Indoor pseudo satellite signal carrier-to-noise ratio positioning method and system based on commercial smart phone
By employing Bayesian inference and alternating optimization iterative estimation strategies, the problem of carrier-to-noise ratio observations in indoor pseudo-satellite positioning by commercial smartphones being affected by base station density and equipment performance was solved, achieving high-precision and high-reliability indoor positioning services.
Patent Information
- Application Number
- CN202511562113.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-10-29
- Publication Date
- 2026-03-20
AI Technical Summary
In existing technologies, commercial smartphones suffer from low positioning accuracy in indoor pseudo-satellite positioning due to the susceptibility of carrier-to-noise ratio observations to differences in base station deployment density and equipment performance. Traditional methods also fail or lack sufficient positioning reliability in indoor environments.
An alternating optimization iterative estimation strategy based on Bayesian inference is adopted. By establishing an observation model between the target and the base station, and utilizing the free space attenuation model and the normal distribution assumption, combined with maximum a posteriori estimation and alternating optimization iterative algorithm, the carrier-to-noise ratio observation is optimized to improve positioning accuracy.
It effectively mitigates performance differences between devices, improves indoor positioning accuracy by 30%, provides highly reliable location services, is suitable for smartphones with different performance characteristics, and reduces the time synchronization requirements for pseudo-satellite system deployment.
Smart Images

Figure CN121703757A_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the field of pseudo-satellite positioning technology, and particularly relates to an indoor pseudo-satellite signal carrier-to-noise ratio positioning method and system based on commercial smartphones. Background Technology
[0002] Pseudosatellite positioning technology, as a positioning technique compatible with existing Global Navigation Satellite Systems (GNSS), has been applied in various scenarios with poor GNSS observation conditions to provide regional-level location enhancement services. However, in complex indoor topological environments and with significant differences in the performance of commercial smartphone receivers, it suffers from core drawbacks in practical applications, including the lack of reliable ranging-related observations and the inability of outdoor GNSS positioning methods to be used in indoor scenarios. Specific problems are as follows:
[0003] Due to the difficulty in synchronizing clocks in pseudosatellite systems, the significantly lower stability of clock sources compared to atomic clocks on GNSS satellite payloads, and the impact of indoor multipath propagation and environmental noise on signals, using pseudosatellite pseudorange observations for single-point positioning is unreliable. Achieving high-precision pseudorange measurement often requires solving special signal processing algorithms such as multipath error elimination, which poses a serious challenge to low-cost GNSS chips and antennas used in smartphones. Furthermore, carrier phase observations themselves lack ranging capabilities, and the LAMBDA (Least-squares Ambiguity Decorrelation Adjustment) algorithm for converging integer ambiguities requires the satellite to be in high-speed motion. However, pseudosatellite base stations are typically located in fixed indoor positions, making them unsuitable for solving integer ambiguities. Additionally, some smartphones on the market no longer support carrier phase observation output. Therefore, pseudosatellite pseudorange and carrier phase observations obtained from smartphones are limited by various factors and are unsuitable for indoor positioning.
[0004] The carrier-to-noise ratio (C / N0) observation output by a smartphone is used to reflect the signal quality of outdoor GNSS. Similar to the Received Signal Strength Index (RSSI) observation of wireless signals such as Wi-Fi and Bluetooth in traditional indoor positioning technologies, it reflects the distance relationship between the smartphone and pseudo-satellite base stations in indoor scenarios, i.e., the C / N0 attenuation characteristics. Furthermore, the C / N0 observation is only related to the signal strength received by the smartphone and does not strictly require time synchronization of each pseudo-satellite base station. However, due to differences in signal systems and corresponding signal processing methods, the strength of pseudo-satellite signals received by the smartphone is severely affected by co-channel interference. This leads to problems such as varying deployment densities affecting the C / N0 attenuation coefficient when facing multiple pseudo-satellite base stations, and severe damage to attenuation characteristics in near-far effect scenarios. These problems cause traditional signal strength attenuation models to fail or lose their universality, and result in low positioning accuracy using traditional weighted centroid positioning methods. While fingerprint positioning technology achieves location matching by constructing a signal feature database and has shown high-precision positioning potential in some controlled environments, its core shortcomings lie in insufficient environmental adaptability, high maintenance costs, and significant performance differences between different devices, making it similarly lacking in universality.
