WO2011138773A1

WO2011138773A1 - Smart fence

Info

Publication number: WO2011138773A1
Application number: PCT/IL2011/000056
Authority: WO
Inventors: Beni Magal
Original assignee: Beni Magal
Priority date: 2010-05-03
Filing date: 2011-01-19
Publication date: 2011-11-10

Abstract

A sensing system for a fence having string wires and support pillars that comprises a plurality of sensors coupled to the string wires and configured to allow generating data representing movement of the wire coupled thereto. It comprises also spring elements connecting the string wires to the support pillars and a central communication unit configured to allow simultaneously obtaining data representing a portion of the data of the sensors as well as a storage unit configured to allow repeatedly storing additional data representing a portion of the data over a predefined period of time. Real-time analysis of data is processed by attributing portions of data to the string wires, identifying an event occuring in the predefined period on the fence based on the data, identifying the string wires at which the event occurs in the predefined period, and generating alerts regarding the event.

Description

SMART FENCE

FIELD OF THE INVENTION

The present invention relates to fences. In particular, the invention relates to intrusion detection systems that may be integrated with or retrofitted into fences.

BACKGROUND

Wire fences are typically used as easily erectable and relatively inexpensive means to secure property, animals and people within a bounded perimeter. Wire fences may be used for escape discouragement, for example in human prisons or cattle yards. Alternatively, fences may be used for intrusion detection as penetration prevention means. Wire fences are also a major feature of fortifications in trench warfare.

Wire fences may be constructed as a mesh of metal strips, requiring only fence posts, wire, and fixing devices such as staples. Barbed wire comprising sharp edges or points arranged at intervals along the strands may be used in a fence construction. A person or animal trying to pass through or over the barbed wire will suffer discomfort and possible injury. People, however, may climb over or through a regular barbed wire fence, for example by stretching the gaps between the wires using non-barbed sections of the wire as hand holds.

Razor wire is a barbed wire variant which instead of occasional barbs features near-continuous cutting surfaces, more likely to injure an intruder. Several European countries such as England and France have begun restricting the use of barbed wire variants due to the risk of injury they pose to animals and trespassers.

Wire fences are often integrated with intrusion detection and / or prevention systems. Intrusion detection focuses on monitoring the fence and detecting malicious activities, and intrusion prevention further attempts to stop such activities. Such systems typically identify security violations, log information about them, attempt to stop them, and report them to a management station.

Typical fence intrusion detection systems comprise a plurality of sensors capable of monitoring events such as vibrations and heat signals and communicating them to a central control station for evaluation. Alternative approaches such as mentioned in US patent No. 7,450,006 to Doyle et al. titled "Distributed perimeter security threat confirmation" use smart sensor systems capable of performing threat evaluation and communicating with each other, thus reducing processing resources at the central evaluation station.

Such sensor systems may be able to provide some differentiation between environmental factors (such as wind or rain) and human intervention, but may nevertheless be prone to generating false alarms. Unless further equipped with cameras, such sensor systems are not capable of providing an ongoing image of the fence's state at real-time.

The need remains therefore, for a simple, accurate, affordable, real-time intrusion detection system which can be integrated into wire fences. Embodiments described hereinbelow address this need.

SUMMARY OF THE INVENTION

According to one aspect, an online sensing system for a fence having string wires and support pillars is provided, the system comprising:

a plurality of sensors, wherein each sensor of the plurality of sensors is coupled to one of the string wires, and wherein each sensor is configured to allow generating first data representing movement of the wire coupled thereto; spring elements connecting at least some of the string wires to the support pillars; at least one central communication unit configured to allow simultaneously obtaining second data representing at least a portion of the first data of the sensors; at least one storage unit configured to allow repeatedly storing third data representing at least a portion of the second data over a predefined period of time; and at least one processing unit configured to allow performing real-time analysis of fourth data representing at least a portion of the third data, wherein the analysis comprises:

attributing portions of the fourth data to the string wires, wherein each portion represents first data from a sensor coupled to the attributed string wire; identifying at least one event occuring in the predefined period on the fence based on the fourth data; identifying the string wires at which the events occur in the predefined period, and generating alerts regarding at least some of the events, wherein the alerts each comprises informing the identity of the alerted event and string wires at which each event occurred, with the proviso that the central communication unit and the storage unit are not configured to identify the events.

In some embodiments, the processing unit is further configured to allow generation of a substantially real-time image representing the fence, the image comprises indicators of the string wires at which the events occurred and a type of each event.

In some embodiments, the processing unit is further configured to allow generation of a substantially real-time audio signal, the audio signal representing indicators of the string wires at which the events occurred and a type of each event.

In some embodiments, the sensors are each capable of measuring movement of the string wire coupled thereto. In some embodiments, the measurement of movement is independent measurement of movement in two or three axes. In preferred embodiments, the measurement of movement is independent measurement of movement in three axes.

In some embodiments, the plurality of sensors are selected from a group consisting of: accelerometers, speedometers, gyroscopes, and combinations thereof. Some embodiments further comprise at least one display unit configured to display the substantially real-time image of the fence. The display unit may be selected from a group consisting of: computer screens, laptops, PDAs, cellular phone screens, printed sheets, integrated LCD screens , Thin Film Transistors, touch screens and combinations thereof.

In some embodiments, the system is further configured to allow generation of alerts for one or more of the events. In some embodiments, the display unit is configured to display these alerts. In some embodiments, the real-time image comprises dynamic representation of occurrence of events. In preferred embodiments, the system is configured to allow a user selection of event types to be dynamically represented.

