CN102455421B - Sound positioning system and method without time synchronization - Google Patents
Sound positioning system and method without time synchronization Download PDFInfo
- Publication number
- CN102455421B CN102455421B CN201010521203.7A CN201010521203A CN102455421B CN 102455421 B CN102455421 B CN 102455421B CN 201010521203 A CN201010521203 A CN 201010521203A CN 102455421 B CN102455421 B CN 102455421B
- Authority
- CN
- China
- Prior art keywords
- tdoa
- receiver
- node
- td2s
- sound source
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 58
- 238000005070 sampling Methods 0.000 claims description 19
- 230000002159 abnormal effect Effects 0.000 claims description 14
- 238000004364 calculation method Methods 0.000 claims description 8
- 238000012545 processing Methods 0.000 claims description 8
- 230000004807 localization Effects 0.000 claims description 6
- 238000002474 experimental method Methods 0.000 description 13
- 238000004891 communication Methods 0.000 description 8
- 238000010586 diagram Methods 0.000 description 7
- 238000004458 analytical method Methods 0.000 description 4
- 230000001186 cumulative effect Effects 0.000 description 4
- 230000001934 delay Effects 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 230000005236 sound signal Effects 0.000 description 3
- 230000005540 biological transmission Effects 0.000 description 2
- 238000012937 correction Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000007613 environmental effect Effects 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 239000011159 matrix material Substances 0.000 description 2
- 238000012544 monitoring process Methods 0.000 description 2
- 102100028082 Tapasin Human genes 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 239000003245 coal Substances 0.000 description 1
- 230000002950 deficient Effects 0.000 description 1
- 230000009977 dual effect Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000008447 perception Effects 0.000 description 1
- 229920001690 polydopamine Polymers 0.000 description 1
- 230000005855 radiation Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 108010059434 tapasin Proteins 0.000 description 1
- 239000013598 vector Substances 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Images
Landscapes
- Measurement Of Velocity Or Position Using Acoustic Or Ultrasonic Waves (AREA)
Abstract
The invention discloses a sound positioning system and a sound positioning method without time synchronization, which are designed for solving the problems of dependency of the conventional positioning system on the time synchronization and errors and extra overhead caused by the time synchronization. The sound positioning system comprises a node to be positioned, more than two receivers, a server and a wireless access point (AP), wherein the node to be positioned is used for producing sound source signals to the outside; the more than two receivers are divided into a standard receiver and a beacon receiver and used for recording the sound source signals emitted by the node to be positioned and successive sound source signals emitted to the outside by the standard receiver and outputting the record data after the recording is completed to the server; the server is used for receiving the record data output by each receiver, calculating the position of the node to be positioned according to the record data and displaying the position of the node to be positioned on an interface of the server; and the wireless AP is used for constructing a wireless local area network consisting of the node to be positioned, the receivers and the server. According to the system and the method, the time synchronization in the positioning system is removed, and good positioning precision is obtained.
Description
Technical Field
The invention relates to the field of wireless sensor networks, in particular to a sound positioning system and a sound positioning method without time synchronization in a wireless sensor network.
Background
With the popularity of wireless mobile devices, many context-aware based applications have come into existence. In the context in which these applications are based, location is often one of the most important factors, such as: the specific position for acquiring the environmental information needs to be known in the environmental monitoring; the area where enemy personnel and equipment move needs to be known on the battlefield; in coal mine management it is necessary to know the exact location of the accident etc. In these applications, monitors often use wireless sensor networks to monitor events. In this case, to obtain the location of the occurrence of the monitoring event, the geographical location of the sensor node itself is first known. Since sensor networks are often deployed in restricted or dangerous areas, such as nuclear radiation areas, battlefields or mountains, people usually randomly deploy sensor nodes and let these nodes work in a self-organizing manner. A common example is the broadcast of sensor nodes into a designated area by an airplane. The randomly played nodes cannot know the positions of the nodes in advance, and therefore self-positioning after distribution must be achieved. A small number of nodes in the sensor network, which are called beacons (beacons), are initially known to their own locations, and often carry some special devices, such as GPS, etc., for knowing their own locations. The positioning of an unknown node is realized according to some positioning mechanism according to the beacon nodes.
Researchers have proposed a variety of positioning methods that can be broadly classified into two categories, distance-based and distance-independent. The former needs to measure the absolute distance or the azimuth between the nodes and calculate the position of an unknown node according to the actual distance between the nodes; the latter does not need to derive the absolute distance or orientation between nodes, but uses estimates. Of the two methods, the distance-based method can achieve higher precision and has wide application.
