JP2003095101A - Platform fall detecting device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To avoid influence of weather conditions such as change in ambient temperature, rain and snow. SOLUTION: When an extracted fluctuation pattern extracted from a received spread spectrum signal received by a leakage transmission line 14 for reception coincides with a pattern of a setting fluctuation pattern signal stored in a setting pattern storing means 23 and the fluctuation level of the extracted fluctuation pattern signal is higher than a prescribed value, an arithmetic means 24 decides that an obstacle 15 is in existence.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a home fall detection device for detecting a fallen person from a home or an obstacle such as a fallen object by utilizing a leaky transmission line such as a leaky coaxial cable or a leaky waveguide. Is.

[0002]

2. Description of the Related Art When detecting an obstacle such as a fallen person from a home or a fallen object, it is necessary to detect an object existing within the detection range without making a blind spot by setting a long area along the home as a detection range. is there. In such a situation, a leaky transmission line such as a leaky coaxial cable or a leaky waveguide is used. In the leaky transmission line, slots are provided at predetermined intervals in the longitudinal direction of the conductor, and radio waves are leaked from the slots. Since the leaky coaxial cable and the leaky waveguide are the same in principle, the leaky coaxial cable (hereinafter referred to as LCX)
The thing using is explained. FIG. 4 is a configuration diagram of a conventional home fall detection device on a line, which is described in, for example, Japanese Patent Laid-Open No. 10-95338. In FIG. 4, the pulse-like signal is input from the signal generating means 1 to the transmission LCX 3 arranged along the home 2. The pulse signal is transmitted from each slot (not shown) provided in the longitudinal direction of the transmission LCX3 toward the reception LCX5 across the line 4.
Sequentially emitted as radio waves. The signal incident from each slot (not shown) of the reception LCX 5 is received by the receiver 6. The signal received by the receiver 6 is a low pass filter 7
Take out the envelope (hereinafter referred to as the envelope). On the other hand, the envelope of the signal waveform when there is no fallen person (obstacle) is previously stored in the storage means 8 as a reference waveform. Then, each time the calculator 9 obtains the envelope of the signal waveform taken out by the low-pass filter 7, it compares it with the reference waveform of the storage means 8 to obtain a difference waveform. Fallen person 1
When there is an obstacle such as 0, the radio wave is blocked at the position where it is present.
The reception strength at the position of the position is reduced by a predetermined ratio. That is, the position of the fallen person 10 appears in the difference waveform.

[0003]

Since the conventional home fall detection device is constructed as described above, when the obstacle is small like a fallen person, the received signal generated when the obstacle cuts off the radio wave. Since the level fluctuation is small, it often reaches only a fraction of the fluctuation of the received signal level due to a change in ambient temperature, a change in radio wave propagation environment due to rain and snow.
Therefore, if the threshold value is set to a small value in order to detect the received signal level with little fluctuation, erroneous detection will occur when the received signal level changes due to changes in the radio wave propagation environment due to changes in ambient temperature or weather conditions such as rain and snow. There was a problem that it might be done. The present invention has been made to solve the above problems, and an object of the present invention is to provide a home fall detection device that can avoid the influence of weather conditions such as changes in ambient temperature and rain and snow. It is a thing.

[0004]

A home fall detection device according to the present invention is a home fall detection device for detecting an obstacle which has fallen on a track adjacent to a home and within a detection range of a clearance space between the track and the platform. , A leakage transmission line for transmission arranged on one side of the detection range along the home, a leakage transmission line for reception arranged on the other side of the detection range so as to face the leakage transmission line for transmission, and this reception Transmission spectrum spread signal generation means for radiating a transmission spectrum spread signal from the transmission leaky transmission path toward the transmission leaky transmission path, and from one end to the other end of the transmission leaky transmission path to which the transmission spread spectrum signal is input, A measuring distance setting means for setting a measuring distance from one end to a measuring point for detecting an obstacle, a delay means for generating a delay time according to a delay amount, and a phase timing corresponding to the delay time. And a reference spread spectrum signal generating means for outputting a reference spread spectrum signal having a reference spread spectrum signal, and the received spread spectrum signal received by the reception leaky transmission line and the reference spread spectrum signal are correlated to obtain a received signal level at a measurement point as a signal level output signal. , A memory means for accumulating the signal level output signal in correspondence with the measurement distance, and a variation of the received signal level extracted from the data sequence for each measurement distance accumulated in the memory means as a variation pattern signal Variation pattern extraction means, setting pattern storage means for storing a preset variation pattern signal, the extracted variation pattern signal and the variation pattern setting pattern, the variation level of the extraction variation pattern signal is a predetermined value In the above case, it is provided with a calculation means for determining that an obstacle exists.

