CN105515685B - A kind of white Gaussian noise power measurement method and device - Google Patents

Abstract

The invention provides a Gaussian white noise power measurement method, including: judging whether there is effective Gaussian white noise power; if there is effective Gaussian white noise power, then make the current Gaussian white noise power equal to the effective Gaussian white noise power, and according to The current Gaussian white noise power obtains an equalization coefficient, and equalizes a received signal according to the equalization coefficient to obtain an equalized chip-level signal; descrambles and spreads the equalized chip-level signal The code is despread to obtain a detection signal; the equalized symbol-level SINR is obtained according to the detection signal; and the chip-level Gaussian white noise power is obtained according to the equalized symbol-level SINR.

Description

A Gaussian white noise power measurement method and device

technical field

The invention relates to the technical field of communications, in particular to a Gaussian white noise power measurement method and device.

Background technique

The WCDMA (Wideband Code Division Multiple Access) system is a broadband mobile communication system that uses orthogonal spreading codes to distinguish services and users. Because its transmission bandwidth is much larger than the coherent bandwidth of the space physical channel. Therefore, signal transmission in a WCDMA system is often accompanied by frequency selective fading, and the embodiment of frequency selective fading in the time domain is multipath transmission.

RAKE is a common receiver in the WCDMA system. It combines the energy of the multipath components in the system transmission to improve the receiving signal-to-noise ratio, thereby estimating the transmission vector. This kind of receiver has a simple principle and does not consider the elimination of inter-path interference, so it is no longer competent in high-speed transmission scenarios such as packet data services. The equalized receiver is a new receiving method proposed after the RAKE receiver. It obtains the equalized coefficient G _MMSE through the LMMSE (Linear Minimum Mean Square Error) criterion, and uses this coefficient to process the received signal r Filtering, so as to obtain the estimated result of the transmitted vector s

The above process can be expressed as:

Among them, the equalization coefficient G _MMSE is usually obtained by the following formula:

G _MMSE = (H ^H H + σ ² I) ^-1 H ^H (2)

It can be seen that the calculation of the equalization coefficient requires two estimation results, one is the estimation result H of the channel impulse response, and the other is the measured value σ ² of the Gaussian white noise power.

In WCDMA systems, special modules are usually required to measure Gaussian white noise power σ ² . The measurement accuracy of this value directly affects the performance of the equalization algorithm.

The currently commonly used Gaussian white noise power measurement method is as follows: by using the pilot channel, the channel impulse response estimation results of each path position are obtained. Taking single-antenna transmission as an example, the pilot channel transmits fixed symbols, which are set to A. Therefore, by selecting a certain root path, the channel estimation result is subtracted two by two from the front and back symbols, and the expectation of the modulus square is calculated, then the symbol itself can be subtracted, and the noise power can be reserved. The specific process can be described as follows:

Suppose the received signals corresponding to the two symbols before and after the pilot channel are r _CPICH,1 and r _CPICH,2 , there are:

r _CPICH,1 = A·h ₁ +n ₁ (3)

r _CPICH,2 = A·h ₂ +n ₂

Wherein, h ₁ , h ₂ , n ₁ , and n ₂ are the channel impulse response and noise experienced by the two pilot symbols before and after, respectively.

Since the time interval between two symbols before and after is relatively short, it can be assumed that the channel impulse response does not change, that is, h ₁ =h ₂ . And because there is no correlation between Gaussian white noise superimposed on the two symbols before and after, the noise power can be obtained by calculating the expectation of the modulus square of the subtraction result of the two symbols before and after.

The dual-antenna transmission mode is similar to the single-antenna transmission. The first transmission antenna still transmits a fixed symbol A, while the second transmission antenna alternately transmits A and -A according to certain rules. Therefore, a certain root path is selected, and the channel estimation result of the first transmitting antenna is subtracted pairwise from the preceding and following symbols, and the channel estimation result of the second transmitting antenna is recombined, and the pairwise subtraction of the preceding and following symbols is performed. , the noise power result can also be obtained.

However, due to the non-ideal characteristics of the scrambling code, the channel estimation obtained by using the pilot contains a large influence of inter-path interference, and the channel estimation result of each path has the components of the autocorrelation sidelobe of the scrambling code of other paths , so the noise power calculated by this method will also include a part of the inter-path interference, which will cause the noise measurement to be too large and affect the performance of the LMMSE equalizer.

Contents of the invention

The purpose of the present invention is to provide a Gaussian white noise power measurement method and device to solve the problem of inaccurate noise power measurement caused by inter-path interference.

In order to solve the above technical problems, the present invention provides a Gaussian white noise power measurement method, comprising:

Determine whether there is effective Gaussian white noise power;

If there is an effective Gaussian white noise power, then make the current Gaussian white noise power equal to the effective Gaussian white noise power;

Obtain an equalization coefficient according to the current Gaussian white noise power;

Equalize a received signal according to the equalization coefficient to obtain an equalized chip-level signal;

performing descrambling and spreading code despreading on the equalized chip-level signal to obtain a detection signal;

obtaining an equalized symbol-level signal-to-interference-noise ratio according to the detection signal;

The chip-level Gaussian white noise power is obtained according to the equalized symbol-level SINR.

Further, in the Gaussian white noise power measurement method, in the step of judging whether there is effective Gaussian white noise power, if the time difference between the current measurement moment and the previous measurement moment is less than the preset time, it is considered that there is an effective Gaussian white noise power. Gaussian white noise power.

Further, in the Gaussian white noise power measurement method, if there is no effective Gaussian white noise power, the Gaussian white noise power is obtained by a difference method.

Further, in the Gaussian white noise power measurement method, in the step of obtaining the equalization coefficient according to the current Gaussian white noise power, the equalization coefficient is obtained by the following formula: G _MMSE =(H ^H H+σ ² I) ^-1 H ^H ; where G _MMSE represents the equalization coefficient, H represents the estimation result of the channel impulse response, σ ² represents the current Gaussian white noise power, and I represents the identity matrix.

Further, in the Gaussian white noise power measurement method, in the step of equalizing a received signal according to the equalization coefficient to obtain an equalized chip-level signal, the chip-level signal is obtained by the following formula: in, Represents a chip-level signal, G _MMSE represents an equalization coefficient, and r represents a received signal.

Further, in the Gaussian white noise power measurement method, in the steps of descrambling and spreading code despreading of the equalized chip-level signal, the spreading code used for the despreading is a common Pilot channel spreading code.

Further, in the Gaussian white noise power measurement method, in the step of obtaining the chip-level Gaussian white noise power according to the equalized symbol-level SINR, the chip-level Gaussian white noise is obtained by the following formula power: Among them, P _{CPICH_BeforeEQ} represents the power of the common pilot channel, 256 represents the spreading factor of the common pilot channel, and SINR _{CPICH_AfterEQ_Symbol} represents the symbol-level SINR after equalization.

Correspondingly, the present invention also provides a Gaussian white noise power measurement system, comprising:

Judgment module, for judging whether there is effective Gaussian white noise power;

The equalization coefficient module is used to make the current Gaussian white noise power equal to the effective Gaussian white noise power when there is effective Gaussian white noise power, and obtain the equalization coefficient according to the current Gaussian white noise power;

A chip-level signal module, configured to equalize a received signal according to the equalization coefficient, to obtain an equalized chip-level signal;

A detection signal module, configured to perform descrambling and spreading code despreading on the equalized chip-level signal to obtain a detection signal;

A symbol-level SINR module, configured to obtain an equalized symbol-level SINR according to the detection signal;

A chip-level Gaussian white noise power module, configured to obtain chip-level Gaussian white noise power according to the equalized symbol-level SINR.

Further, in the Gaussian white noise power measurement system, in the judging module, if the time difference between the current measurement moment and the previous measurement moment is less than the preset time, it is considered that there is effective Gaussian white noise power.