Distance-based methods include Time of Arrival (TOA) based positioning, Time Difference of Arrival (TDOA) based positioning, and Received Signal Strength (RSS) based positioning. These methods employ different ranging approaches, with TOA and TDOA having better accuracy. However, both methods require the mutual operation of time stamps (time stamps) on two different nodes, and therefore require the clocks of the nodes in the network to be highly consistent. The process of bringing the otherwise asynchronous local clocks of the nodes to congruence through some mechanism is called time synchronization (time synchronization), which is an indispensable step in the TOA and TDOA methods.
In order to make the error of time synchronization smaller and the cost lower, researchers start with a simple method and continuously improve the time synchronization algorithm. Currently, commonly used Synchronization methods include Reference Broadcast Synchronization (RBS), sensor network Time Synchronization Protocol (TPSN), Flooding Time Synchronization Protocol (FTSP), Gradient Time Synchronization Protocol (GTSP), and the like. Generally, time synchronization is accomplished through a series of information exchanges between beacons. When a node generates its own timestamp and sends it to another node seeking synchronization, the transmitted packet carrying the timestamp tends to experience a delay before reaching the receiving node. This delay makes it difficult for two nodes to accurately align their clocks. Fikret et al have roughly divided this delay into the following components:
(1) the time of transmission. I.e. the time a transmit packet is generated at the transmitting end.
(2) The time of entry. The time for which the packet stays in the Medium Access Control (MAC) sublayer before being transmitted.
(3) The propagation time. The time a packet is sent from leaving the sender to reaching the receiver.
(4) The time of reception. After receiving the packet, the receiving end needs to go through a decoding process and transmit the decoded data to the upper layer of the network.
Some synchronization algorithms may remove one or both of the delays, but may not remove all of the delays. Even with a method that can eliminate all the delays, time synchronization still needs to be performed multiple times due to clock skew (clock skew and drift), which greatly increases the overhead of the system.
Therefore, the time synchronization introduces additional communication overhead, consumes the originally limited energy of the sensor node, and leaves enough time error to bring large distance error to a positioning system using high-speed signals (such as radio frequency signals), so that the positioning accuracy is remarkably reduced.
Disclosure of Invention
Aiming at the problems that the existing positioning system depends on time synchronization, and the time synchronization brings errors and extra expenses, the invention provides a sound positioning system and a sound positioning method based on TDOA and without time synchronization.
To achieve the above object, the sound localization system without time synchronization according to the present invention comprises:
the node to be positioned externally sends out a sound source signal;
the two or more receivers are divided into a reference receiver and a beacon receiver, record the sound source signal sent by the node to be positioned and the subsequent sound source signal sent by the reference receiver to the outside, and output the recorded recording data to the server;
the server receives the recording data output by each receiver, calculates the position of the node to be positioned according to the recording data and displays the position on a server interface; and the number of the first and second groups,
and the wireless AP builds a wireless local area network consisting of the node to be positioned, the receiver and the server.
In particular, the receiver is constituted by a speaker, a microphone and a wireless connector.
In particular, the server comprises a communication and transceiving data module and a data processing and computing module; wherein,
the communication and data receiving and transmitting module is used for sending a control instruction to the receivers and receiving the recording data output by each receiver;
and the data processing and calculating module is used for analyzing the recording data transmitted to the server and obtaining the TDOA and the final positioning result through correlation calculation and peak value taking calculation.
The invention relates to a sound positioning method without time synchronization, which comprises the following steps:
(1) when the receiver starts recording, a node to be positioned sends a sound source signal outwards at the moment t;
(2) the reference receiver sends out subsequent sound source signals outwards at the moment t';
(3) after the recording is finished, the receiver sends the recorded recording data consisting of the recorded sound source signal and the subsequent sound source signal to the server;
(4) and after receiving the recording data, the server performs data processing on the recording data and calculates a positioning result of the node to be positioned.
Wherein, the step (4) comprises the following steps:
4.1 after the server receives the recording data, searching a corresponding sampling point i when the sound source signal reaches the receiver and a corresponding sampling point j when the subsequent sound source signal reaches the receiver based on a sampling point counting method, and calculating a TD2S value of the receiver;
4.2 based on the majority method, removing abnormal TD2S value;
4.3 calculating TDOA value according to the TD2S value;
and 4.4, calculating a positioning result of the node to be positioned according to the TDOA value.
Further, the majority method described in step 4.2 is implemented as follows:
4.2.1, a receiver A and a receiver B are set to record a sound source signal externally sent by a node to be positioned at the moment t and a subsequent sound source signal externally sent by a reference receiver at the moment t';
4.2.2 calculate the TD2S values for receiver a and receiver B, respectively, and establish the TD2S value relation for receiver a and receiver B:
|△TD|=|TD2SA-TD2SB|≤2dAB/V；
4.2.3 given the maximum distance D between receivers, there is |. DELTA |TD|≤2D/V；
4.2.4 groups the TD2S values of all receivers into a set S;
4.2.5 assuming that the TD2S values for most receivers are correct, the elements in set S are classified into the following three categories:
one type is as follows: any two elements a, b in the class have | a-b | < 2D/V.