[0005] Therefore, developing an indoor pseudosatellite signal carrier-to-noise ratio (CNR) positioning method for commercial smartphones has become an urgent need to overcome existing technological bottlenecks. In current indoor environments, the density of different base stations is limited by objective environmental deployment conditions, and there are complex factors such as performance differences between different smartphones. Therefore, there is a pressing need for an innovative ranging and positioning solution that can respond in real-time to changes in the attenuation system and mitigate differences between devices.
[0006] Based on the above analysis, the problems and shortcomings of the existing technology are as follows:
[0007] Current pseudo-satellite carrier-to-noise ratio (CNR) observations acquired by smartphones are susceptible to co-channel interference caused by differences in base station deployment density, as well as variations in attenuation coefficients due to differences in device performance. In indoor positioning methods based on smartphone wireless signal RSSI, traditional positioning methods that construct attenuation models have significant limitations. While traditional weighted centroid positioning methods are relatively universal, their positioning accuracy and reliability are difficult to guarantee, often requiring high-density base station deployment to improve accuracy. Fingerprint positioning methods require independent database construction for each scenario, resulting in insufficient scenario adaptability and high costs for database construction and long-term maintenance. Existing technology verification systems lack adaptability testing for general scenarios and general equipment, leading to insufficient reliability in practical positioning applications. Summary of the Invention
[0008] To address the problems existing in the prior art, this invention provides an indoor pseudo-satellite signal carrier-to-noise ratio positioning method based on commercial smartphones.
[0009] This invention is implemented as follows: An indoor pseudo-satellite signal carrier-to-noise ratio positioning method based on a commercial smartphone includes:
[0010] Step 1: Establishment of the observation model and joint probability density function between the target and the base station;
[0011] Step 2: Bayesian inference and maximum a posteriori estimation;
[0012] Step 3: Alternate optimization iterative estimation strategy.
[0013] Furthermore, the observation model and joint probability density function between the target and the base station are established as follows:
[0014] First, an observation model is established between the distance between the target location and the base station location and the carrier-to-noise ratio. Here, it is assumed that this model always conforms to a free-space attenuation model with specific parameters. The target user location is assumed to be... , No. The location of each base station is The Euclidean distance between the two can be written as For ease of writing, let the corresponding carrier-to-noise ratio observation be... , can be written as
[0015]
[0016] In the formula and The meaning is the same as in the previous formula. To observe the noise, assume For a given and It follows a mean and variance of 0. The normal distribution, i.e. It should be emphasized that only the given... and ,for This is a constant, representing the maximum output value of the carrier-to-noise ratio (CNR) observation from the smartphone, which can be obtained during indoor or outdoor measurements; therefore, the likelihood function for a single CNR observation can be written as...
[0017]
[0018]
[0019] Assuming that what is observed at this time The carrier-to-noise ratio (CNR) observations of each pseudosatellite base station are independent of each other. The joint likelihood is the product of the likelihoods of each observation. Considering that the variance increases with distance at different distances, the variances from different base stations are taken into account when multiplying the likelihoods of each observation. , written as
[0020]
[0021]
[0022] In the formula Represents what the smartphone received A list of carrier-to-noise ratios. When reflecting the variance of measurement noise, it is considered to be related to the carrier-to-noise ratio observation. It is negatively correlated and can be designed as
[0023]
[0024] In the formula Represented as the baseline variance; This represents a smoothing parameter used to prevent the denominator from being zero; therefore, it is a number greater than 0. This is the lower limit of intensity.
[0025] Furthermore, the Bayesian inference and maximum a posteriori estimation:
[0026] Based on the likelihood function derived from the above equation, and using Bayesian inference, the joint posterior distribution can be written as:
[0027]
[0028] In the above formula, it is assumed that the target location is prior. Prior path loss coefficient The prior terms are flat and independent; therefore, for the joint posterior distribution, we find... The parameters maximize the posterior probability, i.e., the Maximum A Posteriori (MAP) estimate. By taking the logarithm of the posterior probability while maintaining monotonicity, and ignoring the constant term and the flat prior part, we obtain the objective function, which can be simplified as follows:
[0029]
[0030] That is, it is equivalent to minimizing the negative log-likelihood, where
[0031]
[0032] Therefore, MAP can ultimately be transformed into the form of minimizing the squared error, after simplification
[0033]
[0034] As the objective function.
