WO2012024836A1

WO2012024836A1 - Narrow band interference suppression method and device

Info

Publication number: WO2012024836A1
Application number: PCT/CN2010/076377
Authority: WO
Inventors: 周旭武
Original assignee: 中兴通讯股份有限公司
Priority date: 2010-08-26
Filing date: 2010-08-26
Publication date: 2012-03-01

Abstract

A narrow band interference suppression method is disclosed. The method comprises: performing channel estimation using a received pilot symbol in a time-frequency unit; calculating the noise-interference power of the pilot symbol based on the channel estimation result, and taking the noise-interference power of the pilot symbol as the noise-interference power of the data symbol in the time-frequency unit; performing combination estimation under the noise-interference of multiple antennae using the channel estimation result and the noise-interference power of the data symbol; performing data decoding with the bottom noise power when it is determined that the estimation result of the noise-interference power is less than the setting threshold value, otherwise performing data decoding with the current estimation result of the noise-interference power. A device for realizing such interference suppression method is also disclosed. Such solutions can implement better interference suppression on account of the solutions perform the interference estimation for the noise signals of the data signals in advance. And, the calculation of the method is simple.

Description

Narrowband interference suppression method and device

The present invention relates to a narrowband interference suppression technique, and more particularly to a narrowband interference suppression method and apparatus for data decoding in an Orthogonal Frequency Division Multiplexing (OFDM) system. Background technique

A new generation of wireless communication systems requires higher transmission rates, and OFDM technology has emerged. It divides the serial data into N different parallel data streams, transmits them in parallel on N carriers, has no interference with each other, greatly improves the transmission rate of the system, and the data stream of each subcarrier has a lower bit rate. Improve the reliability of transmission.

OFDM encodes and modulates the data as frequency domain information, transforms it into the time domain by Inverse Discrete Fourier Transform (IDFT), and transmits it on the channel, and passes the discrete Fourier transform (DFT, Discrete Fourier Transform) at the receiving end. ), the original modulated data after the channel is obtained.

Figure 1 is a schematic diagram of the carrier of the OFDM system being subjected to the dry 4, as shown in Figure 1, the OFDM system becomes very fragile under external interference, and the interference is very difficult to suppress, so the interference is very difficult. Accurate detection, including interference location and frequency detection, is a prerequisite for interference cancellation and is necessary to ensure OFDM system performance.

In the traditional interference detection method, when the interference is strong, setting a certain threshold can detect the interference, but more is only detecting the power of the interference and the number of interferences. When the interference is weak, it cannot be detected. At this time, the interference cancellation will not work. Summary of the invention

In view of the above, it is a primary object of the present invention to provide a narrowband interference suppression method and apparatus that can accurately estimate the power of a noise signal so that accurate decoding can be performed during decoding.

In order to achieve the above object, the technical solution of the present invention is achieved as follows:

A narrowband interference suppression method includes:

Channel estimation is performed by using pilot symbols in the received time-frequency unit, and noise interference (NI, which is simply referred to as NI power) of the pilot symbols is calculated according to the channel estimation result, and the noise interference power of the pilot symbols is used as Noise interference power of data symbols in the time-frequency unit;

Multi-antenna noise interference power combining processing is performed on the currently received signal by using the received antenna channel estimation result and the noise interference power of the data symbol;

When it is judged that the noise interference power estimation result is less than the set threshold value, the data is decoded by using the noise floor power as the noise interference power of the currently received signal. Otherwise, the data is decoded by the current noise interference power estimation result.

Preferably, the time-frequency unit includes at least two pilot symbols located on at least two subcarriers in the same OFDM symbol; or, the time-frequency unit includes at least two OFDM symbols located in the same subcarrier. a pilot symbol; or, the time-frequency unit includes pilot symbols located on at least two subcarriers in the same OFDM, and pilot symbols located on at least two OFDM symbols in the same subcarrier.