The second type is as follows: any element c in the class and any element D in the class (1) have | c-D | > 2D/V.
Three types are as follows: other elements;
4.2.6 it is assumed that both elements are abnormal values.
Further, step 4.3 is specifically implemented as follows:
4.3.1 if C is the reference receiver and D is the beacon receiver, then the source signal S is at t respectivelyC1And tD1To C and D, C at tC3Emitting successive sound source signals S 'from the horn at times, and then S' at t respectivelyC4And tD3The arrival of time C and D yields:
TD2SC=tC4-tC1，TD2SD=tD3-tD1
for TDOA values between C and D:
TDOACD=tC1-tD1=dCD/V-TD2SD+TD2SC–dCC／V
wherein d isCDAnd dCCRespectively representing the distance from the loudspeaker of C to the microphone of D and the distance from the loudspeaker of C to the microphone of C, and V is the propagation speed of sound in the air under the scene: v =331.3+0.6 theta, where theta is the temperature at the scene;
4.3.2 if neither C, D is a reference receiver, then the TDOA value between C and D has: TDOACD=TDOACE-TDOADEWherein, TDOACEFor the TDOA value between C and reference receiver E, the calculation is performed in the same way as in step 4.3.1, TDOADEThe TDOA value between D and reference receiver E is calculated as in step 4.3.1.
The time-free sound positioning system and the time-free sound positioning method have the following beneficial effects that:
first, no time synchronization is required. The present invention fundamentally changes the TDOA scheme by eliminating this step, which is necessary in other TDOA methods, and avoids many errors caused by time synchronization.
And secondly, high positioning precision. On the premise of adopting 44.1kHz sampling frequency, the invention can obtain the time precision of 0.023ms, and can obtain high positioning precision under the time precision. According to the experimental result in a 3D area of 9m 4m, the system can obtain the precision of 10-20 cm, and the requirement of a large number of applications can be met.
Thirdly, the invention can be deployed on Commercial off-the-shelf (COTS) devices, such as mobile phones, PDAs, MP3 players, etc. The system is easier to use, lower in deployment cost and easy to popularize.
Drawings
FIG. 1 is a schematic diagram of a sound localization system without time synchronization according to the present invention;
FIG. 2 is a diagram of an exemplary time series of two nodes A and B;
FIG. 3 is a diagram illustrating the correlation results of an exemplary signal;
FIG. 4 is a layout diagram of nodes in a multi-environment experimental 2D scene according to the present invention;
FIG. 5 is a layout diagram of nodes in a multi-environment experimental 3D scene according to the present invention;
FIG. 6 is a projection view of each node of a 3D scene on an xy plane;
FIG. 7 is a projection view of nodes of a 3D scene on the yz plane;
FIG. 8 is a schematic diagram of the number of outliers of the experimental results in the 2D scenario of the present invention;
FIG. 9 is a diagram illustrating the number of outliers of the experimental results in a 3D scenario according to the present invention;
FIG. 10 is a cumulative distribution plot of positioning errors for experimental results in a 2D scenario in accordance with the present invention;
FIG. 11 is a cumulative distribution plot of positioning errors for experimental results in a 3D scenario in accordance with the present invention; .
Detailed Description
The following detailed description of embodiments of the invention refers to the accompanying drawings.
The basic system architecture of the present invention is shown in fig. 1, and mainly includes hardware: a node to be located, a wireless AP2, a notebook 3 as a server and several receivers 1. Wherein the several receivers 1 are divided into a beacon receiver and a reference receiver. The reference receiver and the beacon receiver are deployed at known determined locations. And the position of the node to be positioned is to be solved. The working principle of each part of the sound positioning system without time synchronization is as follows:
the node to be positioned externally sends out a sound source signal;
the receiver is divided into a reference receiver and a beacon receiver, records the sound source signal sent by the node to be positioned and the subsequent sound source signal sent by the reference receiver to the outside, and outputs the recorded recording data to the server;
the server is used for receiving the recording data output by the beacon receiver, calculating the position of the node to be positioned according to the recording data and displaying the position on a server interface;
and the wireless AP builds a wireless local area network formed by the node to be positioned, the reference receiver, the beacon receiver and the server.
The receivers all have a set of basic hardware, including speakers, microphones, and wireless connectors (e.g., bluetooth or WiFi). The receivers form a local area network with the server through the wireless AP.
The flow of system work is roughly: the receiver sends the recording file to the server for processing data to obtain the positioning result of the node to be positioned.