In various embodiments, the system further comprises at least one sensor pole comprising: a protective housing configured to accommodate at least a portion of the plurality of sensors; at least one driving unit; at least one sensor pole communication unit; and at least one power source; wherein the power source is configured to supply power to the portion of said plurality of sensors, driving unit and sensor pole communication unit, and wherein said driving unit is configured to collect first data from said portion of the plurality of sensors, and wherein the sensor pole communication unit is configured to transmit second data to a central communication unit.

In some embodiments, the portion of the plurality of sensors are vertically aligned within the sensor pole. In some embodiments, at least portion of the plurality of sensors comprises sensors having different sensitivity levels.

In preferred embodiments, the second data comprises: a sensor pole ID representing the sensor pole communication unit from which the first data was received; a sensor ID within the sensor pole, representing the string wire to which each of the plurality of sensors is coupled to; and movement parameters of the sensor independently in each of three axes. In preferred embodiments, the movement parameters are selected from a group consisting of: acceleration, velocity, position and combinations thereof. In some embodiments, the system is configured to retrofit with existing fences. In other embodiments, the system is integrated within a fence having string wires.

Other embodiments teach a method for a fence having sections comprising string wires and sensors configured to produce data representing a time-dependent pattern of movement of the string wires in the sections, the method comprising: collecting the data from the plurality of sensors; sending at least a portion of the data to a central communication unit configured to receive the portion of data; processing in concert the portion of the data, generating a dynamic virtual image of the fence, identifying events occurring on the fence sections and generating alerts to a user regarding at least part of the events.

In some embodiments, the processing comprises mapping each of the plurality of sensors to a specific string wire and a specific fence section and interpreting the movement parameters obtained from each of the plurality of sensors to movements of specific string wires in specific fence sections. In some embodiments identifying events occurring on the fence further comprises differentiating between human generated events and non-human generated events.

In preferred embodiments, events are selected from a group comprising: penetration, intrusion, cutting at least one of the string wires, spreading string wires apart, climbing upon string wires, collision with the string wires, wind, rain, and combinations thereof.

In some embodiments, collecting the data from the plurality of sensors comprises periodic readings obtained at a sample speed and sensitivity rate set by a user. Optionally, the sample speed and sensitivity rate change according to pre- defined thresholds. Optionally, the sample speed and sensitivity rate change according to previously detected events.

In some embodiments, interpreting the movement parameters further comprises determining a location within a fence section where forces are applied upon the string wires. In some embodiments, the fence further comprises at least two sensor poles within each fence section, a portion of said plurality of sensors being vertically aligned within each of said sensor poles, determining a location within a fence section where forces are applied upon the string wires further comprises: obtaining movement data in at least two axes from two sensors coupled to a single string wire laid between two sensor poles comprising each one of the two sensors, translating the movement data to a movement vector having a movement angle, and using the movement angle in respect to the direction of the string wire as an indication of the location on the wire between the sensor poles of the force being applied to the string wire.

In preferred embodiments, the time-dependent pattern is compared to predefined patterns. Optionally, the time-dependent patterns are studied and added over time. Optionally, time-dependent patterns are compared with data obtained from external sources. Optionally, such external sources are selected from a group comprising: humidity detection sensors, cameras, and weather stations.

In some embodiments, a pattern characterized by a continual wavy movement across a plurality of fence sections indicates a wind event. In some embodiments, a pattern characterized by strong movement of wires in the Z direction indicates a collision event.

In preferred embodiments, multiple interpretations for event types may be suggested to a user in one alert.

BRIEF DESCRIPTION OF THE FIGURES

For a better understanding of the invention and to show how it may be carried into effect, reference will now be made, purely by way of example, to the accompanying drawings.

With specific reference now to the drawing in detail, it is stressed that the particulars shown are by way of example and for purposes of illustrative discussion of the preferred embodiments of the present invention only, and are presented in the cause of providing what is believed to be the most useful and readily understood description of the principles and conceptual aspects of the invention. In this regard, no attempt is made to show structural details of the invention in more detail than is necessary for a fundamental understanding of the invention; the description taken with the drawing making apparent to those skilled in the art how the several forms of the invention may be embodied in practice.

In the accompanying drawings:

Figures 1A illustrates an embodiment of prior-art fence construction section;

Figures IB illustrates another embodiment of prior-art fence construction section;

Figure 2A schematically illustrates a single sensor hanging from a string wire in an exemplary embodiment of the present invention;

Figure 2B schematically illustrates another single sensor coupled to a string wire in another exemplary embodiment;

Figure 3A schematically illustrates a cross-sectional view of a multi-sensor pole in an exemplary embodiment;

Figure 3B schematically illustrates an exterior view of a multi-sensor pole in an exemplary embodiment;

Figure 3C illustrates an interior view of a multi-sensor pole in an exemplary embodiment;

Figure 4A schematically illustrates a fence section having two sensor poles positioned between two support pillars in an exemplary embodiment;

Figure 4B schematically illustrates another fence section having a sensor pole position between two support pillars in another exemplary embodiment;

Figure 4C schematically illustrates a string wire connected to support pillars via spring elements; is a block diagram illustrating the main components of a sensor pole in an exemplary embodiment; is a block diagram illustrating the main components of a central station in an exemplary embodiment ; is a schematic block diagram illustration of the main components of a fence sensing system in an exemplary embodiment;

illustrates the fence section of Figure 4 where some string wires are pulled upward and downward in an exemplary embodiment;

is a close up view of the pulled string wires of Figure 6a in an exemplary embodiment;

is a table showing information obtained from a single sensor pole integrated into a fence construction in an exemplary embodiment; is an exemplary graphic representation of information regarding acceleration, velocity and position received from a sensor in an exemplary embodiment;

is an exemplary graphic representation of acceleration information received from a sensor;

is an exemplary graphic representation of velocity information received from a sensor;

is an exemplary graphic representation of position information received from a sensor;

illustrates a snapshot of a climbing pattern synthesized from input data received from a plurality of sensors; and

illustrates a flow chart of a method for generating a virtual image of a fence, identifying events and generating alerts regarding the events. DETAILED DESCRIPTION OF THE SELECTED EMBODIMENTS

Various embodiments of a sensing system and method for generating alerts based on events which were identified by creating an artificial image of a string wire fence are disclosed hereinbelow. The system can be used for example and without limitation for intrusion detection. By generating a substantially real-time virtual image of the fence and its movements, the system is capable of performing security evaluations and identifying possible threats to the fence or the perimeter it protects. The system typically uses acceleration sensors capable of measuring string wire movements in 3 axes. Movement data obtained from a plurality of sensors is used by a processing unit to create for example a virtual dynamic time-dependent image of the fence and to visually simulate the movement of the string wires comprising the fence.

