carrier system, the equalized symbol signal-to-noise ratio is generally adopted as a performance parameter; in OFDM systems, various signal-to-noise ratios are typically employed as performance indicators. The system BER or BLER performance can be obtained by the signal-to-noise ratio.

Research shows that in an Adaptive Modulation Coding (AMC) system aiming at underwater sound OFDM, for decoding performance evaluation, the effective signal-to-noise ratio ESNR has better consistency than the input signal-to-noise ratio ISNR and the pilot signal-to-noise ratio PSNR, but the effective signal-to-noise ratio ESNR can be obtained after successful decoding.

Disclosure of Invention

The invention aims to provide an OFDM system performance analysis method based on OMP channel estimation, which can effectively estimate each signal-to-noise ratio performance index of a system by estimating system parameters.

The purpose of the invention is realized as follows:

(1) signal pre-processing

1) The received passband signal is subjected to down-conversion and low-pass filtering to obtain the average energy of Gaussian white noise in the time domain, which is equivalent to the average noise energy of the frequency domain subcarrier

2) Synchronizing, namely estimating Doppler frequency shift by using cyclic prefix, and then compensating for consistent and inconsistent Doppler;

3) estimating the interference between the average subcarriers by using the null subcarriers and the time domain average noise;

4) channel estimation is carried out by adopting an OMP algorithm to obtain the number N of pilot frequencies_PAnd channel parameters,The method comprises the following steps: channel multipath number S, delay tau of the ith path_lAnd amplitude coefficient beta l]And channel energy

(2) Estimating system performance from system parameters

1) Using formulas

Estimating the effective SNR gamma of a system_E，V_ICIIs the inter-subcarrier average interference;

2) using formulas

Estimating the pilot signal-to-noise ratio gamma of a system_P。

Inter-subcarrier average interference V of the present invention_ICIBy inter-subcarrier interference V on the pilot_kObtaining:

wherein the content of the first and second substances,

representing the energy of the ith path of the channel, s representing the transmitted frequency domain vector, K representing the number of subcarriers contained in an OFDM symbol, alpha representing the residual Doppler factor, f_cRepresenting the carrier frequency, T is the duration of one OFDM symbol.

The technical scheme of the invention is mainly characterized in that:

1. the method estimates the performance index of the signal-to-noise ratio of the pilot frequency and the effective signal-to-noise ratio of the system by estimating the parameters of the communication system, such as the number of multipath channel paths, the interference among subcarriers, the number of the pilot frequencies used for OMP channel estimation, the input signal-to-noise ratio of a receiving end time domain, the channel energy and the like, and breaks the limitation that the effective signal-to-noise ratio can be calculated only by correct decoding in the prior art.

2. If a statistical underwater channel model is given, under a certain AMC scheme, the method can be used for establishing a mapping relation between a pilot signal-to-noise ratio (PSNR) or an effective signal-to-noise ratio (ESNR) and an input signal-to-noise ratio (ISNR), wherein the mapping relation is related to the BER performance or the BLER performance of a system. And then, signal-to-noise ratio threshold values of target BER or BLER requirements under different scenes can be obtained immediately, so that the most suitable AMC scheme is designed or selected.

3. The effective snr ESNR performance prediction error is determined by the OMP channel estimation error when the input snr ISNR and oversampling factor in the OMP channel estimation are sufficiently large. The method can be used for obtaining the prediction error of the effective signal-to-noise ratio (ESNR) performance, thereby finding a more accurate OMP channel estimation iteration stop condition.

4. The estimated effective signal-to-noise ratio ESNR can be used for channel condition characterization in underwater acoustic communications.

5. Better system performance indicators may be selected for different environments.

6. And estimating the interference between the average subcarriers by subtracting the time domain average noise energy from the average energy of the null subcarriers.

7. The multipath channel energy estimated by using the OMP is equivalent to the pilot frequency average energy.

The invention aims at the problem that the decoding performance cannot be accurately guided by the input signal-to-noise ratio caused by the complexity of the underwater acoustic channel condition, and utilizes the known typical underwater acoustic communication system and the channel setting to simplify the problem. The correlation between the equivalent signal-to-noise ratio ESNR and the pilot signal-to-noise ratio PSNR and the input signal-to-noise ratio ISNR under specific conditions (a typical channel estimation module, a typical multipath distribution condition, a typical Doppler frequency offset and a specific input signal-to-noise ratio) is deduced, and the estimation of the effective signal-to-noise ratio in an actual communication system is realized.

The invention provides an Orthogonal Frequency Division Multiplexing (OFDM) communication system performance analysis method based on Orthogonal Matching Pursuit (OMP) channel estimation, in particular to a pilot signal-to-noise ratio (PSNR) and effective signal-to-noise ratio (ESNR) estimation method of an OFDM communication system based on OMP channel estimation. The invention relates to a performance analysis method of an underwater sound OFDM system, which is different from the traditional method of obtaining signal-to-noise ratio by utilizing signals obtained in the signal processing process, and establishes a model of the relation between each signal-to-noise ratio and system parameters. The limitation that the effective signal-to-noise ratio can only be obtained after successful decoding is broken through, the effective signal-to-noise ratio and the pilot signal-to-noise ratio are obtained through system parameters (such as pilot frequency quantity, ICI, channel multipath number, input signal-to-noise ratio and the like), and the relation between the signal-to-noise ratios is obtained.

The invention has the advantages that:

(1) mathematically describing the relationship between each signal-to-noise ratio of the system and the system parameters, and explaining the difference of the performance of the input signal-to-noise ratio ISNR, the pilot signal-to-noise ratio PSNR and the effective signal-to-noise ratio ESNR which are taken as system performance indexes.

(2) Under different environments, the optimal system performance signal-to-noise ratio index is selected according to the relationship between the input signal-to-noise ratio ISNR, the pilot signal-to-noise ratio PSNR, the effective signal-to-noise ratio ESNR and the system parameters.

(3) The method is different from the traditional method for obtaining the pilot signal-to-noise ratio by utilizing the frequency domain subcarrier symbol and obtaining the effective signal-to-noise ratio by utilizing the decoded modulation symbol, breaks the limitation that the effective signal-to-noise ratio can only be obtained after the successful decoding, and estimates the pilot signal-to-noise ratio and the effective signal-to-noise ratio by estimating the system parameters.

Drawings

FIG. 1 is a flow chart of the present invention.

Fig. 2 is a channel used in a simulation experiment.

FIG. 3 is a channel scattering function in the offshore test (YS 17).

Fig. 4 is a time domain signal received in a simulation experiment.

Fig. 5 is a received time domain signal in the marine test (YS 17).

Detailed Description

The invention is described in more detail below by way of example.

In order to verify the effectiveness of the algorithm, a simulation experiment and sea test data are respectively adopted to verify whether the pilot signal-to-noise ratio and the effective signal-to-noise ratio obtained by estimation are close to actual values.

In the simulation experiment, 1024 sub-carriers were used, which included 704 data sub-carriers and 256 uniform pilots with a spacing of 4. The signal adopts QPSK modulation, the bandwidth is 6KHz, and the center frequencyThe rate is 9KHz and the CP duration is 50 ms. In simulation, a channel model assumes that time delay differences of adjacent multipath of an underwater acoustic channel obey exponential distribution, multipath amplitudes obey Rayleigh distribution, and the average power of the multipath amplitudes is negatively exponentially attenuated along with time delay, and 3 channel conditions are adopted in total: 1)5 multipath and no Doppler; 2)15 multipath and no doppler; 3)15 multi-paths, the Doppler scale on each path is randomly distributed uniformly

On the basis of the simulation result, a performance result based on an experimental data set is given, and the analysis of the signal-to-noise ratio index is further verified. The experimental setup is as follows.

Offshore test (YS17) in 2017, 10 months, the test was carried out in the yellow sea near Qingdao, China. During the experiment, the transmitter was about 3.5 km from the receiver, and the transmitter was submerged under water 3 meters deep. The channel length is about 7ms, contains different clusters, and has a small common doppler shift. The OFDM symbol uses 1024 subcarriers, and uses 128, 256, 512 uniform pilots. The signal is BPSK modulated, the bandwidth is 6KHz, the center frequency is 12KHz, and the CP duration is 100 ms.

The following is a specific implementation scheme for obtaining the performance of the system pilot signal-to-noise ratio (PSNR) and the effective signal-to-noise ratio (ESNR) by taking experimental data as an example.

(1) Signal pre-processing

① the received passband signal is first down converted and low pass filtered to obtain the average energy of white Gaussian noise in the time domain, which is assumed to be equivalent to the average noise energy of the subcarriers in the frequency domain

② synchronization, estimates the Doppler shift using the cyclic prefix, and then compensates for non-uniform, uniform Doppler.

③ FFT demodulation for estimating inter-subcarrier average interference V using null subcarriers_ICI。

④ adopts OMP algorithm to realize channel estimation and obtain pilot frequency quantity N_PAnd channel parameters: channel multipath number S, delay tau of the ith path_lAnd amplitude coefficient beta l]1, …, S and channel energy

(2) And acquiring an actual calculated value and a theoretical estimated value of the system performance.

① estimating the effective signal-to-noise ratio (ESNR) of the system based on system parameters

Wherein, V_ICIBy inter-subcarrier interference V on the pilot_kObtaining:

wherein the content of the first and second substances,

representing the energy of the ith path of the channel, s representing the transmitted frequency domain vector, K representing the number of subcarriers contained in an OFDM symbol, alpha representing the residual Doppler factor, f_cRepresenting the carrier frequency, T is the duration of one OFDM symbol. In actual data processing, V is estimated by subtracting the average noise energy of the time domain from the average energy of the null sub-carriers_ICI。

② estimating the pilot signal-to-noise ratio (PSNR) of the system according to a formula

③ obtaining the effective signal-to-noise ratio (ESNR) gamma of the system by definition_E。

Wherein the content of the first and second substances,representing the estimated channel vector, z represents the frequency domain received vector,representing a set of data subcarriers.

④ deriving the pilot signal-to-noise ratio (PSNR) gamma of the system by definition_P。

Wherein the content of the first and second substances,

representing a set of pilot subcarriers.

Finally, the comparison is made

And gamma_E，

And gamma_PAnd verifying the validity of the model.