The sound positioning method without time synchronization comprises the following specific steps:
and 4, after receiving the sound recording file, the server processes the sound recording file and calculates a positioning result of the node to be positioned.
4.1 after the server receives the recording data, based on the sampling point counting method, finding out the corresponding sampling point i when the sound source signal reaches the beacon receiver and the corresponding sampling point j when the subsequent sound source signal reaches the beacon receiver, and calculating the TD2S value of the beacon receiver.
Fig. 2 is a typical time sequence for two nodes a and B (a node refers to a reference receiver or a beacon receiver, the same applies below). In the figure, t is used respectivelyA1And tB1Representing the times at which sound source S arrives at the microphones of a and B. If the clocks of the two nodes are completely identical, then tB1-tA1Are the TDOA values for a and B. However, if only such a simple operation is performed, a clock synchronization step is necessarily required.
In the present invention, TD2S is used to indicate the time that elapses from the time when a node receives S to the time when it receives S'. For each node, calculating this value only involves its own clock and does not require synchronization with other nodes. From the TD2S value of each node and the coordinates of the anchor node (i.e., beacon receiver), the TDOA between any two nodes can be calculated.
However, it still takes a while for the sound signal to reach the microphone until it is detected by the node. In this case, tB2-tA2Is the TDOA value calculated by the system, and tB2-tA2And tB1-tA1Are generally unequal. To eliminate this uncertainty, the present invention uses a method of sample point counting. Since the node is always in the recording state when S and S 'reach the node, and the sound signal is sampled at a certain recording frequency f, if the corresponding sample point i when S reaches the node and the corresponding sample point j when S' reaches the node in the sample data can be found, the TD2S value of the node, that is, TD2S = (j-i)/f, can be obtained. Obviously, the sampling frequency of the device affects the accuracy of TD2S, and the higher the sampling frequency, the higher the time accuracy. The acoustic sampling frequency employed by the present invention is 44.1kHz, which is achievable by most COTS devices.
Then, how to accurately find the sampling points corresponding to S and S'? Here, the present invention employs a conventional correlation method. This method is only valid for signals that can be mathematically described in advance. Therefore, in the experiment, in consideration of the autocorrelation characteristic requirement of the signal, a Linear chirp signal having a frequency range of 2 to 6kHz and a length of 50ms is used as S, and S' is set to be the same. In the ideal case, the correlation result is shown in fig. 3, where the former peak corresponds to the arrival time of S and the latter peak corresponds to the arrival of S'. In practical applications, however, the two peaks do not represent S and S', respectively, due to multipath effects (multi-path) caused by reflections and other factors. Therefore, the invention changes the highest peak into the front length of the highest peak as omega0The earliest peak, ω, within the sliding window of (c)0Is an empirical parameter. A threshold may be set from the height and average slope to determine if a point can be calculated as a peak.
4.2 based on the majority method, removing abnormal TD2S value;
through step 4.1, TD2S is obtained, but the recording of the node is often unstable, and the obstacle or background noise between the sound source and the node may cause defective recording. The TD2S value obtained from such a sound recording file is an abnormal value (outlier).
Since both the time and place of S emission are unknown, it is difficult to specify a range in which TD2S should be. The present invention provides a majority decision method to determine which values are outliers.
Assuming that the sound source M emits an S signal at time t and the reference node N emits an S 'signal at time t', both nodes a and B record the two utterances. Then TD2S for a and B satisfies the following relationship:
|△TD|=|TD2SA-TD2SB|
=|((t’+dAN/v)-(t+dAM/v))-((t’+dBN/v)-(t+dBM/v))|
=|(dAN–dBN)–(dAM–dBM)|/v
≤|dAN–dBN|/v+|dAM–dBM|/v
≤2dAB/v
assuming D is the maximum distance between all nodes, then for any pair of nodes, there is |. DELTATDThe | is less than or equal to 2D/V. Therefore, if TD2S of all nodes are grouped into a set S, and no outliers exist in the set, the difference between the largest element and the smallest element in the set will not be greater than 2D/V.
If most of the nodes have correct TD2S values and a few have wrong values, the elements in S can be classified into the following three categories:
(1) any two elements a, b in the class have | a-b | < 2D/V.
(2) Any element c in the class and any element D in the class (1) have | a-b | > 2D/V.
(3) And (3) other elements.
In this case, it is considered that all the elements of the class (2) are abnormal values, and the next operation is performed using only the elements of the class (1). This classification is based on the assumption that most TD2S are not outliers, which is very close to the fact. This method of classifying and removing outliers performed very well in the experiment.