PRIOR ART

Reference is now made to Figure 1A showing an embodiment 10 of prior-art fence construction section. In the figure, a section is defined by two support pillars 12a and 12b, vertically attached to a horizontal surface 14. The figure illustrates a string barbed wire fence 18 stretched across support pillars 12a and 12b.

Reference is now made to Figure IB illustrating a schematic view of a typical prior-art fence construction 20 comprising two consecutive sections 10a and 10b defined by three support pillars 12a, 12b and 12c vertically attached to a substantially horizontal surface 14, for example the ground. Support pillar 12b is common to the two consecutive sections 10a and 10b. The figure further depicts two taut wire fences 22a and 22b, fence 22a stretched from support pillar 12a to support pillar 22b and fence 22b stretched from support pillar 12b to support pillar 12c. Each of string wire fences 22a and 22b comprises a plurality of string wires 24 horizontally stretching across support pillars such that each string wire 24 is parallel to horizontal surface 14. FENCE IMAGE GENERATION SYSTEM CONSTRUCTION

For the purpose of illustration and without limiting the scope of the disclosure, embodiments described hereinbelow will be demonstrated via a specific example of retrofitting the fence image-generation system into existing fence constructions.

A preferred embodiment will be disclosed hereinbelow, configured to be retrofitted into a string wire fence construction as illustrated in Figures 1A and IB. The wires used to construct the fence may be made of from a plurality of materials for example and without limitation metal such as steel, iron, aluminum, and combinations thereof. The fence may be constructed from regular wires, barbed wires, or any other wires which suit fencing requirements.

Reference is now made to Figure 2A illustrating a single sensor 40 hanging from a wire 30 stretched substantially in parallel to a horizontal surface 14. Surface 14 may be, for example and without limitation the ground. In this embodiment, wire 30 is a barbed wire having spikes (two spikes, 32a and 32b are seen for drawing clarity). Connecting element 34 is used for connecting sensor 40 to wire 30 in a manner that enables sensor 40 to move freely while still connected to wire 30. Connecting element 34 may be for example and without limitation a hook, a clip, a magnet, a string knot, or combinations thereof. In the embodiment shown in Figure 2, sensor 40 hangs from wire 30 by optional wire 36 attached to connecting element 34, thus creating a state in which a movement of wire 30 triggers movement of sensor 40.

Vectorial movement of sensor 40 is directly affected by movement and tension of wire 30, and defined by both magnitude and three-dimensional direction: · side movement (left and right, X axis) roughly parallel to surface 14 and parallel to the direction in which string wire 30 stretches;

• elevation movement (up and down, Y axis), roughly perpendicular to surface 14, capable of measuring changes in elevation of string wire 30 relative to surface 14; and • forward movement (in and out, Z axis), roughly parallel to surface 14 and perpendicular to the direction in which string wire 30 stretches.

Sensor 40 may be, for example and without limitation gyroscopes, speedometers or accelerometers capable of measuring acceleration of the triggered movement. Other sensors and sensor combinations may also be used, which suit requirements of measuring movement-related parameters such as but not limited to wire tension, speed, acceleration, rotation and direction. In preferred embodiments, the sensor may be a known sensor, such as but not limited to Freescale's MMA7455L, which is a Three-Axis Low-g Digital Output Accelerometer. This low power, low profile capacitive micromachined accelerometer features signal conditioning, a low pass filter, temperature compensation, and self-testing. It is configurable to detect Og through interrupt pins, and pulse detect for quick motion detection. Og offset and sensitivity are factory set and require no external devices. The Og offset can be customer calibrated using assigned Og registers and g-Select which allows for command selection for 3 acceleration ranges (2g/4g/8g).

Further features of the sensor include:

• Digital Output (I2C/SPI)

• Small size: 3mm x 5mm x 1mm LGA-14 Package

• Self-Test for Z-Axis

• Low Voltage Operation: 2.4 V - 3.6 V

• User Assigned Registers for Offset Calibration

• Programmable Threshold Interrupt Output

• Level Detection for Motion Recognition (Shock, Vibration, Freefall)

• Dual mode operation (accelerometer / gyroscope)

• Pulse Detection for Single or Double Pulse Recognition

• Sensitivity (64 LSB/g @ 2g and @ 8g in 10-Bit Mode) • Selectable Sensitivity (+2g, ±4g, ±8g) for 8-bit Mode

• Sample rate of up to 120 samples per second

• Robust Design, High Shocks Survivability (5,000g)

• RoHS Compliant Reference is further made to Figure 2B illustrating a connecting element 34' threaded through string wire 30 and accommodating a single sensor 40'. Movement in string wire 30 is transferred to movement in connecting element 34' which is capable of moving in three axes as seen in the Figure. In preferred embodiments, movement data is recorded for three axes, but other embodiments may record data for two axes, for example and without limitation the X and Y axes only.