[0035] Furthermore, the alternating optimization iterative estimation strategy:
[0036] Since the objective function of the above equation is about and Since it is non-convex, direct joint optimization is difficult. Therefore, this technique adopts an alternating optimization strategy:
[0037] (1) For fixed ,optimization ; and then through the Calculate the gradient by taking the derivative, and then optimize using the gradient descent method. , written as
[0038]
[0039]
[0040] actually For learning rate, Represents the iteration count;
[0041] (2) By fixing ,optimization ; and then through the By taking the derivative and setting it to zero, we can obtain the closed-form solution as follows:
[0042]
[0043]
[0044] Based on the above formula, firstly... and Perform initialization, and iterate to the... In this process, with the current fixed... From gradient descent Updated to and fix Updated via closed-form solution until satisfied
[0045]
[0046] For alternating optimization iterative estimation of output and As a result, respectively through and A threshold is set to determine whether the iterative process has converged. This indicates the maximum number of iterations, limiting the total number of iterations to avoid getting stuck in a loop on non-converged results.
[0047] Another objective of this invention is to provide an indoor pseudo-satellite signal carrier-to-noise ratio positioning system based on a commercial smartphone, comprising:
[0048] The function establishment module is used to establish the observation model and joint probability density function between the target and the base station;
[0049] The estimation module is used for Bayesian inference and maximum a posteriori estimation;
[0050] The optimization module is used to alternately optimize the iterative estimation strategy.
[0051] Another object of the present invention is to provide a computer device including a memory and a processor, the memory storing a computer program, which, when executed by the processor, causes the processor to perform the steps of the indoor pseudo-satellite signal carrier-to-noise ratio positioning method based on a commercial smartphone.
[0052] Another object of the present invention is to provide a computer-readable storage medium storing a computer program, which, when executed by a processor, causes the processor to perform the steps of the indoor pseudo-satellite signal carrier-to-noise ratio positioning method based on a commercial smartphone.
[0053] Another objective of this invention is to provide an information data processing terminal for implementing the indoor pseudo-satellite signal carrier-to-noise ratio positioning system based on a commercial smartphone.
[0054] Based on the above technical solutions and the technical problems solved, the advantages and positive effects of the technical solution to be protected by this invention are as follows:
[0055] This invention addresses the aforementioned core issues through innovative design. It proposes a positioning technique based on Bayesian inference and alternating optimization dynamic estimation. Based on the attenuation relationship between the observed carrier-to-noise ratio (CNR) and the distance to the base station, the maximum a posteriori (MAP) estimate is calculated using Bayesian inference. An alternating estimation strategy is then employed to iteratively optimize the attenuation coefficient and location until they converge. This effectively addresses variations in the attenuation coefficient and mitigates performance differences between smartphones, resulting in a 30% improvement in average positioning accuracy compared to the traditional weighted centroid method. The proposed technique is applicable to smartphones with varying performance levels and can be used in the same way. This technology provides a universal solution for high-reliability positioning of smartphones under co-channel interference conditions.
[0056] (1) The expected benefits and commercial value of the technical solution of this invention after transformation are as follows:
[0057] The technology of this invention can effectively realize reliable indoor location services. Since the positioning terminal is a smartphone, it also has the characteristics of high availability. After the technical solution is transformed, it can fully provide location services for a series of outdoor GNSS-inaccessible scenarios such as precise navigation marketing in shopping malls and car finding in underground parking garages, thereby improving user experience and generating commercial value.
[0058] (2) The technical solution of this invention fills a technical gap in the industry both domestically and internationally:
[0059] It abandons the pseudo-satellite technology solutions based on repeater strategies commonly used in existing commercial applications both domestically and internationally, reduces the time synchronization accuracy requirements for the entire pseudo-satellite system deployment, and improves the accuracy of indoor positioning and navigation. Attached Figure Description
[0060] Figure 1 This is a flowchart of an indoor pseudo-satellite signal carrier-to-noise ratio positioning method based on a commercial smartphone, provided in an embodiment of the present invention.
[0061] Figure 2 This is a block diagram of an indoor pseudo-satellite signal carrier-to-noise ratio positioning system based on a commercial smartphone, provided in an embodiment of the present invention.
[0062] Figure 3 This is a flowchart of the technical solution provided in the embodiments of the present invention.
[0063] Figure 4 This is a schematic diagram of the underground parking garage passage scene and location acquisition provided in an embodiment of the present invention.
[0064] Figure 5 This is a statistical result graph showing the carrier-to-noise ratio attenuation characteristics of different smartphones under different deployment distances, as provided in the embodiments of the present invention.
[0065] Figure 6 This is a scenario experiment design diagram provided in an embodiment of the present invention.
[0066] Figure 7 This is a comparison chart of positioning errors of different smartphones using different algorithms under different deployment densities, provided by an embodiment of the present invention. Detailed Implementation
[0067] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.