4.3 calculating the TDOA values of the two nodes according to the TD2S value;
the TDOA values for any two nodes are further derived based on step 4.2 to derive a normal TD2S value (i.e., the TD2S value for each node with insufficient error to make it an outlier). Fig. 2 is a typical time sequence of two nodes a and B in a system, where a is the reference node. In the figure, S is respectively at tA1And tB1To reach A and B, A is at tA3At a moment S 'is emitted from the horn, and then S' is respectively at tA4And tB3The moments arrive at the microphones of a and B. Notably, TD2SA=tA4-tA1,TD2SB=tB3-tB1. For the TDOA value between A and B, the following method is used for obtaining:
TDOAAB=tB1-tA1
=(tB3-tA3）-（tB3–tB1)+(tA3–tA1)
=(tB3-tA3)-（tB3-tB1)+(tA4-tA1)-（tA4-tA3)
=dAB/V-（tB3-tB1)+(tA4–tA1）-dAA／V
=dAB/V-TD2SB+TD2SA-dAA／V
wherein d isABAnd dAARespectively representing the distance from the horn of a to the microphone of B and the distance from the horn of a to the microphone of a. Since the positions of a and B are both fixedly measurable, both values are invariant; two TD2S values mayObtained by the method mentioned above; v is the speed of sound in air in the scene, which can be calculated by the formula V =331.3+0.6 theta, where theta is the temperature in the scene. This results in TDOA values at asynchronous nodes.
If neither of the A and B nodes is the reference node, their TDOA values can be obtained by simple operation, and the formula is: TDOAAB=TDOAAC–TDOABCWherein, TDOAACAs TDOA value between A and reference node C, TDOABCIs the TDOA value between B and reference node C.
4.4 calculating the positioning result of the node to be positioned according to the TDOA value;
the position is obtained by using the TDOA value, and the method is a process of solving a nonlinear equation system in general. There are many corresponding methods, such as a method of directly solving equations, a method of applying Taylor's formula, a method of dividing and dividing, etc. The invention selects an algorithm proposed by Chan through the analysis of the algorithms, and the algorithm has better balance on the precision and the computational complexity. These algorithms typically require at least 4 nodes.
The Chan algorithm first uses a method similar to Taylor expansion to obtain an initial solution, then finds a covariance matrix of error vectors, and further corrects the initial solution by using the matrix to obtain a final result. Through theoretical analysis, the method only carries out one correction, but has the precision equivalent to that of the Taylor expansion method of multiple iterations. In addition, different correction methods can be adopted in the areas close to the beacon nodes and the areas far away from the beacon nodes according to the difference of the areas where the sound sources are located, so that the result is more accurate, and the application scene is wider.
Through the steps, the invention starts from the recorded data of a plurality of nodes, and finally calculates the position of the sound source, namely the position of the node to be positioned.
In summary, the sound positioning system and method without time synchronization of the present invention mainly use the following technologies:
(1) dual signal sensing. At system start-up, several receivers with different clocks record the sound source signal S and the subsequent signal S' from the reference receiver.
(2) The sampling points are counted. After the recording is finished, the system counts the number of sampling points between the two signals on each receiver, so as to obtain the receiving time difference of the two signals on each receiver. Through further analysis and calculation, the differences in the reception times of the source signals at the different receivers can be obtained, and the position can then be calculated using the principles of TDOA. Since the signal sampling can generally reach a higher frequency, the calculation of the time difference can reach a higher accuracy. In addition, the system avoids the error of time synchronization, so that the positioning of the system can be more accurate.
The sound localization system and method of the present invention without time synchronization is further described below in conjunction with a specific embodiment.
Firstly, set up experiment system
In the embodiment, 8 mobile phones are used, wherein 6 pod mobile phones are beacon receivers and are used for recording; a SonyEricsson mobile phone is used as a node to be positioned; an O2XDA handset is used as the reference receiver. The 8 mobile phones, a PC and an AP (Access Point) are connected to a star network, the AP is used as the center of the network, all the recording data recorded by the mobile phones are sent to the PC, and the PC processes and calculates the recording data to finally obtain a position result.
In the experimental system constructed in this embodiment, more than 4 beacons are used, because this can ensure that when an abnormal point occurs, the available TD2S value still exceeds 4; secondly, the positioning accuracy can be improved because more nodes are provided.
Secondly, operating positioning program software
The system software of the embodiment is divided into two parts, namely software deployed on a server and software deployed on a receiver. The software deployed on the server mainly serves as system control and data processing, and the software deployed on the receiver mainly serves as recording and sound production control. Both of which contain communication modules for sending and receiving commands to and from each other.
The software on the server is divided into two independent modules: a communication and data receiving and transmitting module and a data processing and calculating module. The communication and data receiving and transmitting module is used for sending control instructions to the receivers and receiving the recording data of the receivers; and the data processing and calculating module is used for analyzing the recording data transmitted to the server, and obtaining the TDOA and a final result through correlation and peak value calculation.