Preferably, calculating the noise interference power of the pilot symbol according to the channel estimation result is: calculating, by the pilot symbols on any two subcarriers in the same OFDM, the noise interference power, and/or And calculating pilot noise interference power by using a pilot symbol on any two OFDM symbols in the same subcarrier in the time-frequency unit as a group;

The average value of the calculated noise interference power is used as the noise interference power of the pilot symbol. Preferably, the guides on any two subcarriers in the same OFDM in the time-frequency unit The frequency symbol is a set of calculated noise interference power, and/or a pilot symbol on any two OFDM symbols in the same subcarrier in the time-frequency unit is calculated as a set of noise interference power, specifically: The channel estimation values of the two pilot symbols in the group calculate the total channel power of each pilot symbol as:

Calculating the channel power of each pilot symbol based on channel estimates of two pilot symbols in a group is:

Where H _w is a channel estimation value of the first pilot symbol in the group, H ₂ is a channel estimation value of the second pilot symbol in the group, * indicates a conjugate, and real ( ) indicates a real part, I represents a modulo operation;

The noise interferes with power ^ = - .

Preferably, before performing noise estimation on the currently received signal, the method further includes: smoothing noise interference power of the data symbol according to an OFDM symbol position where the data symbol is located, where: + wherein 0<"<1, σ ² is the noise value of the currently received symbol on the nth OFDM symbol, is the noise value of the received symbol on the n-1th OFDM symbol; the received symbol on the nth OFDM symbol The received symbols on the n-1 OFDM symbols are on the same subcarrier.

Preferably, all the received antenna channel estimation results and the noise interference power of the data symbols are used to perform multi-antenna noise interference power combining processing on the currently received signal, specifically:

Suppose the transmitted signal is ^, and the received signal is Y _k = H _k X _{k +} n _k , which is a noise signal; the above equation has = ^+; the further transform has 3⁄4 = ^+^3⁄4, where (indicating conjugate transpose, 1.1 means modulo operation; 11.1 means vector taking norm operation; The noise interference power is specifically: σΐ , ;

^() denotes an autocorrelation operation, denotes approximately equal to, representing the channel estimation value of carrier k on the ith antenna, _έ denotes the noise interference power value of carrier k on the ith antenna; for all receiving antennas in the communication system at carrier k A matrix composed of channel estimation values, 3⁄4c represents the number of receiving antennas.

Preferably, the set threshold value is ηιχ σ^ , where 1.5≤m≤3; is the noise floor power. A narrowband interference suppression apparatus includes a channel estimation unit, a noise interference power calculation unit, a noise interference estimation combining unit, a determining unit, and a decoding unit;

a channel estimation unit, configured to perform channel estimation by using pilot symbols in the received time-frequency unit; a noise interference power calculation unit, configured to calculate a noise interference power of the pilot symbol according to the channel estimation result, and perform noise interference of the pilot symbol Power as noise interference power of data symbols in the time-frequency unit;

a noise interference estimation combining unit, configured to perform multi-antenna noise interference power combining processing on the currently received signal by using all receiving antenna channel estimation results and noise interference power of the data symbol;

a determining unit, configured to determine a relationship between the noise interference estimation result and the set threshold; the decoding unit, configured to trigger the bottom noise power as the noise interference of the currently received signal when the noise interference estimation result is less than the set threshold The power is decoded by the data. When the noise estimation result is greater than or equal to the set threshold, the data is decoded by the current noise interference power estimation result.

Preferably, the noise interference power calculation unit further performs median of the time-frequency unit The pilot symbols on any two subcarriers in the same OFDM are a set of calculated noise interference power, and/or the pilot symbols on any two OFDM symbols in the same subcarrier in the time-frequency unit are One group calculates the noise interference power; and uses the average of the calculated noise interference power as the noise interference power of the pilot symbols.

Preferably, the noise interference power calculation unit further calculates a channel total power PI of each pilot symbol according to channel estimation values of two pilot symbols in a group, which is: =tk^i ² ; The channel estimation values of the pilot symbols calculate the channel power estimate i P _{sl of} each pilot symbol, which is:

Where H _w is a channel estimation value of the first pilot symbol in the group, H ₂ is a channel estimation value of the second pilot symbol in the group, and * indicates a conjugate, real

( ) table takes the real part, |.| indicates the modulo operation;

The noise interference power P = - .

Preferably, the device further includes:

Smoothing processing means for noise power prior to output data symbols to the combining unit estimation noise, the OFDM data symbols according to the symbol position where the noise smoothing processing power of the data symbols: (1- X_ _{i + t1⁄27n} ² ; where 0<"<1, σ„ ² is the noise value of the currently received symbol on the nth OFDM symbol, which is the noise value of the received symbol on the n- _1th OFDM symbol; The received symbols on the nth OFDM symbol are in the same subcarrier as the received symbols on the n-1th OFDM symbol.

Preferably, the noise interference estimation combining unit performs noise estimation on the currently received signal by using the channel estimation result and the noise interference power of the data symbol. Further,

Suppose the transmitted signal is ^, the received signal is) the current channel is Y _k = H _k X _k + n _k , which is the noise signal; the above equation is ^ ^ ^ + ;

(indicating conjugate transposition, bu I represents modulo operation; 11.1 represents vector taking norm operation;

E{H _k ^H n _k n _k ^H H _k )

Noise interference power is specifically: σ

() indicates an autocorrelation operation, indicating approximately equal to, representing the channel estimate of carrier k on the ith antenna, _έ indicating the noise interference power value of carrier k on the ith antenna; for all antennas in the communication system on carrier k A matrix consisting of channel estimates, 3⁄4c representing the number of receive antennas.

Preferably, the set threshold value is ηιχ σ^ , where 1.5≤m≤3; is the noise floor power. In the present invention, channel estimation is performed on pilot symbols in a certain time-frequency unit in the received symbols, and the noise interference power of each data symbol is determined by using the estimated channel value, and the estimated channel value and data are used. The noise interference power of the symbol, the noise estimation of the data signal transmitted by the multiple antennas in the OFDM system, thereby accurately determining the noise interference power of the data signal in the multi-antenna system; according to the estimated data signal in the multi-antenna system The noise interference power is compared with the set threshold to determine the signal-to-noise ratio parameter for decoding, thereby achieving accurate decoding of the data signal of the multi-antenna system. Since the present invention performs noise estimation on the noise signal of the data signal in advance, better noise suppression can be achieved, and the method of the present invention is simple in calculation. DRAWINGS

FIG. 1 is a schematic diagram of a carrier of an OFDM system being subjected to interference;

2 is a schematic diagram of a structure of a data symbol codec in an OFDM system;

3 is a schematic structural diagram of a time-frequency unit in an uplink partial shared channel (PUSC) mode;

4 is a flowchart of a narrowband interference suppression method according to the present invention;

Figure 5 is a schematic diagram of system simulation of the present invention;

6 is another schematic diagram of system simulation of the present invention; Fig. 7 is a schematic view showing the structure of a narrowband interference suppression device of the present invention. detailed description

2 is a schematic diagram of a structure of a data symbol codec in an OFDM system. As shown in FIG. 2, in the encoding and decoding process of an OFDM system, the binary information needs to be encoded first, and then modulated, and the data symbol to be transmitted passes through the antenna system to be transmitted. The symbol is transmitted. Since the wireless channel is affected by various interference sources, the noise symbol ^ and the data symbol are transmitted in the above-mentioned data symbols. Since the data symbol transmission and reception decoding technology belongs to the prior art, the description thereof will not be repeated here. For implementation details, see the relevant protocol standards. The received signal can be expressed as: Set to indicate the transmission signal on the ^:th carrier, which is the channel response in the frequency domain, indicating additive white noise, interference, combined noise and interference signal, ^ NI _k二 N _k +I _k .