[0068] like Figure 1 As shown in the figure, an indoor pseudo-satellite signal carrier-to-noise ratio positioning method based on a commercial smartphone provided by an embodiment of the present invention includes the following steps:
[0069] S101: Establishment of the observation model and joint probability density function between the target and the base station;
[0070] S102: Bayesian inference and maximum a posteriori estimation;
[0071] S103: Alternating optimization iterative estimation strategy;
[0072] S104: Output of estimated location results.
[0073] The observation model establishment and joint probability density function between the target and the base station provided in this embodiment of the invention are as follows:
[0074] First, an observation model is established between the distance between the target location and the base station location and the carrier-to-noise ratio. Here, it is assumed that this model always conforms to a free-space attenuation model with specific parameters. The target user location is assumed to be... , No. The location of each base station is The Euclidean distance between the two can be written as For ease of writing, let the corresponding carrier-to-noise ratio observation be... , can be written as
[0075]
[0076] In the formula and The meaning is the same as in the previous formula. To observe the noise, assume For a given and It follows a mean and variance of 0. The normal distribution, i.e. It should be emphasized that only the given... and ,for This is a constant, representing the maximum output value of the carrier-to-noise ratio (CNR) observation from the smartphone, which can be obtained during indoor or outdoor measurements; therefore, the likelihood function for a single CNR observation can be written as...
[0077]
[0078]
[0079] Assuming that what is observed at this time The carrier-to-noise ratio (CNR) observations of each pseudosatellite base station are independent of each other. The joint likelihood is the product of the likelihoods of each observation. Considering that the variance increases with distance at different distances, the variances from different base stations are taken into account when multiplying the likelihoods of each observation. , written as
[0080]
[0081]
[0082] In the formula Represents what the smartphone received A list of carrier-to-noise ratios. When reflecting the variance of measurement noise, it is considered to be related to the carrier-to-noise ratio observation. It is negatively correlated and can be designed as
[0083]
[0084] In the formula Represented as the baseline variance; This represents a smoothing parameter used to prevent the denominator from being zero; therefore, it is a number greater than 0. This is the lower limit of intensity.
[0085] Bayesian inference and maximum a posteriori estimation provided in this embodiment of the invention:
[0086] Based on the likelihood function derived from the above equation, and using Bayesian inference, the joint posterior distribution can be written as:
[0087]
[0088] In the above formula, it is assumed that the target location is prior. Prior path loss coefficient The prior terms are flat and independent; therefore, for the joint posterior distribution, we find... The parameters maximize the posterior probability, i.e., the Maximum A Posteriori (MAP) estimate. By taking the logarithm of the posterior probability while maintaining monotonicity, and ignoring the constant term and the flat prior part, we obtain the objective function, which can be simplified as follows:
[0089]
[0090] That is, it is equivalent to minimizing the negative log-likelihood, where
[0091]
[0092] Therefore, MAP can ultimately be transformed into the form of minimizing the squared error, after simplification
[0093]
[0094] As the objective function.
[0095] The alternating optimization iterative estimation strategy provided in this embodiment of the invention:
[0096] Since the objective function of the above equation is about and Since it is non-convex, direct joint optimization is difficult. Therefore, this technique adopts an alternating optimization strategy:
[0097] (1) For fixed ,optimization ; and then through the Calculate the gradient by taking the derivative, and then optimize using the gradient descent method. , written as
[0098]
[0099]
[0100] actually For learning rate, Represents the iteration count;
[0101] (2) By fixing ,optimization ; and then through the By taking the derivative and setting it to zero, we can obtain the closed-form solution as follows:
[0102]
[0103]
[0104] Based on the above formula, firstly... and Perform initialization, and iterate to the... In this process, with the current fixed... From gradient descent Updated to and fix Updated via closed-form solution until satisfied
[0105]
[0106] For alternating optimization iterative estimation of output and As a result, respectively through and A threshold is set to determine whether the iterative process has converged. This indicates the maximum number of iterations, limiting the total number of iterations to avoid getting stuck in a loop on non-converged results.
[0107] like Figure 2 As shown in the figure, an indoor pseudo-satellite signal carrier-to-noise ratio positioning system based on a commercial smartphone provided by an embodiment of the present invention includes:
[0108] The function establishment module is used to establish the observation model and joint probability density function between the target and the base station;
[0109] The estimation module is used for Bayesian inference and maximum a posteriori estimation;
[0110] The optimization module is used to alternately optimize the iterative estimation strategy.