The software deployed on the mobile phone and the communication module of the server are both realized by Visual C + + programming, the computing module of the server is realized by Matlab, and a COM component capable of being called by the communication module is generated.
In a complete positioning process, firstly, all receivers, APs and a server PC are connected in a network to ensure the intercommunication among the devices; then, a user of the server issues a positioning command, and the system automatically executes the works of recording, sounding of a sound source and transmission of recording data of all receivers until all the recording data are transmitted to the server; and finally, the server calls a calculation module, analyzes all the sound recording files by combining the coordinate information of each receiver, and finally calculates the position of the sound source and displays the position on a server interface. Next, the next positioning can be started.
Three, multi environment experiment
In the above-described specific steps, besides some negligible factors, such as fluctuation of the speed of sound, there are still some factors that can affect the accuracy of positioning. This mainly includes:
(1) Signal-to-Noise Ratio of Signal (Signal to Noise Ratio, SNR)
In the experiment, the ambient noise was always recorded. If the energy of the transmitted signal is too weak or the frequency of the noise is close to the frequency of the transmitted signal, it may be difficult for the system to find the arrival time of the transmitted signal.
(2) Multipath effect
Due to the presence of sound reflections, the same sound signal may reach the same receiver several times through different paths. Although the experiment adopts a reasonable method of selecting the first peak as the arrival time, errors still occur.
(3) Solution equation
When the error of TDOA is small, the method of Chan can obtain higher precision. But this method also performs worse and worse as the error increases. In experiments, most TDOA errors can be kept below 0.4ms, and the Chan method can obtain better results in the case. But still a small part of the error is larger, so this method brings a larger error.
According to the analysis of the error sources, in the 2D and 3D scenes of running the experiment, the system constructed by the embodiment is subjected to the experiment in three different environments. These three environments are:
(1) outdoor quiet environments.
(2) Outdoor noisy environments. The set position is the same as (1), but a loudspeaker for playing music is arranged beside the loudspeaker.
(3) A quiet indoor environment. An open lobby, about 9m by 4m in size, is quieter.
For simplicity of representation, the three environments (1), (2), and (3) are referred to herein using "Normal", "noise", and "Inside", respectively. The purpose of setting up these three experimental environments is: attempts were made to reveal the effect of a single interference factor on the experimental results. In each experimental environment, the sound source was placed at 16 different points and sampling was repeated 10 times at each point. Therefore, there are 960 sets of recorded results in 2 scenarios and 3 environments.
In 2D and 3D scenarios, node placement is identical under various circumstances. The node placement for the 2D scene is shown in fig. 4, and the node placement for the 3D scene is shown in fig. 5, 6 and 7. Wherein ■ represents the position of the sound source emitted by the different nodes to be positioned, and ● represents the position of the receiver. Since the receivers are symmetrically arranged, the sound source position only needs to occupy 1/4.
Fourth, experimental results
1. Number of abnormal points
If the number of the beacon nodes with abnormal values exceeds 2 in a 3D scene and exceeds 3 in a 2D scene, a positioning result with a great error can be obtained. Such results greatly increase the average error of the positioning results, are substantially worthless, and are easily distinguishable from other results.
Therefore, when calculating the average error of the positioning, these abnormal results are removed first, and then the average of the remaining results is calculated. However, in order to study the stability of the system, the number of abnormal positioning results is recorded in the embodiment.
Fig. 8 and 9 show the number of the positioning results of the anomaly in the 2D scene and the 3D scene respectively in the whole experiment. As can be seen from the figure, the abnormal results obtained by the experiment in the 2D scene are much less than those in the 3D scene, which shows that more redundant nodes can increase the stability of the system.
2. Error in positioning
Fig. 10, 11 depict the cumulative density function (cumulative density function) of the localization error in 2D and 3D scenes, respectively. Wherein, 4 is a "Normal" environment, 5 is a "Noisy" environment, and 6 is an "Inside" environment. Table 1, table 2 records the mean value of the positioning error and some thresholds. As can be seen from the graph, the average error of localization in 3D scene is 10-20 cm, while the average error of 2D is 10-21 cm. This demonstrates that the system of the present invention achieves better accuracy in a variety of environments.
TABLE 12D positioning error
| Index (cm) | Normal | Noisy | Inside |
| Mean error | 20.0 | 20.3 | 13.7 |
| 50% error | 12.7 | 18.8 | 10.9 |
| Error of 90% | 38.1 | 33.7 | 29.9 |
| Standard deviation of | 12.8 | 10.9 | 12.1 |
| Proportion of abnormal points | 1.25% | 5% | 0.6% |
TABLE 23D positioning error
| Index (cm) | Normal | Noisy | Inside |
| Mean error | 14.7 | 14.4 | 15.2 |
| 50% error | 13.7 | 13.8 | 12.8 |
| Error of 90% | 22.8 | 21.9 | 28.2 |
| Standard deviation of | 6.2 | 7.4 | 9.9 |
| Proportion of abnormal points | 2.5% | 5% | 12.5% |
It is observed from the above table that the results of positioning in outdoor quiet environments are not necessarily the best. Although it can be seen from fig. 8 and 9 that noise and multipath increase the number of outliers, they do not significantly change the total average error. This also shows that the system of the present invention has strong anti-interference capability.
