CN-122017895-A - GNSS receiver interference sensitivity threshold calibration method and device based on loss function

CN122017895ACN 122017895 ACN122017895 ACN 122017895ACN-122017895-A

Abstract

The invention discloses a GNSS receiver interference sensitivity threshold calibration method and device based on a loss function, wherein the method comprises the following steps of S1, giving an interference signal type set, a frequency set and a power value set to be tested; S2, measuring a reference frequency spectrum and a measurement frequency spectrum, S3, calculating single-frequency cosine similarity between the measurement frequency spectrum and the reference frequency spectrum, introducing a weighted similarity concept, combining frequency weights to give a power sensitive characteristic index fused with multi-frequency point information, S4, evaluating the performance of a GNSS receiver, S5, constructing a joint loss function, S6, and optimizing to obtain a final interference sensitivity threshold. The invention can carry out iterative calibration on the interference sensitivity threshold based on the classification loss and the threshold alignment loss, and obtains the final interference sensitivity threshold when the joint loss function converges, thereby realizing the calibration of the interference sensitivity threshold and obtaining the accurate interference sensitivity threshold for judging the performance of the GNSS receiver.

Inventors

Yan Puti
ZHOU KAI
Yan mengxi
CAI SHAOXIONG

Assignees

江苏启云飞行科技有限公司

Dates

Publication Date: 20260512
Application Date: 20260130

Claims (8)