[0111] Another object of the present invention is to provide a computer device including a memory and a processor, the memory storing a computer program, which, when executed by the processor, causes the processor to perform the steps of the indoor pseudo-satellite signal carrier-to-noise ratio positioning method based on a commercial smartphone.
[0112] Another object of the present invention is to provide a computer-readable storage medium storing a computer program, which, when executed by a processor, causes the processor to perform the steps of the indoor pseudo-satellite signal carrier-to-noise ratio positioning method based on a commercial smartphone.
[0113] Another objective of this invention is to provide an information data processing terminal for implementing the indoor pseudo-satellite signal carrier-to-noise ratio positioning system based on a commercial smartphone.
[0114] Specific implementation of the present invention:
[0115] A technical solution flow for an indoor pseudo-satellite signal carrier-to-noise ratio positioning method based on commercial smartphones Figure 3 .
[0116] 1. Technical solutions of embodiments of the present invention
[0117] (1) Platform construction
[0118] Considering the significant differences in pseudosatellite deployment density and smartphone device performance in real-world environments, both of which lead to variations in carrier-to-noise ratio attenuation coefficients, three pseudosatellite system platforms with spacings (15m / 25m / 75m) were constructed in a 75m passageway scenario within a large parking lot. Data was collected at 5m intervals. The data collection environment for the 75m spacing deployment scenario is shown below.
[0119] like Figure 4 As shown, in a passageway scene inside a large parking lot, two smartphones with different performance were used for data collection. The data collection points were spaced 5 meters apart, and each data collection point took 3 minutes to collect data.
[0120] (2) Test statistics
[0121] Using separate 15m / 25m / 75m pseudosatellite system platforms, the results collected by two smartphones with different performance characteristics were summarized, analyzed, and plotted. Figure 5 Compare them.
[0122] like Figure 5As shown, the statistical results of the carrier-to-noise ratio (CNR) observations for two smartphones, Device 1 and Device 2, are plotted separately. For observations at a 75-meter interval, Device 1 and Device 2 are represented by solid blue circles and solid orange pentagrams, respectively, to compare the performance differences between different smartphones. For results with denser deployments, the same legend is used: statistical results at a 25-meter interval are represented by yellow dots and dashed lines, and statistical results at a 15-meter interval are represented by purple dots and dashed lines. The variance of the results collected at each point is also represented by rectangles plotted at the mean. It should be noted that to prevent the variance results from being overwritten, the distance of the independent variable in the 25-meter interval results is misaligned, which is not a statistical error.
[0123] from Figure 5 It can be seen that as the deployment density of base stations increases, the relationship between carrier-to-noise ratio (CNR) and distance also changes, with the decreasing trend of CNR and distance becoming increasingly apparent. Figure 5 In the results using device 1 (a) as an example, the attenuation differences between the 75-meter and 25-meter spacing are quite significant. At 1 meter, the difference is small and almost equal, but at 10 meters, the carrier-to-noise ratio (CNR) difference reaches 3-4 dBHz; at 20 meters, the difference reaches 6-7 dBHz; the difference then gradually increases until it disappears after 40-50 meters, after which the randomness increases at further distances, generally remaining above 10 dBHz. In the results for the 25-meter and 15-meter spacings, the difference is not particularly significant, with the attenuation decreasing slightly less at 15 meters than at 25 meters. It is also worth noting that the 15-meter spacing results show a significant decrease at 15 and 30 meters. This is because the CNR decreases considerably after passing through intermediate base stations 2 and 3, disrupting the established attenuation characteristics. Figure 5 (b) The phenomenon represented by device 2 is very similar to that of device 1, but because it processes the signal better, its carrier-to-noise ratio gain is greater, resulting in different distance resolution capabilities. This proves that a single attenuation coefficient cannot be used in these scenarios.
[0124] (3) Experimental design
[0125] Motion experiments were designed for deployment scenarios with spacings of 15 meters and 25 meters, respectively. Figure 6 As shown.
[0126] for Figure 6In scenario 1 of (a), starting from position TX1, the solid line represents moving 10 meters eastward, the dashed line represents moving westward back to the starting point, and then moving eastward again. Each movement is 10 meters longer than the previous one, and this back-and-forth movement is repeated 6 times. The lengths of the last two movements are 55 meters and 65 meters, respectively. Figure 6 In scenario 2 of (b), starting from position TX1, the solid line indicates alternating movements of 5 meters and 10 meters eastward, while the dashed line indicates movements westward back to the starting point, followed by another eastward movement. This back-and-forth movement is repeated 6 times, resulting in subsequent movement distances of 5, 10, 20, 25, 35, and 40 meters. After 6 back-and-forth movements, the user eventually returns to the starting point.