Reference is now made to Figure 3A showing a schematic cross-sectional view of a fence section including a multi-sensor pole 50 comprising a housing 52 for a plurality of sensors 40a - 40f hanging from wires 30a - 30f respectively. Wires 30a - 30f are substantially parallel to each other and stretch horizontally from support pillars (not shown). Wires 30a - 30f are also substantially parallel to substantially horizontal surface 14. Sensors 40a - 40f are lined vertically, one on top of the other within the housing 52. It should be noted that number of wires 30 may vary, and optionally number of sensors 40 may differ from the number of wires within the scope of the current invention.

Sensor pole 50 typically serves as a main data aggregation unit configured to collect data from a plurality of sensors on a plurality of string wires. A sensor pole 50 typically comprises a communication and control unit 60 capable of receiving and optionally storing data from the sensors 40a - 40f within the sensor pole 50 and transmitting it to an external processing unit (not shown).

Sensor pole 50 comprises a housing serving as an external envelope for the sensors 40a - 40f. The housing may be constructed from a plurality of materials, natural and synthetic, such as but not limited to metal, nylon or plastic. In preferred embodiments, the sensor pole is made of relatively inexpensive material. Preferably, the housing is made of material capable of protecting residing sensors from environmental hazards which may be caused by excess exposure to sun, wind, rain or the like. The housing typically assists in tamper proofing, making deliberate, undetected interference with the sensors within the sensor pole nearly impossible. In some embodiments, the housing may be partially or completely camouflaged, such that it mixes with the surrounding environment and is relatively unrecognizable to an untrained eye.

Sensors within a pole may be identical to each other. Alternatively, different sensors having different sensitivity levels may be used in a single pole, for example more sensitive sensors connected to middle wires where most security breaches are likely to occur.

In some embodiments, the bottommost part of a sensor pole 50 is embedded into the ground, such as shown in Figure 3A. In this case, sensor pole 50 may further serve as a support pillar (12a and 12b in Figures 1A - 1C). In other embodiments, the sensor pole may not be attached to the floor but rather to a "lintel" connecting two support pillars to each other. Alternatively, the sensor pole may be constructed to float upon fence wires 30 without a structural support element other than the wires themselves.

Reference is now made to Figure 3B showing a schematic exterior view of multi-sensor pole 50 comprising front apertures 52 and back apertures (not shown). The figure shows frontal apertures 52a-f for threading wires through sensor pole 50. For example, wire 30c may be threaded through multi-sensor pole 50 via front aperture 52c and a respective back aperture (not shown).

Figure 3C illustrates an interior view of an embodiment of a sensor pole 50 in an open configuration. The figure illustrates pole 50, three connecting elements 34a, 34b and 34c coupled with acceleration sensors 40a, 40b and 40c and threaded through wires 30a, 30b and 30c respectively. The Figure illustrates connecting element 34a slightly tilted as a result of a pulling action in string wire 30a. This tilting movement is recognized by coupled sensor 40a. Acceleration data from sensor 40 is recorded and transferred via a communication and control unit within the pole (not shown) to a central station (not shown).

Reference is now made to Figure 4A showing a fence section having two sensor poles 50a and 50b positioned between two support pillars 12a and 12b in some embodiments. Sensor poles may be placed at a variable distance from support pillars 12a and 12b and from each other. A different number of sensor poles may be used within a single fence section 19 defined by two support pillars 12a and 12b. In some embodiments some or all of support pillars 12 comprise sensors as well.

Reference is now made to Figure 4B showing another fence section 19' defined by two support pillars 12c and 12d in other embodiments. In these embodiments, the distance between support pillars 12c and 12d may be fifty meters.

String wires horizontally stretch from support pillar 12c to support pillar 12d.

Multiple string wires are positioned in parallel to the ground and vertically spaced aprt up to a height of nearly three meters. The vertical distance between one string wire to another may be seven to ten centimeters. A single sensor pole 50c is preferably positioned in the middle of the fence section, at a distance of roughly 25 meters from each of support pillars 12c and 12d. Alternatively, other configurations using different measurements may also be used.

Reference is now made to Figure 4C showing a string wire 30 connected to support pillars 12 via spring elements 26. In preferred embodiments, some or all of string wires 30 may be connected to support pillars 12 via spring elements 26. Spring elements contribute to greater sensitivity of string wires 30 when compared to string wires connected directly to support pillars. The springs may cause stronger acceleration of the string wire and acceleration sensors coupled to it in response to events occurring in the proximity of the fence section.

Reference is now made to the block diagram of Figure 5A representing the main components of a sensor pole 50 having a plurality of sensors 40 and a communication and control unit 60 in some embodiments. Communication and control unit 60 typically comprises at least one power source 62, at least one driver unit 64 capable of communicating with sensors 40 and at least one sensor pole communication unit 66 capable of transmitting data collected from sensors 40 to an external processing unit (not shown).

Power source 62 is typically capable of supplying power to sensors 40, driver unit 64 and sensor pole communication unit 66. Power source 62 may be wired to a main grid or alternatively sources such as but not limited to a replaceable and rechargeable battery or a solar panel with a rechargeable battery. Sensor pole communication unit 66 may use a multitude of technologies which suit requirements, wired or unwired, such as but not limited to radio or cellular communication link 69. Communication link 69 may be unidirectional, transferring sensor data from sensor pole 50, or bi-directional, further transferring control signals to sensor pole 50. In some cases, wired data cables may be inappropriate and data transmission via wireless means may be preferred, for example via radio waves using protocols such as Wi-Fi, Bluetooth or the like. Reference is now made to the block diagram of Figure 5B representing the main components of a central station 70 used in a general fence sensing system embodiment. A central station 70 is capable of receiving information collected from a plurality of sensor poles (seen in details in Figure 5A) via a central communication unit 72. Data is collected and optionally stored in Data aggregation and storage unit 74. Processing unit 76 is capable of analyzing the collected data as will be described hereinbelow. Processing output may include events which were analyzed from the collected data. The majority or all of the processing and analysis is typically conducted processing unit 76. This output may be displayed to a user for example by using an optional display unit 78. Optionally, the system may further utilize central communication unit 72 to send analyzed data to a remote station, optionally for further processing and display.