1. The GNSS receiver interference sensitivity threshold calibration method based on the loss function is characterized by comprising the following steps: S1, giving an interference signal type set, a frequency set and a power value set to be tested; s2, generating an interference signal to be tested, and performing signal reception and spectrum measurement through a GNSS receiver when the interference generator is turned off and transmits the interference signal to be tested to obtain a reference spectrum and a measurement spectrum; S3, carrying out normalization processing on the interference frequency spectrum and the measurement frequency spectrum, and calculating single-frequency cosine similarity between the measurement frequency spectrum and the non-interference reference frequency spectrum based on a normalization processing result; S4, based on the power sensitive characteristic index, constructing a continuous exponential model to evaluate the performance of the GNSS receiver; S5, constructing a joint loss function based on classification loss, monotonicity loss and threshold alignment loss; s6, carrying out iterative calibration on the interference sensitivity threshold based on the classification loss and the threshold alignment loss, and obtaining a final interference sensitivity threshold when the joint loss function converges.
2. The GNSS receiver performance evaluation method based on joint loss and frequency weight of claim 1, wherein the step S1 comprises: Given that the set of interference signal types to be tested is m= The frequency set is: The power value set is In the power value set, each power is increased from front to back according to the same power step; Wherein N m 、N f 、N P represents the total number of types, the total number of frequencies and the total number of power values of the interference signals to be tested, respectively.
3. The GNSS receiver performance evaluation method based on joint loss and frequency weight of claim 1, wherein the step S2 comprises: Under the conditions of the interference signal type M epsilon M, the interference signal power P epsilon P and the interference signal frequency F epsilon F, generating an interference signal to be tested, transmitting the interference signal through an interference generator, and receiving and measuring a frequency spectrum of a GNSS signal by a GNSS receiver under the interference signal to obtain a measured frequency spectrum ; When the interference generator is turned off, the GNSS receiver receives signals at a plurality of continuous moments and performs spectrum measurement to obtain a spectrum vector under the 'no-interference baseline', which is recorded as a reference spectrum 。
4. The GNSS receiver performance evaluation method based on joint loss and frequency weight of claim 1, wherein the step S3 comprises: S301, measuring frequency spectrum by using L2 norm pair Reference spectrum Normalization processing is carried out to obtain: ; Wherein, the Representing the measured spectrum The result of the normalization process is that, Representing the reference spectrum Carrying out normalization treatment to obtain a result; Represents an L2 norm; S302, calculating single-frequency cosine similarity between a measurement spectrum and a non-interference reference spectrum based on a normalization processing result : ; Wherein, the Representing that the single-frequency cosine similarity between the normalized measurement spectrum and the reference spectrum at the frequency f is equivalent to the inner product thereof; s303, introducing a weighted similarity concept, and combining frequency weights to give a power sensitive characteristic index fused with the multi-frequency point information: In order to reflect the sensitivity of the device to power, firstly, introducing an index penalty term into the single-frequency cosine similarity to obtain single-frequency weighted similarity : ; Order the Representing a power penalty term, wherein An attenuation factor indicating a control attenuation rate; as the power increases, the penalty term grows exponentially, and the weighted similarity decays exponentially, which indicates that for electromagnetic sensitive devices, the performance drops sharply once the power exceeds a threshold; integrating the multi-frequency information by using frequency weighting to obtain a power sensitive characteristic index : ; Wherein, the The value range is [0,1] as the weighting coefficient, and the frequency weight Is that , In order to measure Euclidean distance difference degree between the frequency spectrum and the reference frequency spectrum under the conditions of the interference signal type m, the interference power p and the interference frequency f: 。
5. The GNSS receiver performance evaluation method based on joint loss and frequency weight of claim 1, wherein the step S4 comprises: S401, repeating the steps S1-S3 when the power P is valued and the power set P is traversed, and obtaining corresponding power values ; S402, constructing a continuous index model Using different power p and corresponding Fitting a scaling factor by least squares And attenuation coefficient ; S403, inputting the measured interference signal power p into a continuous exponential model obtained by fitting to obtain Then compared to an interference sensitivity threshold, and if so, the receiver is sensitive.
6. The GNSS receiver performance evaluation method based on joint loss and frequency weight of claim 1, wherein the step S5 comprises: constructing a joint loss function based on classification loss, monotonicity loss, and threshold alignment loss ; Wherein, the 、、 Respectively representing classification loss, monotonicity loss and threshold alignment loss; Weighting coefficients for monotonicity loss; Wherein, the classification loss The method comprises the following steps: ; The score is the GNSS performance evaluation result in the step S4, when the score=1, the evaluation result is that the GNSS receiver is normal, and when the score=0, the evaluation result is that the GNSS receiver is sensitive/faulty; when the interference signal type m, the interference frequency f and the interference power p are traversed through an interference signal type set, a frequency set and a power value set respectively, superposition is carried out according to a formula in the classification loss according to a GNSS performance evaluation result and an actual measurement result, so that the classification loss calculation is completed; Loss of monotonicity The method comprises the following steps: ; In the formula, Representing the expected similarity reduction per 1 dBm increases in power; Threshold alignment loss The method comprises the following steps: Wherein, the As a similarity of the key power points, 、 Similarity of the left side and the right side of the key point power point; is a sensitivity threshold, wherein the critical power point refers to a critical interference power point for sensitivity/failure of the GNSS receiver.
7. The GNSS receiver performance evaluation method based on joint loss and frequency weight according to claim 6, wherein in the step S6, before each iteration, the steps S1-S5 are repeated to obtain a joint loss function, a classification loss and a threshold alignment loss, and the updating process of the t-th iteration is as follows: s601, setting the joint loss obtained by the t-th iteration as At the t-1 th iteration, the joint loss is obtained as Firstly, judging whether the joint loss is converged or not in the t-th iteration, wherein the convergence condition is as follows: ; Representing a convergence threshold; If the joint loss converges, taking the interference sensitivity threshold at that time as an optimized interference sensitivity threshold, and if the joint loss does not meet the convergence condition, advancing step S602; s602, determining the updated value of the interference sensitivity threshold as based on the classification loss and the threshold alignment loss : Wherein, the An updated value representing the interference sensitivity threshold at iteration t-1 is obtained, when t=1, Taking an initial value of 0; is a damping coefficient; the learning rate is used for controlling the step length of single step updating and ensuring convergence stability; s603, on the basis of the current interference sensitivity threshold value, adding The next iteration is then entered until the joint loss converges.
8. The GNSS receiver interference sensitivity threshold calibration device based on the loss function, which is characterized by comprising the following steps: The system comprises a spectrum measuring module, a reference spectrum measuring module, a power value setting module, a frequency setting module and a power value setting module, wherein the spectrum measuring module is used for giving an interference signal type set, a frequency set and a power value set to be tested, generating an interference signal to be tested, and carrying out signal receiving and spectrum measuring through a GNSS receiver when an interference generator is closed and transmits the interference signal to be tested to obtain a reference spectrum and a measuring spectrum; The power sensitive characteristic index calculation module performs normalization processing on the interference frequency spectrum and the measurement frequency spectrum, and calculates single-frequency cosine similarity between the measurement frequency spectrum and the non-interference reference frequency spectrum based on a normalization processing result; the performance evaluation module is used for constructing a continuous exponential model to evaluate the performance of the GNSS receiver based on the power sensitive characteristic index; The loss construction module is used for constructing a joint loss function based on the classification loss, the monotonicity loss and the threshold alignment loss; and the interference sensitivity threshold calibration module is used for carrying out iterative calibration on the interference sensitivity threshold based on the classification loss and the threshold alignment loss, and obtaining a final interference sensitivity threshold when the joint loss function converges.