[0127] 2. Technical Effects of the Embodiments of the Invention
[0128] The invention examples compare and evaluate the proposed method using traditional weighted centroid methods with different smartphones in two scenarios. By analyzing the positioning error of an indoor pseudo-satellite carrier-to-noise ratio positioning method based on commercial smartphones, the cumulative error distribution function (CDF) is statistically analyzed and plotted. Figure 7 As shown.
[0129] like Figure 7 As shown, the sub Figure 7 (a) and (c) represent the CDF (Cut-Off Value) in Scenario 1 using different algorithms on different mobile phones. Different smartphones using the optimal weighted positioning algorithm can achieve positioning accuracy within 67% of 3.9 meters and 4.3 meters, and within 95% of 8.4 meters and 9.1 meters, respectively. The proposed algorithm, however, can achieve positioning accuracy within 67% of 2.4 meters and 3.5 meters, and within 95% of 5.8 meters and 6.2 meters, respectively, showing improvements in Scenario 1. The sub-... Figure 7 (b) and (d) represent the CDF of different mobile phones using different algorithms in scenario 2. The optimal weighted positioning algorithm can provide positioning accuracy within 67% of 3.4 meters and 3.0 meters, and within 95% of 8.0 meters and 6.5 meters. The proposed algorithm can provide positioning accuracy within 67% of 2.0 meters and 2.2 meters, and within 95% of 4.9 meters and 4.1 meters, respectively, which is also improved in scenario 2.
[0130] Example 1: Indoor Pseudo-Satellite Signal Carrier-to-Noise Ratio Positioning Method Based on Commercial Smartphones
[0131] In this embodiment, the observation model between the target and the base stations adopts a free-space attenuation model. First, the carrier-to-noise ratio (C / N0) of pseudosatellite signals from multiple base stations is collected using a smartphone. Based on the distance between the target and the base stations, a free-space path loss model is used to model the distance between the target location Y and each base station, and the corresponding C / N0. Next, based on this observation model, assuming that the observation noise follows a normal distribution, a joint likelihood function is further derived. The likelihood function is used to calculate the estimated values of the target location and path loss coefficient, and the optimal solution is determined through Bayesian inference and maximum a posteriori estimation (MAP) methods.
[0132] To further optimize positioning accuracy, this embodiment employs an alternating optimization strategy to iteratively optimize the target position and path loss coefficient. During the iteration process, the path loss coefficient α is first fixed, and the target position Y is optimized using gradient descent, calculating the gradient for each iteration and updating the value of Y. Subsequently, the target position Y is fixed, and the path loss coefficient α is solved using a closed-form solution. This process is repeated until the changes in the target position Y and the path loss coefficient α are less than a set threshold, at which point convergence is determined, and the optimal position estimate of the target is finally output.
[0133] Example 2: Application of Bayesian Inference and Maximum A posteriori Estimation
[0134] This embodiment addresses the indoor positioning problem based on Bayesian inference. First, based on the carrier-to-noise ratio (CNR) data of multi-base station pseudosatellite signals received in an indoor environment, the likelihood function for each CNR observation is calculated using an observation model of the target and base stations. Then, by combining the prior distributions of the target location and path loss coefficients, a joint posterior distribution of the target location and path loss coefficients is obtained. Based on this, the negative log-likelihood function is minimized using the maximum a posteriori (MAP) estimation method to obtain the optimal target location estimate and path loss coefficient estimate.
[0135] To enhance positioning accuracy, this embodiment employs the Weighted Squared Error (WSE) method as the objective function. In each iteration, the differences in observation noise between different base stations are considered, and the target position Y and path loss coefficient α are progressively updated by combining the carrier-to-noise ratio observations of each base station. Ultimately, by maximizing the posterior probability, a more accurate positioning result is obtained. Furthermore, combined with real indoor test data, it is demonstrated that this method can effectively reduce positioning errors and improve the robustness of the positioning system.
[0136] Example 3: Implementation of the Alternating Optimization Algorithm
[0137] This embodiment implements an alternating optimization algorithm to solve for the target position Y and the path loss coefficient α. In practical applications, the relationship between the target position Y and the path loss coefficient α is non-linear, making direct joint optimization difficult. Therefore, an alternating optimization strategy is adopted for iterative optimization. First, the path loss coefficient α is fixed, and the target position Y is optimized. By optimizing the error function of the target position using gradient descent, the gradient is calculated in each round, and the target position is updated, gradually approaching the optimal solution. Then, the target position Y is fixed, and the closed-form solution of the path loss coefficient α is calculated to further improve the estimation accuracy of the path loss coefficient.
[0138] In each iteration, a dynamically adjusted learning rate η is used to control the optimization pace and ensure algorithm convergence. In real-world scenarios, convergence is determined when the changes in target position and path loss coefficients are less than preset thresholds. Furthermore, a maximum number of iterations T_max is set to prevent the algorithm from getting stuck in a non-convergent loop. Experimental results show that alternating optimization can effectively improve positioning accuracy, especially in indoor environments with weak signal strength, where the positioning error is smaller.
[0139] Example 4: Base station layout and data acquisition in an indoor positioning system
[0140] This embodiment describes how to deploy pseudosatellite signal base stations in an indoor environment and collect the carrier-to-noise ratio (CNR) data required for positioning, targeting an indoor smartphone. A multi-base station deployment is employed to ensure sufficient signal coverage in different locations. The signal strength of each base station is optimized through indoor environment modeling to ensure optimal signal quality between each base station and the smartphone. The smartphone, as the terminal device, collects the received pseudosatellite signal CNR data in real time and transmits the data wirelessly to a server for further processing.
[0141] During data processing, the smartphone performs preliminary processing on the collected carrier-to-noise ratio (CNR) data using local algorithms to remove noise interference and calculates the distance between each base station and the target. Based on this data, the target location and path loss coefficient are estimated using the aforementioned positioning algorithm. Throughout this process, considering signal attenuation and noise characteristics, the system can dynamically assess signal quality and optimize positioning accuracy.
[0142] Example 5: Positioning System in Real-World Application Scenarios
[0143] This embodiment describes the deployment of indoor pseudosatellite signal carrier-to-noise ratio (CNR) positioning technology based on commercial smartphones in a practical application scenario. Considering the complex indoor environment and limited signal propagation, a multi-base station layout is adopted to cover the pseudosatellite signal. By modeling the relationship between the base station CNR observation data and the target location, a Bayesian inference method is used to estimate the target location, and the path loss coefficient is optimized using a maximum a posteriori (MAP) estimation algorithm. The system acquires signal data in real time via smartphones, uses optimization algorithms for positioning estimation, and ultimately outputs an accurate target location.
[0144] This technology can be widely applied in indoor spaces such as smart buildings, museums, and hospitals. Since smartphones themselves have positioning and signal reception capabilities, and no additional hardware support is required, the deployment cost of the system is significantly reduced. Furthermore, because the system can adapt to changes in the indoor environment, it exhibits strong robustness and accuracy. Real-world case studies have demonstrated that the system performs well in various complex indoor environments, with a positioning error of less than 1 meter, meeting the requirements for high-precision positioning.
[0145] Example 6: Estimation of target location and path loss coefficient
[0146] This embodiment focuses on an indoor positioning system based on a commercial smartphone, estimating the target location and path loss coefficient through an optimized algorithm. In the initial data acquisition phase, the smartphone acquires carrier-to-noise ratio (CNR) observation data via its built-in sensors and received pseudosatellite signals. Based on the distance between the target and the base station and the CNR observations, the system uses Bayesian inference to establish a joint probability distribution model of the target location and path loss coefficient, and obtains the optimal solution using the maximum a posteriori (MAP) estimation method.
[0147] Subsequently, an alternating optimization strategy was employed to optimize the target location and path loss coefficient. During the iteration process, with the path loss coefficient fixed, gradient descent was used to optimize the target location and update the estimated target location; then, with the target location fixed, the closed-form solution was used to update the path loss coefficient. Finally, accurate path loss coefficients and corresponding target location estimates were obtained. Experimental results show that this method can effectively reduce positioning errors and is particularly suitable for indoor environments with severe signal attenuation.
[0148] It should be noted that embodiments of the present invention can be implemented in hardware, software, or a combination of both. The hardware portion can be implemented using dedicated logic; the software portion can be stored in memory and executed by a suitable instruction execution system, such as a microprocessor or dedicated-design hardware. Those skilled in the art will understand that the above-described devices and methods can be implemented using computer-executable instructions and / or included in processor control code, for example, such code provided on a carrier medium such as a disk, CD, or DVD-ROM, a programmable memory such as read-only memory (firmware), or a data carrier such as an optical or electronic signal carrier. The devices and modules of the present invention can be implemented by hardware circuitry such as very large-scale integrated circuits or gate arrays, semiconductors such as logic chips, transistors, or programmable hardware devices such as field-programmable gate arrays, programmable logic devices, etc., or by software executed by various types of processors, or by a combination of the above-described hardware circuitry and software, such as firmware.