A transmission leaky transmission line arranged on one side of the detection range along the home and a reception leaky transmission line arranged on the other side of the detection range so as to face the transmission leaky transmission line. , A transmission spread spectrum signal generating means for radiating a transmission spread spectrum signal from the transmission leaky transmission path toward the reception leaky transmission path, and one end to the other end of the transmission leaky transmission path to which the transmission spread spectrum signal is input. Direction, a measurement distance setting means for setting measurement distances from one end to a plurality of measurement points for detecting an obstacle, a delay means for generating a delay time according to each measurement distance, and a phase timing according to each delay time And a reference spread spectrum signal generating means for outputting a reference spread spectrum signal having the following, and the received spread spectrum signal received by the reception leaky transmission line are distributed corresponding to each delay time, and distributed. Receiving signal distribution means for outputting a spectrum spread signal, correlation means for correlating each distributed spread spectrum signal corresponding to the delay time and each reference spread spectrum signal, and outputting a received signal level for each delay time, and each reception Memory means for accumulating the signal level in association with each measurement distance; variation pattern extraction means for extracting the variation of each received signal level as an extraction variation pattern signal from the data sequence for each measurement distance accumulated in the memory means; A set pattern storage unit that stores a preset set fluctuation pattern signal and the patterns of the extracted fluctuation pattern signals and the set fluctuation pattern signals corresponding to the measured distances match, and the fluctuation level of the extracted fluctuation pattern signal is a predetermined value. In the above case, it is provided with a calculating means for determining that an obstacle exists. Further, the calculating means is adapted to calculate the position of the obstacle from the center position of the pattern of the extracted variation pattern signal.

[0006]

BEST MODE FOR CARRYING OUT THE INVENTION Embodiment 1. FIG. 1 is a configuration diagram of the first embodiment. In FIG. 1, 11 is a home and 12
Is a track adjacent to the platform 11. The space on the track 12 and between the track 12 and the platform 11 is the detection range. Reference numeral 13 denotes a leaky transmission line for transmission, which is arranged on one side of the detection range along the home 11, and is a leaky coaxial cable or a leaky waveguide. The later-described transmission spread spectrum signal 16a input from one end of the transmission leaky transmission line 13 is radiated toward the reception leaky transmission line 14 described later so as to pass through the routes 13a and 13b, for example. Reference numeral 14 denotes a reception leaky transmission path arranged on the other side of the detection range so as to face the transmission leaky transmission path 13, and is a leaky coaxial cable, a leaky waveguide, or the like. The reception leaky transmission line 14 is a transmission spread spectrum signal 1 to be described later radiated from the transmission leakage transmission line 13.
6a is received and the received spread spectrum signal 14a is output. An obstacle 15 is a fallen person, an object, or the like that has fallen into the detection range. Reference numeral 16 is a transmission spread spectrum signal generating means, which spread spectrum modulates a carrier with a pseudo noise code to create a transmission spread spectrum signal 16a, and a plurality of transmission spread spectrum transmission lines 13 are arranged from one end to the other end in the longitudinal direction. Radiation is sequentially directed to the leaky transmission path for reception 14 from the slot via the paths 13a and 13b.

Reference numeral 17 denotes a measurement distance setting means, which is a transmission leakage transmission line 13 from which a transmission spread spectrum signal 16a is radiated.
At each slot position, that is, from one end to the other end in the longitudinal direction of the transmission leakage transmission line 13, the measurement distance of the measurement point for detecting the obstacle 15 is set. A delay unit 18 generates a delay time according to the measured distance. Reference numeral 19 is a reference spread spectrum signal generating means for outputting the reference spread spectrum signal 19a at a phase timing according to the delay time. 20
Is a correlating means, which correlates the reference spread spectrum signal 19a and the received spread spectrum signal 14a, and outputs the received signal level at the measurement point at each measurement distance as a signal level output signal 20a. Reference numeral 21 denotes a memory means, which divides the signal level output signal 20a into a plurality for each measurement distance and stores data corresponding to each measurement distance. Reference numeral 22 denotes a fluctuation pattern extraction means, which detects fluctuations in the received signal level in the distance direction from a plurality of data strings accumulated in the memory means 21 and extracts an extracted fluctuation pattern signal. Reference numeral 23 is a setting pattern storage means, which stores a setting variation pattern signal in the case where a preset obstacle 15 exists. An arithmetic means 24 compares the extracted variation pattern signal with the set variation pattern signal to detect the degree of coincidence, and when the variation level is equal to or higher than a predetermined value, determines that the obstacle 15 exists and outputs an obstacle detection signal 24a. To do.

Next, the operation will be described. In FIG. 1, the transmission spread spectrum signal 16 having a predetermined repetition period generated by the transmission spread spectrum signal generating means 16.
a is radiated from the transmission leaky transmission line 13 to the reception leaky transmission line 14 through the detection range by the paths 13a and 13b having a predetermined measurement distance. On the other hand, the delay means 18 generates a delay time according to the measurement distance set by the measurement distance setting means 17. Then, the reference spread spectrum signal generation means 19 has a reference spread spectrum signal 19a having a phase timing corresponding to the delay time created by the delay means 18.
Is output and input to the correlation means 20. In the correlating means 20, the reference spread spectrum signal 19a and the reception leaky transmission line 14
Is correlated with the received spread spectrum signal 14a, and the received signal level at the measurement point at each measurement distance via each of the paths 13a and 13b is used as the signal level output signal 20.
Output as a. The memory means 21 divides the signal level output signal 20a into a plurality for each measurement distance, and stores data corresponding to the measurement distance. Variation pattern extraction means 2
Reference numeral 2 detects variations in the received signal level for each measurement distance from a plurality of data strings stored in the memory means 21 and extracts an extracted variation pattern signal.

Then, the calculating means 24 associates the extracted variation pattern signal extracted by the variation pattern extracting means 22 and the set variation pattern signal stored in the setting pattern storage means 23 with each measurement point corresponding to the measurement distance. Compare. In this case, the extracted variation pattern signal extracted from the reception spread spectrum signal 14a received on the path 13a where the obstacle 15 exists, for example, has the same pattern as the setting variation pattern stored in the setting pattern storage means 23.
If the fluctuation level is equal to or higher than a predetermined value, it is determined that the obstacle 15 exists within the detection range, and the obstacle detection signal 24a.
Is output. Further, the position of the obstacle 15 is calculated from the center position of the pattern of the extracted fluctuation pattern signal. Since the extracted variation pattern extracted from the received spread spectrum signal 14a received on the path 13b does not match the setting variation pattern stored in the setting pattern storage means 23, the obstacle is not detected. Here, the level fluctuation due to the obstacle 15 existing within the detection range will be described. Figure 2 is published by The Institute of Electronics, Information and Communication Engineers, "LCX
Communication system ", 3rd edition (July 10, 1991), 38th
FIG. 7 is an explanatory diagram showing level fluctuations of the electric field distribution due to the metal cylinder (obstacle) described on page 3.7 (a). Fig. 2 shows the electric field distribution around the metal cylinder obtained by computer simulation in order to investigate the influence of the metal cylinder on the electric field distribution when the radio wave is cut off when a train radio LCX is laid along the Shinkansen. . Note that λ is the wavelength of the radio wave. For the sake of simplicity, it is assumed that the metal cylinder cuts off a radio wave that arrives as a plane wave from infinity.

In FIG. 2, the metal cylinder is replaced by the obstacle 1 shown in FIG.
5, the train radio LCX is replaced with the reception leaky transmission line 14 laid along the platform 11, so that the reception leaky transmission line 1 is provided when the obstacle 15 exists in the detection range.
4. A specific fluctuation pattern due to the level fluctuation of the electric field distribution generated along 4 can be explained. When an electric field fluctuation of a specific pattern occurs along the distance direction (longitudinal direction) of the reception leaky transmission line 14, the reception level received by the reception leaky transmission line 14 also changes in the same pattern. That is, when the obstacle 15 exists in the detection range, the obstacle 1 is placed behind the obstacle 15.
With a width almost equal to the width of 5, a drop of the reception level of 10 dB or more occurs. Then, on both sides of the depression, the reception level changes sinusoidally around the position of the obstacle 15 and the electric field is distributed. It can be inferred from FIG. 2 that the period of this electric field change is close to one wavelength of the radio wave arriving at the reception leaky transmission line 14, and the change of the reception level is several dB.

Therefore, it is possible to determine that the obstacle 15 exists in the detection range by detecting a specific variation pattern of the reception level in the distance direction which occurs only when the obstacle 15 exists. The center position of the fluctuation pattern is
Since the reception level is much lower than that of other portions, the position of the obstacle 15 can be specified by capturing this phenomenon. Along the longitudinal direction of the reception leaky transmission line 14, measurement points are set at intervals of one-half wavelength of radio waves or less,
It is clear from the sampling theorem that the variation pattern of the reception level can be identified from the accumulated series of data by measuring and accumulating the reception level while shifting the measurement point. Specifically, for example, by subjecting a series of data strings to FFT processing, the frequency components included in the series of data strings can be read with high accuracy and the distribution of the magnitudes of the respective frequency components can be identified. Variations in the received level can occur to identify a particular variation pattern. When the obstacle 15 does not exist, a specific fluctuation does not occur in the reception level, so that the reception level can be clearly distinguished. Also, when the reception level fluctuates due to changes in temperature, rain and snow, etc., the fluctuation may reach from 10 dB to 20 dB, but it is a gradual change with time and Since there is no possibility of periodic fluctuations, it can be clearly distinguished from the case where the obstacle 15 exists.

As described above, the extracted fluctuation pattern extracted from the received spread spectrum signal received by the reception leaky transmission line 14 and the setting fluctuation pattern signal stored in the setting pattern storage means 23 match, When the fluctuation level of the extracted fluctuation pattern signal is equal to or higher than a predetermined value, the calculation means 24 determines that the obstacle 15 exists, thereby avoiding the influence of the change in ambient temperature and the weather condition such as rain and snow, and the obstacle 15 is removed. It is possible to prevent erroneous detection of. Furthermore, the position of the obstacle 15 within the detection range can be detected by calculating the position of the obstacle 15 from the center position of the pattern of the extracted fluctuation pattern signal in the calculating means 24.

Embodiment 2. FIG. 3 is a configuration diagram of the second embodiment. In FIG. 2, 11 to 17 and 23 are the same as those in the first embodiment. 25 and 26 are delay means, which are delay times t corresponding to the paths 13a and 13b, respectively.
1 and t2 are generated. Reference spread spectrum signal generating means 27 and 28 are provided for the reference spread spectrum signals 27a and 27a at phase timings corresponding to the delay times t1 and t2, respectively.
28a is output. Reference numerals 29 and 30 denote correlating means, which respectively correlate the corresponding reference spread spectrum signals 27a and 28a with the distributed spread spectrum signals 31a and 31b output from the reception signal distributing means 31, which will be described later, to determine the measurement point at each measurement distance. The received signal level is output as signal level output signals 29a and 30a. Reference numeral 31 is a reception signal distribution means, which is a distribution spread spectrum signal 3 obtained by halving the reception spread spectrum signal 14a received by the reception leaky transmission line 14.
1a and 31b are input to the respective correlating means 29 and 30. 32
Is a memory means for storing data in which the signal level output signals 29a and 30a are associated with the measured distances. 33
Is a variation pattern extraction means, which detects variations in the received signal level in the distance direction from a plurality of data strings accumulated in the memory means 32, and extracts the extracted variation pattern signal.

Reference numeral 34 denotes a computing means which compares the extracted variation pattern extracted by the variation pattern extraction means 22 with the setting variation pattern of the setting pattern storage means 23 to detect the degree of coincidence, and when the variation level is a predetermined value or more. If there is, it is determined that the obstacle 15 exists within the detection range, and the obstacle detection signal 34a is output. Next, the operation will be described. In FIG.
Similar to the first embodiment, the transmission spread spectrum signal 16a radiated from the transmission leaky transmission line 13 has the paths 13a, 13
The signal is received by the reception leaky transmission line 14 via b and output as the reception spread spectrum signal 14a. Received spread spectrum signal 14a input to received signal distribution means 31
Is divided into two and input to the correlating means 29 and 30, respectively.

On the other hand, the delay means 25 and 26 are provided with a delay time t corresponding to the measurement distance set by the measurement distance setting means 17.
1 and t2 are generated. Then, the reference spread spectrum signal generating means 27, 28 generate reference spread spectrum signals 27a, 28a having phase timings corresponding to the delay times t1, t2, respectively, and input them to the correlating means 29, 30. The correlating means 29 and 30 respectively correlate the corresponding reference spread spectrum signals 27a and 28a and the distributed spread spectrum signals 31a and 31b, and the received signal level via the paths 13a and 13b at the measurement points corresponding to the respective measurement distances. Are output as signal level output signals 29a and 30a. The correlating means 29 and 30 perform a series of processes in parallel and store a series of data strings in the memory means 32. The fluctuation pattern extraction means 33 detects fluctuations in the received signal level for each measurement distance from the plurality of data strings accumulated in the memory means 32, and extracts the extracted fluctuation pattern signal. Then, the calculation means 34 is the fluctuation pattern extraction means 33.
The extracted variation pattern signal extracted by is compared with the setting variation pattern signal stored in the setting pattern storage means 23 at each measurement point corresponding to the measurement distance. in this case,
When the obstacle 15 exists, the extracted variation pattern signal extracted from the received spread spectrum signal 14a received on the path 13a, for example, has the same pattern as the variation pattern stored in the setting pattern storage unit 23. If the fluctuation level is equal to or higher than the predetermined value, it is determined that the obstacle 15 exists within the detection range, and the obstacle detection signal 34a is output. Further, the position of the obstacle 15 is calculated from the center position of the pattern of the extracted fluctuation pattern signal. Embodiment 1
Similarly, for example, by subjecting a series of data strings to FFT processing, the frequency components included in the series of data strings can be read with high accuracy and the distribution of the size of each frequency component can be identified. Level fluctuations can occur to identify specific fluctuation patterns.

As described above, the extraction fluctuation pattern signal extracted from the reception spread spectrum signal received by the reception leaky transmission line 14 is distributed to the plurality by the reception signal distribution means 31, and the extracted extraction fluctuation pattern signal is distributed. When the correlation with the reference spell spread signal corresponding to the delay time is processed in parallel by the plurality of correlating means 29 and 30, and the patterns match and the fluctuation level of the extracted fluctuation pattern signal is equal to or more than a predetermined value,
By determining that the obstacle 15 is present in the calculation unit 34, it is possible to shorten the processing time, avoid the influence of weather conditions such as changes in ambient temperature and rain / snow, and prevent erroneous detection of the obstacle 15. it can. Furthermore, the position of the obstacle 15 within the detection range can be detected by calculating the position of the obstacle 15 from the center position of the pattern of the extracted variation pattern signal in the calculation means 34.

In the second embodiment, in order to extract a specific variation pattern of the reception level in the distance direction, a series of data strings is subjected to FFT processing, and the distribution of the magnitude of each frequency component is identified to identify a specific pattern. Although the case of determining the occurrence of a fluctuation pattern has been described, the same effect can be obtained by determining the occurrence of a specific fluctuation pattern by identifying the coincidence of waveform patterns formed by arranging a series of data using the pattern recognition method. Can be expected. In the first and second embodiments, the spread spectrum signal for transmission 1
Although 6a detects the obstacle 15 from the two measurement points via the routes 13a and 13b, the same effect can be expected even if the measurement points are three or more. Further, in the second embodiment, the received signal spread means 31 divides the received spread spectrum signal 14a into two equal parts, and the spread spectrum signals 31a and 31b, and the delay time t.
The reference spread spectrum signals 27a and 27b having the phase timings corresponding to 1 and t2 are input to the correlating means 29 and 30 for correlation, but the received spread spectrum signal 14a is divided into three or more equal parts. Similar effects can be expected by inputting the spread spectrum signal and the reference spread spectrum signal of the phase timing corresponding to the delay time of 3 or more to the correlation means of 3 or more to take the correlation. .

[0018]

According to the present invention, the extracted fluctuation pattern extracted from the received spread spectrum signal received by the reception leaky transmission line and the setting fluctuation pattern signal stored in the setting pattern storage means match each other. When the fluctuation level of the extracted fluctuation pattern signal is equal to or higher than a predetermined value, the operation means determines that there is an obstacle, thereby avoiding the influence of weather conditions such as changes in ambient temperature and rain and snow, and Detection can be prevented. Further, the extracted fluctuation pattern signal extracted from the received spread spectrum signal received by the reception leaky transmission line is distributed to a plurality by the received signal distribution means, and the reference fluctuation spectrum signal corresponding to the distributed extracted fluctuation pattern signal and delay time is distributed. Correlation with the signal is processed in parallel by a plurality of correlating means, and when the patterns match and the fluctuation level of the extracted fluctuation pattern signal is above a predetermined value, the calculating means judges that an obstacle exists, thereby shortening the processing time. At the same time, it is possible to avoid influences of changes in ambient temperature and weather conditions such as rain and snow, and prevent erroneous detection of obstacles. Further, the position of the obstacle within the detection range can be detected by calculating the position of the obstacle from the central position of the pattern of the extracted variation pattern signal in the calculation means.

[Brief description of drawings]

FIG. 1 is a configuration diagram of a first embodiment of the present invention.

FIG. 2 is an explanatory diagram showing a level variation of an electric field distribution due to a metal cylinder.

FIG. 3 is a configuration diagram of a second embodiment of the present invention.

FIG. 4 is a configuration diagram of a conventional home fall detection device on a track.

[Explanation of symbols]

11 homes, 12 lines, 13 leaky transmission lines for transmission,
14 reception leaky transmission line, 15 obstacle, 16 transmission spread spectrum signal generating means, 17 measurement distance setting means,
18, 25, 26 delay means, 19, 27, 28 reference spread spectrum signal generation means, 20, 29, 30 correlation means, 21, 32 memory means, 22, 33 fluctuation pattern extraction means, 23 setting pattern storage means, 24, Three
4 computing means, 31 received signal distributing means.

Claims (3)

[Claims] 1. A home fall detection device for detecting an obstacle that has fallen on a track adjacent to a home and within a detection range of a clearance space between the track and the platform, and is arranged on one side of the detection range along the home. A leaky transmission line for transmission,
A reception leaky transmission path arranged opposite to the transmission leaky transmission path on the other side of the detection range, and a transmission spread spectrum signal is radiated from the transmission leaky transmission path toward the reception leaky transmission path. And a measuring distance from the one end to the measuring point for detecting the obstacle, in the direction from one end to the other end of the transmission spread spectrum signal inputting means for transmitting the spread spectrum signal Measuring distance setting means, delay means for generating a delay time according to the measuring distance, reference spread spectrum signal generating means for outputting a reference spread spectrum signal having a phase timing according to the delay time, and for receiving the The received signal level at the measurement point is obtained by correlating the received spread spectrum signal received by the leaky transmission line with the reference spread spectrum signal. Correlation means for outputting as a bell output signal, memory means for accumulating the signal level output signal in correspondence with the measuring distance, and fluctuation of the received signal level from the data string for each measuring distance accumulated in the memory means. A variation pattern extraction means for extracting as an extraction variation pattern signal, a setting pattern storage means for storing a preset variation pattern signal, and a pattern of the extracted variation pattern signal and the setting variation pattern signal match, A home fall detection device comprising: a computing unit that determines that the obstacle exists when the variation level of the extracted variation pattern signal is equal to or greater than a predetermined value. 2. A home fall detection device for detecting an obstacle that has fallen on a track adjacent to a home and within a detection range of a clearance space between the track and the platform, wherein the home fall detection device is arranged along the home on one side of the detection range. A leaky transmission line for transmission,
A reception leaky transmission path arranged opposite to the transmission leaky transmission path on the other side of the detection range, and a transmission spread spectrum signal is radiated from the transmission leaky transmission path toward the reception leaky transmission path. Transmitting spread spectrum signal generating means, and the measurement distance from one end to the other end of the transmitting leaky transmission path to which the transmitting spread spectrum signal is input, from the one end to a plurality of measurement points for detecting the obstacle A measuring distance setting means for setting, a delay means for generating a delay time according to each of the measuring distances, and a reference spread spectrum signal generating means for outputting a reference spread spectrum signal having a phase timing according to each delay time, , A distributed spread spectrum signal for distributing the received spread spectrum signal received by the reception leaky transmission line corresponding to each delay time is output. Signal distribution means, correlation means for correlating each of the distributed spread spectrum signals and each of the reference spread spectrum signals, and outputting a received signal level for each of the delay times, and each of the received signal levels for each of the measurement distances. Memory means for storing in correspondence with the above, a variation pattern extracting means for extracting the variation of each received signal level as an extracted variation pattern signal from the data sequence for each of the measured distances accumulated in the memory means, and preset setting Setting pattern storage means for storing the fluctuation pattern signal, the patterns of the extracted fluctuation pattern signal and the setting fluctuation pattern signal corresponding to the measurement distance match, the fluctuation level of the extraction fluctuation pattern signal is a predetermined value or more At this time, a home fall detection device comprising a calculation means for determining that the obstacle is present. 3. The home fall detection device according to claim 1, wherein the calculation means calculates the position of the obstacle from the center position of the pattern of the extracted fluctuation pattern signal.