Further, in the Gaussian white noise power measurement system, in the equalization coefficient module, the equalization coefficient is obtained by the following formula: G _MMSE =(H ^H H+σ ² I) ^-1 H ^H ; wherein, G _MMSE represents Equalization coefficient, H represents the estimated result of channel impulse response, σ ² represents the current Gaussian white noise power, and I represents the identity matrix.

Further, in the Gaussian white noise power measurement system, in the chip-level signal module, the chip-level signal is obtained by the following formula: in, Represents a chip-level signal, G _MMSE represents an equalization coefficient, and r represents a received signal.

Further, in the Gaussian white noise power measurement system, in the chip-level signal module, the spreading code used for the despreading is a common pilot channel spreading code.

Further, in the described Gaussian white noise power measurement system, in the chip-level Gaussian white noise power module, the chip-level Gaussian white noise power is obtained by the following formula: Among them, P _{CPICH_BeforeEQ} represents the power of the common pilot channel, 256 represents the spreading factor of the common pilot channel, and SINR _{CPICH_AfterEQ_Symbol} represents the symbol-level SINR after equalization.

The Gaussian white noise power measurement method and device provided by the present invention have the following beneficial effects: obtain the equalized signal-to-interference-noise ratio according to the equalized data, and use the signal-to-interference-noise ratio to obtain the Gaussian white noise power. Therefore, the final Gaussian white noise power does not include inter-path interference, and the accuracy is significantly improved. In addition, within a certain time interval, the Gaussian white noise power obtained by this method can be used for the next equalization operation, which improves the performance of data equalization.

Description of drawings

Fig. 1 is a schematic flow chart of the Gaussian white noise power measurement method of the present invention.

Detailed ways

The method and device for measuring Gaussian white noise power proposed by the present invention will be further described in detail below in conjunction with the accompanying drawings and specific embodiments. Advantages and features of the present invention will be apparent from the following description and claims. It should be noted that all the drawings are in a very simplified form and use imprecise scales, and are only used to facilitate and clearly assist the purpose of illustrating the embodiments of the present invention.

Please refer to FIG. 1 , which is a schematic flow chart of the Gaussian white noise power measurement method of the present invention. As shown in Figure 1, the present invention provides a kind of Gaussian white noise power measurement method, comprises:

Step 101, judging whether there is effective Gaussian white noise power Since it was obtained during equalization processing at the last measurement moment, when the time difference between the current measurement moment and the last measurement moment is less than the preset time Thred _time , it is considered that there is effective Gaussian white noise power

Step 102, when there is no effective Gaussian white noise power , the Gaussian white noise power is obtained through the difference rule, that is, obtained through the formula (4) in the prior art.

Step 103, when there is effective Gaussian white noise power Then make the current Gaussian white noise power σ ² equal to the effective Gaussian white noise power which is

Step 104, according to the current Gaussian white noise power Obtain the equalization coefficient G _MMSE ; specifically, the equalization coefficient is obtained by the formula G _MMSE =(H ^H H+σ ² I) ^-1 H ^H , wherein, G _MMSE represents the equalization coefficient, H represents the estimated result of the channel impulse response, and σ ² represents the current Gaussian white noise power, and I represents the identity matrix.

Step 105: Equalize a received signal r according to the equalization coefficient G _MMSE to obtain an equalized chip-level signal Specifically, through the formula The equalized chip-level signal is obtained.

Step 106, for the equalized chip-level signal Perform descrambling and spreading code despreading to obtain the detection signal In the process of despreading, the spreading code used for despreading is common pilot channel (CPICH channel) spreading code.

Step 107, according to the detection signal Obtain the equalized symbol-level SINR _{CPICH_AfterEQ_Symbol} ;

Step 108, obtain chip-level white Gaussian noise power σ ² according to the equalized symbol-level SINR _{CPICH_AfterEQ_Symbol} . Specifically, through the formula To obtain chip-level Gaussian white noise power, where P _{CPICH_BeforeEQ} represents the power of the common pilot channel, 256 represents the spreading factor of the common pilot channel, and SINR _{CPICH_AfterEQ_Symbol} represents the equalized symbol-level SINR.

further, The establishment of SNR is based on the following assumption: SINR _{CPICH_AfterEQ_Chip} ≈SNR _{CPICH_AfterEQ_Chip} ≈SNR _{ReceiveSignal} ; wherein, the above description of SNR is based on a unified standard (both chip-level SNR). Because, for the received signal of the WCDMA system, the interference it receives mainly comes from inter-path interference, supplemented by adjacent cell interference, synchronization signal interference, etc., and an equalizer with excellent performance can effectively eliminate inter-path interference , and at the same time, it can also suppress the interference of adjacent cells and synchronization signals. Therefore, the interference item in the output signal of the equalizer can be ignored, and it is approximately considered that SINR _{CPICH_AfterEQ_Chip} ≈ SNR _{CPICH_AfterEQ_Chip} .

On the other hand, if the interference is not considered, it is approximately considered that the signal-to-noise ratio SNR _{CPICH_AfterEQ_Chip} ≈SNR _{ReceiveSignal} output by the MMSE equalizer. This is because, taking a one-path channel as an example, the signal-to-noise ratio of the signal before equalization is:

Wherein, for a certain equalization, the channel estimation result h is constant, and due to the effect of normalization of the transmission power, E(x ^H x)=1, and x is the transmission signal.

The signal-to-noise ratio of the signal after equalization is:

It can be seen that for a channel with only one path, the chip-level SNR before and after equalization is completely equal. For multipath channels, although the SNR before and after equalization cannot be completely equal, experiments have found that the deviation is completely acceptable for the accuracy requirements of noise power measurement.

So when calculating the noise power, consider SNR _{CPICH_AfterEQ_Chip} ≈ SNR _{ReceiveSignal} .

On the other hand, due to the effect of spread spectrum, the chip-level noise power and symbol-level noise power of the equalized signal have the following relationship: Among them, 256 is the spreading factor of the CPICH channel. F:

Wherein, P _{CPICH_BeforeEQ} represents the power of the CPICH channel, which can usually be replaced by a measurement result of CPICH_RSCP (Received Signal Code Power).

Step 109, retain the noise power ^σ2 calculated in step 108 for subsequent use.

Correspondingly, the present invention also provides a Gaussian white noise power measurement system, comprising:

Judgment module, used to judge whether there is effective Gaussian white noise power Since it was obtained during equalization processing at the last measurement moment, when the time difference between the current measurement moment and the last measurement moment is less than the preset time Thred _time , it is considered that there is effective Gaussian white noise power

The equalization coefficient module is used when there is effective Gaussian white noise power , let the current Gaussian white noise power σ ² be equal to the effective Gaussian white noise power which is and according to the current white Gaussian noise power Obtain the equalization coefficient G _MMSE ; specifically, the equalization coefficient is obtained by the formula G _MMSE =(H ^H H+σ ² I) ^-1 H ^H , wherein, G _MMSE represents the equalization coefficient, H represents the estimated result of the channel impulse response, and σ ² represents the current Gaussian white noise power, and I represents the identity matrix. When there is no effective white Gaussian noise power , the Gaussian white noise power is obtained through the difference method, that is, obtained through the formula (4) in the prior art.

A chip-level signal module, configured to equalize a received signal r according to the equalization coefficient G _MMSE to obtain an equalized chip-level signal Specifically, through the formula The equalized chip-level signal is obtained.

A signal detection module, configured to detect the equalized chip-level signal Perform descrambling and spreading code despreading to obtain the detection signal In the process of despreading, the spreading code used for despreading is common pilot channel (CPICH channel) spreading code.

A symbol-level signal-to-interference-noise ratio module for detecting signals based on the Obtain the equalized symbol-level SINR _{CPICH_AfterEQ_Symbol} ;

A chip-level white Gaussian noise power module, configured to obtain chip-level white Gaussian noise power σ ² according to the equalized symbol-level SINR _{CPICH_AfterEQ_Symbol} . Specifically, through the formula To obtain chip-level Gaussian white noise power, where P _{CPICH_BeforeEQ} represents the power of the common pilot channel, 256 represents the spreading factor of the common pilot channel, and SINR _{CPICH_AfterEQ_Symbol} represents the equalized symbol-level SINR.

The above description is only a description of the preferred embodiments of the present invention, and does not limit the scope of the present invention. Any changes and modifications made by those of ordinary skill in the field of the present invention based on the above disclosures shall fall within the protection scope of the claims.

Claims (3)

1. a Gaussian white noise power measurement method, is characterized in that, comprises: Judging whether there is effective Gaussian white noise power, if the time difference between the current measurement moment and the previous measurement moment is less than the preset time, it is considered that there is effective Gaussian white noise power; If there is an effective Gaussian white noise power, then make the current Gaussian white noise power equal to the effective Gaussian white noise power; The equalization coefficient is obtained according to the current Gaussian white noise power, and the equalization coefficient is obtained by the following formula: G _MMSE =(H ^H H+σ ² I) ^-1 H ^H ; wherein, G _MMSE represents the equalization coefficient, and H represents the channel impulse response Estimation result, σ ² represents the current Gaussian white noise power, I represents the identity matrix; A received signal is equalized according to the equalization coefficient to obtain an equalized chip-level signal, and the chip-level signal is obtained by the following formula: in, Represents a chip-level signal, G _MMSE represents an equalization coefficient, and r represents a received signal; Descrambling and spreading code despreading are performed on the equalized chip-level signal to obtain a detection signal, and the spreading code used in the despreading is a common pilot channel spreading code; obtaining an equalized symbol-level signal-to-interference-noise ratio according to the detection signal; Obtain the chip-level Gaussian white noise power according to the equalized symbol-level signal-to-interference-noise ratio, and obtain the chip-level Gaussian white noise power by the following formula: Among them, P _{CPICH_BeforeEQ} represents the power of the common pilot channel, 256 represents the spreading factor of the common pilot channel, and SINR _{CPICH_AfterEQ_Symbol} represents the symbol-level SINR after equalization. 2. Gaussian white noise power measuring method as claimed in claim 1 is characterized in that, if there is no effective Gaussian white noise power, then obtains current Gaussian white noise power by differential method. 3. A Gaussian white noise power measurement system, characterized in that, comprising: Judging module, used to judge whether there is effective Gaussian white noise power, if the time difference between the current measurement moment and the last measurement moment is less than the preset time, then it is considered that there is effective Gaussian white noise power; The equalization coefficient module is used to make the current Gaussian white noise power equal to the effective Gaussian white noise power when there is effective Gaussian white noise power, and obtain the equalization coefficient according to the current Gaussian white noise power, and obtain the equalization coefficient by the following formula: G _MMSE =(H ^H H+σ ² I) ^-1 H ^H ; Wherein, G _MMSE represents the equalization coefficient, H represents the estimated result of the channel impulse response, σ ² represents the current Gaussian white noise power, and I represents the identity matrix; The chip-level signal module is used to equalize a received signal according to the equalization coefficient to obtain an equalized chip-level signal, and obtain the chip-level signal by the following formula: in, Represents a chip-level signal, G _MMSE represents an equalization coefficient, and r represents a received signal; A detection signal module, configured to perform descrambling and despreading by a spreading code on the equalized chip-level signal to obtain a detection signal, and the spreading code used for the despreading is a common pilot channel spreading code; A symbol-level SINR module, configured to obtain an equalized symbol-level SINR according to the detection signal; The chip-level Gaussian white noise power module is used to obtain the chip-level Gaussian white noise power according to the equalized symbol-level SINR, and obtain the chip-level Gaussian white noise power by the following formula: Among them, P _{CPICH_BeforeEQ} represents the power of the common pilot channel, 256 represents the spreading factor of the common pilot channel, and SINR _{CPICH_AfterEQ_Symbol} represents the symbol-level SINR after equalization.