Other embodiments of fence sensing systems can be designed for scale, aiming to monitor taut wire fences and to dynamically simulate fence images representing a plurality of fence constructions, optionally located in remote sites. Such embodiments may include a plurality of fences each comprising a plurality of sensor poles 50 transmitting data to at least one central station 70. Additionally and alternatively, a single central station may be used to monitor data obtained from a plurality of fences.

Power may be supplied to sensor poles from a plurality of sources such as batteries and power lines, multiple processors may be used for calculation and analysis of the data, which may be stored in a plurality of data storage units. Display systems may vary in size, number and configuration, and optionally include computer screens, laptops, PDAs, cellular phone screens, printed sheets, integrated LCD screens (e.g. Thin Film Transistors, touch screens) and the like. Reference is now made to the schematic block diagram of Figure 5C illustrating a Fence sensing system 100 comprising a plurality of sensor poles 50a to 50n in communication with a common central station 70 . For simplicity, four sensor poles are seen in the figure, but the number may vary. Optionally, a single sensor pole may communicate with more than one central station.

FENCE IMAGE GENERATION - PROCESSING

The concerted processing of data obtained from a plurality of sensor poles enables automatic monitoring of fence movements at a relatively high sample rate, analyzing and interpreting the movement according to pre-defined movement patterns, indentifying the location of the movement on the fence and displaying an artificial dynamic, time-dependent image of the moving fence.

Optionally, system 100 is capable of identifying specific events such as but not limited to intrusion or escape attempts. Optionally, the system is capable of reacting to some or all of detected events, for example and without limitation by generating alerts to fence operators, electrifying sections of the fence to scare intruders, activate or direct other sensors such as video or thermal cameras to the suspected intrusion, turn on lights or direct search light, sound an alarm, or other actions which suit requirements.

Event detection will be demonstrated hereinbelow via a specific example of an attempted intrusion event, where a thief tries to pass through the wired fence by bending two or more substantially horizontal string wires and broadening the gap between them, making the gap large enough to enable a person to pass through the fence.

Reference is now made to Figure 6A illustrating of a fence section having two sensor poles 50a and 50b essentially as illustrated in Figure 4A. The figure illustrates string wires 30a - 30f threaded through sensor poles 50a and 50b. Gap 80 between string wire 30e and string wire 30f is widened, due to the pulling of string wire 30e upwards towards string wire 30d, and the pulling of string wires 30f and 30g downwards towards string wire 30h. This is demonstrated in the schematic close-up view in Figure 6B.

Figure 6B also illustrates sensors 40d, 40e, 40f, 40g, and 40h in sensor pole 50a and sensors 40d', 40e\ 40f, 40g'₍ and 40h' in sensor pole 50b. Although not shown in the figure, each of the sensors hangs on a wire string essentially as schematically shown in Figure 3A. For example,

• sensors 40d and 40d' are connected to string wire 30d;

• sensors 40e and 40e' are connected to string wire 30e;

• sensors 40f and 40f are connected to string wire 30f;

• sensors 40g and 40g' are connected to string wire 30g; and · sensors 40h and 40h' are connected to string wire 30h.

The sensors are connected to the wires in a way that a movement of a wire triggers a movement in the sensor attached to it. In the embodiments shown in Figures 6A and 6B, string wires 24d and 24h remain essentially stationary, and thus no substantial movement is detected by either of the sensors attached to them. String wire 24e is pulled upwards, and its corresponding sensors 40e and 40e' in sensor poles 50a and 50b respectively detect significant movement in the Y axis. Similarly, sensors 40f and 40 corresponding to string wire 24f record a significant movement in the Y axis as string wire 24f is pulled downwards. String wire 24g is pulled downwards only slightly, and thus sensors 40g and 40g' attached to it detect a slight movement in the Y axis. Most likely, some movement will be made in the other axes as well.

Periodic readings are typically obtained at a sample speed and sensitivity rate which may be set by a user, for example and without limitation 2 readings per second. In some embodiments, the sample speed and sensitivity rate may change once a threshold is reached or an event is detected. For example, the sampling rate of a specific sensor may be doubled if strong motion is detected in either of the axes, or the sampling rate of a set of sensors residing on a single sensor pole may change when an intrusion event is detected.

The sample rate should be relatively high, to enable detection of events which may occur within a very limited time interval. For example, climbing a two- meter barbed wire fence may take as little as eight or ten seconds. Thus, it is important that the sample rate will sufficiently high, such that multiple readings from the sensors will be transmitted to a central station in this time interval, and allow real-time construction of a movement pattern sufficiently detailed to allow real-time characterization of the event.

Information obtained from sensors about wire state changes such as illustrated above is obtained from a plurality of sensors optionally arranged on a plurality of sensor poles. For example and without limitation, the information obtained from the sample illustrated in Figure 6B may take the form of table 90 of Figure 7. Figure 6B represents a situation in which the fence section is defined by two sensor poles 50A and 50B, and in which forces applied upon the string wires 30e, 30f, and 30g are applied in the middle of the fence section.

In the table, column 91 represents the sensor pole's ID, column 92 represents a sensor ID within the sensor pole of column 91, column 93 represents the sensor's detected movement in the X axis, column 94 represents the sensor's detected movement in the Y axis, and column 95 represents the sensor's detected movement in the Z axis. The sensor pole's ID and the sensor IDs typically serve as indications of the location of the sensors on the fence. The sensor pole's ID is typically used to map data received from sensors within the sensor poles to a specific fence section, and the sensor's ID is typically used to map data obtained from a specific sensor to a specific string wire within the fence section.

In some embodiments where data regarding movement of string wires within a section is obtained from two sensor poles simultaneously, the data can be used to determine a more precise location of the force applied upon the string wires within a fence section. For example, movement data in two or three axes obtained from a sensor can be translated to a movement vector having a movement angle. A larger movement angle in respect to the direction of the wire next to a first sensor pole, and a smaller such movement angle next to a second sensor pole may indicate that the force being applied to the string wires is being applied relatively close to the first sensor pole and relatively far from the second sensor pole. An initial calibration of pre-defined forces and sensor measurements may assist in interpreting data obtained from multiple sensors residing on multiple sensor poles to a more precise location of the forces applied upon the string wires within the section.

Positive and negative values may represent the movement direction relative to a baseline position of the sensor on the wire represented by zero values. Larger absolute numbers may represent a more forceful or intense movement, for example a value of -20 in the Y axis may represent a strong downward movement in the Y axis towards the ground, and a value of +5 in the Y axis may represent a weak upward movement in the Y axis . The baseline position representing 0 values in each of the axes may be determined randomly, or alternatively initialized and calibrated by electrical, mechanical, or any other means which suit requirements. Calibration may take into account surrounding factors such as typical weather and terrain conditions.

Other measurable values may be obtained from the sensors, for example and without limitation the duration of detected movement, movement intensity, acceleration, and vibrations along the axes. In its resting position, the typical coordinates, speed and acceleration measurements received from a sensor regarding all three axes are zeros representing a baseline position. When moved, acceleration information received over the period of the movement from an acceleration sensor in each of the three axes is used to calculate the change in the sensor's speed and coordinates in the three axes.

When moved, data readings from the sensor in three axes may include the following information:

• the location of the sensor as a result of the movement, optionally reflected by new coordinates in three axes;

• the intensity (strength) of the movement, typically reflected by numbers whose absolute value represents the acceleration of the string wire movement in each of the three axes; and

• the direction of the movement in each of the axes, reflected by positive and negative values.

A processor can calculate the difference between data received from each of the sensors and data received from the sensor in its resting position. Typical data received from a sensor in its baseline resting position comprises zero values for coordinates and intensity.

A processor may use the acceleration of the string wire and the initial speed of the string wire as received by an accelerometer sensor to determine the speed of the string wire. By measuring the acceleration and integrating it with the initial speed of the string wire, the new speed of the string wire can be determined, as well as the displacement of that string wire relative to its baseline position or its previous position.

Data obtained continuously from a plurality of sensors over a period of time is integrated to calculate the movement of each of the string wires in that time. Wire movement types differ from one another according to the interaction with external factors such as humans, wild animals or weather hazards. Wire movement types may be sorted for example and without limitation according to the speed in which the sensor is reset to its zero position and the initial tension of the string wire. An intentional movement of the fence by a person will typically be characterized by different intensity and duration than that of a wild animal or a weather change.

An optional graphic representation of information received from a sensor is presented in Figure 8A. For clarity, the figure illustrates data received from a sensor for the X axis only, but similar data may be obtained by the sensor for the Y and Z axes as well. In the Figure, curves 110, 120 and 130 illustrate movement parameters (acceleration, speed and position respectively) recorded simultaneously in a single sensor over a period of time. In the graph, the horizontal axis represents elapsing time, and the vertical axis represents the change in the parameters.

For clarity, the curves have been separated and shown in Figures 8B, 8C and

8D. Figure 8B shows the change in acceleration as measured by the sensor during the time period in units of meters per second². Figure 8C shows the change in speed as measured by the sensor during the time period in units of meters per second. Figure 8D shows the change in position as measured by the sensor during the time period in units of meters.

All changes recorded in the graphs are relative to baseline (zero) values (zero acceleration, zero speed and zero coordinates). Figure 8B illustrates strong positive acceleration 112 followed by strong negative acceleration 114. Figure 8C demonstrates the velocity increase 122 up to 1.5 meters per second and decrease 124 back to 0 meters per second as a result of the acceleration changes 112 and 114 in Figure 8B. Figure 8C represents the changing position of the sensor 132 following the acceleration and speed changes, until it stabilizes at a new position 134 roughly 0.1 meters from the baseline coordinates. Graphic representations of data received from a plurality of sensors simultaneously over a pre-defined time period may be analyzed, optionally by graphical algorithms. Analysis of the continuous data obtained from the sensors and presented in a tabular or a graphical form, along with the location indication of the sensors (the mapping of every sensor to a specific string wire in a specific fence section) is used to interpret the data obtained from the sensors to movement data of the respective string wires and to generate a virtual image of the fence and its movements. Further analysis of the movement of the string wires helps detect events occurring on the fence, for example penetration events.

Information continuously obtained from a plurality of sensors on a fence section is typically aggregated over time in a data aggregation and storage unit, analyzed and compared to pre-defined patterns. Alternatively, patterns may be studied and added over time, optionally using machine learning.

Movement patterns may be used as an indication for events occurring on the fence. Environmental factors such as wind and rain should typically be filtered out, to determine whether a security breach occurred.

A "wind" event may be detected by a continual wavy movement behavior of sensors across a plurality of sensor poles. A "rain" event may be detected according to other patterns, and compared from data obtained from other sensors such as sensors configured to detect humidity. Additionally and alternatively, the system could validate wire movement data with weather stations.

Weather events are usually continuous events detected across a plurality of sensor poles, unlike security breaches which are typically detected upon one or two poles. Differentiation is preferably made between human generated events and non- human generated events, to determine malicious intent.

The ability to discern between animals and humans may be demonstrated by a wild boar example: when a wild boar attempts to cross a string wire fence, it tends to collide with the lowest wires, creating a strong movement in the Z axis upon these wires. Human attacks on a fence tend to involve stretching of wires and spreading them apart such as illustrated above, causing movement in the Y axis. Cutting of wires may be characterized by a movement which never returns to the initial baseline position. Such patterns may be used to determine whether a security breach occurs.

Output from more than one sensor is generally used for pattern recognition. A climbing pattern illustrated in Figure 9 can only be identified when data from a plurality of sensors is obtained and analyzed to detect two gaps 81 and 82 which may be caused by a person climbing upon the fence. Climbing the fence typically involves the person's legs pushing downwards string wires to create lower gap 81, and the person's hands hanging on to string wires and creating the higher gap 82. In addition, a climbing pattern may be characterized by staggered pulling of wires about 2 seconds between the pullings.

Further analysis may depend on patterns indicating whether string wires ever return to their baseline position. For example, when a string wire is cut, the sensor may eventually cease to report velocity changes indicating that there is no further movement in the string wire, but the coordinates of the sensor will be different than the sensor's baseline position. Another example could be a "wind" event, in which continuous movement is recorded over a long period of time and from a plurality of sensor poles.

The analysis of movement patterns can then be displayed on a display system to a user as an image, illustrating the changes in string wire positions relative to their baseline position. The presentation may be, for example and without limitation, somewhat similar to the schematic illustrations in Figure 6a or Figure 9.

The system may generate alerts and alarms to fence personnel, issue silent or noisy alarms on specific fence sections, operate cameras at specific fence sections, attempt to scare intruders by remotely operating means such as electrifying the fence, or exercise other means which suit requirements. By generating alarms after the smart analysis of the fence, fences can be monitored from afar and security personnel may be reduced. Energy can be saved, as energy consumption means such as electric fences and cameras may be operated only on a need basis rather than continuously.

Alarms and alerts generated by the system may be audial, visual, or tactile. For example, an auditory signal may be sent to patrol personnel regarding events that may compromise the integrity of the fence.

Integrating the system with cameras may help with a continuous calibration process, contributing to the stability of the system and a reduction in false alarms.

Alerts generated by the system may comprise information extracted from analysis of sensor signals, optionally augmented by other sensors known in the art and external input received from a variety of sources. The alerts may comprise information consisting of one or more of the following:

• location of event: for example the fence section on which the event was detected, the sensors / multi-sensor poles from which the data was synthesized; · type of event: for example intrusion event, escape event, climbing event, extreme weather event or the like. A plurality of interpretations for event types may be suggested to a user in one alert;

• recurrence data for consecutive related events; · previous events resembling the detected events in the past, in other fence sections or in other fences; and

• any other data which may be relevant to a user.

Fences may be constructed with a built-in artificial image generation system such that the system is integrated into the fence construction. Alternatively, the artificial image generation system may be retrofitted into existing fences. A system may be configured to act as a standalone system, or share resources such as but not limited to processing resources and data analysis patterns with other systems built on other fences in remote sites. Reference is now made to Figure 10 illustrating a flow chart 200 of a method for generating a virtual image of a fence having string wires and a plurality of sensors each coupled to a string wire within the fence. Each of the sensors is configured to produce data representing a time-dependent pattern of movement of the string wire it is coupled to.

The method comprises collecting data from the sensors 210 and sending at least part of the data 220 to a central communication unit configured to receive the data and optionally store it in a storage unit.

The collection of data may require a selection phase in which data from the sensors which may be immaterial to the state of the fence is not transmitted to the central communication unit. For example, the driving unit in each of the sensor poles may be configured to filtering out data obtained by the sensors if that data represents insignificant changes in the state of the string wires.

Part or all of the data may be collected in its original form, in a converted form or in a partially processed form. Data conversion may be achieved by using any device that converts a signal from one form to another, such as a transducer.

The method further comprises processing the data 230. Preferably, processing comprises analyzing the data received from all the sensors in a central processing unit. Processing may include, for example and without limitation mapping the data received from each of the sensors to a specific string wire and a specific sensor pole in a specific fence section, and interpreting the time-dependent movement parameters obtained from each of the sensors to a movement of that specific string wire. Generating a virtual dynamic image of the fence 240 is then made possible, as well as identifying events occurring on the fence 250 for example according to pre-defined movement patterns, differentiating between human generated events and non-human generated events, and generating alerts 260 regarding selected events.

The selection of events for which alerts will be generated may be according to the potential threat they pose to the fence or according to malicious intent associated with the identified event. For example, no alerts will be generated for weather-related events such as wind or rain, but alerts will be generated for events such as climbing, cutting of at least one string wire, or strong collisions events which may endanger the integrity of the fence. The scope of the present invention is defined by the appended claims and includes both combinations and sub combinations of the various features described hereinabove as well as variations and modifications thereof, which would occur to persons skilled in the art upon reading the foregoing description.

In the claims, the word "comprise", and variations thereof such as "comprises", "comprising" and the like indicate that the components listed are included, but not generally to the exclusion of other components.

Claims

A sensing system for a fence having string wires and support pillars, the system comprising:

a plurality of sensors, wherein each sensor of said plurality of sensors is coupled to one of the string wires, and wherein each sensor is configured to allow generating first data representing movement of the wire coupled thereto; spring elements connecting at least some of the string wires to the support pillars; at least one central communication unit configured to allow simultaneously obtaining second data representing at least a portion of said first data of the sensors;

at least one storage unit configured to allow repeatedly storing third data representing at least a portion of said second data over a predefined period of time; and

at least one processing unit configured to allow performing real-time analysis of fourth data representing at least a portion of said third data, wherein the analysis comprises:

attributing portions of the fourth data to the string wires, wherein each portion represents first data from a sensor coupled to the attributed string wire;

identifying at least one event occuring in the predefined period on the fence based on the fourth data;

identifying the string wires at which the events occur in the predefined period, and

generating alerts regarding at least some of said events, wherein the alerts each comprises informing the identity of the alerted event and string wires at which each event occurred, with the proviso that said central communication unit and said storage unit are not configured to identify said events.

2. The system of claim 1, the processing unit further configured to allow generation of a substantially real-time image representing the fence, wherein the image comprises indicators of the string wires at which the events occurred and a type of each event.

3. The system of claim 1, the processing unit further configured to allow generation of a substantially real-time audio signal, the audio signal representing indicators of the string wires at which the events occurred and a type of each event.

4. The system of claim 1 or 2, wherein said plurality of sensors are each capable of measuring movement of said string wire coupled thereto.

5. The sensing system of claim 4, wherein the measurement of movement is independent measurement of movement in two or three axes.

6. The sensing system of claim 5, wherein the measurement of movement is independent measurement of movement in three axes.

7. The system of claim 6, wherein said plurality of sensors are selected from a group consisting of: accelerometers, speedometers, gyroscopes, and combinations thereof.

8. The system of claim 2 or 3, further comprising at least one display unit configured to display said substantially real-time image of the fence.

9. The system of claim 6, wherein said at least one display unit is selected from a group consisting of: computer screens, laptops, PDAs, cellular phone screens, printed sheets, integrated LCD screens , Thin Film Transistors, touch screens and combinations thereof.

10. The system of claim 1, further configured to allow generation of alerts for one or more of the events.

11. The system of claim 7 wherein said at least one display unit is configured to display said alerts.

12. The system of claim 1 wherein said at least one display unit is configured to display said alerts.

13. The system of claim 2, wherein the real-time image comprises dynamic representation of occurrence of events.

14. The system of claim 11, wherein the system is configured to allow a user selection of event types to be dynamically represented.

15. The system of claim 1, further comprising at least one sensor pole, said sensor pole comprising: a protective housing configured to accommodate at least a portion of said plurality of sensors;

at least one driving unit;

at least one sensor pole communication unit; and

at least one power source;

wherein said power source is configured to supply power to said portion of said plurality of sensors, driving unit and sensor pole communication unit, and wherein said driving unit is configured to collect first data y said portion of said plurality of sensors,

and wherein said sensor pole communication unit is configured to transmit second data to said central communication unit.

16. The system of claim 15, wherein said portion of said plurality of sensors are vertically aligned within said sensor pole.

17. The system of claim 15 wherein said at least portion of said plurality of sensors comprises sensors having different sensitivity levels.

18. The system of claim 15, wherein said second data comprises: a sensor pole ID representing the sensor pole communication unit from which the first data was received; a sensor ID within said sensor pole, representing the string wire to which said each of said plurality of sensors is coupled to; and movement parameters of said sensor independently in each of three axes.

19. The system of claim 15 wherein said movement parameters are selected from a group consisting of: acceleration, velocity, position and combinations thereof.

20. The system of any one of claims 1, 2, 3, 5, 8, 10 to 17, configured to retrofit with existing fences.

21. The system of any one of claims 1, 2, 3, 5, 8, 10 to 17, integrated within a fence having string wires.

22. A method for a fence having sections comprising string wires and plurality of sensors configured to produce data representing a time-dependent pattern of movement of the string wires in the sections, the method comprising: collecting the data from the sensors;

sending at least a portion of the data to a central communication unit configured to receive said portion of data; processing in concert said at least portion of data; generating a dynamic virtual image of the fence; identifying events occurring on the fence sections; and generating alerts to a user regarding at least part of said events.

23. The method of claim 22, wherein said processing comprises: mapping each of the plurality of sensors to a specific string wire and a specific fence section; and

interpreting the movement parameters obtained from said each of the plurality of sensors to movements of said specific string wires in said specific fence sections.

24. The method of claim 23, wherein said identifying events occurring on the fence further comprises differentiating between human generated events and non-human generated events.

25. The method of claim 24, wherein said events are selected from a group comprising: penetration, intrusion, cutting at least one of the string wires, spreading string wires apart, climbing upon string wires, collision with the string wires, wind, rain, .and combinations thereof.

26. The method of claim 22 wherein said collecting the data from the plurality of sensors comprises periodic readings obtained at a sample speed and sensitivity rate set by a user.

27. The method of claim 26 wherein said sample speed and sensitivity rate change according to pre-defined thresholds.

28. The method of claim 26 wherein said sample speed and sensitivity rate change according to previously detected events.

29. The method of claim 22, wherein said interpreting the movement parameters further comprises determining a location within a fence section where forces are applied upon the string wires.

30. The method of claim 29, the fence further comprising at least two sensor poles within each fence section, a portion of said plurality of sensors being vertically aligned within each of said sensor poles, wherein said determining a location within a fence section where forces are applied upon the string wires further comprises: obtaining movement data in at least two axes from two sensors coupled to a single string wire laid between two sensor poles comprising each one of the two sensors; translating said movement data to a movement vector having a movement angle; and

using said movement angle in respect to the direction of the string wire as an indication of the location on the wire between the sensor poles, of the force being applied to the string wire.

31. The method of claim 22 wherein the time-dependent pattern is compared to pre-defined patterns.

32. The method of claim 22 wherein the time-dependent patterns are studied and added over time.

33. The method of claim 22 wherein time-dependent patterns are compared with data obtained from external sources.

34. The method of claim 33 wherein said external source comprises: humidity detection sensors, cameras, weather stations.

35. The method of claim 22 wherein a pattern characterized by a continual wavy movement across a plurality of fence sections indicates a wind event.

36. The method of claim 22 wherein a pattern characterized by strong movement of wires in the Z direction indicates a collision event.

37. The method of claim 22 wherein a plurality of interpretations for event types is suggested to a user in one alert.