[0149] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any modifications, equivalent substitutions, and improvements made by those skilled in the art within the scope of the technology disclosed in the present invention, and within the spirit and principles of the present invention, should be covered within the scope of protection of the present invention.
Claims
1. A method for indoor pseudo-satellite signal carrier-to-noise ratio positioning based on commercial smartphones, characterized in that, Includes the following steps: Establish an observation model between the target and the base stations. Let the location of the target user be a two-dimensional coordinate point and the location of the i-th base station be a fixed coordinate point. Calculate the Euclidean distance between the target and each base station. Using the carrier-to-noise ratio (CNR) observations received by each base station as input data, a functional relationship between CNR and distance is established. This relationship includes a signal power constant term, a path loss coefficient term, and an observation noise term that follows a zero-mean normal distribution. Based on the above model, a joint probability density function for target location and path loss coefficient is constructed; By employing Bayesian inference and maximum a posteriori estimation, the optimal estimates of the target location and path loss coefficient are obtained by maximizing the posterior probability. An alternating optimization iterative strategy is adopted to optimize the target position and path loss coefficient separately until the convergence condition is met, and then the target position result is output.
2. The positioning method as described in claim 1, characterized in that, The observation model conforms to the free space signal attenuation law. The observed carrier-to-noise ratio has a logarithmic attenuation relationship with the distance from the target to the base station. The signal power constant term is a fixed constant, the path loss coefficient is a parameter to be estimated, and the observation noise is a normally distributed random variable with a mean of zero.
3. The positioning method as described in claim 1, characterized in that, In the process of establishing the joint probability density function, it is assumed that the prior distributions of the target position and the path loss coefficient are independent of each other, and the joint posterior distribution is represented by the product of the observation likelihood function and their respective prior probabilities.
4. The positioning method as described in claim 1, characterized in that, The maximum a posteriori estimation is achieved by minimizing the weighted sum of squared errors between the observed data and the model output, and is used to simultaneously solve for the optimal combination of target location and path loss coefficient.
5. The positioning method as described in claim 1, characterized in that, The alternating optimization iterative strategy includes: (1) With the path loss coefficient fixed, the target position parameters are updated using the gradient descent method, so that the objective function is corrected along the error descent direction; (2) When the target position is fixed, the path loss coefficient is updated by analytical closed-form solution to minimize the sum of squared errors; (3) Repeat the above two update operations until the changes in the target location and path loss coefficient are both less than the preset threshold.
6. The positioning method as described in claim 1, characterized in that, The convergence condition determination includes: when the difference between the target positions in two consecutive iterations is less than a first preset threshold and the difference in the path loss coefficient is less than a second preset threshold, the iteration is stopped and the final positioning result is output.
7. An indoor pseudo-satellite signal carrier-to-noise ratio positioning system based on a commercial smartphone, characterized in that, The system includes: The data acquisition module is used to acquire carrier-to-noise ratio observation data of pseudosatellite signals from smartphones; The function establishment module is used to establish an observation model based on the distance and carrier-to-noise ratio observation data between the target and the base station, and to calculate the joint probability density of the target location and path loss coefficient. The estimation module is used to perform Bayesian inference and maximum a posteriori estimation to obtain initial estimates of the target location and path loss coefficient. The optimization module is used to alternately optimize the estimated values of the target location and path loss coefficient, and iteratively optimizes the solution using gradient descent and closed-form solutions. The convergence determination module is used to determine whether the iterative process meets the preset convergence conditions. If it does, the final estimation result is output.
8. The positioning system as described in claim 7, characterized in that, The function establishment module includes a noise modeling unit, which is used to assume that the carrier-to-noise ratio observation noise follows a normal distribution with a mean of zero and a fixed variance, and to calculate the observation error variance.
9. The positioning system as described in claim 7, characterized in that, The optimization module adopts an alternating optimization mechanism, which updates the target position when the path loss coefficient is fixed, solves the path loss coefficient when the target position is fixed, and sends the parameter change amount to the convergence determination module after each update to determine whether to continue iterating.
10. The positioning system as described in claim 7, characterized in that, The system is deployed in the local processing unit of a commercial smartphone. The carrier-to-noise ratio data is obtained by the phone's built-in sensors or communication interface, and the positioning results are calculated locally and displayed in real time on the terminal interface.