In summary, the sound positioning system and method without time synchronization of the present invention fundamentally changes the scheme of TDOA by removing time synchronization; and high positioning precision is obtained by the technology of double-signal perception and sampling point counting. Experiments show that experiments performed in a 3D region of 9m by 4m can obtain an average accuracy of 10-20 cm. The experimental equipment adopts a commercial mobile phone, so that the cost is reduced and the system deployment is easier.
The above description is only for the preferred embodiment of the present invention, but the scope of the present invention is not limited thereto, and any changes or substitutions that can be easily conceived by those skilled in the art within the technical scope of the present invention are also included in the scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope defined by the claims.
Claims (1)
1. A method of sound localization without time synchronization, comprising the steps of:
(1) when the receiver starts recording, a node to be positioned sends a sound source signal outwards at the moment t;
(2) the reference receiver sends out subsequent sound source signals outwards at the moment t';
(3) after the recording is finished, the receiver sends the recorded recording data consisting of the recorded sound source signal and the subsequent sound source signal to the server;
(4) after receiving the recording data, the server performs data processing on the recording data and calculates a positioning result of a node to be positioned;
wherein, the step (4) comprises the following steps:
4.1 after the server receives the recording data, searching a corresponding sampling point i when the sound source signal reaches the receiver and a corresponding sampling point j when the subsequent sound source signal reaches the receiver based on a sampling point counting method, and calculating a TD2S value of the receiver;
4.2 based on the majority method, removing abnormal TD2S value;
4.3 calculating TDOA value according to the TD2S value;
4.4 calculating the positioning result of the node to be positioned according to the TDOA value;
step 4.3 is specifically realized as follows:
4.3.1 if C is the reference receiver and D is the beacon receiver, then the source signal S is at t respectivelyC1And tD1To C and D, C at tC3Emitting successive sound source signals S 'from the horn at times, and then S' at t respectivelyC4And tD3The arrival of time C and D yields:
TD2SC=tC4-tC1，TD2SD=tD3-tD1
for TDOA values between C and D:
TDOACD=tC1-tD1=dCD/V-TD2SD+TD2SC–dCC／V
wherein d isCDAnd dCCRespectively representing the distance from the loudspeaker of C to the microphone of D and the distance from the loudspeaker of C to the microphone of C, and V is the propagation speed of sound in the air under the scene: v =331.3+0.6 theta, where theta is the temperature at the scene;
4.3.2 if neither C, D is a reference receiver, then the TDOA value between C and D has: TDOACD=TDOACE-TDOADEWherein, TDOACEFor the TDOA value between C and reference receiver E, the calculation is performed in the same way as in step 4.3.1, TDOADEThe TDOA value between D and reference receiver E is calculated as in step 4.3.1.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201010521203.7A CN102455421B (en) | 2010-10-27 | 2010-10-27 | Sound positioning system and method without time synchronization |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201010521203.7A CN102455421B (en) | 2010-10-27 | 2010-10-27 | Sound positioning system and method without time synchronization |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102455421A CN102455421A (en) | 2012-05-16 |
| CN102455421B true CN102455421B (en) | 2014-03-26 |
Family
ID=46038830
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201010521203.7A Active CN102455421B (en) | 2010-10-27 | 2010-10-27 | Sound positioning system and method without time synchronization |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102455421B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103813441B (en) * | 2012-11-08 | 2017-04-26 | 阿里巴巴集团控股有限公司 | Terminal positioning method and apparatus |
Families Citing this family (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| TW201413672A (en) * | 2012-09-17 | 2014-04-01 | Wistron Corp | Warning device and warning method |
| CN102901949B (en) * | 2012-10-13 | 2014-04-16 | 天津大学 | Two-dimensional spatial distribution type relative sound positioning method and device |
| CN103906212B (en) * | 2014-04-14 | 2018-10-19 | 中北大学 | A kind of intelligence suspend mode control electricity low-consumption wireless sensor network acoustic fix ranging node |
| CN104515972B (en) * | 2014-12-31 | 2017-04-05 | 苏州触达信息技术有限公司 | Multimedia equipment group based on sound wave exempts from synchronous localization method to newly added equipment |