OFDM system decoding uses soft decision decoding. In soft decision decoding, the baseband demodulator calculates the Euclidean distance between each received bit and the possible transmitted bits (0 or 1) as a measure of the soft decision Viterbi decoding. For each received symbol 3⁄4, the demodulator calculates the metric for each bit _c . =b(b = 0,l; = l,2,...m), expressed as:

(c _i =b) = \ogp(c _i =b\y _k )

Oc log Z p(y _k \x _k )^max\ogp(y

Λ—

Mm • + const. where) is a function expression, p( ) is a probability function, oc is proportional to, _co "w is a constant, is the modulation symbol set with the first bit equal to b; represents the noise on the first carrier The above formula is only for the purpose of explaining the essence of the technical solution of the present invention. The above formula is the prior art, and can be specifically referred to the relevant provisions of data demodulation in the OFDM system.

The power for the noise floor is the power of the interference noise; when there is no interference, the noise of the bottom noise (white noise) can be considered as a constant, which can be ignored when calculating the metric value; Interference power, each subcarrier is not a constant. If it is ignored, it will cause the metric value to be mismatched. When the interference is strong, the degree of mismatch will be very serious, which will greatly reduce the performance of the system.

The optimal decoder, which needs to accurately calculate the metric value of each received bit, must know the underlying noise interference power, the interference noise interference power, and the location of the interference. In order to achieve a near-optimal decoding effect, the present invention first needs to obtain the above-mentioned necessary information to estimate the information of the unknown interference. And through the interference information, adjust the metric value to achieve the effect of interference suppression.

4 is a flowchart of a narrowband interference suppression method according to the present invention. As shown in FIG. 4, the specific steps of the narrowband interference suppression method of the present invention are as follows:

Step S401, performing interference noise interference power estimation according to a time-frequency unit in the received signal;

As shown in FIG. 3, which is a schematic structural diagram of a time-frequency unit in an 802.16e uplink PUSC mode, a schematic structural diagram of a time-frequency unit of a tile is shown, and the present invention is not limited to the illustrated time-frequency unit, but Time-frequency unit for any situation. As shown in Figure 3, a tile is a time-frequency unit, and a tile contains a total of 12 subcarriers, of which there are 4 pilot subcarriers and 8 data subcarriers. The first OFDM symbol and the third one contain pilot subcarriers, and all the subcarriers in the second OFDM symbol are data subcarriers.

As can be seen from FIG. 3, and ₂ are pilot subcarriers in the first OFDM symbol, and /^ are pilot subcarriers in the third OFDM symbol, and the remaining subcarriers are data subcarriers.

The four pilot subcarriers in one tile are analyzed, that is, ₂ , and the corresponding frequency domain channel responses are H _{1 P} H _{1 2} , H _{3 1} , H _{3 2 , respectively} . Due to the time-varying characteristics of the channel and the frequency selective fading characteristics The effects of H _w , Η _ί2 , ^^^ are different from each other. First calculate the first OFDM symbol in the tile, and the total power of the pilot subcarrier groups ^ and ₂ at different frequency points is:

Wherein, the representation is the estimated channel value.

The estimated signal power in ^ and is:

P _sl = 2xreal(H _ul xHl ₂ ), where * denotes a conjugate, and the real ( ) table takes the real part;

Therefore, the estimated interference plus noise interference power in ^ and /^ is:

In the same way, the power of the actual interference plus noise in the pilot subcarrier groups /^ and /^ on the third OFDM symbol in the tile can be calculated as:

p =p -P

Step 2 calculates, in units of pilot subcarrier groups, interference plus noise interference in frequency domain response estimates corresponding to two pairs of pilot subcarriers located in different OFDM symbols and at the same frequency point in the selected pilot subcarrier group. power. The calculation method is the same as above, and only the conclusion is given here.

p =p -P

p = p -p then the interference noise power of this time slot unit is:

P = P + P + P +

Since the above method is the sum of the powers of the four pilots on the entire tile, if the power of each pilot is to be understood, the average is taken, that is, /4; 8 data subcarriers in the entire tile are considered to be each data sub The carrier is equal to the pilot noise (NI) power, which is = /4.

After obtaining the NI _k power value of the continuous carrier k on each symbol, it is necessary to record the corresponding physical position according to the subcarrier mapping relationship f, and set the corresponding physical position to be _/· = f{k). This completes the interference information acquisition process. Those skilled in the art should understand the mapping of subcarriers and carriers. The relationship is set by the system in advance, and the details of determining the carrier and its corresponding physical location are not described here. Step S402, smoothing noise interference power of different OFDM symbols; smoothing processing of different OFDM symbols can improve the accuracy of the NI estimation to a certain extent, which is more suitable for the change of the actual communication system. Here, the following smoothing method is used: The NI power value of each subcarrier of the current symbol takes part of the NI power value of the same physical position of the previous symbol, that is, updated to: where 0 <"<1; the smoothing coefficient can be based on The actual communication conditions are set. In the simulation, the golden ratio is "= 0.618." The value is not limited to 0.618, and can also be taken as 0.5, depending on the design of the system and the actual noise interference mode.

Step S403, estimating noise interference (ΝΙ) power of the plurality of receiving antennas; since in the actual communication system, a plurality of receiving antennas are mostly used, thereby improving the overall system energy. After using the ΝΙ power value of each antenna obtained by the above steps, a certain criterion is needed to calculate the ΝΙ power combination value of the plurality of antennas. Prepare for subsequent interference suppression. The merge method is as follows:

For multi-antenna systems, the MRC combining method is used to set the signal to be transmitted in the frequency domain, H _k is the channel response of all antennas on the first carrier in the frequency domain, and Λ ^ is the additive white noise, which is interference. Is the carrier serial number. Combine noise and interference signals,

Receive signals in the frequency domain! Indicates that there are:

Y _k = H _k X _{k +} n _k ,

These include noise and interference, and channel compensation is:

H ^H Y, The combined interference and noise interference power is

E{H n H _k k))

Wherein, and the channel estimation value and the interference noise interference power corresponding to the first receiving antenna data respectively for the first carrier, (indicating conjugate transposition, |.| indicates modulo operation, . indicates that the vector takes a norm operation, () Representing an autocorrelation operation, ^ denotes approximately equal to, a matrix of channel estimates for all antennas in the communication system on carrier k, 3⁄4c representing the number of receive antennas.

It can be seen from the foregoing step S401 that the above ^ and have been determined, and ^ is a matrix of channel estimation values on the kth carrier on all antennas, and therefore, the above can be determined.

Step S404, comparing the calculated interference threshold with the set interference threshold to determine a signal to noise ratio for decoding.

Since the pilot is used for estimation and smoothed, the estimation is more accurate. When there is no interference, it is close to the bottom noise variance. It is only necessary to set the threshold value to several times the noise variance. When the threshold is exceeded, the interference is considered. Take the current calculated value. The method for detecting interference using the threshold is as follows:

The threshold here is taken as m times the noise variance, ie π3⁄41⁄2/έ = «σ^ , where 1.5≤m≤3, m can be any real number in the above interval.

At this time, 5^^ is

Step S405, performing interference suppression at the time of decoding.

The decoder needs to use the probability of the log likelihood ratio LR calculated by the demodulator to translate The code, and each bit HR needs to have a reliability measurement weight. Since the traditional decoder cannot detect the interference information, the weight of the measurement under the interference is not accurate enough, resulting in a decrease in decoding performance. By acquiring the interference information, the corresponding 5% of the interference information containing the above-mentioned adjustment is multiplied by the HR as a weight, and the result of the last multiplication is sent to the decoder. Then the HR that is finally sent to the decoder is:

LLR = SINR _k xLLR

The maximum likelihood decoder utilizes this information to effectively suppress interference. Effectively improve system performance without increasing computational complexity.

The simulation in Figure 5 uses Quadrature Phase Shift Keying (QPSK) modulation, and the Convolutional Turbo Code (CTC) is 1/2 coded. In one transmit antenna, four receive antennas. Under the channel ITU VA60 Km / h, the interference intensity INR = 20dB, using the method of the invention, the performance is close to the interference-free performance, and the performance is improved by nearly 10dB compared with the interference performance; Figure 6 simulation 16QAM (Quadature Amplitude Modulation) modulation method, CTC is 3/4 coded, under one transmit antenna and four receive antennas, through the channel ITU VA60 Km/h, with interference intensity INR=20dB, using the method of the present invention Compared with the performance without interference, the performance is improved by nearly 12dB compared with the performance when there is interference; its robustness is strong, and it can effectively suppress the narrowband interference, which is more than 10dB compared with the performance in the presence of interference, and the effect is very significant. Even in the absence of interference, the performance loss is small with the method of the present invention.

The method of the present invention is equally applicable to a MIMO (Multiple Input Multiple Output) system to suppress narrowband interference.

7 is a schematic structural diagram of a narrowband dry 4 especially suppressing apparatus according to the present invention. As shown in FIG. 7, the narrowband interference suppressing apparatus of the present invention includes a channel estimating unit 70, a noise interference power calculating unit 71, a noise interference estimating combining unit 72, and a determining unit. 73 and decoding unit 74; wherein

Channel estimation unit 70, configured to perform channel estimation by using pilot symbols in the received time-frequency unit Count

The noise interference power calculation unit 7 1 is configured to calculate a noise interference power of the pilot symbol according to the channel estimation result, and use the noise interference power of the pilot symbol as the noise interference power of the data symbol in the time-frequency unit;

The noise interference estimation combining unit 72 is configured to perform multi-antenna noise interference power combining processing on the currently received signal by using the received antenna channel estimation result and the noise interference power of the data symbol;

The determining unit 73 is configured to determine a relationship between the noise interference power estimation result and the set threshold; the decoding unit 74 is configured to trigger the bottom noise power as the current location when the noise interference power estimation result is less than the set threshold The noise interference power of the received signal is used for data decoding. When the noise estimation result is greater than or equal to the set threshold, the data is decoded by the current noise interference power estimation result.

The time-frequency unit includes at least two pilot symbols located on at least two subcarriers in the same OFDM symbol; or, the time-frequency unit includes at least pilot symbols located on at least two OFDM symbols in the same subcarrier; Or, the time-frequency unit includes pilot symbols located on at least two subcarriers in the same OFDM, and pilot symbols located on at least two OFDM symbols in the same subcarrier.

The noise interference power calculation unit 71 further calculates pilot noise interference power of any one of the time-frequency units located on any two subcarriers in the same OFDM, and/or, in the time-frequency unit The pilot symbols located on any two OFDM symbols in the same subcarrier are a set of calculated noise interference powers; and the average of the calculated noise interference powers is used as the noise interference power of the pilot symbols.

The noise interference power calculation unit 71 further calculates a total channel power of each pilot symbol according to channel estimation values of two pilot symbols in a group, which is: = k^ i ² ;

j=l

Calculating the channel power estimate of each pilot symbol based on channel estimation values of two pilot symbols in a group For example, P _sl =2x _rea H _u xH; _>2 ) where H _w is the channel estimate of the first pilot symbol in the group, and H ₂ is the channel estimate of the second pilot symbol in the group Value, * indicates conjugate;

The noise interference power P = - .

Based on the device shown in FIG. 7, the interference suppression device of the present invention further includes:

And a smoothing processing unit (not shown), configured to smooth the noise interference power of the data symbol according to the OFDM symbol position where the data symbol is located before outputting the noise interference power of the data symbol to the noise interference estimation combining unit: (1-

Where 0<"<1, σ„ ² is the noise value of the currently received symbol on the nth OFDM symbol, which is the noise value of the received symbol on the n-1th OFDM symbol; the nth OFDM The received symbol on the symbol is in the same subcarrier as the received symbol on the n-1th OFDM symbol. The above-mentioned smoothing coefficient can be set according to the actual communication conditions. In the simulation, the golden ratio is "0.618". "The value is not limited to 0.618, but can also be taken as 0.5, depending on the design of the system and the actual noise interference.

The noise interference estimation combining unit 72 performs multi-antenna noise interference power combining processing on the currently received signal by using the received antenna channel estimation result and the noise interference power of the data symbol. Further,

Suppose the transmitted signal is ^, and the received signal is the current channel, then there is Y _k = H _k X _k + n _k , which is a noise signal;

Medium, (for conjugate transpose, |.| for modulo operation, 11.1 for vector fetch operation; noise interference power for: σ

() indicates an autocorrelation operation, indicating approximately equal to, representing the channel of carrier k on the ith antenna The estimated value represents the noise interference power value of the carrier k on the i-th antenna; H _k is a matrix composed of channel estimation values of all antennas on the carrier k in the Yixin system, and 3⁄4c represents the number of receiving antennas.

The above set threshold is ηιχ σ^ , where 1.5≤m≤3; is the noise floor power.

It should be understood by those skilled in the art that the interference suppression apparatus of the present invention is designed to implement the foregoing method of the interference suppression method of the present invention, and the implementation functions of the above-described processing units and sub-processing units can be understood by referring to the related description of the foregoing methods. The functions of each of the above processing units and sub-processing units may be implemented by a corresponding integrated circuit, or by a corresponding computer program on a processor having a program execution function.

The above is only the preferred embodiment of the present invention and is not intended to limit the scope of the present invention.

Claims

Claim

A narrowband interference suppression method, the method comprising:

Using the pilot symbols in the received time-frequency unit for channel estimation, calculating the noise interference power of the pilot symbols according to the channel estimation result, and using the noise interference power of the pilot symbols as the noise interference of the data symbols in the time-frequency unit Power

2. The method according to claim 1, wherein the time-frequency unit includes at least pilot symbols located on at least two subcarriers in the same orthogonal frequency division multiplexing (OFDM) symbol; or The time-frequency unit includes at least two pilot symbols located on at least two OFDM symbols in the same subcarrier; or, the time-frequency unit includes pilot symbols located on at least two subcarriers in the same OFDM, and Pilot symbols located on at least two OFDM symbols in the same subcarrier.

3. The method according to claim 2, wherein calculating the noise interference power of the pilot symbol according to the channel estimation result is specifically:

Generating pilot symbols on any two subcarriers in the same OFDM in the time-frequency unit into a group to calculate noise interference power, and/or any two OFDMs in the same subcarrier in the time-frequency unit The pilot symbols on the symbol are a set of calculated noise interference powers;

The average value of the calculated noise interference power is used as the noise interference power of the pilot symbol.

The method according to claim 3, wherein pilot symbols on any two subcarriers in the same OFDM in the time-frequency unit are used to calculate noise interference power, and/or a pilot symbol on any two OFDM symbols in the same subcarrier in the time-frequency unit The number is a set of calculation noise interference power, specifically:

Calculating the total channel power of each pilot symbol based on channel estimation values of two pilot symbols in a group is:

Calculating a channel power estimate of each pilot symbol according to channel estimation values of two pilot symbols in a group, where: P _s , = 2 x _rea l (i, 0 where H _w is the first pilot symbol in the group Channel estimation value, H ₂ is the channel estimation value of the second pilot symbol in the group, * indicates conjugate, real

( ) table takes the real part, |.| indicates the modulo operation;

The noise interference power = - .

The method according to claim 1, wherein before the noise estimation of the currently received signal, the method further includes:

The noise interference power of the data symbol is smoothed according to the OFDM symbol position where the data symbol is located, as: (1 - "Χ^ + σ^; where 0<"<1, σ„ ² is the nth OFDM symbol The noise value of the currently received symbol is the noise value of the received symbol on the n-1th OFDM symbol; the received symbol on the nth OFDM symbol and the received on the n-1th OFDM symbol The symbols are in the same subcarrier.

The method according to claim 1, wherein all the received antenna channel estimation results and the noise interference power of the data symbols are used to perform multi-antenna noise interference power combining processing on the currently received signal, specifically:

Medium, (for conjugate transpose, |.| for modulo operation, 11.1 for vector fetch operation; The noise interference power is specifically: σ

^() denotes an autocorrelation operation, denotes approximately equal, h _ik denotes a channel estimation value of carrier k on the i-th antenna, denotes a noise interference power value of carrier k on the i-th antenna; for all antennas in the communication system on carrier k A matrix composed of channel estimation values, 3⁄4c represents the number of receiving antennas.

The method according to any one of claims 1 to 6, wherein the set width is mx, wherein 1.5 ≤ m ≤ 3; is the noise floor power.

A narrowband interference suppression apparatus, comprising: a channel estimation unit, a noise interference power calculation unit, a noise interference estimation combining unit, a determining unit, and a decoding unit;

a determining unit, configured to determine a relationship between the noise estimation result and the set threshold;

a decoding unit, configured to: when the noise estimation result is less than a set threshold, trigger data decoding by using the bottom noise power as the noise interference power of the currently received signal, and when the noise estimation result is greater than or equal to the set threshold, the current noise estimation The result is data decoding.

The device according to claim 8, wherein the time-frequency unit includes at least two pilot symbols located on at least two subcarriers in the same OFDM symbol; or, the time-frequency unit includes at least Pilot symbols on at least two OFDM symbols in the same subcarrier; or, the time-frequency unit includes at least two subcarriers located in the same OFDM Pilot symbols, and pilot symbols located on at least two OFDM symbols in the same subcarrier.

The apparatus according to claim 9, wherein the noise interference power calculation unit further calculates a pilot symbol of any two subcarriers located in the same OFDM in the time-frequency unit as a group Noise interference power, and/or, calculating pilot noise interference power by using pilot symbols on any two OFDM symbols in the same subcarrier in the time-frequency unit; and calculating an average value of the noise interference power Noise interference power as a pilot symbol.

The device according to claim 10, wherein the noise interference power calculation unit further calculates a total channel power of each pilot symbol according to channel estimation values of two pilot symbols in a group, which is: = k^i ² ;

7=1 '

Calculating a channel power estimate of each pilot symbol according to channel estimation values of two pilot symbols in a group, where: P _s , = 2 x _rea l (i, 0 where H _w is the first pilot symbol in the group Channel estimation value, H ₂ is the channel estimation value of the second pilot symbol in the group, * indicates that the conjugate real ( ) table takes the real part, and |. | represents the modulo operation;

The noise interference power P = - .

The device according to claim 8, wherein the device further comprises: a smoothing processing unit, configured to press the data symbol before outputting the noise interference power of the data symbol to the noise interference estimation combining unit The OFDM symbol position smoothes the noise interference power of the data symbol: (1 - ^X_ _{i + t1⁄27n} ² ; where 0<"<1, σ„ ² is the currently received symbol on the nth OFDM symbol a noise value, which is a noise value of the received symbol on the n-1th OFDM symbol; the received symbol on the nth OFDM symbol is located in the same subcarrier as the received symbol on the n-1th OFDM symbol .

The device according to claim 8, wherein the noise interference estimation combining unit is configured to perform multi-antenna noise interference power on the currently received signal by using the received antenna channel estimation result and the noise interference power of the data symbol. Merge processing, further, Suppose the transmitted signal is the received signal for the current channel.

Y _k =H _k X _k +n _k , is a noise signal; transforming the above equation has its

Medium, (for conjugate transpose, |.| for modulo operation, 11.1 for vector fetch operation; noise interference power _σ ² for: 2 ₌ F(H ^H nn _k n η ^Η nΗ _k ) ;

^() denotes an autocorrelation operation, representing approximately equal to, representing the channel estimate of carrier k on the ith antenna, representing the noise interference power value of carrier k on the ith antenna; for all antennas in the communication system on carrier k A matrix consisting of channel estimates, 3⁄4c representing the number of receive antennas.

The device according to any one of claims 8 to 13, characterized in that the set threshold value is mx, wherein 1.5 ≤ m ≤ 3; is the noise floor power.