Description

GNSS receiver interference sensitivity threshold calibration method and device based on loss function Technical Field The present invention relates to the field of GNSS signal reception, and in particular, to a method and apparatus for calibrating an interference sensitivity threshold of a GNSS receiver based on a loss function. Background Global Navigation Satellite Systems (GNSS), represented by the Global Positioning System (GPS), have become a central component of modern infrastructure, playing an irreplaceable role in the key fields of aerospace navigation, autopilot, precision agriculture, emergency response, etc. However, the sensitivity of satellite navigation receivers to electromagnetic interference (EMI) poses a serious challenge for their functional performance. In particular, with the popularity of wireless devices and the increasing popularity of jamming, GPS L1 band signals are threatened by a variety of jamming, including Continuous Wave (CW), pulse modulation and BPSK modulation, which can significantly affect the performance of the receiver, and thus, the assessment of GNSS receiver performance is very important. However, there are some key problems in the response mechanism of the receiver rf front end in the complex electromagnetic environment, firstly, the conventional signal-to-interference-and-noise ratio (SINR) evaluation framework is difficult to quantify the spectrum threshold distribution modes of different modulation types of the intermediate frequency output related to the navigation fault, which lacks a general theoretical framework, limits the general design of the anti-interference receiver, and for the interference sensitivity threshold, which is generally a determined value, lacks a flexible calibration scheme, and is unfavorable for high-precision performance evaluation of the GNSS receiver. Disclosure of Invention The invention aims to overcome the defects of the prior art, and provides a GNSS receiver interference sensitivity threshold calibration method and device based on a loss function, which can perform iterative calibration on the interference sensitivity threshold based on classification loss and threshold alignment loss, and obtain a final interference sensitivity threshold when a joint loss function converges, thereby realizing the calibration of the interference sensitivity threshold and obtaining the interference sensitivity threshold with high precision. The invention aims at realizing the technical scheme that the GNSS receiver interference sensitivity threshold calibration method based on the loss function comprises the following steps: S1, giving an interference signal type set, a frequency set and a power value set to be tested; s2, generating an interference signal to be tested, and performing signal reception and spectrum measurement through a GNSS receiver when the interference generator is turned off and transmits the interference signal to be tested to obtain a reference spectrum and a measurement spectrum; S3, carrying out normalization processing on the interference frequency spectrum and the measurement frequency spectrum, and calculating single-frequency cosine similarity between the measurement frequency spectrum and the non-interference reference frequency spectrum based on a normalization processing result; S4, based on the power sensitive characteristic index, constructing a continuous exponential model to evaluate the performance of the GNSS receiver; S5, constructing a joint loss function based on classification loss, monotonicity loss and threshold alignment loss; s6, carrying out iterative calibration on the interference sensitivity threshold based on the classification loss and the threshold alignment loss, and obtaining a final interference sensitivity threshold when the joint loss function converges. A loss function based GNSS receiver interference sensitivity threshold calibration apparatus comprising: The system comprises a spectrum measuring module, a reference spectrum measuring module, a power value setting module, a frequency setting module and a power value setting module, wherein the spectrum measuring module is used for giving an interference signal type set, a frequency set and a power value set to be tested, generating an interference signal to be tested, and carrying out signal receiving and spectrum measuring through a GNSS receiver when an interference generator is closed and transmits the interference signal to be tested to obtain a reference spectrum and a measuring spectrum; The power sensitive characteristic index calculation module performs normalization processing on the interference frequency spectrum and the measurement frequency spectrum, and calculates single-frequency cosine similarity between the measurement frequency spectrum and the non-interference reference frequency spectrum based on a normalization processing result; the performance evaluation module is used for constructing a continuous exponential mode