| TWI590666B (en) | 2015-02-12 | 2017-07-01 | 絡達科技股份有限公司 | Voice enhancement method for distributed system |
| CN105323772B (en) * | 2015-09-23 | 2018-09-07 | 浙江大学 | A kind of method for self-locating sensor network node based on smart mobile phone |
| CN106559870A (en) * | 2015-09-30 | 2017-04-05 | 华为技术有限公司 | Localization method and positioner based on wireless network |
| CN106657852B (en) * | 2016-12-21 | 2019-04-16 | 智威富（北京）科技有限公司 | Time synchronization method, equipment and system |
| CN113311392B (en) * | 2021-06-03 | 2024-01-23 | 苏州触达信息技术有限公司 | Error compensation method for sound wave positioning under unsynchronized network |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP1806952A2 (en) * | 2006-01-06 | 2007-07-11 | Agilent Technologies, Inc. | Acoustic location and acoustic signal enhancement |
| CN101044371A (en) * | 2004-10-04 | 2007-09-26 | 株式会社拓普康 | Time difference measuring device, measuring method, distance measuring device, and distance measuring method |
| CN101320090A (en) * | 2008-07-15 | 2008-12-10 | 浙江大学 | Wireless sensor network ranging method facing round-trip time of time asynchronous node |
| CN101394679A (en) * | 2007-09-17 | 2009-03-25 | 深圳富泰宏精密工业有限公司 | Sound source positioning system and method |
-
2010
- 2010-10-27 CN CN201010521203.7A patent/CN102455421B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101044371A (en) * | 2004-10-04 | 2007-09-26 | 株式会社拓普康 | Time difference measuring device, measuring method, distance measuring device, and distance measuring method |
| EP1806952A2 (en) * | 2006-01-06 | 2007-07-11 | Agilent Technologies, Inc. | Acoustic location and acoustic signal enhancement |
| CN101394679A (en) * | 2007-09-17 | 2009-03-25 | 深圳富泰宏精密工业有限公司 | Sound source positioning system and method |
| CN101320090A (en) * | 2008-07-15 | 2008-12-10 | 浙江大学 | Wireless sensor network ranging method facing round-trip time of time asynchronous node |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103813441B (en) * | 2012-11-08 | 2017-04-26 | 阿里巴巴集团控股有限公司 | Terminal positioning method and apparatus |
Also Published As
| Publication number | Publication date |
|---|---|
| CN102455421A (en) | 2012-05-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102455421B (en) | Sound positioning system and method without time synchronization | |
| Hashem et al. | WiNar: RTT-based sub-meter indoor localization using commercial devices | |
| Wang et al. | RSSI-based bluetooth indoor localization | |
| CN107631732B (en) | Bluetooth, Wi-Fi and sound wave based hybrid indoor positioning method and system | |
| Peng et al. | BeepBeep: A high-accuracy acoustic-based system for ranging and localization using COTS devices | |
| US8837316B2 (en) | RTT based ranging system and method | |
| Höflinger et al. | Acoustic self-calibrating system for indoor smartphone tracking (assist) | |
| Yu et al. | TOA-based distributed localisation with unknown internal delays and clock frequency offsets in wireless sensor networks | |
| Lopes et al. | Accurate smartphone indoor positioning using a WSN infrastructure and non-invasive audio for TDoA estimation | |
| CN107566065A (en) | TOF localization methods based on UWB | |
| Lazik et al. | Ultrasonic time synchronization and ranging on smartphones | |
| Flueratoru et al. | High-accuracy ranging and localization with ultrawideband communications for energy-constrained devices | |
| JP2004514376A (en) | Clock synchronization by radio | |
| US8620355B2 (en) | Method and apparatus for determining the position of a wireless terminal based on propagation delay taps of base stations | |
| US8634405B2 (en) | System and method for synchronizing phases and frequencies of devices in multi-user, wireless communications systems | |
| Janson et al. | Self-localization application for iphone using only ambient sound signals | |
| WO2015089232A1 (en) | Positioning system ranging measurement | |
| JP2006080681A (en) | System and method for position detection | |
| US20160041251A1 (en) | Ultrasound-based location determination and inertial navigation with accuracy improvement in determining client device location | |
| Nagy et al. | Time-based localisation in unsynchronized wireless LAN for industrial automation systems | |
| Qiu et al. | BLE-based collaborative indoor localization with adaptive multi-lateration and mobile encountering | |
| Schwalowsky et al. | System integration of an IEEE 802.11 based TDoA localization system | |
| Elikplim et al. | Wi-fi fine time measurement: Is it a viable alternative to ultrawideband for ranging in industrial environments? | |
| WO2018121439A1 (en) | Method and apparatus for determining direct path | |
| Wan et al. | Time delay estimation of co-frequency signals in TDOA localization based on WSN |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |